Ability-to-Repay and Qualified Mortgage Standards Under the Truth in Lending Act (Regulation Z)

AGENCY:

Bureau of Consumer Financial Protection.

ACTION:

Final rule; official interpretations.

SUMMARY:

The Bureau of Consumer Financial Protection (Bureau) is amending Regulation Z, which implements the Truth in Lending Act (TILA). Regulation Z currently prohibits a creditor from making a higher-priced mortgage loan without regard to the consumer's ability to repay the loan. The final rule implements sections 1411 and 1412 of the Dodd-Frank Wall Street Reform and Consumer Protection Act (Dodd-Frank Act), which generally require creditors to make a reasonable, good faith determination of a consumer's ability to repay any consumer credit transaction secured by a dwelling (excluding an open-end credit plan, timeshare plan, reverse mortgage, or temporary loan) and establishes certain protections from liability under this requirement for “qualified mortgages.” The final rule also implements section 1414 of the Dodd-Frank Act, which limits prepayment penalties. Finally, the final rule requires creditors to retain evidence of compliance with the rule for three years after a covered loan is consummated.

DATES:

The rule is effective January 10, 2014.

FOR FURTHER INFORMATION CONTACT:

Joseph Devlin, Gregory Evans, David Friend, Jennifer Kozma, Eamonn K. Moran, or Priscilla Walton-Fein, Counsels; Thomas J. Kearney or Mark Morelli, Senior Counsels; or Stephen Shin, Managing Counsel, Office of Regulations, at (202) 435-7700.

SUPPLEMENTARY INFORMATION:

I. Summary of the Final Rule

The Consumer Financial Protection Bureau (Bureau) is issuing a final rule to implement laws requiring mortgage lenders to consider consumers' ability to repay home loans before extending them credit. The rule will take effect on January 10, 2014.

The Bureau is also releasing a proposal to seek comment on whether to adjust the final rule for certain community-based lenders, housing stabilization programs, certain refinancing programs of the Federal National Mortgage Association (Fannie Mae) or the Federal Home Loan Mortgage Corporation (Freddie Mac) (collectively, the GSEs) and Federal agencies, and small portfolio creditors. The Bureau expects to finalize the concurrent proposal this spring so that affected creditors can prepare for the January 2014 effective date.

Background

During the years preceding the mortgage crisis, too many mortgages were made to consumers without regard to the consumer's ability to repay the loans. Loose underwriting practices by some creditors—including failure to verify the consumer's income or debts and qualifying consumers for mortgages based on “teaser” interest rates that would cause monthly payments to jump to unaffordable levels after the first few years—contributed to a mortgage crisis that led to the nation's most serious recession since the Great Depression.

In response to this crisis, in 2008 the Federal Reserve Board (Board) adopted a rule under the Truth in Lending Act which prohibits creditors from making “higher-price mortgage loans” without assessing consumers' ability to repay the loans. Under the Board's rule, a creditor is presumed to have complied with the ability-to-repay requirements if the creditor follows certain specified underwriting practices. This rule has been in effect since October 2009.

In the 2010 Dodd-Frank Wall Street Reform and Consumer Protection Act, Congress required that for residential mortgages, creditors must make a reasonable and good faith determination based on verified and documented information that the consumer has a reasonable ability to repay the loan according to its terms. Congress also established a presumption of compliance for a certain category of mortgages, called “qualified mortgages.” These provisions are similar, but not identical to, the Board's 2008 rule and cover the entire mortgage market rather than simply higher-priced mortgages. The Board proposed a rule to implement the new statutory requirements before authority passed to the Bureau to finalize the rule.

Summary of Final Rule

The final rule contains the following key elements:

Ability-to-Repay Determinations. The final rule describes certain minimum requirements for creditors making ability-to-repay determinations, but does not dictate that they follow particular underwriting models. At a minimum, creditors generally must consider eight underwriting factors: (1) Current or reasonably expected income or assets; (2) current employment status; (3) the monthly payment on the covered transaction; (4) the monthly payment on any simultaneous loan; (5) the monthly payment for mortgage-related obligations; (6) current debt obligations, alimony, and child support; (7) the monthly debt-to-income ratio or residual income; and (8) credit history. Creditors must generally use reasonably reliable third-party records to verify the information they use to evaluate the factors.

The rule provides guidance as to the application of these factors under the statute. For example, monthly payments must generally be calculated by assuming that the loan is repaid in substantially equal monthly payments during its term. For adjustable-rate mortgages, the monthly payment must be calculated using the fully indexed rate or an introductory rate, whichever is higher. Special payment calculation rules apply for loans with balloon payments, interest-only payments, or negative amortization.

The final rule also provides special rules to encourage creditors to refinance “non-standard mortgages”—which include various types of mortgages which can lead to payment shock that can result in default—into “standard mortgages” with fixed rates for at least five years that reduce consumers' monthly payments.

Presumption for Qualified Mortgages. The Dodd-Frank Act provides that “qualified mortgages” are entitled to a presumption that the creditor making the loan satisfied the ability-to-repay requirements. However, the Act did not specify whether the presumption of compliance is conclusive (i.e., creates a safe harbor) or is rebuttable. The final rule provides a safe harbor for loans that satisfy the definition of a qualified mortgage and are not “higher-priced,” as generally defined by the Board's 2008 rule. The final rule provides a rebuttable presumption for higher-priced mortgage loans, as described further below.

The line the Bureau is drawing is one that has long been recognized as a rule of thumb to separate prime loans from subprime loans. Indeed, under the existing regulations that were adopted by the Board in 2008, only higher-priced mortgage loans are subject to an ability-to-repay requirement and a rebuttable presumption of compliance if creditors follow certain requirements. The new rule strengthens the requirements needed to qualify for a rebuttable presumption for subprime loans and defines with more particularity the grounds for rebutting the presumption. Specifically, the final rule provides that consumers may show a violation with regard to a subprime qualified mortgage by showing that, at the time the loan was originated, the consumer's income and debt obligations left insufficient residual income or assets to meet living expenses. The analysis would consider the consumer's monthly payments on the loan, loan-related obligations, and any simultaneous loans of which the creditor was aware, as well as any recurring, material living expenses of which the creditor was aware. Guidance accompanying the rule notes that the longer the period of time that the consumer has demonstrated actual ability to repay the loan by making timely payments, without modification or accommodation, after consummation or, for an adjustable-rate mortgage, after recast, the less likely the consumer will be able to rebut the presumption based on insufficient residual income.

With respect to prime loans—which are not currently covered by the Board's ability-to-repay rule—the final rule applies the new ability-to-repay requirements but creates a strong presumption for those prime loans that constitute qualified mortgages. Thus, if a prime loan satisfies the qualified mortgage criteria described below, it will be conclusively presumed that the creditor made a good faith and reasonable determination of the consumer's ability to repay.

General Requirements for Qualified Mortgages. The Dodd-Frank Act sets certain product-feature prerequisites and affordability underwriting requirements for qualified mortgages and vests discretion in the Bureau to decide whether additional underwriting or other requirements should apply. The final rule implements the statutory criteria, which generally prohibit loans with negative amortization, interest-only payments, balloon payments, or terms exceeding 30 years from being qualified mortgages. So-called “no-doc” loans where the creditor does not verify income or assets also cannot be qualified mortgages. Finally, a loan generally cannot be a qualified mortgage if the points and fees paid by the consumer exceed three percent of the total loan amount, although certain “bona fide discount points” are excluded for prime loans. The rule provides guidance on the calculation of points and fees and thresholds for smaller loans.

The final rule also establishes general underwriting criteria for qualified mortgages. Most importantly, the general rule requires that monthly payments be calculated based on the highest payment that will apply in the first five years of the loan and that the consumer have a total (or “back-end”) debt-to-income ratio that is less than or equal to 43 percent. The appendix to the rule details the calculation of debt-to-income for these purposes, drawing upon Federal Housing Administration guidelines for such calculations. The Bureau believes that these criteria will protect consumers by ensuring that creditors use a set of underwriting requirements that generally safeguard affordability. At the same time, these criteria provide bright lines for creditors who want to make qualified mortgages.

The Bureau also believes that there are many instances in which individual consumers can afford a debt-to-income ratio above 43 percent based on their particular circumstances, but that such loans are better evaluated on an individual basis under the ability-to-repay criteria rather than with a blanket presumption. In light of the fragile state of the mortgage market as a result of the recent mortgage crisis, however, the Bureau is concerned that creditors may initially be reluctant to make loans that are not qualified mortgages, even though they are responsibly underwritten. The final rule therefore provides for a second, temporary category of qualified mortgages that have more flexible underwriting requirements so long as they satisfy the general product feature prerequisites for a qualified mortgage and also satisfy the underwriting requirements of, and are therefore eligible to be purchased, guaranteed or insured by either (1) the GSEs while they operate under Federal conservatorship or receivership; or (2) the U.S. Department of Housing and Urban Development, Department of Veterans Affairs, or Department of Agriculture or Rural Housing Service. This temporary provision will phase out over time as the various Federal agencies issue their own qualified mortgage rules and if GSE conservatorship ends, and in any event after seven years.

Rural Balloon-Payment Qualified Mortgages. The final rule also implements a special provision in the Dodd-Frank Act that would treat certain balloon-payment mortgages as qualified mortgages if they are originated and held in portfolio by small creditors operating predominantly in rural or underserved areas. This provision is designed to assure credit availability in rural areas, where some creditors may only offer balloon-payment mortgages. Loans are only eligible if they have a term of at least five years, a fixed-interest rate, and meet certain basic underwriting standards; debt-to-income ratios must be considered but are not subject to the 43 percent general requirement.

Creditors are only eligible to make rural balloon-payment qualified mortgages if they originate at least 50 percent of their first-lien mortgages in counties that are rural or underserved, have less than $2 billion in assets, and (along with their affiliates) originate no more than 500 first-lien mortgages per year. The Bureau will designate a list of “rural” and “underserved” counties each year, and has defined coverage more broadly than originally had been proposed. Creditors must generally hold the loans on their portfolios for three years in order to maintain their “qualified mortgage” status.

Other Final Rule Provisions. The final rule also implements Dodd-Frank Act provisions that generally prohibit prepayment penalties except for certain fixed-rate, qualified mortgages where the penalties satisfy certain restrictions and the creditor has offered the consumer an alternative loan without such penalties. To match with certain statutory changes, the final rule also lengthens to three years the time creditors must retain records that evidence compliance with the ability-to-repay and prepayment penalty provisions and prohibits evasion of the rule by structuring a closed-end extension of credit that does not meet the definition of open-end credit as an open-end plan.

Summary of Concurrent Proposal

The concurrent proposal seeks comment on whether the general ability-to-repay and qualified mortgage rule should be modified to address potential adverse consequences on certain narrowly-defined categories of lending programs. Because those measures were not proposed by the Board originally, the Bureau believes additional public input would be helpful. Specifically, the proposal seeks comment on whether it would be appropriate to exempt designated non-profit lenders, homeownership stabilization programs, and certain Federal agency and GSE refinancing programs from the ability-to-repay requirements because they are subject to their own specialized underwriting criteria.

The proposal also seeks comment on whether to create a new category of qualified mortgages, similar to the one for rural balloon-payment mortgages, for loans without balloon-payment features that are originated and held on portfolio by small creditors. The new category would not be limited to lenders that operate predominantly in rural or underserved areas, but would use the same general size thresholds and other criteria as the rural balloon-payment rules. The proposal also seeks comment on whether to increase the threshold separating safe harbor and rebuttable presumption qualified mortgages for both rural balloon-payment qualified mortgages and the new small portfolio qualified mortgages, in light of the fact that small creditors often have higher costs of funds than larger creditors. Specifically, the Bureau is proposing a threshold of 3.5 percentage points above APOR for first-lien loans.

II. Background

For over 20 years, consumer advocates, legislators, and regulators have raised concerns about creditors originating mortgage loans without regard to the consumer's ability to repay the loan. Beginning in about 2006, these concerns were heightened as mortgage delinquencies and foreclosure rates increased dramatically, caused in part by the loosening of underwriting standards. See 73 FR 44524 (July 30, 2008). The following discussion provides background information, including a brief summary of the legislative and regulatory responses to the foregoing concerns, which culminated in the enactment of the Dodd-Frank Act on July 21, 2010, the Board's May 11, 2011, proposed rule to implement certain amendments to TILA made by the Dodd-Frank Act, and now the Bureau's issuance of this final rule to implement sections 1411, 1412, and 1414 of that act.

A. The Mortgage Market

Overview of the Market and the Mortgage Crisis

The mortgage market is the single largest market for consumer financial products and services in the United States, with approximately $9.9 trillion in mortgage loans outstanding. During the last decade, the market went through an unprecedented cycle of expansion and contraction that was fueled in part by the securitization of mortgages and creation of increasingly sophisticated derivative products. So many other parts of the American financial system were drawn into mortgage-related activities that, when the housing market collapsed in 2008, it sparked the most severe recession in the United States since the Great Depression.

The expansion in this market is commonly attributed to both particular economic conditions (including an era of low interest rates and rising housing prices) and to changes within the industry. Interest rates dropped significantly—by more than 20 percent—from 2000 through 2003. Housing prices increased dramatically—about 152 percent—between 1997 and 2006. Driven by the decrease in interest rates and the increase in housing prices, the volume of refinancings increased rapidly, from about 2.5 million loans in 2000 to more than 15 million in 2003.

In the mid-2000s, the market experienced a steady deterioration of credit standards in mortgage lending, with evidence that loans were made solely against collateral, or even against expected increases in the value of collateral, and without consideration of ability to repay. This deterioration of credit standards was particularly evidenced by the growth of “subprime” and “Alt-A” products, which consumers were often unable to repay. Subprime products were sold primarily to consumers with poor or no credit history, although there is evidence that some consumers who would have qualified for “prime” loans were steered into subprime loans as well. The Alt-A category of loans permitted consumers to take out mortgage loans while providing little or no documentation of income or other evidence of repayment ability. Because these loans involved additional risk, they were typically more expensive to consumers than “prime” mortgages, although many of them had very low introductory interest rates. In 2003, subprime and Alt-A origination volume was about $400 billion; in 2006, it had reached $830 billion.

So long as housing prices were continuing to increase, it was relatively easy for consumers to refinance their existing loans into more affordable products to avoid interest rate resets and other adjustments. When housing prices began to decline in 2005, however, refinancing became more difficult and delinquency rates on subprime and Alt-A products increased dramatically. More and more consumers, especially those with subprime and Alt-A loans, were unable or unwilling to make their mortgage payments. An early sign of the mortgage crisis was an upswing in early payment defaults—generally defined as borrowers being 60 or more days delinquent within the first year. Prior to 2006, 1.1 percent of mortgages would end up 60 or more days delinquent within the first two years. Taking a more expansive definition of early payment default to include 60 days delinquent within the first two years, this figure was double the historic average during 2006, 2007 and 2008. In 2006, 2007, and 2008, 2.3 percent, 2.1 percent, and 2.3 percent of mortgages ended up 60 or more days delinquent within the first two years, respectively. By the summer of 2006, 1.5 percent of loans less than a year old were in default, and this figure peaked at 2.5 percent in late 2007, well above the 1.0 percent peak in the 2000 recession. First payment defaults—mortgages taken out by consumers who never made a single payment—exceeded 1.5 percent of loans in early 2007. In addition, as the economy worsened, the rates of serious delinquency (90 or more days past due or in foreclosure) for the subprime and Alt-A products began a steep increase from approximately 10 percent in 2006, to 20 percent in 2007, to more than 40 percent in 2010.

The impact of this level of delinquencies was severe on creditors who held loans on their books and on private investors who purchased loans directly or through securitized vehicles. Prior to and during the bubble, the evolution of the securitization of mortgages attracted increasing involvement from financial institutions that were not directly involved in the extension of credit to consumers and from investors worldwide. Securitization of mortgages allows originating creditors to sell off their loans (and reinvest the funds earned in making new ones) to investors who want an income stream over time. Securitization had been pioneered by what are now called government-sponsored enterprises (GSEs), including the Federal National Mortgage Association (Fannie Mae) and the Federal Home Loan Mortgage Corporation (Freddie Mac). But by the early 2000s, large numbers of private financial institutions were deeply involved in creating increasingly complex mortgage-related investment vehicles through securities and derivative products. The private securitization-backed subprime and Alt-A mortgage market ground to a halt in 2007 in the face of the rising delinquencies on subprime and Alt-A products.

Six years later, the United States continues to grapple with the fallout. The fall in housing prices is estimated to have resulted in about $7 trillion in household wealth losses. In addition, distressed homeownership and foreclosure rates remain at unprecedented levels.

Response and Government Programs

In light of these conditions, the Federal government began providing support to the mortgage markets in 2008 and continues to do so at extraordinary levels today. The Housing and Economic Recovery Act of 2008, which became effective on October 1, 2008, provided both new safeguards and increased regulation for Fannie Mae and Freddie Mac, as well as provisions to assist troubled borrowers and to the hardest hit communities. Fannie Mae and Freddie Mac, which supported the mainstream mortgage market, experienced heavy losses and were placed in conservatorship by the Federal government in 2008 to support the collapsing mortgage market. Because private investors have withdrawn from the mortgage securitization market and there are no other effective secondary market mechanisms in place, the GSEs' continued operations help ensure that the secondary mortgage market continues to function and to assist consumers in obtaining new mortgages or refinancing existing mortgages. The Troubled Asset Relief Program (TARP), created to implement programs to stabilize the financial system during the financial crisis, was authorized through the Emergency Economic Stabilization Act of 2008 (EESA), as amended by the American Recovery and Reinvestment Act of 2009, and includes programs to help struggling homeowners avoid foreclosure. Since 2008, several other Federal government efforts have endeavored to keep the country's housing finance system functioning, including the Treasury Department's and the Federal Reserve System's mortgage-backed securities (MBS) purchase programs to help keep interest rates low and the Federal Housing Administration's (FHA's) increased market presence. As a result, mortgage credit has remained available, albeit with more restrictive underwriting terms that limit or preclude some consumers' access to credit. These same government agencies together with the GSEs and other market participants have also undertaken a series of efforts to help families avoid foreclosure through loan-modification programs, loan-refinance programs and foreclosure alternatives.

Size and Volume of the Current Mortgage Origination Market

Even with the economic downturn and tightening of credit standards, approximately $1.28 trillion in mortgage loans were originated in 2011. In exchange for an extension of mortgage credit, consumers promise to make regular mortgage payments and provide their home or real property as collateral. The overwhelming majority of homebuyers continue to use mortgage loans to finance at least some of the purchase price of their property. In 2011, 93 percent of all home purchases were financed with a mortgage credit transaction.

Consumers may obtain mortgage credit to purchase a home, to refinance an existing mortgage, to access home equity, or to finance home improvement. Purchase loans and refinancings together produced 6.3 million new first-lien mortgage loan originations in 2011. The proportion of loans that are for purchases as opposed to refinances varies with the interest rate environment and other market factors. In 2011, 65 percent of the market was refinance transactions and 35 percent was purchase loans, by volume. Historically the distribution has been more even. In 2000, refinances accounted for 44 percent of the market while purchase loans comprised 56 percent; in 2005, the two products were split evenly.

With a home equity transaction, a homeowner uses his or her equity as collateral to secure consumer credit. The credit proceeds can be used, for example, to pay for home improvements. Home equity credit transactions and home equity lines of credit resulted in an additional 1.3 million mortgage loan originations in 2011.

The market for higher-priced mortgage loans remains significant. Data reported under the Home Mortgage Disclosure Act (HMDA) show that in 2011 approximately 332,000 transactions, including subordinate liens, were reportable as higher-priced mortgage loans. Of these transactions, refinancings accounted for approximately 44 percent of the higher-priced mortgage loan market, and 90 percent of the overall higher-priced mortgage loan market involved first-lien transactions. The median first-lien higher-priced mortgage loan was for $81,000, while the interquartile range (quarter of the transactions are below, quarter of the transactions are above) was $47,000 to $142,000.

GSE-eligible loans, together with the other federally insured or guaranteed loans, cover the majority of the current mortgage market. Since entering conservatorship in September 2008, the GSEs have bought or guaranteed roughly three of every four mortgages originated in the country. Mortgages guaranteed by FHA make up most of the rest. Outside of the securitization available through the Government National Mortgage Association (Ginnie Mae) for loans primarily backed by FHA, there are very few alternatives in place today to assume the secondary market functions served by the GSEs.

Continued Fragility of the Mortgage Market

The current mortgage market is especially fragile as a result of the recent mortgage crisis. Tight credit remains an important factor in the contraction in mortgage lending seen over the past few years. Mortgage loan terms and credit standards have tightened most for consumers with lower credit scores and with less money available for a down payment. According to CoreLogic's TrueStandings Servicing, a proprietary data service that covers about two-thirds of the mortgage market, average underwriting standards have tightened considerably since 2007. Through the first nine months of 2012, for consumers that have received closed-end first-lien mortgages, the weighted average FICO score was 750, the loan-to-value (LTV) ratio was 78 percent, and the debt-to-income (DTI) ratio was 34.5 percent. In comparison, in the peak of the housing bubble in 2007, the weighted average FICO score was 706, the LTV was 80 percent, and the DTI was 39.8 percent.

In this tight credit environment, the data suggest that creditors are not willing to take significant risks. In terms of the distribution of origination characteristics, for 90 percent of all the Fannie Mae and Freddie Mac mortgage loans originated in 2011, consumers had a FICO score over 700 and a DTI less than 44 percent. According to the Federal Reserve's Senior Loan Officer Opinion Survey on Bank Lending Practices, in April 2012 nearly 60 percent of creditors reported that they would be much less likely, relative to 2006, to originate a conforming home-purchase mortgage to a consumer with a 10 percent down payment and a credit score of 620—a traditional marker for those consumers with weaker credit histories. The Federal Reserve Board calculates that the share of mortgage borrowers with credit scores below 620 has fallen from about 17 percent of consumers at the end of 2006 to about 5 percent more recently. Creditors also appear to have pulled back on offering these consumers loans insured by the FHA, which provides mortgage insurance on loans made by FHA-approved creditors throughout the United States and its territories and is especially structured to help promote affordability.

The Bureau is acutely aware of the high levels of anxiety in the mortgage market today. These concerns include the continued slow pace of recovery, the confluence of multiple major regulatory and capital initiatives, and the compliance burdens of the various Dodd-Frank Act rulemakings (including uncertainty on what constitutes a qualified residential mortgage (QRM), which, as discussed below, relates to the Dodd-Frank Act's credit risk retention requirements and mortgage securitizations). These concerns are causing discussion about whether creditors will consider exiting the business. The Bureau acknowledges that it will likely take some time for the mortgage market to stabilize and that creditors will need to adjust their operations to account for several major regulatory and capital regimes.

B. TILA and Regulation Z

In 1968, Congress enacted the Truth in Lending Act (TILA), 15 U.S.C. 1601 et seq., based on findings that the informed use of credit resulting from consumers' awareness of the cost of credit would enhance economic stability and competition among consumer credit providers. One of the purposes of TILA is to promote the informed use of consumer credit by requiring disclosures about its costs and terms. See 15 U.S.C. 1601. TILA requires additional disclosures for loans secured by consumers' homes and permits consumers to rescind certain transactions secured by their principal dwellings when the required disclosures are not provided. 15 U.S.C. 1635, 1637a. Section 105(a) of TILA directs the Bureau (formerly directed the Board of Governors of the Federal Reserve System) to prescribe regulations to carry out TILA's purposes and specifically authorizes the Bureau, among other things, to issue regulations that contain such additional requirements, classifications, differentiations, or other provisions, or that provide for such adjustments and exceptions for all or any class of transactions, that in the Bureau's judgment are necessary or proper to effectuate the purposes of TILA, facilitate compliance thereof, or prevent circumvention or evasion therewith. See 15 U.S.C. 1604(a).

General rulemaking authority for TILA transferred to the Bureau in July 2011, other than for certain motor vehicle dealers in accordance with the Dodd-Frank Act section 1029, 12 U.S.C. 5519. Pursuant to the Dodd-Frank Act and TILA, as amended, the Bureau published for public comment an interim final rule establishing a new Regulation Z, 12 CFR part 1026, implementing TILA (except with respect to persons excluded from the Bureau's rulemaking authority by section 1029 of the Dodd-Frank Act). 76 FR 79768 (Dec. 22, 2011). This rule did not impose any new substantive obligations but did make technical and conforming changes to reflect the transfer of authority and certain other changes made by the Dodd-Frank Act. The Bureau's Regulation Z took effect on December 30, 2011. The Official Staff Interpretations interpret the requirements of the regulation and provides guidance to creditors in applying the rules to specific transactions. See 12 CFR part 1026, Supp. I.

C. The Home Ownership and Equity Protection Act (HOEPA) and HOEPA Rules

In response to evidence of abusive practices in the home-equity lending market, in 1994 Congress amended TILA by enacting the Home Ownership and Equity Protection Act (HOEPA) as part of the Riegle Community Development and Regulatory Improvement Act of 1994. Public Law 103-325, 108 Stat. 2160. HOEPA was enacted as an amendment to TILA to address abusive practices in refinancing and home-equity mortgage loans with high interest rates or high fees. Loans that meet HOEPA's high-cost triggers are subject to special disclosure requirements and restrictions on loan terms, and consumers with high-cost mortgages have enhanced remedies for violations of the law.

The statute applied generally to closed-end mortgage credit, but excluded purchase money mortgage loans and reverse mortgages. Coverage was triggered where a loan's annual percentage rate (APR) exceeded comparable Treasury securities by specified thresholds for particular loan types, or where points and fees exceeded eight percent of the total loan amount or a dollar threshold.

For high-cost loans meeting either of those thresholds, HOEPA required creditors to provide special pre-closing disclosures, restricted prepayment penalties and certain other loan terms, and regulated various creditor practices, such as extending credit without regard to a consumer's ability to repay the loan. HOEPA also provided a mechanism for consumers to rescind covered loans that included certain prohibited terms and to obtain higher damages than are allowed for other types of TILA violations. Finally, HOEPA amended TILA section 131, 15

U.S.C. 1641, to provide that purchasers of high-cost loans generally are subject to all claims and defenses against the original creditor with respect to the mortgage, including a creditor's failure to make an ability-to-repay determination before making the loan. HOEPA created special substantive protections for high-cost mortgages, such as prohibiting a creditor from engaging in a pattern or practice of extending a high-cost mortgage to a consumer based on the consumer's collateral without regard to the consumer's repayment ability, including the consumer's current and expected income, current obligations, and employment. TILA section 129(h); 15 U.S.C. 1639(h).

In addition to the disclosures and limitations specified in the statute, HOEPA expanded the Board's rulemaking authority, among other things, to prohibit acts or practices the Board found to be unfair and deceptive in connection with mortgage loans.

In 1995, the Board implemented the HOEPA amendments at §§ 226.31, 226.32, and 226.33 of Regulation Z. See 60 FR 15463 (Mar. 24, 1995). In particular, § 226.32(e)(1) implemented TILA section 129(h)'s ability-to-repay requirements to prohibit a creditor from engaging in a pattern or practice of extending a high-cost mortgage based on the consumer's collateral without regard to the consumer's repayment ability, including the consumer's current income, current obligations, and employment status.

In 2001, the Board published additional significant changes to expand both HOEPA's protections to more loans by revising the annual percentage rate (APR) threshold for first-lien mortgage loans, expanded the definition of points and fees to include the cost of optional credit insurance and debt cancellation premiums, and enhanced the restrictions associated with high-cost loans. See 66 FR 65604 (Dec. 20, 2001). In addition, the ability-to-repay provisions in the regulation were revised to provide for a presumption of a violation of the rule if the creditor engages in a pattern or practice of making high-cost mortgages without verifying and documenting the consumer's repayment ability.

D. 2006 and 2007 Interagency Supervisory Guidance

In December 2005, the Federal banking agencies ⁴³ responded to concerns about the rapid growth of nontraditional mortgages in the previous two years by proposing supervisory guidance. Nontraditional mortgages are mortgages that allow the consumer to defer repayment of principal and sometimes interest. The guidance advised institutions of the need to reduce “risk layering” with respect to these products, such as by failing to document income or lending nearly the full appraised value of the home. The final guidance issued in September 2006 specifically advised creditors that layering risks in nontraditional mortgage loans to consumers receiving subprime credit may significantly increase risks to consumers as well as institutions. See Interagency Guidance on Nontraditional Mortgage Product Risks, 71 FR 58609 (Oct. 4, 2006) (2006 Nontraditional Mortgage Guidance).

The Federal banking agencies addressed concerns about the subprime market in March 2007 with proposed supervisory guidance addressing the heightened risks to consumers and institutions of adjustable-rate mortgages with two- or three-year “teaser” interest rates followed by substantial increases in the rate and payment. The guidance, finalized in June of 2007, set out the standards institutions should follow to ensure consumers in the subprime market obtain loans they can afford to repay. Among other steps, the guidance advised creditors: (1) To use the fully indexed rate and fully-amortizing payment when qualifying consumers for loans with adjustable rates and potentially non-amortizing payments; (2) to limit stated income and reduced documentation loans to cases where mitigating factors clearly minimize the need for full documentation of income; and (3) to provide that prepayment penalty clauses expire a reasonable period before reset, typically at least 60 days. See Statement on Subprime Mortgage Lending, 72 FR 37569 (July 10, 2007) (2007 Subprime Mortgage Statement). The Conference of State Bank Supervisors (CSBS) and the American Association of Residential Mortgage Regulators (AARMR) issued parallel statements for state supervisors to use with state-supervised entities, and many states adopted the statements.

E. 2008 HOEPA Final Rule

After the Board finalized the 2001 HOEPA rules, new consumer protection issues arose in the mortgage market. In 2006 and 2007, the Board held a series of national hearings on consumer protection issues in the mortgage market. During those hearings, consumer advocates and government officials expressed a number of concerns, and urged the Board to prohibit or restrict certain underwriting practices, such as “stated income” or “low documentation” loans, and certain product features, such as prepayment penalties. See 73 FR 44527 (July 30, 2008). The Board was also urged to adopt additional regulations under HOEPA, because, unlike the Interagency Supervisory Guidance, the regulations would apply to all creditors and would be enforceable by consumers through civil actions. As discussed above, in 1995 the Board implemented TILA section 129(h)'s ability-to-repay requirements for high-cost mortgage loans. In 2008, the Board exercised its authority under HOEPA to extend certain consumer protections concerning a consumer's ability to repay and prepayment penalties to a new category of “higher-priced mortgage loans” (HPMLs) with APRs that are lower than those prescribed for high-cost loans but that nevertheless exceed the average prime offer rate by prescribed amounts. This new category of loans was designed to include subprime credit. Specifically, the Board exercised its authority to revise HOEPA's restrictions on high-cost loans based on a conclusion that the revisions were necessary to prevent unfair and deceptive acts or practices in connection with mortgage loans. 73 FR 44522 (July 30, 2008) (2008 HOEPA Final Rule). The Board determined that imposing the burden to prove “pattern or practice” on an individual consumer would leave many consumers with a lesser remedy, such as those provided under some State laws, or without any remedy for loans made without regard to repayment ability. In particular, the Board concluded that a prohibition on making individual loans without regard for repayment ability was necessary to ensure a remedy for consumers who are given unaffordable loans and to deter irresponsible lending, which injures individual consumers. The 2008 HOEPA Final Rule provides a presumption of compliance with the higher-priced mortgage ability-to-repay requirements if the creditor follows certain procedures regarding underwriting the loan payment, assessing the debt-to-income ratio or residual income, and limiting the features of the loan, in addition to following certain procedures mandated for all creditors. See § 1026.34(a)(4)(iii) and (iv). However, the 2008 HOEPA Final Rule makes clear that even if the creditor follows the required and optional criteria, the creditor has merely obtained a presumption of compliance with the repayment ability requirement. The consumer can still rebut or overcome that presumption by showing that, despite following the required and optional procedures, the creditor nonetheless disregarded the consumer's ability the loan.

F. The Dodd-Frank Act

In 2007, Congress held numerous hearings focused on rising subprime foreclosure rates and the extent to which lending practices contributed to them. Consumer advocates testified that certain lending terms or practices contributed to the foreclosures, including a failure to consider the consumer's ability to repay, low- or no-documentation loans, hybrid adjustable-rate mortgages, and prepayment penalties. Industry representatives, on the other hand, testified that adopting substantive restrictions on subprime loan terms would risk reducing access to credit for some consumers. In response to these hearings, the House of Representatives passed the Mortgage Reform and Anti-Predatory Lending Act, both in 2007 and again in 2009. H.R. 3915, 110th Cong. (2007); H.R. 1728, 111th Cong. (2009). Both bills would have amended TILA to provide consumer protections for mortgages, including ability-to-repay requirements, but neither bill was passed by the Senate. Instead, both houses shifted their focus to enacting comprehensive financial reform legislation.

In December 2009, the House passed the Wall Street Reform and Consumer Protection Act of 2009, its version of comprehensive financial reform legislation, which included an ability-to-repay and qualified mortgage provision. H.R. 4173, 111th Cong. (2009). In May 2010, the Senate passed its own version of ability-to-repay requirements in its own version of comprehensive financial reform legislation, called the Restoring American Financial Stability Act of 2010. S. 3217, 111th Cong. (2010). After conference committee negotiations, the Dodd-Frank Act was passed by both houses of Congress and was signed into law on July 21, 2010. Public Law 111-203, 124 Stat. 1376 (2010).

In the Dodd-Frank Act, Congress established the Bureau and, under sections 1061 and 1100A, generally consolidated the rulemaking authority for Federal consumer financial laws, including TILA and RESPA, in the Bureau. Congress also provided the Bureau, among other things, with supervision authority for Federal consumer financial laws over certain entities, including insured depository institutions and credit unions with total assets over $10 billion and their affiliates, and mortgage-related non-depository financial services providers. In addition, Congress provided the Bureau with authority, subject to certain limitations, to enforce the Federal consumer financial laws, including the 18 enumerated consumer laws. Title X of the Dodd-Frank Act, and rules thereunder. The Bureau can bring civil actions in court and administrative enforcement proceedings to obtain remedies such as civil penalties and cease-and-desist orders.

At the same time, Congress significantly amended the statutory requirements governing mortgage practices with the intent to restrict the practices that contributed to the crisis. Title XIV of the Dodd-Frank Act contains a modified version of the Mortgage Reform and Anti-Predatory Lending Act. The Dodd-Frank Act requires the Bureau to propose consolidation of the major federal mortgage disclosures, imposes new requirements and limitations to address a wide range of consumer mortgage issues, and imposes credit risk retention requirements in connection with mortgage securitization.

Through the Dodd-Frank Act, Congress expanded HOEPA to apply to more types of mortgage transactions, including purchase money mortgage loans and home-equity lines of credit. Congress also amended HOEPA's existing high-cost triggers, added a prepayment penalty trigger, and expanded the protections associated with high-cost mortgages.

In addition, sections 1411, 1412, and 1414 of the Dodd-Frank Act created new TILA section 129C, which establishes, among other things, new ability-to-repay requirements and new limits on prepayment penalties. Section 1402 of the Dodd-Frank Act states that Congress created new TILA section 129C upon a finding that “economic stabilization would be enhanced by the protection, limitation, and regulation of the terms of residential mortgage credit and the practices related to such credit, while ensuring that responsible, affordable mortgage credit remains available to consumers.” TILA section 129B(a)(1), 15 U.S.C. 1639b(a)(1). Section 1402 of the Dodd-Frank Act further states that the purpose of TILA section 129C is to “assure that consumers are offered and receive residential mortgage loans on terms that reasonably reflect their ability to repay the loans.” TILA section 129B(a)(2), 15 U.S.C. 1639b(a)(2).

Specifically, TILA section 129C:

• Expands coverage of the ability-to-repay requirements to any consumer credit transaction secured by a dwelling, except an open-end credit plan, credit secured by an interest in a timeshare plan, reverse mortgage, or temporary loan.
• Prohibits a creditor from making a mortgage loan unless the creditor makes a reasonable and good faith determination, based on verified and documented information, that the consumer has a reasonable ability to repay the loan according to its terms, and all applicable taxes, insurance, and assessments.
• Provides a presumption of compliance with the ability-to-repay requirements if the mortgage loan is a “qualified mortgage,” which does not contain certain risky features and does not exceed certain thresholds for points and fees on the loan and which meets such other criteria as the Bureau may prescribe.
• Prohibits prepayment penalties unless the mortgage is a fixed-rate qualified mortgage that is not a higher-priced mortgage loan, and the amount and duration of the prepayment penalty are limited.

The statutory ability-to-repay standards reflect Congress's belief that certain lending practices (such as low- or no-documentation loans or underwriting loans without regard to principal repayment) led to consumers having mortgages they could not afford, resulting in high default and foreclosure rates. Accordingly, new TILA section 129C generally prohibits a creditor from making a residential mortgage loan unless the creditor makes a reasonable and good faith determination, based on verified and documented information, that the consumer has a reasonable ability to repay the loan according to its terms.

To provide more certainty to creditors while protecting consumers from unaffordable loans, the Dodd-Frank Act provides a presumption of compliance with the ability-to-repay requirements for certain “qualified mortgages.” TILA section 129C(b)(1) states that a creditor or assignee may presume that a loan has met the repayment ability requirement if the loan is a qualified mortgage. Qualified mortgages are prohibited from containing certain features that Congress considered to increase risks to consumers and must comply with certain limits on points and fees.

The Dodd-Frank Act creates special remedies for violations of TILA section 129C. As amended by section 1416 of the Dodd-Frank Act, TILA provides that a consumer who brings a timely action against a creditor for a violation of TILA section 129C(a) (the ability-to-repay requirements) may be able to recover special statutory damages equal to the sum of all finance charges and fees paid by the consumer, unless the creditor demonstrates that the failure to comply is not material. TILA section 130(a). This recovery is in addition to: (1) Actual damages; (2) statutory damages in an individual action or class action, up to a prescribed threshold; and (3) court costs and attorney fees that would be available for violations of other TILA provisions. In addition, the statute of limitations for a violation of TILA section 129C is three years from the date of the occurrence of the violation (as compared to one year for most other TILA violations, except for actions brought under section 129 or 129B, or actions brought by a State attorney general to enforce a violation of section 129, 129B, 129C, 129D, 129E, 129F, 129G, or 129H, which may be brought not later than 3 years after the date on which the violation occurs, and private education loans under 15 U.S.C. 1650(a), which may be brought not later than one year from the due date of first regular payment of principal). TILA section 130(e). Moreover, as amended by section 1413 of the Dodd-Frank Act, TILA provides that when a creditor, or an assignee, other holder or their agent initiates a foreclosure action, a consumer may assert a violation of TILA section 129C(a) “as a matter of defense by recoupment or setoff.” TILA section 130(k). There is no time limit on the use of this defense and the amount of recoupment or setoff is limited, with respect to the special statutory damages, to no more than three years of finance charges and fees. For high-cost loans an assignee generally continues to be subject to all claims and defenses, not only in foreclosure, with respect to that mortgage that the consumer could assert against the creditor of the mortgage, unless the assignee demonstrates, by a preponderance of evidence, that a reasonable person exercising ordinary due diligence, could not determine that the mortgage was a high-cost mortgage. TILA section 131(d).

In addition to the foregoing ability-to-repay provisions, the Dodd-Frank Act established other new standards concerning a wide range of mortgage lending practices, including compensation of mortgage originators, Federal mortgage disclosures, and mortgage servicing. Those and other Dodd-Frank Act provisions are the subjects of other rulemakings by the Bureau. For additional information on those other rulemakings, see the discussion below in part III.C.

G. Qualified Residential Mortgage Rulemaking

Section 15G of the Securities Exchange Act of 1934, added by section 941(b) of the Dodd-Frank Act, generally requires the securitizer of asset-backed securities (ABS) to retain not less than five percent of the credit risk of the assets collateralizing the ABS. 15 U.S.C. 78o-11. The Dodd-Frank Act's credit risk retention requirements are aimed at addressing weaknesses and failures in the securitization process and the securitization markets. By requiring that the securitizer retain a portion of the credit risk of the assets being securitized, the Dodd-Frank Act provides securitizers an incentive to monitor and ensure the quality of the assets underlying a securitization transaction. Six Federal agencies (not including the Bureau) are tasked with implementing this requirement. Those agencies are the Board, Office of the Comptroller of the Currency (OCC), Federal Deposit Insurance Corporation (FDIC), Securities and Exchange Commission (SEC), Federal Housing Finance Agency (FHFA), and Department of Housing and Urban Development (HUD) (collectively, the QRM agencies).

Section 15G of the Securities Exchange Act of 1934 provides that the credit risk retention requirements shall not apply to an issuance of ABS if all of the assets that collateralize the ABS are “qualified residential mortgages” (QRMs). See 15 U.S.C. 78o-11(c)(1)(C)(iii), (4)(A) and (B). Section 15G requires the QRM agencies to jointly define what constitutes a QRM, taking into consideration underwriting and product features that historical loan performance data indicate result in a lower risk of default. See 15 U.S.C. 78o-11(e)(4). Notably, section 15G also provides that the definition of a QRM shall be “no broader than” the definition of a “qualified mortgage,” as the term is defined under TILA section 129C(b)(2), as amended by the Dodd-Frank Act, and regulations adopted thereunder. 15 U.S.C. 78o-11(e)(4)(C).

On April 29, 2011, the QRM agencies issued joint proposed risk retention rules, including a proposed QRM definition (2011 QRM Proposed Rule). See 76 FR 24090 (Apr. 29, 2011). The proposed rule has not been finalized. Among other requirements, the 2011 QRM Proposed Rule incorporates the qualified mortgage restrictions on negative amortization, interest-only, and balloon payments, limits points and fees to three percent of the loan amount, and prohibits prepayment penalties. The proposed rule also establishes underwriting standards designed to ensure that QRMs have high credit quality, including:

• A maximum “front-end” monthly debt-to-income ratio (which looks at only the consumer's mortgage payment relative to income, but not at other debts) of 28 percent;
• A maximum “back-end” monthly debt-to-income ratio (which includes all of the consumer's debt, not just the mortgage payment) of 36 percent;
• A maximum loan-to-value (LTV) ratio of 80 percent in the case of a purchase transaction (with a lesser combined LTV permitted for refinance transactions);
• A 20 percent down payment requirement in the case of a purchase transaction; and
• Credit history verification and documentation requirements.

The proposed rule also includes appraisal requirements, restrictions on the assumability of the mortgage, and requires the creditor to commit to certain servicing policies and procedures regarding loss mitigation. See 76 FR at 24166-67.

To provide clarity on the definitions, calculations, and verification requirements for the QRM standards, the 2011 QRM Proposed Rule incorporates certain definitions and key terms established by HUD and required to be used by creditors originating FHA-insured residential mortgages. See 76 FR at 24119. Specifically, the 2011 QRM Proposed Rule incorporates the definitions and standards set out in the HUD Handbook 4155.1 (New Version), Mortgage Credit Analysis for Mortgage Insurance, as in effect on December 31, 2010, for determining and verifying the consumer's funds and the consumer's monthly housing debt, total monthly debt, and monthly gross income.

The qualified mortgage and QRM definitions are distinct and relate to different parts of the Dodd-Frank Act with different purposes, but both are designed to address problems that had arisen in the mortgage origination process. The qualified mortgage standard provides creditors with a presumption of compliance with the requirement in TILA section 129C(a) to assess a consumer's ability to repay a residential mortgage loan. The purpose of these provisions is to ensure that consumers are offered and receive residential mortgage loans on terms that reasonably reflect their ability to repay the loans. See TILA section 129B(a)(2). The Dodd-Frank Act's credit risk retention requirements are intended to address problems in the securitization markets and in mortgage markets by requiring that securitizers, as a general matter, retain an economic interest in the credit risk of the assets they securitize. The QRM credit risk retention requirement was meant to incentivize creditors to make more responsible loans because they will need to keep some skin in the game.

Nevertheless, as discussed above, the Dodd-Frank Act requires that the QRM definition be “no broader than” the qualified mortgage definition. Therefore, in issuing the 2011 QRM Proposed Rule, the QRM agencies sought to incorporate the statutory qualified mortgage standards, in addition to other requirements, into the QRM definition. 76 FR at 24118. This approach was designed to minimize the potential for conflicts between the QRM standards in the proposed rule and the qualified mortgage definition that the Bureau would ultimately adopt in a final rule.

In the 2011 QRM Proposed Rule, the QRM agencies stated their expectation to monitor the rules adopted by the Bureau under TILA to define a qualified mortgage and to review those rules to ensure that the definition of QRM in the final rule is “no broader” than the definition of a qualified mortgage and to appropriately implement the Dodd-Frank Act's credit risk retention requirement. See 76 FR at 24118. In preparing this final rule, the Bureau has consulted regularly with the QRM agencies to coordinate the qualified mortgage and qualified residential mortgage definitions. However, while the Bureau's qualified mortgage definition will set the outer boundary of a QRM, the QRM agencies have discretion under the Dodd-Frank Act to define QRMs in a way that is stricter than the qualified mortgage definition.

III. Summary of the Rulemaking Process

A. The Board's Proposal

In 2011, the Board published for public comment a proposed rule amending Regulation Z to implement the foregoing ability-to-repay amendments to TILA made by the Dodd-Frank Act. See 76 FR 27390 (May 11, 2011) (2011 ATR Proposal, the Board's proposal or the proposal). Consistent with the Dodd-Frank Act, the Board's proposal applied the ability-to-repay requirements to any consumer credit transaction secured by a dwelling (including vacation home loans and home equity loans), except an open-end credit plan, extension of credit secured by a consumer's interest in a timeshare plan, reverse mortgage, or temporary loan with a term of 12 months or less.

The Board's proposal provided four options for complying with the ability-to-repay requirement, including by making a “qualified mortgage.” First, the proposal would have allowed a creditor to meet the general ability-to-repay standard by originating a covered mortgage loan for which the creditor considered and verified eight underwriting factors in determining repayment ability, and, for adjustable rate loans, the mortgage payment calculation is based on the fully indexed rate. Second, the proposal would have allowed a creditor to refinance a “non-standard mortgage” into a “standard mortgage.” Under this option, the proposal would not have required the creditor to verify the consumer's income or assets. Third, the proposal would have allowed a creditor to originate a qualified mortgage, which provides special protection from liability for creditors. Because the Board determined that it was unclear whether that protection is intended to be a safe harbor or a rebuttable presumption of compliance with the repayment ability requirement, the Board proposed two alternative definitions of a qualified mortgage. Finally, the proposal would have allowed a small creditor operating predominantly in rural or underserved areas to originate a balloon-payment qualified mortgage if the loan term is five years or more, and the payment calculation is based on the scheduled periodic payments, excluding the balloon payment. The Board's proposal also would have implemented the Dodd-Frank Act's limits on prepayment penalties, lengthened the time creditors must retain evidence of compliance with the ability-to-repay and prepayment penalty provisions, and prohibited evasion of the rule by structuring a closed-end extension of credit that does not meet the definition of an open-end plan. As discussed above, rulemaking authority under TILA generally transferred from the Board to the Bureau in July 2011, including the authority under Dodd-Frank Act section 1412 to prescribe regulations to carry out the purposes of the qualified mortgage rules. 12 U.S.C. 5512; 12 U.S.C. 5581; 15 U.S.C. 1639c. As discussed above, TILA section 105(a) directs the Bureau to prescribe regulations to carry out the purposes of TILA. Except with respect to the substantive restrictions on high-cost mortgages provided in TILA section 129, TILA section 105(a) authorizes the Bureau to prescribe regulations that may contain additional requirements, classifications, differentiations, or other provisions, and may provide for such adjustments and exceptions for all or any class of transactions that the Bureau determines are necessary or proper to effectuate the purposes of TILA, to prevent circumvention or evasion thereof, or to facilitate compliance therewith.

B. Comments and Post-Proposal Outreach

The Board received numerous comments on the proposal, including comments regarding the criteria for a “qualified mortgage” and whether a qualified mortgage provides a safe harbor or a presumption of compliance with the repayment ability requirements. As noted above, in response to the proposed rule, the Board received approximately 1,800 letters from commenters, including members of Congress, creditors, consumer groups, trade associations, mortgage and real estate market participants, and individual consumers. As of July 21, 2011, the Dodd-Frank Act generally transferred the Board's rulemaking authority for TILA, among other Federal consumer financial laws, to the Bureau. Accordingly, all comment letters on the proposed rule were also transferred to the Bureau. Materials submitted were filed in the record and are publicly available at http://www.regulations.gov.

Through various comment letters and the Bureau's own collection of data, the Bureau received additional information and new data pertaining to the proposed rule. Accordingly, in May 2012, the Bureau reopened the comment period in order to solicit further comment on data and new information, including data that may assist the Bureau in defining loans with characteristics that make it appropriate to presume that the creditor complied with the ability-to-repay requirements or assist the Bureau in assessing the benefits and costs to consumers, including access to credit, and covered persons, as well as the market share covered by, alternative definitions of a “qualified mortgage.” The Bureau received approximately 160 comments in response to the reopened comment period from a variety of commenters, including creditors, consumer groups, trade associations, mortgage and real estate market participants, individuals, small entities, the SBA's Office of Advocacy, and FHA. As discussed in more detail below, the Bureau has considered these comments in adopting this final rule.

C. Other Rulemakings

In addition to this final rule, the Bureau is adopting several other final rules and issuing one proposal, all relating to mortgage credit to implement requirements of title XIV of the Dodd-Frank Act. The Bureau is also issuing a final rule jointly with other Federal agencies to implement requirements for mortgage appraisals in title XIV. Each of the final rules follows a proposal issued in 2011 by the Board or in 2012 by the Bureau alone or jointly with other Federal agencies. Collectively, these proposed and final rules are referred to as the Title XIV Rulemakings.

• Ability to Repay: Simultaneously with this final rule (the 2013 ATR Final Rule), the Bureau is issuing a proposal to amend certain provisions of the final rule, including by the addition of exemptions for certain nonprofit creditors and certain homeownership stabilization programs and a definition of a “qualified mortgage” for certain loans made and held in portfolio by small creditors (the 2013 ATR Concurrent Proposal). The Bureau expects to act on the 2013 ATR Concurrent Proposal on an expedited basis, so that any exceptions or adjustments can take effect simultaneously with this final rule.
• Escrows: The Bureau is finalizing a rule, following a March 2011 proposal issued by the Board (the Board's 2011 Escrows Proposal), to implement certain provisions of the Dodd-Frank Act expanding on existing rules that require escrow accounts to be established for higher-priced mortgage loans and creating an exemption for certain loans held by creditors operating predominantly in rural or underserved areas, pursuant to TILA section 129D as established by Dodd-Frank Act sections 1461. 15 U.S.C. 1639d. The Bureau's final rule is referred to as the 2013 Escrows Final Rule.

• HOEPA: Following its July 2012 proposal (the 2012 HOEPA Proposal), the Bureau is issuing a final rule to implement Dodd-Frank Act requirements expanding protections for “high-cost mortgages” under the Homeownership and Equity Protection Act (HOEPA), pursuant to TILA sections 103(bb) and 129, as amended by Dodd-Frank Act sections 1431 through 1433. 15 U.S.C. 1602(bb) and 1639. The Bureau also is finalizing rules to implement certain title XIV requirements concerning homeownership counseling, including a requirement that creditors provide lists of homeownership counselors to applicants for federally related mortgage loans, pursuant to RESPA section 5(c), as amended by Dodd-Frank Act section 1450. 12 U.S.C. 2604(c). The Bureau's final rule is referred to as the 2013 HOEPA Final Rule.

• Servicing: Following its August 2012 proposals (the 2012 RESPA Servicing Proposal and 2012 TILA Servicing Proposal), the Bureau is adopting final rules to implement Dodd-Frank Act requirements regarding force-placed insurance, error resolution, information requests, and payment crediting, as well as requirements for mortgage loan periodic statements and adjustable-rate mortgage reset disclosures, pursuant to section 6 of RESPA and sections 128, 128A, 129F, and 129G of TILA, as amended or established by Dodd-Frank Act sections 1418, 1420, 1463, and 1464. 12 U.S.C. 2605; 15 U.S.C. 1638, 1638a, 1639f, and 1639g. The Bureau also is finalizing rules on early intervention for troubled and delinquent consumers, and loss mitigation procedures, pursuant to the Bureau's authority under section 6 of RESPA, as amended by Dodd-Frank Act section 1463, to establish obligations for mortgage servicers that it finds to be appropriate to carry out the consumer protection purposes of RESPA, and its authority under section 19(a) of RESPA to prescribe rules necessary to achieve the purposes of RESPA. The Bureau's final rule under RESPA with respect to mortgage servicing also establishes requirements for general servicing standards policies and procedures and continuity of contact pursuant to its authority under section 19(a) of RESPA. The Bureau's final rules are referred to as the 2013 RESPA Servicing Final Rule and the 2013 TILA Servicing Final Rule, respectively.

• Loan Originator Compensation: Following its August 2012 proposal (the 2012 Loan Originator Proposal), the Bureau is issuing a final rule to implement provisions of the Dodd-Frank Act requiring certain creditors and loan originators to meet certain duties of care, including qualification requirements; requiring the establishment of certain compliance procedures by depository institutions; prohibiting loan originators, creditors, and the affiliates of both from receiving compensation in various forms (including based on the terms of the transaction) and from sources other than the consumer, with specified exceptions; and establishing restrictions on mandatory arbitration and financing of single premium credit insurance, pursuant to TILA sections 129B and 129C as established by Dodd-Frank Act sections 1402, 1403, and 1414(a). 15 U.S.C. 1639b, 1639c. The Bureau's final rule is referred to as the 2013 Loan Originator Final Rule.

• Appraisals: The Bureau, jointly with other Federal agencies, is issuing a final rule implementing Dodd-Frank Act requirements concerning appraisals for higher-risk mortgages, pursuant to TILA section 129H as established by Dodd-Frank Act section 1471. 15 U.S.C. 1639h. This rule follows the agencies' August 2012 joint proposal (the 2012 Interagency Appraisals Proposal). The agencies' joint final rule is referred to as the 2013 Interagency Appraisals Final Rule. In addition, following its August 2012 proposal (the 2012 ECOA Appraisals Proposal), the Bureau is issuing a final rule to implement provisions of the Dodd-Frank Act requiring that creditors provide applicants with a free copy of written appraisals and valuations developed in connection with applications for loans secured by a first lien on a dwelling, pursuant to section 701(e) of the Equal Credit Opportunity Act (ECOA) as amended by Dodd-Frank Act section 1474. 15 U.S.C. 1691(e). The Bureau's final rule is referred to as the 2013 ECOA Appraisals Final Rule.

The Bureau is not at this time finalizing proposals concerning various disclosure requirements that were added by title XIV of the Dodd-Frank Act, integration of mortgage disclosures under TILA and RESPA, or a simpler, more inclusive definition of the finance charge for purposes of disclosures for closed-end mortgage transactions under Regulation Z. The Bureau expects to finalize these proposals and to consider whether to adjust regulatory thresholds under the Title XIV Rulemakings in connection with any change in the calculation of the finance charge later in 2013, after it has completed quantitative testing, and any additional qualitative testing deemed appropriate, of the forms that it proposed in July 2012 to combine TILA mortgage disclosures with the good faith estimate (RESPA GFE) and settlement statement (RESPA settlement statement) required under the Real Estate Settlement Procedures Act, pursuant to Dodd-Frank Act section 1032(f) and sections 4(a) of RESPA and 105(b) of TILA, as amended by Dodd-Frank Act sections 1098 and 1100A, respectively (the 2012 TILA-RESPA Proposal). Accordingly, the Bureau already has issued a final rule delaying implementation of various affected title XIV disclosure provisions. The Bureau's approaches to coordinating the implementation of the Title XIV Rulemakings and to the finance charge proposal are discussed in turn below.

Coordinated Implementation of Title XIV Rulemakings

As noted in all of its foregoing proposals, the Bureau regards each of the Title XIV Rulemakings as affecting aspects of the mortgage industry and its regulations. Accordingly, as noted in its proposals, the Bureau is coordinating carefully the Title XIV Rulemakings, particularly with respect to their effective dates. The Dodd-Frank Act requirements to be implemented by the Title XIV Rulemakings generally will take effect on January 21, 2013, unless final rules implementing those requirements are issued on or before that date and provide for a different effective date. See Dodd-Frank Act section 1400(c), 15 U.S.C. 1601 note. In addition, some of the Title XIV Rulemakings are to take effect no later than one year after they are issued. Id.

The comments on the appropriate effective date for this final rule are discussed in detail below in part VI of this notice. In general, however, consumer advocates requested that the Bureau put the protections in the Title XIV Rulemakings into effect as soon as practicable. In contrast, the Bureau received some industry comments indicating that implementing so many new requirements at the same time would create a significant cumulative burden for creditors. In addition, many commenters also acknowledged the advantages of implementing multiple revisions to the regulations in a coordinated fashion. Thus, a tension exists between coordinating the adoption of the Title XIV Rulemakings and facilitating industry's implementation of such a large set of new requirements. Some have suggested that the Bureau resolve this tension by adopting a sequenced implementation, while others have requested that the Bureau simply provide a longer implementation period for all of the final rules.

The Bureau recognizes that many of the new provisions will require creditors to make changes to automated systems and, further, that most administrators of large systems are reluctant to make too many changes to their systems at once. At the same time, however, the Bureau notes that the Dodd-Frank Act established virtually all of these changes to institutions' compliance responsibilities, and contemplated that they be implemented in a relatively short period of time. And, as already noted, the extent of interaction among many of the Title XIV Rulemakings necessitates that many of their provisions take effect together. Finally, notwithstanding commenters' expressed concerns for cumulative burden, the Bureau expects that creditors actually may realize some efficiencies from adapting their systems for compliance with multiple new, closely related requirements at once, especially if given sufficient overall time to do so.

Accordingly, the Bureau is requiring that, as a general matter, creditors and other affected persons begin complying with the final rules on January 10, 2014. As noted above, section 1400(c) of the Dodd-Frank Act requires that some provisions of the Title XIV Rulemakings take effect no later than one year after the Bureau issues them. Accordingly, the Bureau is establishing January 10, 2014, one year after issuance of this final rule and the Bureau's 2013 Escrows and HOEPA Final Rules (i.e., the earliest of the title XIV final rules), as the baseline effective date for most of the Title XIV Rulemakings. The Bureau believes that, on balance, this approach will facilitate the implementation of the rules' overlapping provisions, while also affording creditors sufficient time to implement the more complex or resource-intensive new requirements.

The Bureau has identified certain rulemakings or selected aspects thereof, however, that do not present significant implementation burdens for industry. Accordingly, the Bureau is setting earlier effective dates for those final rules or certain aspects thereof, as applicable. Those effective dates are set forth and explained in the Federal Register s notices for those final rules.

More Inclusive Finance Charge Proposal

As noted above, the Bureau proposed in the 2012 TILA-RESPA Proposal to make the definition of finance charge more inclusive, thus rendering the finance charge and annual percentage rate a more useful tool for consumers to compare the cost of credit across different alternatives. 77 FR 51116, 51143 (Aug. 23, 2012). Because the new definition would include additional costs that are not currently counted, it would cause the finance charges and APRs on many affected transactions to increase. This in turn could cause more such transactions to become subject to various compliance regimes under Regulation Z. Specifically, the finance charge is central to the calculation of a transaction's “points and fees,” which in turn has been (and remains) a coverage threshold for the special protections afforded “high-cost mortgages” under HOEPA. Points and fees also will be subject to a 3-percent limit for purposes of determining whether a transaction is a “qualified mortgage” under this final rule. Meanwhile, the APR serves as a coverage threshold for HOEPA protections as well as for certain protections afforded “higher-priced mortgage loans” under § 1026.35, including the mandatory escrow account requirements being amended by the 2013 Escrows Final Rule. Finally, because the 2013 Interagency Appraisals Final Rule uses the same APR-based coverage test as is used for identifying higher-priced mortgage loans, the APR affects that rulemaking as well. Thus, the proposed more inclusive finance charge would have had the indirect effect of increasing coverage under HOEPA and the escrow and appraisal requirements for higher-priced mortgage loans, as well as decreasing the number of transactions that may be qualified mortgages—even holding actual loan terms constant—simply because of the increase in calculated finance charges, and consequently APRs, for closed-end mortgage transactions generally.

As noted above, these expanded coverage consequences were not the intent of the more inclusive finance charge proposal. Accordingly, as discussed more extensively in the Escrows Proposal, the HOEPA Proposal, the ATR Proposal, and the Interagency Appraisals Proposal, the Board and subsequently the Bureau (and other agencies) sought comment on certain adjustments to the affected regulatory thresholds to counteract this unintended effect. First, the Board and then the Bureau proposed to adopt a “transaction coverage rate” for use as the metric to determine coverage of these regimes in place of the APR. The transaction coverage rate would have been calculated solely for coverage determination purposes and would not have been disclosed to consumers, who still would have received only a disclosure of the expanded APR. The transaction coverage rate calculation would exclude from the prepaid finance charge all costs otherwise included for purposes of the APR calculation except charges retained by the creditor, any mortgage broker, or any affiliate of either. Similarly, the Board and Bureau proposed to reverse the effects of the more inclusive finance charge on the calculation of points and fees; the points and fees figure is calculated only as a HOEPA and qualified mortgage coverage metric and is not disclosed to consumers. The Bureau also sought comment on other potential mitigation measures, such as adjusting the numeric thresholds for particular compliance regimes to account for the general shift in affected transactions' APRs.

The Bureau's 2012 TILA-RESPA Proposal sought comment on whether to finalize the more inclusive finance charge proposal in conjunction with the Title XIV Rulemakings or with the rest of the TILA-RESPA Proposal concerning the integration of mortgage disclosure forms. 77 FR 51116, 51125 (Aug. 23, 2012). Upon additional consideration and review of comments received, the Bureau decided to defer a decision whether to adopt the more inclusive finance charge proposal and any related adjustments to regulatory thresholds until it later finalizes the TILA-RESPA Proposal. 77 FR 54843 (Sept. 6, 2012); 77 FR 54844 (Sept. 6, 2012). Accordingly, this final rule and the 2013 Escrows, HOEPA, and Interagency Appraisals Final Rules all are deferring any action on their respective proposed adjustments to regulatory thresholds.

IV. Legal Authority

The final rule was issued on January 10, 2013, in accordance with 12 CFR 1074.1. The Bureau issued this final rule pursuant to its authority under TILA and the Dodd-Frank Act. See TILA section 105(a), 15 U.S.C. 1604(a). On July 21, 2011, section 1061 of the Dodd-Frank Act transferred to the Bureau the “consumer financial protection functions” previously vested in certain other Federal agencies, including the Board. The term “consumer financial protection function” is defined to include “all authority to prescribe rules or issue orders or guidelines pursuant to any Federal consumer financial law, including performing appropriate functions to promulgate and review such rules, orders, and guidelines.” TILA is defined as a Federal consumer financial law. Accordingly, the Bureau has authority to issue regulations pursuant to TILA.

A. TILA Ability-to-Repay and Qualified Mortgage Provisions

As discussed above, the Dodd-Frank Act amended TILA to generally prohibit a creditor from making a residential mortgage loan without a reasonable and good faith determination that, at the time the loan is consummated, the consumer has a reasonable ability to repay the loan, along with taxes, insurance, and assessments. TILA section 129C(a), 15 U.S.C. 1639c(a). As described below in part IV.B, the Bureau has authority to prescribe regulations to carry out the purposes of TILA pursuant to TILA section 105(a). 15 U.S.C. 1604(a). In particular, it is the purpose of TILA section 129C, as amended by the Dodd-Frank Act, to assure that consumers are offered and receive residential mortgage loans on terms that reasonably reflect their ability to repay the loans and that are understandable and not unfair, deceptive, and abusive. TILA section 129B(a)(2); 15 U.S.C. 1639b(a)(2).

The Dodd-Frank Act also provides creditors originating “qualified mortgages” special protection from liability under the ability-to-repay requirements. TILA section 129C(b), 15 U.S.C. 1639c(b). TILA generally defines a “qualified mortgage” as a residential mortgage loan for which: the loan does not contain negative amortization, interest-only payments, or balloon payments; the term does not exceed 30 years; the points and fees generally do not exceed three percent of the loan amount; the income or assets are considered and verified; and the underwriting is based on the maximum rate during the first five years, uses a payment schedule that fully amortizes the loan over the loan term, and takes into account all mortgage-related obligations. TILA section 129C(b)(2), 15 U.S.C. 1639c(b)(2). In addition, to constitute a qualified mortgage a loan must meet “any guidelines or regulations established by the Bureau relating to ratios of total monthly debt to monthly income or alternative measures of ability to pay regular expenses after payment of total monthly debt, taking into account the income levels of the borrower and such other factors as the Bureau may determine are relevant and consistent with the purposes described in [TILA section 129C(b)(3)(B)(i)].”

The Dodd-Frank Act also provides the Bureau with authority to prescribe regulations that revise, add to, or subtract from the criteria that define a qualified mortgage upon a finding that such regulations are necessary or proper to ensure that responsible, affordable mortgage credit remains available to consumers in a manner consistent with the purposes of the ability-to-repay requirements; or are necessary and appropriate to effectuate the purposes of the ability-to-repay requirements, to prevent circumvention or evasion thereof, or to facilitate compliance with TILA sections 129B and 129C. TILA section 129C(b)(3)(B)(i), 15 U.S.C. 1639c(b)(3)(B)(i). In addition, TILA section 129C(b)(3)(A) provides the Bureau with authority to prescribe regulations to carry out the purposes of the qualified mortgage provisions, such as to ensure that responsible, affordable mortgage credit remains available to consumers in a manner consistent with the purposes of TILA section 129C. TILA section 129C(b)(3)(A), 15 U.S.C. 1939c(b)(3)(A). As discussed in the section-by-section analysis below, the Bureau is issuing certain provisions of this rule pursuant to its authority under TILA section 129C(b)(3)(B)(i).

The Dodd-Frank Act provides the Bureau with other specific grants of rulewriting authority with respect to the ability-to-repay and qualified mortgage provisions. With respect to the ability-to-repay provisions, TILA section 129C(a)(6)(D)(i) through (iii) provides that when calculating the payment obligation that will be used to determine whether the consumer can repay a covered transaction, the creditor must use a fully amortizing payment schedule and assume that: (1) The loan proceeds are fully disbursed on the date the loan is consummated; (2) the loan is repaid in substantially equal, monthly amortizing payments for principal and interest over the entire term of the loan with no balloon payment; and (3) the interest rate over the entire term of the loan is a fixed rate equal to the fully indexed rate at the time of the loan closing, without considering the introductory rate. 15 U.S.C. 1639c(a)(6)(D)(i) through (iii). However, TILA section 129C(a)(6)(D) authorizes the Bureau to prescribe regulations for calculating the payment obligation for loans that require more rapid repayment (including balloon payments), and which have an annual percentage rate that does not exceed a certain rate threshold. 15 U.S.C. 1639c(a)(6)(D).

With respect to the qualified mortgage provisions, the Dodd-Frank Act contains several specific grants of rulewriting authority. First, as described above, for purposes of defining “qualified mortgage,” TILA section 129C(b)(2)(A)(vi) provides the Bureau with authority to establish guidelines or regulations relating to monthly debt-to-income ratios or alternative measures of ability to pay. Second, TILA section 129C(b)(2)(D) provides that the Bureau shall prescribe rules adjusting the qualified mortgage points and fees limits described above to permit creditors that extend smaller loans to meet the requirements of the qualified mortgage provisions. 15 U.S.C. 1639c(b)(2)(D)(ii). In prescribing such rules, the Bureau must consider their potential impact on rural areas and other areas where home values are lower. Id. Third, TILA section 129C(b)(2)(E) provides the Bureau with authority to include in the definition of “qualified mortgage” loans with balloon payment features, if those loans meet certain underwriting criteria and are originated by creditors that operate predominantly in rural or underserved areas, have total annual residential mortgage originations that do not exceed a limit set by the Bureau, and meet any asset size threshold and any other criteria as the Bureau may establish, consistent with the purposes of TILA. 15 U.S.C. 1639c(b)(2)(E). As discussed in the section-by-section analysis below, the Bureau is issuing certain provisions of this rule pursuant to its authority under TILA sections 129C(a)(6)(D), (b)(2)(A)(vi), (b)(2)(D), and (b)(2)(E).

B. Other Rulemaking and Exception Authorities

This final rule also relies on other rulemaking and exception authorities specifically granted to the Bureau by TILA and the Dodd-Frank Act, including the authorities discussed below.

TILA

TILA section 105(a). As amended by the Dodd-Frank Act, TILA section 105(a), 15 U.S.C. 1604(a), directs the Bureau to prescribe regulations to carry out the purposes of TILA, and provides that such regulations may contain additional requirements, classifications, differentiations, or other provisions, and may provide for such adjustments and exceptions for all or any class of transactions that the Bureau judges are necessary or proper to effectuate the purposes of TILA, to prevent circumvention or evasion thereof, or to facilitate compliance therewith. A purpose of TILA is “to assure a meaningful disclosure of credit terms so that the consumer will be able to compare more readily the various credit terms available to him and avoid the uninformed use of credit.” TILA section 102(a), 15 U.S.C. 1601(a). This stated purpose is informed by Congress's finding that “economic stabilization would be enhanced and the competition among the various financial institutions and other firms engaged in the extension of consumer credit would be strengthened by the informed use of credit[.]” TILA section 102(a). Thus, strengthened competition among financial institutions is a goal of TILA, achieved through the effectuation of TILA's purposes.

Historically, TILA section 105(a) has served as a broad source of authority for rules that promote the informed use of credit through required disclosures and substantive regulation of certain practices. However, Dodd-Frank Act section 1100A clarified the Bureau's section 105(a) authority by amending that section to provide express authority to prescribe regulations that contain “additional requirements” that the Bureau finds are necessary or proper to effectuate the purposes of TILA, to prevent circumvention or evasion thereof, or to facilitate compliance therewith. This amendment clarified the authority to exercise TILA section 105(a) to prescribe requirements beyond those specifically listed in the statute that meet the standards outlined in section 105(a). The Dodd-Frank Act also clarified the Bureau's rulemaking authority over high-cost mortgages under HOEPA pursuant to section 105(a). As amended by the Dodd-Frank Act, TILA section 105(a) authority to make adjustments and exceptions to the requirements of TILA applies to all transactions subject to TILA, except with respect to the substantive provisions of TILA section 129, 15 U.S.C. 1639, that apply to the high-cost mortgages defined in TILA section 103(bb), 15 U.S.C. 1602(bb).

TILA, as amended by the Dodd-Frank Act, states that it is the purpose of the ability-to-repay requirements of TILA section 129C to assure that consumers are offered and receive residential mortgage loans on terms that reasonably reflect their ability to repay the loans and that are understandable and not unfair, deceptive, or abusive. TILA section 129B(a)(2). The Bureau interprets this addition as a new purpose of TILA. Therefore, the Bureau believes that its authority under TILA section 105(a) to make exceptions, adjustments, and additional provisions, among other things, that the Bureau finds are necessary or proper to effectuate the purposes of TILA, to prevent circumvention or evasion thereof, or to facilitate compliance therewith applies with respect to the purpose of section 129C as well as the purpose described in section TILA section 129B(a)(2).

The purpose of TILA section 129C is informed by the findings articulated in section 129B(a) that economic stabilization would be enhanced by the protection, limitation, and regulation of the terms of residential mortgage credit and the practices related to such credit, while ensuring that responsible and affordable mortgage credit remains available to consumers.

As discussed in the section-by-section analysis below, the Bureau is issuing regulations to carry out TILA's purposes, including such additional requirements, adjustments, and exceptions as, in the Bureau's judgment, are necessary and proper to carry out the purposes of TILA, prevent circumvention or evasion thereof, or to facilitate compliance therewith. In developing these aspects of the final rule pursuant to its authority under TILA section 105(a), the Bureau has considered the purposes of TILA, including the purposes of TILA section 129C, and the findings of TILA, including strengthening competition among financial institutions and promoting economic stabilization, and the findings of TILA section 129B(a)(1), that economic stabilization would be enhanced by the protection, limitation, and regulation of the terms of residential mortgage credit and the practices related to such credit, while ensuring that responsible, affordable mortgage credit remains available to consumers. The Bureau believes that ensuring that mortgage credit is offered and received on terms consumers can afford ensures the availability of responsible, affordable mortgage credit.

TILA section 129B(e). Dodd-Frank Act section 1405(a) amended TILA to add new section 129B(e), 15 U.S.C. 1639B(e). That section authorizes the Bureau to prohibit or condition terms, acts, or practices relating to residential mortgage loans that the Bureau finds to be abusive, unfair, deceptive, predatory, necessary or proper to ensure that responsible, affordable mortgage credit remains available to consumers in a manner consistent with the purposes of TILA section 129C, necessary or proper to effectuate the purposes of sections 129B and 129C, to prevent circumvention or evasion thereof, or to facilitate compliance with such sections, or are not in the interest of the consumer. In developing rules under TILA section 129B(e), the Bureau has considered whether the rules are in the interest of the consumer, as required by the statute. As discussed in the section-by-section analysis below, the Bureau is issuing portions of this rule pursuant to its authority under TILA section 129B(e).

The Dodd-Frank Act

Dodd-Frank Act section 1022(b). Section 1022(b)(1) of the Dodd-Frank Act authorizes the Bureau to prescribe rules “as may be necessary or appropriate to enable the Bureau to administer and carry out the purposes and objectives of the Federal consumer financial laws, and to prevent evasions thereof.” 12 U.S.C. 5512(b)(1). TILA and title X of the Dodd-Frank Act are Federal consumer financial laws. Accordingly, the Bureau is exercising its authority under Dodd-Frank Act section 1022(b) to prescribe rules that carry out the purposes and objectives of TILA and title X and prevent evasion of those laws.

V. Section-by-Section Analysis

Section 1026.25 Record Retention

25(a) General Rule

Section 1416 of the Dodd-Frank Act revised TILA section 130(e) to extend the statute of limitations for civil liability for a violation of TILA section 129C, as well as sections 129 and 129B, to three years after the date a violation occurs. Existing § 1026.25(a) requires that creditors retain evidence of compliance with Regulation Z for two years after disclosures must be made or action must be taken. Accordingly, the Board proposed to revise § 226.25(a) to require that creditors retain records that show compliance with proposed § 226.43, which would implement TILA section 129C, for at least three years after consummation. The Board did not propose to alter the regulation's existing clarification that administrative agencies responsible for enforcing Regulation Z may require creditors under the agency's jurisdiction to retain records for a longer period, if necessary to carry out the agency's enforcement responsibilities under TILA section 108, 15 U.S.C. 1607. Under TILA section 130(e), as amended by Dodd-Frank, the statute of limitations for civil liability for a violation of other sections of TILA remains one year after the date a violation occurs, except for private education loans under 15 U.S.C. 1650(a), actions brought under section 129 or 129B, or actions brought by a State attorney general to enforce a violation of section 129, 129B, 129C, 129D, 129E, 129F, 129G, or 129H. 15 U.S.C. 1640(e). Moreover, as amended by section 1413 of the Dodd-Frank Act, TILA provides that when a creditor, an assignee, other holder or their agent initiates a foreclosure action, a consumer may assert a violation of TILA section 129C(a) “as a matter of defense by recoupment or setoff.” TILA section 130(k). There is no time limit on the use of this defense.

As discussed below, the Bureau is adopting minor modifications to § 1026.25(a) and adding in new § 1026.25(c) to reflect section 1416 of the Dodd-Frank Act, in § 1026.25(c)(3) as well as other exceptional record retention requirements related to mortgage loans.

25(c) Records Related to Certain Requirements for Mortgage Loans

The Bureau is adopting the revision proposed in § 226.25(a) to require a creditor to retain records demonstrating compliance with § 1026.43 consistent with the extended statute of limitations for violations of that section, though the Bureau is adopting this requirement in § 1026.25(c)(3) to provide additional clarity. As the 2012 TILA-RESPA Proposal proposed new § 1026.25(c)(1) and the 2012 Loan Originator Proposal proposed new § 1026.25(c)(2), the Bureau concludes that adding new § 1026.25(c)(3) eases compliance burden by placing all record retention requirements that are related to mortgage loans and which differ from the general record retention in one section, § 1026.25(c). Likewise, the Bureau is amending § 1026.25(a) to reflect that certain record retention requirements, such as records related to minimum standards for transactions secured by a dwelling, are governed by § 1026.43(c).

Commenters did not provide the Bureau with significant, specific feedback with respect to proposed § 226.25(a), although industry commenters generally expressed concern with respect to the compliance burden of the 2011 ATR Proposal. Increasing the period a creditor must retain records from two to three years may impose some marginal increase in the creditor's compliance burden in the form of incremental cost of storage. However, the Bureau believes that even absent the rule, responsible creditors will likely elect to retain records of compliance with § 1026.43 for a period of time well beyond three years, given that the statute allows consumers to bring a defensive claim for recoupment or setoff in the event that a creditor or assignee initiates foreclosure proceedings. Indeed, at least one commenter noted this tension and requested that the Bureau provide further regulatory instruction, although the Bureau does not deem it necessary to mandate recordkeeping burdens beyond what is required by section 1416 of the Dodd-Frank Act. Furthermore, the record-keeping burden imposed by the rule is tailored only to show compliance with § 1026.43, and the Bureau believes is justified to protect the interests of both creditors and consumers in the event that an affirmative claim is brought during the first three years after consummation.

The Bureau believes that calculating the record retention period under § 1026.43 from loan consummation facilitates compliance by establishing a single, clear start to the period, even though a creditor will take action (e.g., underwriting the covered transaction and offering a consumer the option of a covered transaction without a prepayment penalty) over several days or weeks prior to consummation. The Bureau is thus adopting the timeframe as proposed to reduce compliance burden.

Existing comment 25(a)-2 clarifies that, in general, a creditor need retain only enough information to reconstruct the required disclosures or other records. The Board proposed, and the Bureau is adopting, amendments to comment 25(a)-2 and a new comment 25(c)(3)-1 to clarify that, if a creditor must verify and document information used in underwriting a transaction subject to § 1026.43, the creditor must retain evidence sufficient to demonstrate having done so, in compliance with § 1026.25(a) and § 1026.25(c)(3). In an effort to reduce compliance burden, comment 25(c)(3)-1 also clarifies that creditors need not retain actual paper copies of the documentation used to underwrite a transaction but that creditors must be able to reproduce those records accurately.

The Board proposed comment 25(a)-7 to provide guidance on retaining records evidencing compliance with the requirement to offer a consumer an alternative covered transaction without a prepayment penalty, as discussed below in the section-by-section analysis of § 1026.43(g)(3) through (5). The Bureau believes the requirement to offer a transaction without a prepayment penalty under TILA section 129C(c)(4) is intended to ensure that consumers who choose an alternative covered transaction with a prepayment penalty do so voluntarily. The Bureau further believes it is unnecessary, and contrary to the Bureau's efforts to streamline its regulations, facilitate regulatory compliance, and minimize compliance burden, for a creditor to document compliance with the requirement to offer an alternative covered transaction without a prepayment penalty when a consumer does not choose a transaction with a prepayment penalty or if the covered transaction is not consummated. Accordingly, the Bureau is adopting as proposed comment 25(a)-7 as comment 25(c)(3)-2, to clarify that a creditor must retain records that document compliance with that requirement if a transaction subject to § 1026.43 is consummated with a prepayment penalty, but need not retain such records if a covered transaction is consummated without a prepayment penalty or a covered transaction is not consummated. See § 1026.43(g)(6).

The Board proposed comment 25(a)-7 also to provide specific guidance on retaining records evidencing compliance with the requirement to offer a consumer an alternative covered transaction without a prepayment penalty when a creditor offers a transaction through a mortgage broker. As discussed in detail below in the section-by-section analysis of § 1026.43(g)(4), the Board proposed that if the creditor offers a covered transaction with a prepayment penalty through a mortgage broker, the creditor must present the mortgage broker an alternative covered transaction without a prepayment penalty. Also, the creditor must provide, by agreement, for the mortgage broker to present to the consumer that transaction or an alternative covered transaction without a prepayment penalty offered by another creditor that has a lower interest rate or a lower total dollar amount of origination points or fees and discount points than the creditor's presented alternative covered transaction. The Bureau did not receive significant comment on this clarification, and is adopting the comment largely as proposed, renumbered as comment 25(c)(3)-2. Comment 25(c)(3)-2 also clarifies that, to demonstrate compliance with § 1026.43(g)(4), the creditor must retain a record of (1) the alternative covered transaction without a prepayment penalty presented to the mortgage broker pursuant to § 1026.43(g)(4)(i), such as a rate sheet, and (2) the agreement with the mortgage broker required by § 1026.34(g)(4)(ii).

Section 1026.32 Requirements for High-Cost Mortgages

32(b) Definitions

32(b)(1)

Points and Fees—General

Section 1412 of the Dodd-Frank Act added TILA section 129C(b)(2)(A)(vii), which defines a “qualified mortgage” as a loan for which, among other things, the total “points and fees” do not exceed 3 percent of the total loan amount. The limits on points and fees for qualified mortgages are implemented in new § 1026.43(e)(3).

TILA section 129C(b)(2)(C) generally defines “points and fees” for qualified mortgages to have the same meaning as in TILA section 103(aa)(4) (renumbered as section 103(bb)(4)), which defines “points and fees” for the purpose of determining whether a transaction qualifies as a high-cost mortgage under HOEPA. TILA section 103(aa)(4) is implemented in current § 1026.32(b)(1). Accordingly, the Board proposed in § 226.43(b)(9) that, for a qualified mortgage, “points and fees” has the same meaning as in § 226.32(b)(1).

The Board also proposed in the 2011 ATR Proposal to amend § 226.32(b)(1) to implement revisions to the definition of “points and fees” under section 1431 of the Dodd-Frank Act. Among other things, the Dodd-Frank Act excluded certain private mortgage insurance premiums from, and added loan originator compensation and prepayment penalties to, the definition of “points and fees” that had previously applied to high-cost mortgage loans under HOEPA. In the Bureau's 2012 HOEPA Proposal, the Bureau republished the Board's proposed revisions to § 226.32(b)(1), with only minor changes, in renumbered § 1026.32(b)(1).

The Bureau noted in its 2012 HOEPA Proposal that it was particularly interested in receiving comments concerning any newly-proposed language and the application of the definition in the high-cost mortgage context. The Bureau received numerous comments from both industry and consumer advocacy groups, the majority of which were neither specific to newly-proposed language nor to the application of the definition to high-cost mortgages. These comments largely reiterated comments that the Board and the Bureau had received in the ATR rulemaking docket. The Bureau is addressing comments received in response to 2012 HOEPA Proposal in the 2013 HOEPA Final Rule. Similarly, comments received in response to the Board's 2011 ATR Proposal are discussed in this final rule. The Bureau is carefully coordinating the 2013 HOEPA and ATR Final Rules to ensure a consistent and cohesive regulatory framework. The Bureau is now finalizing § 1026.32(b)(1), (b)(3), (b)(4)(i), (b)(5), and (b)(6)(i) in this rule in response to the comments received on both proposals. The Bureau is finalizing § 1026.32(b)(2), (b)(4)(ii), and (b)(6)(ii) in the 2013 HOEPA Final Rule.

Existing § 1026.32(b)(1) defines “points and fees” by listing included charges in § 1026.32(b)(1)(i) through (iv). As discussed below, the Board proposed revisions to § 226.32(b)(1)(i) through (iv) and proposed to add new § 226.32(b)(1)(v) and (vi). In the 2012 HOEPA Proposal, the Bureau proposed to add the phrase “in connection with a closed-end mortgage loan” to § 1026.32(b)(1) to clarify that its definition of “points and fees” would have applied only for closed-end mortgages. The Bureau also proposed to define “points and fees” in § 1026.32(b)(3) for purposes of defining which open-end credit plans qualify as “high-cost mortgages” under HOEPA. However, that section is not relevant to this rulemaking because the ability-to-repay requirement in TILA section 129C does not apply to open-end credit. Accordingly, the Bureau is adopting § 1026.32(b)(1) with the clarification that its definition of “points and fees” is “in connection with a closed-end mortgage loan.”

Payable at or before consummation. In the 2011 ATR Proposal, the Board noted that the Dodd-Frank Act removed the phrase “payable at or before closing” from the high-cost mortgage points and fees test in TILA section 103(aa)(1)(B). See TILA section 103(bb)(1)(A)(ii). Prior to the Dodd-Frank Act, fees and charges were included in points and fees for the high-cost mortgage points and fees test only if they were payable at or before closing. The phrase “payable at or before closing” is also not in TILA's provisions on the points and fees cap for qualified mortgages. See TILA section 129C(b)(2)(A)(vii), (b)(2)(C). Thus, the Board stated that, with a few exceptions, the statute provides that any charge that falls within the “points and fees” definition must be counted toward the limits on points and fees for both high-cost mortgages and qualified mortgages, even if it is payable after loan closing. The Board noted that the exceptions are mortgage insurance premiums and charges for credit insurance and debt cancellation and suspension coverage. The statute expressly states that these premiums and charges are included in points and fees only if payable at or before closing. See TILA section 103(bb)(1)(C) (for mortgage insurance) and TILA section 103(bb)(4)(D) (for credit insurance and debt cancellation and suspension coverage).

The Board expressed concern that some fees that occur after closing, such as fees to modify a loan, might be deemed to be points and fees. If so, the Board cautioned that calculating the points and fees to determine whether a transaction is a qualified mortgage may be difficult because the amount of future fees (e.g., loan modification fees) cannot be known prior to closing. The Board noted that creditors might be exposed to excessive litigation risk if consumers were able at any point during the life of a mortgage to argue that the points and fees for the loan exceed the qualified mortgage limits due to fees imposed after loan closing. The Board expressed concern that creditors therefore might be discouraged from making qualified mortgages, which would undermine Congress's goal of increasing incentives for creditors to make more stable, affordable loans. The Board requested comment on whether any other types of fees should be included in points and fees only if they are “payable at or before closing.”

Several industry commenters stated that charges paid after closing should not be included in points and fees and requested that the Bureau clarify whether such charges are included. For example, some industry commenters sought confirmation that charges for a subsequent loan modification would not be included in points and fees. More generally, industry commenters argued that they would have difficulty calculating charges that would be paid after closing and that including such charges in points and fees would create uncertainty and litigation risk. In response to the Bureau's 2012 HOEPA Proposal, one consumer advocate noted that there are inconsistent and confusing standards for when charges must be payable to be included in points and fees. This commenter recommended that the Bureau adopt a “known at or before closing” standard, arguing that this standard would clarify that financed points are included, would prevent creditors from evading the points and fees test by requiring consumers to pay charges after consummation, and would provide certainty to creditors that must know the amount of points and fees at or before closing.

The Bureau appreciates that creditors need certainty in calculating points and fees so they can ensure that they are originating qualified mortgages (or are not exceeding the points and fees thresholds for high-cost mortgages). The Dodd-Frank Act provides that for the points and fees tests for both qualified mortgages and high-cost mortgages, only charges “payable in connection with” the transaction are included in points and fees. See TILA sections 103(bb)(1)(A)(ii) (high-cost mortgages) and 129C(b)(2)(A)(vii) (qualified mortgages). The Bureau interprets this “in connection with” requirement as limiting the universe of charges that need to be included in points and fees. To clarify when charges or fees are “in connection with” a transaction, the Bureau is specifying in § 1026.32(b)(1) that fees or charges are included in points and fees only if they are “known at or before consummation.”

The Bureau is also adding new comment 32(b)(1)-1, which provides examples of fees and charges that are and are not known at or before consummation. The comment explains that charges for a subsequent loan modification generally would not be included in points and fees because, at consummation, the creditor would not know whether a consumer would seek to modify the loan and therefore would not know whether charges in connection with a modification would ever be imposed. Indeed, loan modification fees likely would not be included in the finance charge under § 1026.4, as they would not be charges imposed by creditor as an incident to or a condition of the extension of credit. Thus, this clarification is consistent with the definition of the finance charge. Comment 32(b)(1)-1 also clarifies that the maximum prepayment penalties that may be charged or collected under the terms of a mortgage loan are included in points and fees under § 1026.32(b)(1)(v). In addition, comment 32(b)(1)-1 notes that, under § 1026.32(b)(1)(i)(C)(1) and (iv), premiums or other charges for private mortgage insurance and credit insurance payable after consummation are not included in points and fees. This means that such charges may be included in points and fees only if they are payable at or before consummation. Thus, even if the amounts of such premiums or other charges are known at or before consummation, they are included in points and fees only if they are payable at or before consummation.

32(b)(1)(i)

Points and Fees—Included in the Finance Charge

TILA section 103(aa)(4)(A) specifies that “points and fees” includes all items included in the finance charge, except interest or the time-price differential. This provision is implemented in current § 1026.32(b)(1)(i). Section 1431 of the Dodd-Frank Act added TILA section 103(bb)(1)(C), which excludes from points and fees certain types and amounts of mortgage insurance premiums.

The Board proposed to revise § 226.32(b)(1)(i) to implement these provisions. The Board proposed to move the exclusion of interest or the time-price differential to new § 226.32(b)(1)(i)(A). The Board also proposed to add § 226.32(b)(1)(i)(B) to implement the new exclusion for certain mortgage insurance. In § 226.32(b)(1)(i), the Board proposed to revise the phrase “all items required to be disclosed under § 226.4(a) and 226.4(b)” to read “all items considered to be a finance charge under § 226.4(a) and 226.4(b)” because § 226.4 does not itself require disclosure of the finance charge.

One industry commenter argued that the definition of points and fees was overbroad because it included all items considered to be a finance charge. The commenter asserted that several items that are included in the finance charge under § 1026.4(b) are vague or inapplicable in the context of mortgage transactions or duplicate items specifically addressed in other provisions. Several industry commenters also requested clarification about whether certain types of fees and charges are included in points and fees. At least two commenters asked that the Bureau clarify that closing agent costs are not included in points and fees.

The Bureau is adopting renumbered § 1026.32(b)(1)(i) and (i)(A) substantially as proposed, with certain clarifications in the commentary and in other parts of the rule as discussed below to address commenters' requests for clarification. For consistency with the language in § 1026.4, the Bureau is revising § 1026.32(b)(1)(i) to refer to “items included in the finance charge” rather than “items considered to be a finance charge.”

As noted above, several commenters requested clarification regarding whether certain types of charges would be included in points and fees. With respect to closing agent charges, § 1026.4(a)(2) provides a specific rule for when such charges must be included in the finance charge. If they are not included in the finance charge, they would not be included in points and fees. Moreover, as discussed below and in new comment 32(b)(1)(i)(D)-1, certain closing agent charges may also be excluded from points and fees as bona fide third-party charges that are not retained by the creditor, loan originator, or an affiliate of either.

The Board also proposed to revise comment 32(b)(1)(i)-1, which states that § 226.32(b)(1)(i) includes in the total “points and fees” items defined as finance charges under § 226.4(a) and 226.4(b). The comment explains that items excluded from the finance charge under other provisions of § 226.4 are not included in the total “points and fees” under § 226.32(b)(1)(i), but may be included in “points and fees” under § 226.32(b)(1)(ii) and (iii). The Board proposed to revise this comment to state that items excluded from the finance charge under other provisions of § 226.4 may be included in “points and fees” under § 226.32(b)(1)(ii) through (vi). The proposed revision was intended to reflect the additional items added to the definition of “points and fees” by the Dodd-Frank Act and corrected the previous omission of § 226.32(b)(1)(iv). See proposed § 226.32(b)(1)(v) and (vi).

The proposed comment also would have added an example of how this rule would operate. Under that example, a fee imposed by the creditor for an appraisal performed by an employee of the creditor meets the general definition of “finance charge” under § 226.4(a) as “any charge payable directly or indirectly by the consumer and imposed directly or indirectly by the creditor as an incident to or a condition of the extension of credit.” However, § 226.4(c)(7) expressly provides that appraisal fees are not finance charges. Therefore, under the general rule in proposed § 226.32(b)(1)(i) providing that finance charges must be counted as points and fees, a fee imposed by the creditor for an appraisal performed by an employee of the creditor would not have been counted in points and fees. Proposed § 226.32(b)(1)(iii), however, would have expressly included in points and fees items listed in § 226.4(c)(7) (including appraisal fees) if the creditor receives compensation in connection with the charge. A creditor would receive compensation for an appraisal performed by its own employee. Thus, the appraisal fee in this example would have been included in the calculation of points and fees.

The Bureau did not receive substantial comment on this proposed guidance. The Bureau is adopting comment 32(b)(1)(i)-1, with certain revisions for clarity. As revised, comment 32(b)(1)(i)-1 explains that certain items that may be included in the finance charge under § 1026.32(b)(1)(i) are excluded under § 1026.32(b)(1)(i)(A) through (F).

Mortgage Insurance

Under existing § 1026.32(b)(1)(i), mortgage insurance premiums are included in the finance charge and therefore are included in points and fees if payable at or before closing. As noted above, the Board proposed new § 226.32(b)(1)(i)(B) to implement TILA section 103(bb)(1)(C), which provides that points and fees shall exclude certain charges for mortgage insurance premiums. Specifically, the statute excludes: (1) Any premium charged for insurance provided by an agency of the Federal Government or an agency of a State; (2) any amount that is not in excess of the amount payable under policies in effect at the time of origination under section 203(c)(2)(A) of the National Housing Act, provided that the premium, charge, or fee is required to be refundable on a pro-rated basis and the refund is automatically issued upon notification of the satisfaction of the underlying mortgage loan; and (3) any premium paid by the consumer after closing.

The Board noted that the exclusions for certain premiums could plausibly be interpreted to apply to the definition of points and fees solely for purposes of high-cost mortgages and not for qualified mortgages. TILA section 129C(b)(2)(C)(i) cross-references TILA section 103(aa)(4) (renumbered as 103(bb)(4)) for the definition of “points and fees,” but the provision on mortgage insurance appears in TILA section 103(bb)(1)(C) and not in section 103(bb)(4). The Board also noted that certain provisions in the Dodd-Frank Act's high-cost mortgage section regarding points and fees are repeated in the qualified mortgage section on points and fees. For example, both the high-cost mortgage provisions and the qualified mortgage provisions expressly exclude from points and fees “bona fide third party charges not retained by the mortgage originator, creditor, or an affiliate of the creditor or mortgage originator.” TILA sections 103(bb)(1)(A)(ii) (for high-cost mortgages), 129C(b)(2)(C)(i) (for qualified mortgages). The mortgage insurance provision, however, does not separately appear in the qualified mortgage section.

Nonetheless, the Board concluded that the better interpretation of the statute is that the mortgage insurance provision in TILA section 103(bb)(1)(C) applies to the meaning of points and fees for both high-cost mortgages and qualified mortgages. The Board noted that the statute's structure reasonably supports this view: by its plain language, the mortgage insurance provision prescribes how points and fees should be computed “for purposes of paragraph (4),” i.e., for purposes of TILA section 103(bb)(4). The mortgage insurance provision contains no caveat limiting its application solely to the points and fees calculation for high-cost mortgages. Thus, the Board determined that the cross-reference in the qualified mortgage provisions to TILA section 103(bb)(4) should be read to include provisions that expressly prescribe how points and fees should be calculated under TILA section 103(bb)(4), wherever located.

The Board noted that its proposal to apply the mortgage insurance provision to the meaning of points and fees for both high-cost mortgages and qualified mortgages is also supported by the Board's authority under TILA section 105(a) to make adjustments to facilitate compliance with TILA. The Board also cited its authority under TILA section 129B(e) to condition terms, acts or practices relating to residential mortgage loans that the Board finds necessary or proper to effectuate the purposes of TILA. The purposes of TILA include “assur[ing] that consumers are offered and receive residential mortgage loan on terms that reasonably reflect their ability to repay the loans.” TILA section 129B(a)(2).

The Board also expressed concern about the increased risk of confusion and compliance error if points and fees were to have two separate meanings in TILA—one for determining whether a loan is a high-cost mortgage and another for determining whether a loan is a qualified mortgage. The Board stated that the proposal is intended to facilitate compliance by applying the mortgage insurance provision to the meaning of points and fees for both high-cost mortgages and qualified mortgages.

In addition, the Board expressed concern that market distortions could result due to different treatment of mortgage insurance in calculating points and fees for high-cost mortgages and qualified mortgages. “Points and fees” for both high-cost mortgages and qualified mortgages generally excludes “bona fide third party charges not retained by the mortgage originator, creditor, or an affiliate of the creditor or mortgage originator.” TILA sections 103(bb)(1)(A)(ii), 129C(b)(2)(C)(i). Under this general provision standing alone, premiums for up-front private mortgage insurance would be excluded from points and fees. However, as noted, the statute's specific provision on mortgage insurance (TILA section 103(bb)(1)(C)) imposes certain limitations on the amount and conditions under which up-front premiums for private mortgage insurance are excluded from points and fees. Applying the mortgage insurance provision to the definition of points and fees only for high-cost mortgages would mean that any premium amount for up-front private mortgage insurance could be charged on qualified mortgages; in most cases, none of that amount would be subject to the cap on points and fees for qualified mortgages because it would be excluded as a “bona fide third party fee” that is not retained by the creditor, loan originator, or an affiliate of either. The Board noted that, as a result, consumers who obtain qualified mortgages could be vulnerable to paying excessive up-front private mortgage insurance costs. The Board concluded that this outcome would undercut Congress's clear intent to ensure that qualified mortgages are products with limited fees and more safe features.

For the reasons noted by the Board, the Bureau interprets the mortgage insurance provision in TILA section 103(bb)(1)(C) as applying to the meaning of points and fees for both high-cost mortgages and qualified mortgages. The Bureau is also adopting this approach pursuant to its authority under TILA sections 105(a) and 129C(b)(3)(B)(i). Applying the mortgage insurance provision to the meaning of points and fees for qualified mortgages is necessary and proper to effectuate the purposes of, and facilitate compliance with the purposes of, the ability-to-repay requirements in TILA section 129C. Similarly, the Bureau finds that it is necessary and proper to use its authority under TILA section 129C(b)(3)(B)(i) to revise, add to, or subtract from the criteria that define a qualified mortgage. As noted above, construing the mortgage insurance provision as applying to qualified mortgages will reduce the likelihood that consumers who obtain qualified mortgages will pay excessive private mortgage insurance premiums, and therefore will help ensure that responsible, affordable credit remains available to consumers in a manner consistent with the purposes of TILA section 129C.

Proposed § 226.32(b)(1)(i)(B) tracked the substance of the statute with one exception. The Board interpreted the statute as excluding from points and fees not only up-front mortgage insurance premiums under government programs but also charges for mortgage guaranties under government programs. The Board noted that it was proposing the exclusion from points and fees of both mortgage insurance premiums and guaranty fees under government programs pursuant to its authority under TILA section 105(a) to make adjustments to facilitate compliance with TILA and its purposes and to effectuate the purposes of TILA. The Board also found that the exclusion is further supported by the Board's authority under TILA section 129B(e) to condition terms, acts or practices relating to residential mortgage loans that the Board finds necessary or proper to effectuate the purposes of TILA. The purposes of TILA include “assur[ing] that consumers are offered and receive residential mortgage loan on terms that reasonably reflect their ability to repay the loans.” TILA section 129B(a)(2).

The Board noted that both the U.S. Department of Veterans Affairs (VA) and the U.S. Department of Agriculture (USDA) expressed concerns that, if up-front charges for guaranties provided by those agencies and State agencies were included in points and fees, their loans might exceed high-cost thresholds and exceed the cap for qualified mortgages, thereby disrupting these programs and jeopardizing an important source of credit for many consumers. The Board requested comment on its proposal to exclude up-front charges for any guaranty under a Federal or State government program, as well as any up-front mortgage insurance premiums under government programs.

Several industry commenters argued that premiums for private mortgage insurance should be excluded altogether, even if the premiums do not satisfy the statutory standard for exclusion. These commenters noted that private mortgage insurance provides substantial benefits, allowing consumers who cannot afford a down payment an alternative for obtaining credit. Another commenter noted that the refundability requirement of the rule would make private mortgage insurance more expensive.

One industry commenter asserted that the language in proposed § 226.32(b)(1)(i)(B)(2) was inconsistent with the statutory language and the example in the commentary. The commenter suggested that a literal reading of proposed § 226.32(b)(1)(i)(B)(2) would require exclusion of the entire premium if it exceeded the FHA insurance premium, rather than merely exclusion of that portion of the premium in excess of the FHA premium. Another industry commenter maintained that the term “upfront” is vague and that the Bureau instead should use the phrase “payable at or before closing.”

The Bureau is adopting proposed § 226.32(b)(1)(i)(B) as reunumbered § 1026.32(b)(1)(i)(B) with no substantive changes but with revisions for clarity. The Bureau is dividing proposed § 226.32(b)(1)(i)(B) into two parts. The first part, § 1026.32(b)(1)(i)(B), addresses insurance premiums and guaranty charges under government programs. The second part, § 1026.32(b)(1)(i)(C), addresses premiums for private mortgage insurance.

Consistent with the Board's proposal, § 1026.32(b)(1)(i)(B) excludes from points and fees charges for mortgage guaranties under government programs, as well as premiums for mortgage insurance under government programs. The Bureau concurs with the Board's interpretation that, in addition to mortgage insurance premiums under government programs, the statute also excludes from points and fees charges for mortgage guaranties under government programs. Like the Board, the Bureau believes that this conclusion is further supported by TILA sections 105(a) and 129C(b)(3)(B)(i) and that it is necessary and proper to invoke this authority. The exclusion from points and fees of charges for mortgage guaranties under government programs is necessary and proper to effectuate the purposes of TILA. The Bureau is concerned that including such charges in points and fees could cause loans offered through government programs to exceed high-cost mortgage thresholds and qualified mortgage points and fees limits, potentially disrupting an important source of affordable financing for many consumers. This exclusion helps ensure that loans do not unnecessarily exceed the points and fees limits for qualified mortgages, which is consistent with the purpose, stated in TILA section 129B(a)(2), of assuring that consumers are offered and receive residential mortgage loans on terms that reasonably reflect their ability to repay the loans and with the purpose stated in TILA section 129C(b)(3)(B)(i) of ensuring that responsible, affordable mortgage credit remains available to consumers in a manner consistent with the purposes of TILA section 129C.

Proposed comment 32(b)(1)(i)-2 provided an example of a mortgage insurance premium that is not counted in points and fees because the loan was insured by the FHA. The Bureau is renumbering this comment as 32(b)(1)(i)(B)-1 and revising it to add an additional example to clarify that mortgage guaranty fees under government programs, such as VA and USDA funding fees, are excluded from points and fees. The Bureau is also deleting the reference to “up-front” premiums and charges. Under the statute, premiums for mortgage insurance or guaranty fees in connection with a Federal or State government program are excluded from points and fees whenever paid. The statutory provision excluding premiums or charges paid after consummation applies only to private mortgage insurance.

The Bureau is addressing exclusions for private mortgage insurance in § 1026.32(b)(1)(i)(C). For private mortgage insurance premiums payable after consummation, § 1026.32(b)(1)(i)(C)(1) provides that the entire amount of the premium is excluded from points and fees. For private mortgage insurance premiums payable at or before consummation, § 1026.32(b)(1)(i)(C)(1) provides that the portion of the premium not in excess of the amount payable under policies in effect at the time of origination under section 203(c)(2)(A) of the National Housing Act is excluded from points and fees, provided that the premium is required to be refundable on a pro-rated basis and the refund is automatically issued upon notification of the satisfaction of the underlying mortgage loan.

As noted by one commenter, the language in proposed § 226.32(b)(1)(i)(B) could be read to conflict with the statute and the commentary because it suggested that, if a private mortgage insurance premium payable at or before consummation exceeded the FHA insurance premium, then the entire private mortgage insurance premium would be included in points and fees. The Bureau is clarifying in § 1026.32(b)(1)(i)(C)(2) that only the portion of the private mortgage insurance premium that exceeds the FHA premium must be included in points and fees. With respect to the comments requesting that all private mortgage insurance premiums be excluded from points and fees, the Bureau notes that TILA section 103(bb)(1)(C) prescribes specific and detailed conditions for excluding private mortgage insurance premiums. Under these circumstances, the Bureau does not believe it would be appropriate to exercise its exception authority to reverse Congress's decision.

Proposed comment 32(b)(1)(i)-3 explained that private mortgage insurance premiums payable at or before consummation need not be included in points and fees to the extent that the premium does not exceed the amount payable under policies in effect at the time of origination under section 203(c)(2)(A) of the National Housing Act and the premiums are required to be refunded on a pro-rated basis and the refund is automatically issued upon notification of satisfaction of the underlying mortgage loan. Proposed comment 32(b)(1)(i)-3 also provided an example of this exclusion. Proposed comment 32(b)(1)(i)-4 explained that private mortgage insurance premiums that do not qualify for an exclusion must be included in points and fees whether paid at or before consummation, in cash or financed, whether optional or required, and whether the amount represents the entire premium or an initial payment.

The Bureau did not receive substantial comments on these proposed interpretations. The Bureau is adopting comments 32(b)(1)(i)-3, and -4 with certain revisions for clarity and renumbered as comments 32(b)(1)(i)(C)-1 and -2. Comment 32(b)(1)(i)(C)-1.i is revised to specify that private mortgage insurance premiums paid after consummation are excluded from points and fees. The Bureau also adopts clarifying changes that specify that creditors originating conventional loans—even such loans that are not eligible to be FHA loans (i.e., because their principal balance is too high)—should look to the permissible up-front premium amount for FHA loans, as implemented by applicable regulations and other written authorities issued by the FHA (such as Mortgagee Letters). For example, pursuant to HUD's Mortgagee Letter 12-4 (published March 6, 2012), the allowable up-front FHA premium for single-family homes is 1.75 percent of the base loan amount. Finally, the Bureau clarifies that only the portion of the single or up-front PMI premium in excess of the allowable FHA premium (i.e., rather than any monthly premium or portion thereof) must be included in points and fees. Comments 32(b)(1)(i)(C)-1 and -2 also have both been revised for clarity and consistency. For example, the comments as adopted refer to premiums “payable at or before consummation” rather than “up-front” premiums and to “consummation” rather than “closing.” The Bureau notes that the statute refers to “closing” rather than “consummation.” However, for consistency with the terminology in Regulation Z, the Bureau is using the term “consummation.”

Bona Fide Third-Party Charges and Bona Fide Discount Points

The Dodd-Frank Act amended TILA to add nearly identical provisions excluding certain bona fide third-party charges and bona fide discount points from the calculation of points and fees for both qualified mortgages and high-cost mortgages. Specifically, section 1412 of the Dodd-Frank Act added new TILA section 129C(b)(2)(C), which excludes certain bona fide third-party charges and bona fide discount points from the calculation of points and fees for the qualified mortgage points and fees threshold. Similarly, section 1431 of the Dodd-Frank Act amended TILA section 103(bb)(1)(A)(ii) and added TILA section 103(dd) to provide for nearly identical exclusions in calculating points and fees for the high-cost mortgage threshold.

In the 2011 ATR Proposal, the Board proposed to implement in § 226.43(e)(3)(ii)(A) through (C) the exclusion of certain bona fide third-party charges and bona fide discount points only for the calculation of points and fees for the qualified mortgage points and fees threshold. In the 2012 HOEPA Proposal, the Bureau proposed to implement these exclusions in proposed § 1026.32(b)(5) for the points and fees threshold for high-cost mortgages. The Bureau noted that proposed § 1026.32(b)(5) was generally consistent with the Board's proposed § 226.43(e)(3)(ii)(A) through (C).

The Bureau believes that it is appropriate to consolidate these exclusions in a single provision. The Bureau is now finalizing both rules, and the exclusions are nearly identical for both the qualified mortgage and high-cost mortgage contexts. Moreover, under the Board's ATR Proposal, the points and fees calculation for the qualified mortgage points and fees threshold already would have cross-referenced the definition of points and fees for high-cost mortgages in § 226.32(b)(1). Given that the points and fees calculations for both the qualified mortgage and high-cost mortgage points and fees thresholds will use the same points and fees definition in § 1026.32(b)(1), the Bureau believes it is unnecessary to implement nearly identical exclusions from points and fees in separate provisions for qualified mortgages and high-cost mortgages. Accordingly, the Bureau is consolidating the exclusions for certain bona fide third-party charges and bona fide discount points for both qualified mortgages and high-cost mortgages in new § 1026.32(b)(1)(i)(D) through (F). In addition, the definition of “bona fide discount points” for the purposes of § 1026.32(b)(1)(i)(E) and (F), which the 2011 ATR Proposal would have implemented in § 226.43(e)(3)(iv), is instead being implemented in § 1026.32(b)(3).

Bona fide third-party charges. TILA Section 129C(b)(2)(C)(i) excludes from points and fees “bona fide third party charges not retained by the mortgage originator, creditor, or an affiliate of the creditor or mortgage originator.” Tracking the statute, proposed § 226.43(e)(3)(ii)(A) would have excluded from “points and fees” for qualified mortgages any bona fide third party charge not retained by the creditor, loan originator, or an affiliate of either. Proposed § 226.43(e)(3)(iii) would have specified that the term “loan originator” has the same meaning as in § 226.36(a)(1).

Proposed § 226.43(e)(3)(ii)(A) would also have implemented TILA section 103(bb)(1)(C), which requires that premiums for private mortgage insurance be included in “points and fees” as defined in TILA section 103(bb)(4) under certain circumstances. Applying general rules of statutory construction, the Board concluded that the more specific provision on private mortgage insurance supersedes the more general provision permitting any bona fide third party charge not retained by the creditor, mortgage originator, or an affiliate of either to be excluded from “points and fees.” Thus, proposed § 226.43(e)(3)(ii)(A) would have excluded from points and fees any bona fide third party charge not retained by the creditor, loan originator, or an affiliate of either unless the charges were premiums for private mortgage insurance that were included in points and fees under § 226.32(b)(1)(i)(B).

The Board noted that, in setting the purchase price for specific loans, Fannie Mae and Freddie Mac make loan-level price adjustments (LLPAs) to compensate offset added risks, such as a high LTV or low credit score, among many other risk factors. Creditors may, but are not required to, increase the interest rate charged to the consumer so as to offset the impact of the LLPAs or increase the costs to the consumer in the form of points to offset the lost revenue resulting from the LLPAs. The Board noted that, during outreach, some creditors argued that these points should not be counted in points and fees for qualified mortgages under the exclusion for “bona fide third party charges not retained by the loan originator, creditor, or an affiliate of either” in TILA section 129C(b)(2)(C).

The Board acknowledged creditors' concerns about exceeding the qualified mortgage points and fees thresholds due to LLPAs required by the GSEs. However, the Board questioned whether an exemption for LLPAs would be consistent with congressional intent in limiting points and fees for qualified mortgages. The Board noted that points charged to meet GSE risk-based price adjustment requirements are arguably no different than other points charged on loans sold to any secondary market purchaser to compensate that purchaser for added loan-level risks. Congress clearly contemplated that discount points generally should be included in points and fees for qualified mortgages.

The Board noted that an exclusion for points charged by creditors in response to secondary market LLPAs also would raise questions about the appropriate treatment of points charged by creditors to offset loan-level risks on mortgage loans that they hold in portfolio. The Board reasoned that, under normal circumstances, these points are retained by the creditor, so it would not be appropriate to exclude them from points and fees under the “bona fide third party charge” exclusion. However, the Board cautioned that requiring that these points be included in points and fees, when similar charges on loans sold into the secondary market are excluded, may create undesirable market imbalances between loans sold to the secondary market and loans held in portfolio.

The Board also noted that creditors may offset risks on their portfolio loans (or on loans sold into the secondary market) by charging a higher rate rather than additional points and fees; however, the Board recognized the limits of this approach to loan-level risk mitigation due to concerns such as exceeding high-cost mortgage rate thresholds. Nonetheless, the Board noted that in practice, an exclusion from the qualified mortgage points and fees calculation for all points charged to offset loan-level risks may create compliance and enforcement difficulties. The Board questioned whether meaningful distinctions between points charged to offset loan-level risks and other points and fees charged on a loan could be made clearly and consistently. In addition, the Board observed that such an exclusion could be overbroad and inconsistent with Congress's intent that points generally be counted toward the points and fees threshold for qualified mortgages.

The Board requested comment on whether and on what basis the final rule should exclude from points and fees for qualified mortgages points charged to meet risk-based price adjustment requirements of secondary market purchasers and points charged to offset loan-level risks on mortgages held in portfolio.

Consumer advocates did not comment on this issue. Many industry commenters argued that LLPAs should be excluded from points and fees as bona fide third party charges. The GSE commenters agreed that LLPAs should be excluded as bona fide third party charges, noting that they are not retained by the creditor. One GSE commenter noted that LLPAs are set fees that are transparent and accessible via the GSEs' Web sites. Some industry commenters contended that including LLPAs in points and fees would cause many loans to exceed the points and fees cap for qualified mortgages. Other industry commenters argued that requiring LLPAs to be included in points and fees would force creditors to recover the costs through increases in the interest rate. One of the GSE commenters acknowledged the concern that creditors holding loans in portfolio could be at a disadvantage if LLPAs were excluded from points and fees and suggested that the Bureau consider allowing such creditors to exclude published loan level risk adjustment fees.

One industry commenter urged the Bureau to coordinate with the agencies responsible for finalizing the 2011 QRM Proposed Rule to avoid unintended consequences. The 2011 ARM Proposed Rule, if adopted, would require, in certain circumstances, that sponsors of MBS create premium capture cash reserve accounts to limit sponsors' ability to monetize the excess spread between the proceeds from the sale of the interests and the par value of those interests. See 76 FR 24113. The commenter stated that this would result in any premium in the price of a securitization backed by residential mortgage loans being placed in a first-loss position in the securitization. The commenter argued that this would make premium loans too expensive to originate and that creditors would not be able to recover LLPAs through interest rate adjustments. The commenter maintained that if the LLPAs were included in the calculation for the qualified mortgage points and fees limit, creditors would also be severely constrained in recovering LLPAs through points. The commenter argued that LLPAs therefore should be excluded from the points and fees calculation for qualified mortgages.

The Bureau is adopting § 226.43(e)(3)(ii)(A), with certain revisions, as renumbered § 1026.32(b)(1)(i)(D). As revised, § 1026.32(b)(1)(i)(D) provides that a bona fide third party charge not retained by the creditor, loan originator, or an affiliate of either the general is excluded from points and fees unless the charge is required to be included under § 1026.32(b)(1)(i)(C) (for mortgage insurance premiums), (iii) (for real estate related fees), or (iv) (for credit insurance premiums). As noted above, the Board proposed that the specific provision regarding mortgage insurance, TILA section 103(bb)(1)(C), should govern the exclusion of private mortgage insurance premiums of points and fees, rather than TILA section 129C(b)(2)(C), which provides generally for the exclusion of certain bona fide third-party charges. The Bureau likewise believes that the specific statutory provisions regarding real estate related fees and credit insurance premiums in TILA section 103(bb)(4)(C) and (D) should govern whether these charges are included in points and fees rather than the more general provisions regarding exclusion of bona fide third-party charges, TILA sections 103(bb)(1)(A)(ii) (for high-cost mortgages) or 129C(b)(2)(C) (for qualified mortgages). Thus, § 1026.32(b)(1)(i)(D) provides that the general exclusion for bona fide third-party charges applies unless the charges are required to be included under § 1026.32(b)(1)(i)(C), (iii), or (iv).

The Bureau acknowledges that TILA sections 103(bb)(1)(A)(ii) and 129C(b)(2)(C) could plausibly be read to provide for a two-step calculation of points and fees: first, the creditor would calculate points and fees as defined in TILA section 103(bb)(4); and, second, the creditor would exclude all bona fide third-party charges not retained by the mortgage originator, creditor, or an affiliate of either, as provided in TILA sections 103(bb)(1)(A)(ii) (for high-cost mortgages) and 129C(b)(2)(C) (for qualified mortgages). Under this reading, charges for, e.g., private mortgage insurance could initially, in step one, be included in points and fees but then, in step two, be excluded as bona fide third-party charges under TILA sections 103(bb)(1)(A)(ii) or 129C(b)(2)(C).

To give meaning to the specific statutory provisions regarding mortgage insurance, real estate related fees, and credit insurance, the Bureau believes that the better reading is that these specific provisions should govern whether such charges are included in points and fees, rather than the general provisions excluding certain bona fide third-party charges. For example, Congress added TILA section 103(bb)(1)(C), which prescribes certain conditions under which private mortgage insurance premiums would be included in points and fees. The Bureau believes that the purpose of this provision is to help ensure that consumers with a qualified mortgage are not charged excessive private mortgage insurance premiums. If such premiums could be excluded as bona fide third-party charges under TILA sections 103(bb)(1)(A)(ii) or 129C(b)(2)(C), then the purpose of this provision would be undermined. In further support of its interpretation, the Bureau is invoking its authority under TILA section 105(a) to make such adjustments and exceptions as are necessary and proper to effectuate the purposes of TILA, including that consumers are offered and receive residential mortgage loans on terms that reasonably reflect their ability to repay the loans. Similarly, the Bureau finds that it is necessary, proper and appropriate to use its authority under TILA section 129C(b)(3)(B)(i) to revise and subtract from the statutory language. This use of authority ensures that responsible, affordable mortgage credit remains available to consumers in a manner consistent with the purpose of TILA section 129C, referenced above, as well as effectuating that purpose.

As noted above, several industry commenters argued that points charged by creditors to offset LLPAs should be excluded from points and fees under § 1026.32(b)(1)(i)(D). In setting the purchase price for loans, the GSEs impose LLPAs to offset certain credit risks, and creditors may but are not required to recoup the revenue lost as a result of the LLPAs by increasing the costs to consumers in the form of points. The Bureau believes that the manner in which creditors respond to LLPAs is better viewed as a fundamental component of how the pricing of a mortgage loan is determined rather than as a third party charge. As the Board noted, allowing creditors to exclude points charged to offset LLPAs could create market imbalances between loans sold on the secondary market and loans held in portfolio. While such imbalances could be addressed by excluding risk adjustment fees more broadly, including fees charged by creditors for loans held in portfolio, the Bureau agrees with the Board that this could create compliance and enforcement difficulties. Thus, the Bureau concludes that points charged to offset LLPAs may not be excluded from points and fees under § 1026.32(b)(1)(i)(D). To the extent that creditors offer consumers the opportunity to pay points to lower the interest rate that the creditor would otherwise charge to recover the lost revenue from the LLPAs, such points may, if they satisfy the requirements of § 1026.32(b)(1)(i)(E) or (F), be excluded from points and fees as bona fide discount points.

As noted above, one commenter expressed concern that if the requirements for premium capture cash reserve accounts proposed in the 2011 QRM Proposed Rule were adopted, creditors would have difficulty in recovering the costs of LLPAs through rate and that, because of the points and fees limits for qualified mortgages, creditors would also have trouble recovering the costs of LLPAs through up-front charges to consumers. The Bureau notes that, as proposed, the premium capture cash reserve account requirement would not apply to securities sponsored by the GSEs and would not apply to securities comprised solely of QRMs. See 76 FR 24112, 24120. Thus, it is not clear, that even if it were adopted, the requirement would have as substantial an impact as suggested by the commenter. In any event, the requirement has merely been proposed, not finalized. The Bureau will continue to coordinate with the agencies responsible for finalizing the 2011 QRM Proposed Rule to consider the combined effects of that rule and the instant rule.

The Board proposed comment 43(e)(3)(ii)-1 to clarify the meaning in proposed § 226.43(e)(3)(ii)(A) of “retained by” the loan originator, creditor, or an affiliate of either. Proposed comment 43(e)(3)(ii)-1 provided that if a creditor charges a consumer $400 for an appraisal conducted by a third party not affiliated with the creditor, pays the third party appraiser $300 for the appraisal, and retains $100, the creditor may exclude $300 of this fee from “points and fees” but must count the $100 it retains in “points and fees.”

As noted above, several commenters expressed confusion about the relationship between proposed § 226.43(e)(3)(ii)(A), which would have excluded bona fide third party charges not retained by the loan originator, creditor, or an affiliate of either, and proposed § 226.32(b)(1)(iii), which would have excluded certain real estate related charges if they are reasonable, if the creditor receives no direct or indirect compensation in connection with the charges, and the charges are not paid to an affiliate of the creditor. As explained above, the Bureau interprets the more specific provision governing the inclusion in points and fees of real estate related charges (implemented in § 1026.32(b)(1)(iii)) as taking precedence over the more general exclusion for bona fide third party charges in renumbered § 1026.32(b)(1)(i)(D). Accordingly, the Bureau does not believe that the example in proposed comment 43(e)(3)(ii)-1 is appropriate for illustrating the exclusion for bona fide third party charges because the subject of the example, appraisals, is specifically addressed in § 1026.32(b)(1)(iii).

The Bureau therefore is revising renumbered comment 32(b)(1)(i)(D)-1 by using a settlement agent charge to illustrate the exclusion for bona fide third party charges. By altering this example to address closing agent charges, the Bureau is also responding to requests from commenters that the Bureau provide more guidance on whether closing agent charges are included in points and fees. As noted above, proposed § 226.43(e)(3)(iii) would have specified that the term “loan originator,” as used in proposed § 226.43(e)(3)(ii)(A), has the same meaning as in § 226.36(a)(1). The Bureau is moving the cross-reference to the definition of “loan originator” in § 226.36(a)(1) to comment 32(b)(1)(i)(D)-1.

The Board proposed comment 43(e)(3)(ii)-2 to explain that, under § 226.32(b)(1)(i)(B), creditors would have to include in “points and fees” premiums or charges payable at or before consummation for any private guaranty or insurance protecting the creditor against the consumer's default or other credit loss to the extent that the premium or charge exceeds the amount payable under policies in effect at the time of origination under section 203(c)(2)(A) of the National Housing Act (12 U.S.C. 1709(c)(2)(A)). The proposed comment also would have explained that these premiums or charges would be included if the premiums or charges were not required to be refundable on a pro-rated basis, or the refund is not automatically issued upon notification of the satisfaction of the underlying mortgage loan. The comment would have clarified that, under these circumstances, even if the premiums and charges were not retained by the creditor, loan originator, or an affiliate of either, they would be included in the “points and fees” calculation for qualified mortgages. The comment also would have cross-referenced proposed comments 32(b)(1)(i)-3 and -4 for further discussion of including private mortgage insurance premiums in the points and fees calculation.

The Bureau is adopting proposed comment 43(e)(3)(ii)-2 substantially as proposed, renumbered as comment 32(b)(i)(D)-2. In addition, the Bureau also is adopting new comments 32(b)(i)(D)-3 and -4 to explain that the exclusion of bona fide third party charges under § 1026.32(b)(1)(i)(D) does not apply to real estate-related charges and credit insurance premiums. The inclusion of these items in points and fees is specifically addressed in § 1026.32(b)(iii) and (iv), respectively.

Bona fide discount points. TILA section 129C(b)(2)(C)(ii) excludes up to two bona fide discount points from points and fees under certain circumstances. Specifically, it excludes up to two bona fide discount points if the interest rate before the discount does not exceed the average prime offer rate by more than two percentage points. Alternatively, it excludes up to one discount point if the interest rate before the discount does not exceed the average prime offer rate by more than one percentage point. The Board proposed to implement this provision in proposed § 226.43(e)(3)(ii)(B) and (C).

Proposed § 226.43(e)(3)(ii)(B) would have permitted a creditor to exclude from points and fees for a qualified mortgage up to two bona fide discount points paid by the consumer in connection with the covered transaction, provided that: (1) The interest rate before the rate is discounted does not exceed the average prime offer rate, as defined in § 226.45(a)(2)(ii), by more than one percent; and (2) the average prime offer rate used for purposes of paragraph 43(e)(3)(ii)(B)(1) is the same average prime offer rate that applies to a comparable transaction as of the date the discounted interest rate for the covered transaction is set.

Proposed § 226.43(e)(3)(ii)(C) would have permitted a creditor to exclude from points and fees for a qualified mortgage up to one bona fide discount point paid by the consumer in connection with the covered transaction, provided that: (1) The interest rate before the discount does not exceed the average prime offer rate, as defined in § 226.45(a)(2)(ii), by more than two percent; (2) the average prime offer rate used for purposes of § 226.43(e)(3)(ii)(C)(1) is the same average prime offer rate that applies to a comparable transaction as of the date the discounted interest rate for the covered transaction is set; and (3) two bona fide discount points have not been excluded under § 226.43(e)(3)(ii)(B).

Several industry commenters argued that creditors should be permitted to exclude from points and fees more than two discount points. Some industry commenters maintained that creditors should be permitted to exclude as many discount points as consumers choose to pay. Another commenter contended that creditors should be able to exclude as many as three discount points.

A few industry commenters requested eliminating the requirement that, for the discount points to be bona fide, the interest rate before the discount must be within one or two percentage points of the average prime offer rate. One industry commenter argued that this requirement is too inflexible. Several commenters recommended that this requirement be adjusted for jumbo loans and for second homes. Another commenter claimed that this requirement would limit the options for consumers paying higher interest rates and that these are the consumers for whom it would be most beneficial to pay down their interest rates.

Several commenters argued that the effect of these two limitations for excluding discount points from points and fees—the limit on the number of discount points that could be excluded and the requirement that the pre-discount rate be within one or two points of the average prime offer rate—would have a negative impact on consumers. They maintained that these limitations would prevent consumers from choosing their optimal combination of interest rate and points for their financial circumstances.

One commenter noted that proposed § 226.43(e)(3)(ii)(B) and (C) would require that, for the discount points or point to be excluded from points and fees, the interest rate before the discount must not exceed the average prime offer rate by more than one or two “percent,” respectively. The commenter recommended that, for clarity and consistency with the statute, the requirement should instead require that the interest rate before the discount be within one or two “percentage points” of the average prime offer rate.

The Bureau is adopting proposed § 226.43(e)(3)(ii)(B) and (C), renumbered as § 1026.32(b)(1)(i)(E) and (F), with certain revisions. As suggested by a commenter, the Bureau is revising both § 1026.32(b)(1)(i)(E)(1) and (F)(1) to require that, to exclude the discount points or point, the interest rate must be within one or two “percentage points” (rather than “percent”) of the average prime offer rate. This formulation is clearer and consistent with the statutory language. The Bureau is also adding § 1026.32(b)(1)(i)(E)(2) and (F)(2) to implement TILA section 103(dd)(1)(B) and (C), which specify that, to exclude discount points from points and fees for purposes of determining whether a loan is a high-cost mortgage, the interest rate for personal property loans before the discount must be within one or two percentage points, respectively, of the average rate on a loan in connection with which insurance is provided under title I of the National Housing Act. This provision does not apply to the points and fees limit for qualified mortgages, regardless of whether a loan is a high-cost mortgage. The provision is included in the final rule for completeness. Finally, in § 1026.32(b)(1)(i)(F), the Bureau is clarifying that bona fide discount points cannot be excluded under § 1026.32(b)(1)(i)(F) if any bona fide discount points already have been excluded under § 1026.32(b)(1)(i)(E).

As noted above, several commenters urged the Bureau to alter or eliminate the limitations on how many discount points may be excluded and the requirement that the pre-discount interest rate must be within one or two points of the average prime offer rate. A few industry commenters also requested that the Bureau adjust the limitation on the pre-discount interest rate specifically for jumbo loans and loans for vacation homes. These commenters noted that interest rates for such loans otherwise would often be too high to qualify for the exclusion for bona fide discount points. The Bureau recognizes that these limitations may circumscribe the ability of consumers to purchase discount points to lower their interest rates. Nevertheless, the Bureau does not believe it would be appropriate to exercise its exception authority. Congress apparently concluded that there was a greater probability of consumer injury when consumers purchased more than two discount points or when the consumers were using discount points to buy down higher interest rates. The Bureau also notes that, in other sections of the Dodd-Frank Act, Congress prescribed different thresholds above the average prime offer rate for jumbo loans. See TILA sections 129C(c)(1)(B) (prepayment penalties) and 129H(f)(2) (appraisals). Congress did not do so in the provision regarding exclusion of bona fide discount points.

The Bureau is adding new comment 32(b)(1)(i)(E)-2 to note that the term “bona fide discount point” is defined in § 1026.32(b)(3). To streamline the rule, the Bureau is moving into new comment 32(b)(1)(i)(E)-2 the explanation that the average prime offer rate used for purposes of for both § 1026.32(b)(1)(i)(E) and (F) is the average prime offer rate that applies to a comparable transaction as of the date the discounted interest rate for the covered transaction is set. The Board proposed comment 43(e)(3)(ii)-5 to clarify that the average prime offer rate table indicates how to identify the comparable transaction. The Bureau is adding the language from proposed comment 43(e)(3)(ii)-5 to new comment 32(b)(1)(i)(E)-2, with a revision to the cross-reference for the comment addressing “comparable transaction.”

Proposed comment 43(e)(3)(ii)-3 would have included an example to illustrate the rule permitting exclusion of two bona fide discount points. The example would have assumed a covered transaction that is a first-lien, purchase money home mortgage with a fixed interest rate and a 30-year term. It would also have assumed that the consumer locks in an interest rate of 6 percent on May 1, 2011, that was discounted from a rate of 6.5 percent because the consumer paid two discount points. Finally, assume that the average prime offer rate as of May 1, 2011 for first-lien, purchase money home mortgages with a fixed interest rate and a 30-year term is 5.5 percent. In this example, the creditor would have been able to exclude two discount points from the “points and fees” calculation because the rate from which the discounted rate was derived exceeded the average prime offer rate for a comparable transaction as of the date the rate on the covered transaction was set by only 1 percent.

The Bureau is adopting proposed comment 43(e)(3)(ii)-3 substantially as proposed but renumbered as comment 32(b)(1)(i)(E)-3. The Bureau is also adding new comment 32(b)(1)(i)(F)-1 to explain that comments 32(b)(1)(i)(E)-1 and -2 provide guidance concerning the definitions of “bona fide discount point” and “average prime offer rate,” respectively.

Proposed comment 43(e)(3)(ii)-4 would have provided an example to illustrate the rule permitting exclusion of one bona fide discount point. The example assumed a covered transaction that is a first-lien, purchase money home mortgage with a fixed interest rate and a 30-year term. The example also would have assumed that the consumer locks in an interest rate of 6 percent on May 1, 2011, that was discounted from a rate of 7 percent because the consumer paid four discount points. Finally, the example would have assumed that the average prime offer rate as of May 1, 2011, for first-lien, purchase money home mortgages with a fixed interest rate and a 30-year term is 5 percent.

In this example, the creditor would have been able to exclude one discount point from the “points and fees” calculation because the rate from which the discounted rate was derived (7 percent) exceeded the average prime offer rate for a comparable transaction as of the date the rate on the covered transaction was set (5 percent) by only 2 percent. The Bureau is adopting proposed comment 43(e)(3)(ii)-4 substantially as proposed but renumbered as comment 32(b)(1)(i)(F)-2.

32(b)(1)(ii)

When HOEPA was enacted in 1994, it required that “all compensation paid to mortgage brokers” be counted toward the threshold for points and fees that triggers special consumer protections under the statute. Specifically, TILA section 103(aa)(4) provided that charges are included in points and fees only if they are payable at or before consummation and did not expressly address whether “backend” payments from creditors to mortgage brokers funded out of the interest rate (commonly referred to as yield spread premiums) are included in points and fees. This requirement is implemented in existing § 1026.32(b)(1)(ii), which requires that all compensation paid by consumers directly to mortgage brokers be included in points and fees, but does not address compensation paid by creditors to mortgage brokers or compensation paid by any company to individual employees (such as loan officers who are employed by a creditor or mortgage broker).

The Dodd-Frank Act substantially expanded the scope of compensation included in points and fees for both the high-cost mortgage threshold in HOEPA and the qualified mortgage points and fees limits. Section 1431 of the Dodd-Frank Act amended TILA to require that “all compensation paid directly or indirectly by a consumer or creditor to a mortgage originator from any source, including a mortgage originator that is also the creditor in a table-funded transaction,” be included in points and fees. TILA section 103(bb)(4)(B) (emphasis added). Under amended TILA section 103(bb)(4)(B), compensation paid to anyone that qualifies as a “mortgage originator” is to be included in points and fees. Thus, in addition to compensation paid to mortgage brokerage firms and individual brokers, points and fees also includes compensation paid to other mortgage originators, including employees of a creditor (i.e., loan officers). In addition, as noted above, the Dodd-Frank Act removed the phrase “payable at or before closing” from the high-cost mortgage points and fees test and did not apply the “payable at or before closing” limitation to the points and fees cap for qualified mortgages. See TILA sections 103(bb)(1)(A)(ii) and 129C(b)(2)(A)(vii), (b)(2)(C). Thus, the statute appears to contemplate that even compensation paid to mortgage brokers and other loan originators after consummation should be counted toward the points and fees thresholds.

This change is one of several provisions in the Dodd-Frank Act that focus on loan originator compensation and regulation, in apparent response to concerns that industry compensation practices contributed to the mortgage market crisis by creating strong incentives for brokers and retail loan officers to steer consumers into higher-priced loans. Specifically, loan originators were often paid a commission by creditors that increased with the interest rate on a transaction. These commissions were funded by creditors through the increased revenue received by the creditor as a result of the higher rate paid by the consumer and were closely tied to the price the creditor expected to receive for the loan on the secondary market as a result of that higher rate. In addition, many mortgage brokers charged consumers up-front fees to cover some of their costs at the same time that they accepted backend payments from creditors out of the rate. This may have contributed to consumer confusion about where the brokers' loyalties lay.

The Dodd-Frank Act took a number of steps to address loan originator compensation issues, including: (1) Adopting requirements that loan originators be “qualified” as defined by Bureau regulations; (2) generally prohibiting compensation based on rate and other terms (except for loan amount) and prohibiting a loan originator from receiving compensation from both consumers and other parties in a single transaction; (3) requiring the promulgation of additional rules to prohibit steering consumers to less advantageous transactions; (4) requiring the disclosure of loan originator compensation; and (5) restricting loan originator compensation under HOEPA and the qualified mortgage provisions by including such compensation within the points and fees calculations. See TILA sections 103(bb)(4)(A)(ii), (B); 128(a)(18); 129B(b), (c); 129C(b)(2)(A)(vii), (C)(i).

The Board proposed revisions to § 226.32(b)(1)(ii) to implement the inclusion of more forms of loan originator compensation into the points and fees thresholds. Those proposed revisions tracked the statutory language, with two exceptions. First, proposed § 226.32(b)(1)(ii) did not include the phrase “from any source.” The Board noted that the statute covers compensation paid “directly or indirectly” to the loan originator, and concluded that it would be redundant to cover compensation “from any source.” Second, for consistency with Regulation Z, the proposal used the term “loan originator” as defined in § 226.36(a)(1), rather than the term “mortgage originator” that appears in section 1401 of the Dodd-Frank Act. See TILA section 103(cc)(2). The Board explained that it interpreted the definitions of mortgage originator under the statute and loan originator under existing Regulation Z to be generally consistent, with one exception that the Board concluded was not relevant for purposes of the points and fees thresholds. Specifically, the statutory definition refers to “any person who represents to the public, through advertising or other means of communicating or providing information (including the use of business cards, stationery, brochures, signs, rate lists, or other promotional items), that such person can or will provide” the services listed in the definition (such as offering or negotiating loan terms), while the existing Regulation Z definition does not include persons solely on this basis. The Board concluded that it was not necessary to add this element of the definition to implement the points and fees calculations anyway, reasoning that the calculation of points and fees is concerned only with loan originators that receive compensation for performing defined origination functions in connection with a consummated loan. The Board noted that a person who merely represents to the public that such person can offer or negotiate mortgage terms for a consumer has not yet received compensation for that function, so there is no compensation to include in the calculation of points and fees for a particular transaction.

In the proposed commentary, the Board explained what compensation would and would not have been included in points and fees under proposed § 226.32(b)(1)(ii). The Board proposed to revise existing comment 32(b)(1)(ii)-1 to clarify that compensation paid by either a consumer or a creditor to a loan originator, as defined in § 1026.36(a)(1), would be included in points and fees. Proposed comment 32(b)(1)(ii)-1 also stated that loan originator compensation already included in points and fees because it is included in the finance charge under § 226.32(b)(1)(i) would not be counted again under § 226.32(b)(1)(ii).

Proposed comment 32(b)(1)(ii)-2.i stated that, in determining points and fees, loan originator compensation includes the dollar value of compensation paid to a loan originator for a specific transaction, such as a bonus, commission, yield spread premium, award of merchandise, services, trips, or similar prizes, or hourly pay for the actual number of hours worked on a particular transaction. Proposed comment 32(b)(1)(ii)-2.ii clarified that loan originator compensation excludes compensation that cannot be attributed to a transaction at the time of origination, including, for example, the base salary of a loan originator that is also the employee of the creditor, or compensation based on the performance of the loan originator's loans or on the overall quality of a loan originator's loan files. Proposed comment 32(b)(1)(ii)-2.i also explained that compensation paid to a loan originator for a covered transaction must be included in the points and fees calculation for that transaction whenever paid, whether at or before closing or any time after closing, as long as the compensation amount can be determined at the time of closing. In addition, proposed comment 32(b)(1)(ii)-2.i provided three examples of compensation paid to a loan originator that would have been included in the points and fees calculation.

Proposed comment 32(b)(1)(ii)-3 stated that loan originator compensation includes amounts the loan originator retains and is not dependent on the label or name of any fee imposed in connection with the transaction. Proposed comment 32(b)(1)(ii)-3 offered an example of a loan originator imposing and retaining a “processing fee” and stated that such a fee is loan originator compensation, regardless of whether the loan originator expends the fee to process the consumer's application or uses it for other expenses, such as overhead.

The Board requested comment on the types of loan originator compensation that must be included in points and fees. The Board also sought comment on the appropriateness of specific examples given in the commentary.

Many industry commenters objected to the basic concept of including loan originator compensation in points and fees, urging the Bureau to use its exception authority to exclude loan originator compensation from points and fees altogether. Several industry commenters contended that other statutory provisions and rules, including the Secure and Fair Enforcement for Mortgage Licensing Act of 2008 (SAFE Act), the Board's 2010 Loan Originator Final Rule, and certain Dodd-Frank Act provisions (including those proposed to be implemented in the Bureau's 2012 Loan Originator Proposal), adequately regulate loan originator compensation and prohibit or restrict problematic loan originator compensation practices. Accordingly, they argued it is therefore unnecessary to include loan originator compensation in points and fees.

Many industry commenters also asserted that the amount of compensation paid to loan originators has little or no bearing on a consumer's ability to repay a mortgage, and thus that including loan originator compensation in points and fees under this rulemaking is unnecessary. They further asserted that including loan originator compensation in points and fees would greatly increase compliance burdens on creditors, discourage creditors from making qualified mortgages, and ultimately reduce access to credit and increase the cost of credit.

Several industry commenters argued that, if the Bureau does not exclude all loan originator compensation from points and fees, then the Bureau should at least exclude compensation paid to individual loan originators (i.e., loan officers who are employed by creditors or mortgage brokerage firms). They argued that compensation paid to individual loan originators is already included in the cost of the loan, either in the interest rate or in origination fees. They maintained that including compensation paid to individual loan originators in points and fees would therefore constitute double counting.

Several industry commenters also claimed that they would face significant challenges in determining the amount of compensation for individual loan originators. They noted that creditors need clear, objective standards for determining whether loans satisfy the qualified mortgage standard, and that the complexity of apportioning compensation to individual loans at the time of each closing to determine the amount of loan originator compensation to count toward the points and fees cap would create uncertainty. They also noted that having to track individual loan originators' compensation and allocate that compensation to individual loans would create additional compliance burdens, particularly for compensation paid after closing. Several industry commenters also stated that estimating loan originator compensation in table-funded transactions would prove difficult because the funding assignee may not know the amount paid by the table-funded creditor to the individual loan originator.

Several industry commenters also asserted that including compensation paid to individual loan originators would lead to anomalous results: Otherwise identical loans could have significant differences in points and fees depending on the timing of the mortgage loan or the identity of the loan officer. They noted, for example, that a loan that qualifies a loan officer for a substantial bonus because it enables a loan officer to satisfy a long-term (e.g., annual) origination-volume target or a loan that is originated by a high-performing loan officer could have substantially higher loan originator compensation, and thus substantially higher points and fees, than an otherwise identical loan. Because the consumers would not be paying higher fees or interest rates because of such circumstances, the commenters argued that the result would not further the goals of the statute.

Some industry commenters made a separate argument that the proposed method for including loan originator compensation in points and fees would create an unfair playing field for mortgage brokers. These commenters noted that, since a brokerage firm can be paid by only one source under the Board's 2010 Loan Originator Final Rule and related provisions of the Dodd-Frank Act, a payment by a creditor to a mortgage broker must cover both the broker's overhead costs and the cost of compensating the individual that worked on the transaction. The creditor's entire payment to the mortgage broker is loan originator compensation that is included in points and fees, so that loan originator compensation in a wholesale transaction includes both the compensation received from the creditor to cover the overhead costs of the mortgage broker and the compensation that the broker passes through to the individual employee who worked on the transaction. By contrast, in a loan obtained directly from a creditor, the creditor would have to include in points and fees the compensation paid to the loan officer, but could choose to recover its overhead costs through the interest rate rather than an up-front charge that would count toward the points and fees thresholds. One industry commenter provided examples illustrating that, as a result of this difference, loans obtained through a mortgage broker could have interest rates and fees identical to those in a loan obtained directly through a creditor but could have significantly higher loan originator compensation included in points and fees. Thus, particularly for smaller loan amounts, commenters expressed concern that it would be difficult for loans originated through mortgage brokers to remain under the points and fees limits for qualified mortgages.

A nonprofit loan originator commenter also argued that including loan originator compensation in points and fees could undercut programs that help low and moderate income consumers obtain affordable mortgages. This commenter noted that it relies on payments from creditors to help it provide services to consumers and that counting such payments as loan originator compensation and including them in points and fees could jeopardize its programs. The commenter requested that this problem be addressed by excluding nonprofit organizations from the definition of loan originator or by excluding payments by creditors to nonprofit organizations from points and fees.

Consumer advocates approved of including loan originator compensation in points and fees, regardless of when and by whom the compensation is paid. They asserted that including loan originator compensation would promote more consistent treatment by ensuring that all payments that loan originators receive count toward the points and fees thresholds, regardless of whether the payment is made by the consumer or the creditor and whether it is paid through the rate or through up-front fees. They maintained that the provision was intended to help prevent consumers from paying excessive amounts for loan origination services. More specifically, some consumer advocates argued that the Dodd-Frank Act provision requiring inclusion of loan originator compensation in points and fees is an important part of a multi-pronged approach to address widespread steering of consumers into more expensive mortgage transactions, and in particular, to address the role of commissions funded through the interest rate in such steering. The consumer advocates noted that separate prohibitions on compensation based on terms and on a loan originator's receiving compensation from both the consumer and another party do not limit the amount of compensation a loan originator can receive or prevent a loan originator from inducing consumers to agree to above-market interest rates. They expressed concern that, particularly in the subprime market, loan originators could specialize in originating transactions with above-market interest rates, with the expectation they could arrange to receive above-market compensation for all of their transactions. Consumer advocates argued that counting all methods of loan originator compensation toward the points and fees thresholds was intended to deter such conduct.

Consumer advocates also pointed out that in the wholesale context, the consumer has the option of paying the brokerage firm directly for its services. Such payments have always been included within the calculation of points and fees for HOEPA purposes. The advocates argued that when a consumer elects not to make the up-front payment but instead elects to fund the same amount of money for the brokerage through an increased rate, there is no justification for treating the money received by the brokerage as a result of the consumer's decision any differently.

The Bureau has carefully considered the comments received in light of the concerns about various issues with regard to loan originator compensation practices, the general concerns about the impacts of the ability-to-repay/qualified mortgage rule and revised HOEPA thresholds on a market in which access to mortgage credit is already extremely tight, differences between the retail and wholesale origination channels, and practical considerations regarding both the burdens of day-to-day implementation and the opportunities for evasion by parties who wish to engage in rent-seeking. As discussed further below, the Bureau is concerned about implementation burdens and anomalies created by the requirement to include loan originator compensation in points and fees, the impacts that it could have on pricing and access to credit, and the risks that rent-seekers will continue to find ways to evade the statutory scheme. Nevertheless, the Bureau believes that, in light of the historical record and of Congress's evident concern with loan originator compensation practices, it would not be appropriate to waive the statutory requirement that loan originator compensation be included in points and fees. The Bureau has, however, worked to craft the rule that implements Congress's judgment in a way that is practicable and that reduces potential negative impacts of the statutory requirement, as discussed below. The Bureau is also seeking comment in the concurrent proposal being published elsewhere in today's Federal Register on whether additional measures would better protect consumers and reduce implementation burdens and unintended consequences.

Accordingly, the Bureau in adopting § 1026.32(b)(1)(ii) has generally tracked the statutory language and the Board's proposal in the regulation text, but has expanded the commentary to provide more detailed guidance to clarify what compensation must be included in points and fees. The Dodd-Frank Act requires inclusion in points and fees of “all compensation paid directly or indirectly by a consumer or creditor to a mortgage originator from any source, including a mortgage originator that is also the creditor in a table-funded transaction.” See TILA section 103(bb)(4)(B). Consistent with the Board's proposal, revised § 1026.32(b)(ii) does not include the phrase “from any source.” The Bureau agrees that the phrase is unnecessary because the provision expressly covers compensation paid “directly or indirectly” to the loan originator. Like the Board's proposal, the final rule also uses the term “loan originator” as defined in § 1026.36(a)(1), not the term “mortgage originator” under section 1401 of the Dodd-Frank Act. See TILA section 103(cc)(2). The Bureau agrees that the definitions are consistent in relevant respects and notes that it is in the process of amending the regulatory definition to harmonize it even more closely with the Dodd-Frank Act definition of “mortgage originator.” Accordingly, the Bureau believes use of consistent terminology in Regulation Z will facilitate compliance. Finally, as revised, § 1026.32(b)(1)(ii) also does not include the language in proposed § 226.32(b)(1)(ii) that specified that the provision also applies to a loan originator that is the creditor in a table-funded transaction. The Bureau has concluded that that clarification is unnecessary because a creditor in a table-funded transaction is already included in the definition of loan originator in § 1026.36(a)(1). To clarify what compensation must be included in points and fees, revised § 1026.32(b)(1)(ii) specifies that compensation must be included if it can be attributed to the particular transaction at the time the interest rate is set. These limitations are discussed in more detail below.

In adopting the general rule, the Bureau carefully considered arguments by industry commenters that loan originator compensation should not be included in points and fees because other statutory provisions and rules already regulate loan originator compensation, because loan originator compensation is already included in the costs of mortgage loans, and because including loan originator compensation in points and fees would push many loans over the 3 percent cap on points and fees for qualified mortgages (or even over the points and fees limits for determining whether a loan is a high-cost mortgage under HOEPA), which would increase costs and impair access to credit.

The Bureau views the fact that other provisions within the Dodd-Frank Act address other aspects of loan originator compensation and activity as evidence of the high priority that Congress placed on regulating such compensation. The other provisions pointed to by the commenters address specific compensation practices that created particularly strong incentives for loan originators to “upcharge” consumers on a loan-by-loan basis and particular confusion about loan originators' loyalties. The Bureau believes that the inclusion of loan originator compensation in points and fees has distinct purposes. In addition to discouraging more generalized rent-seeking and excessive loan originator compensation, the Bureau believes that Congress may have been focused on particular risks to consumers. Thus, with respect to qualified mortgages, including loan originator compensation in points and fees helps to ensure that, in cases in which high up-front compensation might otherwise cause the creditor and/or loan originator to be less concerned about long-term sustainability, the creditor is not able to invoke a presumption of compliance if challenged to demonstrate that it made a reasonable and good faith determination of the consumer's ability to repay the loan. Similarly in HOEPA, the threshold triggers additional consumer protections, such as enhanced disclosures and housing counseling, for the loans with the highest up-front pricing.

The Bureau recognizes that the method that Congress chose to effectuate these goals does not ensure entirely consistent results as to whether a loan is a qualified mortgage or a high-cost transaction. For instance, loans that are identical to consumers in terms of up-front costs and interest rate may nevertheless have different points and fees based on the identity of the loan originator who handled the transaction for the consumer, since different individual loan originators in a retail environment or different brokerage firms in a wholesale environment may earn different commissions from the creditor without that translating in differences in costs to the consumer. In addition, there are anomalies introduced by the fact that “loan originator” is defined to include mortgage broker firms and individual employees hired by either brokers or creditors, but not creditors themselves. As a result, counting the total compensation paid to a mortgage broker firm will capture both the firm's overhead costs and the compensation that the firm passes on to its individual loan officer. By contrast, in a retail transaction, the creditor would have to include in points and fees the compensation that it paid to its loan officer, but would continue to have the option of recovering its overhead costs through the interest rate, instead of an up-front charge, to avoid counting them toward the points and fees thresholds. Indeed, the Bureau expects that the new requirement may prompt creditors to shift certain other expenses into rate to stay under the thresholds.

Nevertheless, to the extent there are anomalies from including loan originator compensation in points and fees, these anomalies appear to be the result of deliberate policy choices by Congress to expand the historical definition of points and fees to include all methods of loan originator compensation, whether derived from up-front charges or from the rate, without attempting to capture all overhead expenses by creditors or the gain on sale that the creditor can realize upon closing a mortgage. The Bureau agrees that counting loan originator compensation that is structured through rate toward the points and fees thresholds could cause some loans not to be classified as qualified mortgages and to trigger HOEPA protections, compared to existing treatment under HOEPA and its implementing regulation. However, the Bureau views this to be exactly the result that Congress intended.

In light of the express statutory language and Congress's evident concern with increasing consumer protections in connection with high levels of loan originator compensation, the Bureau does not believe that it is appropriate to use its exception or adjustment authority in TILA section 105(a) or in TILA section 129C(b)(3)(B)(i) to exclude loan originator compensation entirely from points and fees for qualified mortgages and HOEPA. As discussed below, however, the Bureau is attempting to implement the points and fees requirements with as much sensitivity as practicable to potential impacts on the pricing of and availability of credit, anomalies and unintended consequences, and compliance burdens.

The Bureau also carefully considered comments urging it to exclude compensation paid to individual loan originators from points and fees, but ultimately concluded that such a result would be inconsistent with the plain language of the statute and could exacerbate the potential inconsistent effects of the rule on different mortgage origination channels. As noted above, many industry commenters argued that, even if loan originator compensation were not excluded altogether, at least compensation paid to individual loan originators should be excluded from points and fees. Under this approach, only payments to mortgage brokers would be included in points and fees. The commenters contended that it would be difficult to track compensation paid to individual loan originators, particularly when that compensation may be paid after consummation of the loan and that it would create substantial compliance problems. They also argued that including compensation paid to individual loan originators in points and fees would create anomalies, in which identical transactions from the consumer's perspective (i.e., the same interest rate and up-front costs) could nevertheless have different points and fees because of loan originator compensation.

As explained above, the Bureau does not believe it is appropriate to use its exception authority to exclude loan originator compensation from points and fees, and even using that exception authority more narrowly to exclude compensation paid to individual loan originators could undermine Congress's apparent goal of providing stronger consumer protections in cases of high loan originator compensation. Although earlier versions of legislation focused specifically on compensation to “mortgage brokers,” which is consistent with existing HOEPA, the Dodd-Frank Act refers to compensation to “mortgage originators,” a term that is defined in detail elsewhere in the statute to include individual loan officers employed by both creditors and brokers, in addition to the brokers themselves. To the extent that Congress believed that high levels of loan originator compensation evidenced additional risk to consumers, excluding individual loan originators from consideration appears inconsistent with this policy judgment.

Moreover, the Bureau notes that using exception authority to exclude compensation paid to individual loan originators would exacerbate the differential treatment between the retail and wholesale channels concerning overhead costs. As noted above, compensation paid by the consumer or creditor to the mortgage broker necessarily will include amounts for both the mortgage broker's overhead and profit and for the compensation the mortgage broker passes on to its loan officer. Excluding individual loan officer compensation on the retail side, however, would effectively exempt creditors from counting any loan originator compensation at all toward points and fees. Thus, for transactions that would be identical from the consumer's perspective in terms of interest rate and up-front costs, the wholesale transaction could have significantly higher points and fees (because the entire payment from the creditor to the mortgage broker would be captured in points and fees), while the retail transaction might include no loan origination compensation at all in points and fees. Such a result would put brokerage firms at a disadvantage in their ability to originate qualified mortgages and put them at significantly greater risk of originating HOEPA loans. This in turn could constrict the supply of loan originators and the origination channels available to consumers to their detriment.

The Bureau recognizes that including compensation paid to individual loan originators, such as loan officers, with respect to individual transactions may impose additional burdens. For example, creditors will have to track employee compensation for purposes of complying with the rule, and the calculation of points and fees will be more complicated. However, the Bureau notes that creditors and brokers already have to monitor compensation more carefully as a result of the 2010 Loan Originator Final Rule and the related Dodd-Frank Act restrictions on compensation based on terms and on dual compensation. The Bureau also believes that these concerns can be reduced by providing clear guidance on issues such as what types of compensation are covered, when compensation is determined, and how to avoid “double-counting” payments that are already included in points and fees calculations. The Bureau has therefore revised the Board's proposed regulation and commentary to provide more detailed guidance, and is seeking comment in the proposal published elsewhere in the Federal Register today on additional guidance and potential implementation issues among other matters.

As noted above, the Bureau is revising § 1026.32(b)(1)(ii) to clarify that compensation must be counted toward the points and fees thresholds if it can be attributed to the particular transaction at the time the interest rate is set. The Bureau is also revising comment 32(b)(1)(ii)-1 to explain in general terms when compensation qualifies as loan originator compensation that must be included in points and fees. In particular, compensation paid by a consumer or creditor to a loan originator is included in the calculation of points and fees, provided that such compensation can be attributed to that particular transaction at the time the interest rate is set. The Bureau also incorporates part of proposed comment 32(b)(1)(ii)-3 into revised comment 32(b)(1)(ii)-1, explaining that loan originator compensation includes amounts the loan originator retains, and is not dependent on the label or name of any fee imposed in connection with the transaction. However, revised comment 32(b)(1)(ii)-1 does not include the example from proposed comment 32(b)(1)(ii)-3, which stated that, if a loan originator imposes a processing fee and retains the fee, the fee is loan originator compensation under § 1026.32(b)(1)(ii) whether the originator expends the fee to process the consumer's application or uses it for other expenses, such as overhead. That example may be confusing in this context because a processing fee paid to a loan originator likely would be a finance charge under § 1026.4 and would therefore already be included in points and fees under § 1026.32(b)(1)(i).

Revised comment 32(b)(1)(ii)-2.i explains that compensation, such as a bonus, commission, or an award of merchandise, services, trips or similar prizes, must be included only if it can be attributed to a particular transaction. The requirement that compensation is included in points and fees only if it can be attributed to a particular transaction is consistent with the statutory language. The Dodd-Frank Act provides that, for the points and fees tests for both qualified mortgages and high-cost mortgages, only charges that are “in connection with” the transaction are included in points and fees. See TILA sections 103(bb)(1)(A)(ii) (high-cost mortgages) and 129C(b)(2)(A)(vii) (qualified mortgages). Limiting loan originator compensation to compensation that is attributable to the transaction implements the statutory requirement that points and fees are “in connection” with the transaction. This limitation also makes the rule more workable. Compensation is included in points and fees only if it can be attributed to a specific transaction to facilitate compliance with the rule and avoid over-burdening creditors with complex calculations to determine, for example, the portion of a loan officer's salary that should be counted in points and fees. For clarity, the Bureau has moved the discussion of the timing of loan originator compensation into new comment 32(b)(1)(ii)-3, and has added additional examples to 32(b)(1)(ii)-4, to illustrate the types and amount of compensation that should be included in points and fees.

Revised comment 32(b)(1)(ii)-2.ii explains that loan originator compensation excludes compensation that cannot be attributed to a particular transaction at the time the interest rate is set, including, for example, compensation based on the long-term performance of the loan originator's loans or on the overall quality of the loan originator's loan files. The base salary of a loan originator is also excluded, although additional compensation that is attributable to a particular transaction must be included in points and fees. The Bureau has decided to seek further comment in the concurrent proposal regarding treatment of hourly wages for the actual number of hours worked on a particular transaction. The Board's proposal would have included hourly pay for the actual number of hours worked on a particular transaction in loan originator compensation for purposes of the points and fees thresholds, and the Bureau agrees that such wages are attributable to the particular transaction. However, the Bureau is unclear as to whether industry actually tracks compensation this way in light of the administrative burdens. Moreover, while the general rule provides for calculation of loan originator compensation at the time the interest rate is set for the reasons discussed above, the actual hours of hours worked on a transaction would not be known at that time. The Bureau is therefore seeking comment on issues relating to hourly wages, including whether to require estimates of the hours to be worked between rate set and consummation.

New comment 32(b)(1)(ii)-3 explains that loan originator compensation must be included in the points and fees calculation for a transaction whenever the compensation is paid, whether before, at or after closing, as long as that compensation amount can be attributed to the particular transaction at the time the interest rate is set. Some industry commenters expressed concern that it would be difficult to determine the amount of compensation that would be paid after consummation and that creditors might have to recalculate loan originator compensation (and thus points and fees) after underwriting if, for example, a loan officer became eligible for higher compensation because other transactions had been consummated. The Bureau appreciates that industry participants need certainty at the time of underwriting as to whether transactions will exceed the points and fees limits for qualified mortgages (and for high-cost mortgages). To address this concern, the comment 32(b)(1)(ii)-3 explains that loan originator compensation should be calculated at the time the interest rate is set. The Bureau believes that the date the interest rate is set is an appropriate standard for calculating loan originator compensation. It would allow creditors to be able to calculate points and fees with sufficient certainty so that they know early in the process whether a transaction will be a qualified mortgage or a high-cost mortgage.

As noted above, several industry commenters argued that including loan originator compensation in points and fees would result in double counting. They stated that creditors often will recover loan originator compensation costs through origination charges, and these charges are already included in points and fees under § 1026.32(b)(1)(i). However, the underlying statutory provisions as amended by the Dodd-Frank Act do not express any limitation on its requirement to count loan originator compensation toward the points and fees test. Rather, the literal language of TILA section 103(bb)(4) as amended by the Dodd-Frank Act defines points and fees to include all items included in the finance charge (except interest rate), all compensation paid directly or indirectly by a consumer or creditor to a loan originator, “and” various other enumerated items. The use of “and” and the references to “all” compensation paid “directly or indirectly” and “from any source” suggest that compensation should be counted as it flows downstream from one party to another so that it is counted each time that it reaches a loan originator, whatever the previous source.

The Bureau believes the statute would be read to require that loan originator compensation be treated as additive to the other elements of points and fees. The Bureau believes that an automatic literal reading of the statute in all cases, however, would not be in the best interest of either consumers or industry. For instance, the Bureau does not believe that it is necessary or appropriate to count the same payment made by a consumer to a mortgage broker firm twice, simply because it is both part of the finance charge and loan originator compensation. Similarly, the Bureau does not believe that, where a payment from either a consumer or a creditor to a mortgage broker is counted toward points and fees, it is necessary or appropriate to count separately funds that the broker then passes on to its individual employees. In each case, any costs and risks to the consumer from high loan originator compensation are adequately captured by counting the funds a single time against the points and fees cap; thus, the Bureau does not believe the purposes of the statute would be served by counting some or all of the funds a second time, and is concerned that doing so could have negative impacts on the price and availability of credit.

Determining the appropriate accounting rule is significantly more complicated, however, in situations in which a consumer pays some up-front charges to the creditor and the creditor pays loan originator compensation to either its own employee or to a mortgage broker firm. Because money is fungible, tracking how a creditor spends money it collects in up-front charges versus amounts collected through the rate to cover both loan originator compensation and its other overhead expenses would be extraordinarily complex and cumbersome. To facilitate compliance, the Bureau believes it is appropriate and necessary to adopt one or more generalized rules regarding the accounting of various payments. However, the Bureau does not believe it yet has sufficient information with which to choose definitively between the additive approach provided for in the statutory language and other potential methods of accounting for payments in light of the multiple practical and complex policy considerations involved.

The potential downstream effects of different accounting methods are significant. Under the additive approach where no offsetting consumer payments against creditor-paid loan originator compensation is allowed, creditors whose combined loan originator compensation and up-front charges would otherwise exceed the points and fees limits would have strong incentives to cap their up-front charges for other overhead expenses under the threshold and instead recover those expenses by increasing interest rates to generate higher gains on sale. This would adversely affect consumers who prefer a lower interest rate and higher up-front costs and, at the margins, could result in some consumers being unable to qualify for credit. Additionally, to the extent creditors responded to a “no offsetting” rule by increasing interest rates, this could increase the number of qualified mortgages that receive a rebuttable rather than conclusive presumption of compliance.

One alternative would be to allow all consumer payments to offset creditor-paid loan originator compensation. However, a “full offsetting” approach would allow creditors to offset much higher levels of up-front points and fees against expenses paid through rate before the heightened consumer protections required by the Dodd-Frank Act would apply. Particularly under HOEPA, this may raise tensions with Congress's apparent intent. Other alternatives might use a hybrid approach depending on the type of expense, type of loan, or other factors, but would involve more compliance complexity.

In light of the complex considerations, the Bureau believes it is necessary to seek additional notice and comment. The Bureau therefore is finalizing this rule without qualifying the statutory result and is proposing two alternative comments in the concurrent proposal, one of which would explicitly preclude offsetting, and the other of which would allow full offsetting of any consumer-paid charges against creditor-paid loan originator compensation. The Bureau is also proposing comments to clarify treatment of compensation paid by consumers to mortgage brokers and by mortgage brokers to their individual employees. The Bureau is seeking comment on all aspects of this issue, including the market impacts and whether adjustments to the final rule would be appropriate. In addition, the Bureau is seeking comment on whether it would be helpful to provide for additional adjustment of the rules or additional commentary to clarify any overlaps in definitions between the points and fees provisions in this rulemaking and the HOEPA rulemaking and the provisions that the Bureau is separately finalizing in connection with the Bureau's 2012 Loan Originator Compensation Proposal.

Finally, comment 32(b)(1)(ii)-4 includes revised versions of examples in proposed comment 32(b)(1)(ii)-2, as well as additional examples to provide additional guidance regarding what compensation qualifies as loan originator compensation that must be included in points and fees. These examples illustrate when compensation can be attributed to a particular transaction at the time the interest rate is set. New comment 32(b)(1)(ii)-5 adds an example explaining how salary is treated for purposes of loan originator compensation for calculating points and fees.

32(b)(1)(iii)

TILA section 103(aa)(4)(C) provides that points and fees include certain real estate-related charges listed in TILA section 106(e) and is implemented in § 1026.32(b)(1)(iii). The Dodd-Frank Act did not amend TILA section 103(aa)(4)(C) (but did renumber it as section 103(bb)(4)(C)). Although the Board indicated in the Supplementary Information that it was not proposing any changes, proposed § 226.32(b)(1)(iii) would have added the phrase “payable at or before closing of the mortgage” loan and would have separated the elements into three new paragraphs (A) through (C). Thus, proposed § 226.32(b)(1)(iii) would have included in points and fees “all items listed in § 226.4(c)(7) (other than amounts held for future payment of taxes) payable at or before closing of the mortgage loan, unless: (A) The charge is reasonable; (B) the creditor receives no direct or indirect compensation in connection with the charge; and (C) the charge is not paid to an affiliate of the creditor.” The Board noted that the statute did not exclude these charges if they were payable after closing and questioned whether such a limitation was necessary because these charges could reasonably be viewed as charges that by definition are payable only at or before closing. As noted in the section-by-section analysis of § 1026.32(b)(1), the Board requested comment on whether there are any other types of fees that should be included in points and fees only if they are payable at or before closing.

The Board noted that during outreach creditors had raised concerns about including in points and fees real-estate related fees paid to an affiliate of the creditor, such as an affiliated title company. Although these fees always have been included in points and fees for high-cost loans, creditors using affiliated title companies were concerned they would have difficulty meeting the lower threshold for points and fees for qualified mortgages. The Board, however, did not propose to exempt fees paid to creditor-affiliated settlement service providers, noting that Congress appeared to have rejected excluding such fees from points and fees.

Industry commenters criticized the Board's proposed treatment of fees paid to affiliates as overbroad. Industry commenters argued that a creditor's affiliation with a service provider, such as a title insurance agency, does not have any impact on the consumer's ability to repay a loan. They maintained that studies over the past two decades have shown that title services provided by affiliated businesses are competitive in cost compared to services provided by unaffiliated businesses. They contended that the rule should instead focus solely on whether the fee is bona fide.

These commenters also argued that the largest real estate-related charge, title insurance fees, are often either mandated by State law or required to be filed with the relevant state authority and do not vary. Regardless of whether the State sets the rate or requires that the rate be filed, these commenters argued that there are so few insurers that rates tend to be nearly identical among providers.

These commenters also argued that including fees to affiliates would negatively affect consumers. They claimed that the inclusion of fees paid to affiliates would cause loans that would otherwise be qualified mortgages to exceed the points and fees cap, resulting in more expense to the creditor, which would be passed through to consumers in the form of higher interest rates or fees, or in more denials of credit. They also claimed that the proposal would harm consumers by reducing competition among settlement service providers and by eliminating operational efficiencies. One industry trade association reported that some of its members with affiliates would discontinue offering mortgages, which would reduce competition among creditors, especially for creditors offering smaller loans, since these loans would be most affected by the points and fees cap. They claimed that treating affiliated and unaffiliated providers differently would incentivize creditors to use unaffiliated third-party service providers to stay within the qualified mortgage points and fees cap.

Several industry commenters noted that RESPA permits affiliated business arrangements and provides protections for consumers, including a prohibition against requiring that consumers use affiliates, a requirement to disclose affiliation to consumers, and a limitation that compensation include only return on ownership interest. These commenters argued that charges paid to affiliates should be excluded from points and fees as long the RESPA requirements are satisfied. Several industry commenters objected to the requirement that charges be “reasonable” to be excluded from points and fees. They argued that the requirement was vague and that it would be difficult for a creditor to judge whether a third-party charge met the standard. Several commenters also argued that the Dodd-Frank Act provision permitting exclusion of certain bona fide third-party charges should apply rather than the three-part test for items listed in § 1026.4(c)(7). See TILA section 129C(b)(2)(C)(i).

Two consumer advocates commented on this aspect of the proposal. They supported including in points and fees all fees paid to any settlement service provider affiliated with the creditor.

The Bureau is adopting § 226.32(b)(1)(iii) as proposed but renumbered as § 1026.32(b)(1)(iii). TILA section 103(bb)(4) specifically mandates that fees paid to and retained by affiliates of the creditor be included in points and fees. The Bureau acknowledges that including fees paid to affiliates in points and fees could make it more difficult for creditors using affiliated service providers to stay under the points and fees cap for qualified mortgages and that, as a result, creditors could be disincented from using affiliated service providers. This is especially true with respect to affiliated title insurers because of the cost of title insurance. On the other hand, despite RESPA's regulation of fees charged by affiliates, concerns have nonetheless been raised that fees paid to an affiliate pose greater risks to the consumer, since affiliates of a creditor may not have to compete in the market with other providers of a service and thus may charge higher prices that get passed on to the consumer. The Bureau believes that Congress weighed these competing considerations and made a deliberate decision not to exclude fees paid to affiliates. This approach is further reflected throughout title XIV, which repeatedly amended TILA to treat fees paid to affiliates as the equivalent to fees paid to a creditor or loan originator. See, e.g., Dodd-Frank Act sections 1403, 1411, 1412, 1414, and 1431. For example, as noted above, TILA section 129C(b)(2)(C)(i), as added by section 1412 of the Dodd-Frank Act, provides that for purposes of the qualified mortgage points and fees test, bona fide third-party charges are excluded other than charges “retained by * * * an affiliate of the creditor or mortgage originator.” Similarly, TILA section 129B(c)(2)(B)(ii), added by section 1403 of the Dodd-Frank Act, restricts the payment of points and fees but permits the payment of bona fide third-party charges unless those charges are “retained by * * * an affiliate of the creditor or originator.” In light of these considerations, the Bureau does not believe there is sufficient justification to use its exception authority in this instance as the Bureau cannot find, given Congress's clear determination, that excluding affiliate fees from the calculation of points and fees is necessary or proper to effectuate the purposes of TILA, to prevent circumvention or evasion thereof, or to facilitate compliance therewith.

As noted above, some commenters objected to the requirement that charges be “reasonable.” The Bureau notes that a “reasonable” requirement has been in place for many years before the Dodd-Frank Act. TILA section 103(aa)(4)(C) specifically provides that charges listed in TILA section 106(e) are included in points and fees for high-cost mortgages unless, among other things, the charge is reasonable. This requirement is implemented in existing § 1026.32(b)(1)(iii). Similarly, a charge may be excluded from the finance charge under § 1026.4(c)(7) only if it is reasonable. In the absence of any evidence that this requirement has been unworkable, the Bureau declines to alter it. The fact that a transaction for such services is conducted at arms-length ordinarily should be sufficient to make the charge reasonable. The reasonableness requirement is not intended to invite an inquiry into whether a particular appraiser or title insurance company is imposing excessive charges.

Some commenters also maintained that the provision permitting exclusion of certain bona fide third-party charges should apply rather than the three-part test for items listed in § 1026.4(c)(7). See TILA section 129C(b)(2)(C)(i). As discussed in more detail in the section-by-section analysis of § 1026.32(b)(1)(i)(D), the Bureau concludes that § 1026.32(b)(1)(iii), which specifically addresses exclusion of items listed in § 1026.4(c)(7), takes precedence over the more general exclusion in § 1026.32(b)(1)(i)(D).

The Board's proposed comment 32(b)(1)(iii)-1 was substantially the same as existing comment 32(b)(1)(ii)-2. It would have provided an example of the inclusion or exclusion of real-estate related charges. The Bureau did not receive substantial comment on the proposed comment. The Bureau is therefore adopting comment 32(b)(1)(ii)-1 substantially as proposed, with revisions for clarity.

32(b)(1)(iv)

As amended by section 1431 of the Dodd-Frank Act, TILA section 103(bb)(4)(D) includes in points and fees premiums for various forms of credit insurance and charges for debt cancellation or suspension coverage. The Board proposed § 226.32(b)(1)(iv) to implement this provision. The Board also proposed to revise comment 32(b)(1)(iv)-1 to reflect the revised statutory language and to add new comment 32(b)(1)(iv)-2 to clarify that “credit property insurance” includes insurance against loss or damage to personal property such as a houseboat or manufactured home.

Several commenters argued that proposed § 226.32(b)(1)(iv) did not accurately implement the provision in Dodd-Frank Act section 1431 that specifies that “insurance premiums or debt cancellation or suspension fees calculated and paid in full on a monthly basis shall not be considered financed by the creditor.” They argued that comment 32(b)(1)(iv)-1 should be revised so that it expressly excludes monthly premiums for credit insurance from points and fees, including such premiums payable in the first month. At least one industry commenter also argued that voluntary credit insurance premiums should not be included in points and fees. Consumer advocates supported inclusion of credit insurance premiums in points and fees, noting that these services can add significant costs to mortgages.

The Bureau is adopting § 226.32(b)(1)(iv) substantially as proposed, with revisions for clarity, as renumbered § 1026.32(b)(1)(iv). As revised, § 1026.32(b)(1)(iv) states that premiums or other charges for “any other life, accident, health, or loss-of-income insurance” are included in points and fees only if the insurance is for the benefit of the creditor. The Bureau is also adopting proposed comments 32(b)(1)(iv)-1 and -2 substantially as proposed, with revisions for clarity and consistency with terminology in Regulation Z. The Bureau is also adopting new comment 32(b)(1)(iv)-3 to clarify that premiums or other charges for “any other life, accident, health, or loss-of-income insurance” are included in points and fees only if the creditor is a beneficiary of the insurance.

As noted above, several commenters argued that premiums paid monthly, including the first such premium, should not be included in points and fees. The statute requires that premiums “payable at or before closing” be included in points and fees; it provides only that premiums “calculated and paid in full on a monthly basis shall not be considered financed by the creditor.” TILA section 103(bb)(4)(D). Thus, if the first premium is payable at or before closing, that payment is included in points and fees even though the subsequent monthly payments are not.

Another commenter argued that voluntary credit insurance premiums should be excluded from points and fees. However, under the current rule, voluntary credit insurance premiums are included in points and fees. In light of the fact that the Dodd-Frank Act expanded the types of credit insurance that must be included in points and fees, the Bureau does not believe it would be appropriate to reconsider whether voluntary credit insurance premiums should be included in points and fees.

32(b)(1)(v)

As added by the Dodd-Frank Act, new TILA section 103(bb)(4)(E) includes in points and fees “the maximum prepayment penalties which may be charged or collected under the terms of the credit transaction.” The Board's proposed § 226.32(b)(1)(v) closely tracked the statutory language, but it cross-referenced proposed § 226.43(b)(10) for the definition of “prepayment penalty.”

Few commenters addressed this provision. One industry commenter argued that the maximum prepayment penalty should not be included in points and fees because a prepayment that triggers the penalty may never occur and thus the fee may never be assessed.

The Bureau is adopting § 226.32(b)(1)(v) substantially as proposed but renumbered as § 1026.32(b)(1)(v), with a revision to its definitional cross-reference. As revised, § 1026.32(b)(1)(v) refers to the definition of prepayment penalty in § 1026.32(b)(6)(i). With respect to the comment arguing that prepayment penalties should not be included in points and fees, the statute requires inclusion in points and fees of the maximum prepayment penalties that “may be charged or collected.” Thus, under the statutory language, the imposition of the charge need not be certain for the prepayment penalty to be included in points and fees. In this provision (and other provisions added by the Dodd-Frank Act, such as TILA section 129C(c)), Congress sought to limit and deter the use of prepayment penalties, and the Bureau does not believe that it would be appropriate to exercise its exception authority in a manner that could undermine that goal.

32(b)(1)(vi)

New TILA section 103(bb)(4)(F) requires that points and fees include “all prepayment fees or penalties that are incurred by the consumer if the loan refinances a previous loan made or currently held by the same creditor or an affiliate of the creditor.” The Board's proposed § 226.32(b)(1)(vi) would have implemented this provision by including in points and fees the total prepayment penalty, as defined in § 226.43(b)(10), incurred by the consumer if the mortgage loan is refinanced by the current holder of the existing mortgage loan, a servicer acting on behalf of the current holder, or an affiliate of either. The Board stated its belief that this provision is intended in part to curtail the practice of “loan flipping,” which involves a creditor refinancing an existing loan for financial gain resulting from prepayment penalties and other fees that a consumer must pay to refinance the loan—regardless of whether the refinancing is beneficial to the consumer. The Board noted that it departed from the statutory language to use the phrases “current holder of the existing mortgage loan” and “servicer acting on behalf of the current holder” in proposed § 226.32(b)(1)(vi) because, as a practical matter, these are the entities that would refinance the loan and directly or indirectly gain from associated prepayment penalties.

Few commenters addressed this provision. Two consumer groups expressed support for including these prepayment penalties in points and fees, arguing that many consumers were victimized by loan flipping and the resulting fees and charges.

The Bureau is adopting § 226.32(b)(1)(vi) substantially as proposed but renumbered as § 1026.32(b)(1)(vi). In addition to revising for clarity, the Bureau has also revised § 1026.32(b)(1)(vi) to refer to the definition of prepayment penalty in § 1026.32(b)(6)(i). Like the Board, the Bureau believes that it is appropriate for § 1026.32(b)(1)(vi) to apply to the current holder of the existing mortgage loan, the servicer acting on behalf of the current holder, or an affiliate of either. These are the entities that would refinance the loan and gain from the prepayment penalties on the previous loan. Accordingly, the Bureau is invoking its exception and adjustment authority under TILA sections 105(a) and 129C(b)(3)(B)(i). The Bureau believes that adjusting the statutory language to more precisely target the entities that would benefit from refinancing loans with prepayment penalties will more effectively deter loan flipping to collect prepayment penalties and help preserve consumers' access to safe, affordable credit. It also will lessen the compliance burden on other entities that lack the incentive for loan flipping, such as a creditor that originated the existing loan but no longer holds the loan. For these reasons, the Bureau believes that use of its exception and adjustment authority is necessary and proper under TILA section 105(a) to effectuate the purposes of TILA and to facilitate compliance with TILA and its purposes, including the purpose of assuring that consumers are offered and receive residential mortgage loans on terms that reasonably reflect their ability to repay the loans. Similarly, the Bureau finds that it is necessary, proper, and appropriate to use its authority under TILA section 129C(b)(3)(B)(i) to revise and subtract from statutory language. This use of authority ensures that responsible, affordable mortgage credit remains available to consumers in a manner consistent with and effectuates the purpose of TILA section 129C, referenced above, and facilitates compliance with section 129C of TILA.

32(b)(2)

Proposed Provisions Not Adopted

As noted in the section-by-section analysis of § 1026.32(b)(1)(ii) above, section 1431(c) of the Dodd-Frank Act amended TILA to require that all compensation paid directly or indirectly by a consumer or a creditor to a “mortgage originator” be included in points and fees for high-cost mortgages and qualified mortgages. As also noted above, the Board's 2011 ATR Proposal proposed to implement this statutory change in proposed § 226.32(b)(1)(ii) using the term “loan originator,” as defined in existing § 1026.36(a)(1), rather than the statutory term “mortgage originator.” In turn, the Board proposed new § 226.32(b)(2) to exclude from points and fees compensation paid to certain categories of persons specifically excluded from the definition of “mortgage originator” in amended TILA section 103, namely employees of a retailer of manufactured homes under certain circumstances, certain real estate brokers, and servicers.

The Bureau is not adopting proposed § 226.32(b)(2). The Bureau is amending the definition of “loan originator” § 1026.36(a)(1) and the associated commentary to incorporate the statutory exclusion of these persons from the definition. Accordingly, to the extent these persons are excluded from the definition of loan originator compensation, their compensation is not loan originator compensation that must be counted in points and fees, and the exclusions in proposed § 226.32(b)(2) are no longer necessary.

Instead, in the 2013 HOEPA Final Rule, the Bureau is finalizing the definition of points and fees for HELOCs in § 1026.32(b)(2). Current § 1026.32(b)(2), which contains the definition of “affiliate,” is being renumbered as § 1026.32(b)(5).

32(b)(3) Bona Fide Discount Point

32(b)(3)(i) Closed-End Credit

The Dodd-Frank Act defines the term “bona fide discount points” as used in § 1026.32(b)(1)(i)(E) and (F), which, as discussed above, permit exclusion of “bona fide discount points” from points and fees for qualified mortgages. TILA section 129C(b)(2)(C)(iii) defines the term “bona fide discount points” as “loan discount points which are knowingly paid by the consumer for the purpose of reducing, and which in fact result in a bona fide reduction of, the interest rate or time-price differential applicable to the mortgage.” TILA section 129C(b)(2)(C)(iv) limits the types of discount points that may be excluded from “points and fees” to those for which “the amount of the interest rate reduction purchased is reasonably consistent with established industry norms and practices for secondary market transactions.”

Proposed § 226.43(e)(3)(iv) would have implemented these provisions by defining the term “bona fide discount point” as “any percent of the loan amount” paid by the consumer that reduces the interest rate or time-price differential applicable to the mortgage loan by an amount based on a calculation that: (1) Is consistent with established industry practices for determining the amount of reduction in the interest rate or time-price differential appropriate for the amount of discount points paid by the consumer; and (2) accounts for the amount of compensation that the creditor can reasonably expect to receive from secondary market investors in return for the mortgage loan.

The Board's proposal would have required that the creditor be able to show a relationship between the amount of interest rate reduction purchased by a discount point and the value of the transaction in the secondary market. The Board observed that, based on outreach with representatives of creditors and GSEs, the value of a rate reduction in a particular mortgage transaction on the secondary market is based on many complex factors, which interact in a variety of complex ways. The Board noted that these factors may include, among others:

• The product type, such as whether the loan is a fixed-rate or adjustable-rate mortgage, or has a 30-year term or a 15-year term.
• How much the MBS market is willing to pay for a loan at that interest rate and the liquidity of an MBS with loans at that rate.
• How much the secondary market is willing to pay for excess interest on the loan that is available for capitalization outside of the MBS market.
• The amount of the guaranty fee required to be paid by the creditor to the investor.

The Board indicated that it was offering a flexible proposal because of its concern that a more prescriptive interpretation would be operationally unworkable for most creditors and would lead to excessive legal and regulatory risk. In addition, the Board also noted that, due to the variation in inputs described above, a more prescriptive rule likely would require continual updating, creating additional compliance burden and potential confusion.

The Board also noted a concern that small creditors such as community banks that often hold loans in portfolio rather than sell them on the secondary market may have difficulty complying with this requirement. The Board therefore requested comment on whether it would be appropriate to provide any exemptions from the requirement that the interest rate reduction purchased by a “bona fide discount point” be tied to secondary market factors.

Many industry commenters criticized the second prong of the Board's proposal, which would have required that the interest rate reduction account for the amount of compensation that the creditor can reasonably expect to receive from secondary market investors in return for the mortgage loan. Several industry commenters argued that this test would be complex and difficult to apply and that, if challenged, it would be difficult for creditors to prove that the calculation was done properly. Two industry commenters noted that creditors do not always sell or plan to sell loans in the secondary market at the time of origination and so would not know what compensation they would receive on the secondary market. Several industry commenters emphasized that the secondary market test would be impracticable for creditors holding loans in portfolio. Consumer groups did not comment on this issue.

As noted above, the Bureau is consolidating the exclusions for certain bona fide third-party charges and bona fide discount points in § 1026.32(b)(1)(i)(D) through (F). As a result, the Bureau is adopting proposed § 226.43(e)(3)(iv), with the revision discussed below, as renumbered § 1026.32(b)(3)(i). In the 2013 HOEPA Final Rule, the Bureau is adopting a definition of bona fide discount point for open-end credit in § 1026.32(b)(3)(ii).

After carefully considering the comments, the Bureau is modifying the definition of “bona fide discount point.” Specifically, the Bureau believes it would be difficult, if not impossible, for many creditors to account for the secondary market compensation in calculating interest rate reductions. This is particularly true for loans held in portfolio. Therefore, the Board is removing from § 1026.32(b)(3)(i) the requirement that interest rate reductions take into account secondary market compensation. Instead, as revised, § 1026.32(b)(3)(i) requires only that the calculation of the interest rate reduction be consistent with established industry practices for determining the amount of reduction in the interest rate or time-price differential appropriate for the amount of discount points paid by the consumer.

The Bureau finds that removing the secondary market component of the “bona fide” discount point definition is necessary and proper under TILA section 105(a) to effectuate the purposes of and facilitate compliance with TILA. Similarly, the Bureau finds that it is necessary and proper to use its authority under TILA section 129C(b)(3)(B)(i) to revise and subtract from the criteria that define a qualified mortgage by removing the secondary market component from the bona fide discount point definition. It will provide creditors sufficient flexibility to demonstrate that they are in compliance with the requirement that, to be excluded from points and fees, discount points must be bona fide. In clarifying the definition, it also will facilitate the use of bona fide discount points by consumers to help create the appropriate combination of points and rate for their financial situation, thereby helping ensure that consumers are offered and receive residential mortgage loan on terms that reasonably reflect their ability to repay the loans and that responsible, affordable mortgage credit remains available to consumers in a manner consistent with the purposes of TILA as provided in TILA section 129C.

To provide some guidance on how creditors may comply with this requirement, the Bureau is adding new comment 32(b)(3)(i)-1. This comment explains how creditors can comply with “established industry practices” for calculating interest rate reductions. Specifically, comment 32(b)(3)(i)-1 notes that one way creditors can satisfy this requirement is by complying with established industry norms and practices for secondary mortgage market transactions. Comment 32(b)(3)(i)-1 then provides two examples. First a creditor may rely on pricing in the to-be-announced (TBA) market for MBS to establish that the interest rate reduction is consistent with the compensation that the creditor could reasonably expect to receive in the secondary market. Second, a creditor could comply with established industry practices, such as guidelines from Fannie Mae or Freddie Mac that prescribe when an interest rate reduction from a discount point is considered bona fide. However, because these examples from the secondary market are merely illustrations of how a creditor could comply with the “established industry practices” requirement for bona fide interest rate reduction, creditors, and in particular creditors that retain loans in portfolio, will have flexibility to use other approaches for complying with this requirement.

32(b)(4) Total Loan Amount

32(b)(4)(i) Closed-End Credit

As added by section 1412 of the Dodd-Frank Act, TILA section 129C(b)(2)(A)(vii) defines a “qualified mortgage” as a mortgage for which, among other things, “the total points and fees * * * payable in connection with the loan do not exceed 3 percent of the total loan amount.” For purposes of implementing the qualified mortgage provisions, the Board proposed to retain existing comment 32(a)(1)(ii)-1 explaining the meaning of the term “total loan amount,” with certain minor revisions discussed below, while also seeking comment on an alternative approach.

The proposal would have revised the “total loan amount” calculation under current comment 32(a)(1)(ii)-1 to account for charges added to TILA's definition of points and fees by the Dodd-Frank Act. Under Regulation Z for purposes of applying the existing points and fees trigger for high-cost loans, the “total loan amount” is calculated as the amount of credit extended to or on behalf of the consumer, minus any financed points and fees. Specifically, under current comment 32(a)(1)(ii)-1, the “total loan amount” is calculated by “taking the amount financed, as determined according to § 1026.18(b), and deducting any cost listed in § 1026.32(b)(1)(iii) and § 1026.32(b)(1)(iv) that is both included as points and fees under § 1026.32(b)(1) and financed by the creditor.” Section 1026.32(b)(1)(iii) and (b)(1)(iv) pertain to “real estate-related fees” listed in § 1026.4(c)(7) and premiums or other charges for credit insurance or debt cancellation coverage, respectively.

The Board proposed to revise this comment to cross-reference additional financed points and fees described in proposed § 226.32(b)(1)(vi) as well. This addition would have required a creditor also to deduct from the amount financed any prepayment penalties that are incurred by the consumer if the mortgage loan refinances a previous loan made or currently held by the creditor refinancing the loan or an affiliate of the creditor—to the extent that the prepayment penalties are financed by the creditor. As a result, the 3 percent limit on points and fees for qualified mortgages would have been based on the amount of credit extended to the consumer without taking into account any financed points and fees.

The Board's proposal also would have revised one of the commentary's examples of the “total loan amount” calculation. Specifically, the Board proposed to revise the example of a $500 single premium for optional “credit life insurance” used in comment 32(b)(1)(i)-1.iv to be a $500 single premium for optional “credit unemployment insurance.” The Board stated that this change was proposed because, under the Dodd-Frank Act, single-premium credit insurance—including credit life insurance—is prohibited in covered transactions except for certain limited types of credit unemployment insurance. See TILA section 129C(d). The Board requested comment on the proposed revisions to the comment explaining how to calculate the “total loan amount,” including whether additional guidance is needed.

The Board also requested comment on whether to streamline the calculation to ensure that the “total loan amount” would include all credit extended other than financed points and fees. Specifically, the Board solicited comment on whether to revise the calculation of “total loan amount” to be the “principal loan amount” (as defined in § 226.18(b) and accompanying commentary), minus charges that are points and fees under § 226.32(b)(1) and are financed by the creditor. The Board explained that the purpose of using the “principal loan amount” instead of the “amount financed” would be to streamline the calculation to facilitate compliance and to ensure that no charges other than financed points and fees are excluded from the “total loan amount.” In general, the revised calculation would have yielded a larger “total loan amount” to which the percentage points and fees thresholds would have to be applied than would the proposed (and existing) “total loan amount” calculation, because only financed points and fees and no other financed amounts would be excluded. Thus, creditors in some cases would be able to charge more points and fees on the same loan under the alternative outlined by the Board than under either the proposed or existing rule.

In the 2012 HOEPA Proposal, the Bureau proposed the following for organizational purposes: (1) To move the existing definition of “total loan amount” for closed-end mortgage loans from comment 32(a)(1)(ii)-1 to proposed § 1026.32(b)(6)(i); and (2) to move the examples showing how to calculate the total loan amount for closed-end mortgage loans from existing comment 32(a)(1)(ii)-1 to proposed comment 32(b)(6)(i)-1. The Bureau proposed to specify that the calculation applies to closed-end mortgage loans because the Bureau also proposed to define “total loan amount” separately for open-end credit plans. The Bureau also proposed to amend the definition of “total loan amount” in a manner similar to the Board's alternative proposal described above. The Bureau indicated this proposed revision would streamline the total loan amount calculation to facilitate compliance and would be sensible in light of the more inclusive definition of the finance charge proposed in the Bureau's 2012 TILA-RESPA Integration Proposal.

Few commenters addressed the Board's proposal regarding total loan amount. Several industry commenters recommended that the alternative method of calculating total loan amount be used because it would be easier to calculate. At least two industry commenters recommended that, for simplicity, the amount recited in the note be used for calculating the permitted points and fees.

After reviewing the comments, the Bureau is following the 2012 HOEPA Proposal and moving the definition of total loan amount into the text of the rule in § 1026.32(b)(4)(i). In 2013 HOEPA Final Rule, the Bureau is adopting a definition of total loan amount for open-end credit in § 1026.32(b)(4)(ii). The examples showing how to calculate the total loan amount are moved to comment 32(b)(4)(i)-1. However, the Bureau has concluded that, at this point, the current approach to calculating the total loan amount should remain in place. Creditors are familiar with the method from using it for HOEPA points and fees calculations. Moreover, as noted above, the Bureau is deferring action on the more inclusive definition of the finance charge proposed in the Bureau's 2012 TILA-RESPA Integration Proposal. If the Bureau expands the definition of the finance charge, the Bureau will at the same time consider the effect on coverage thresholds that rely on the finance charge or the APR.

32(b)(5)

The final rule renumbers existing § 1026.32(b)(2) defining the term “affiliate” as § 1026.32(b)(5) for organizational purposes.

32(b)(6) Prepayment Penalty

The Dodd-Frank Act's Amendments to TILA Relating to Prepayment Penalties

Sections 1431 and 1432 of the Dodd-Frank Act (relating to high-cost mortgages) and section 1414 of the Dodd-Frank Act (relating to qualified mortgages) amended TILA to restrict and, in many cases, prohibit a creditor from imposing prepayment penalties in dwelling-secured credit transactions. The Dodd-Frank Act restricted prepayment penalties in three main ways.

First, as the Board discussed in its 2011 ATR Proposal, the Dodd-Frank Act added new TILA section 129C(c)(1) relating to qualified mortgages, which generally provides that a covered transaction (i.e., in general, a closed-end, dwelling-secured credit transaction) may include a prepayment penalty only if it; (1) Is a qualified mortgage, to be defined by the Board, (2) has an APR that cannot increase after consummation, and (3) is not a higher-priced mortgage loan. The Board proposed to implement TILA section 129C(c)(1) in § 226.43(g)(1) and to define the term prepayment penalty in § 226.43(b)(10). Under new TILA section 129C(c)(3), moreover, even loans that meet the statutorily prescribed criteria (i.e., fixed-rate, non-higher-priced qualified mortgages) are capped in the amount of prepayment penalties that may be charged, starting at three percent in the first year after consummation and decreasing annually by increments of one percentage point thereafter so that no penalties may be charged after the third year. The Board proposed to implement TILA section 129C(c)(3) in § 226.43(g)(2).

Second, section 1431(a) of the Dodd-Frank Act amended TILA section 103(bb)(1)(A)(iii) to provide that a credit transaction is a high-cost mortgage if the credit transaction documents permit the creditor to charge or collect prepayment fees or penalties more than 36 months after the transaction closing or if such fees or penalties exceed, in the aggregate, more than two percent of the amount prepaid. Moreover, under amended TILA section 129(c)(1), high-cost mortgages are prohibited from having a prepayment penalty. Accordingly, any prepayment penalty in excess of two percent of the amount prepaid on any closed end mortgage would both trigger and violate the rule's high-cost mortgage provisions. The Bureau's 2012 HOEPA Proposal proposed to implement these requirements with several minor clarifications in § 1026.32(a)(1)(iii). See 77 FR 49090, 49150 (Aug. 15, 2012).

Third, both qualified mortgages and most closed-end mortgage loans and open-end credit plans secured by a consumer's principal dwelling are subject to additional limitations on prepayment penalties through the inclusion of prepayment penalties in the definition of points and fees for qualified mortgages and high-cost mortgages. See the section-by-section analysis of proposed § 226.32(b)(1)(v) and (vi); 77 FR 49090, 49109-10 (Aug. 15, 2012).

Taken together, the Dodd-Frank Act's amendments to TILA relating to prepayment penalties mean that most closed-end, dwelling-secured transactions: (1) May provide for a prepayment penalty only if the transaction is a fixed-rate, qualified mortgage that is neither high-cost nor higher-priced under §§ 1026.32 and 1026.35; (2) may not, even if permitted to provide for a prepayment penalty, charge the penalty more than three years following consummation or in an amount that exceeds two percent of the amount prepaid; and (3) may be required to limit any penalty even further to comply with the points and fees limitations for qualified mortgages or to stay below the points and fees trigger for high-cost mortgages.

In the interest of lowering compliance burden and to provide additional clarity for creditors, the Bureau has elected to define prepayment penalty in a consistent manner for purposes of all of the Dodd-Frank Act's amendments. This definition is located in § 1026.32(b)(6). New § 1026.43(b)(10) cross-references this prepayment definition to provide consistency.

TILA establishes certain disclosure requirements for transactions for which a penalty is imposed upon prepayment, but TILA does not define the term “prepayment penalty.” The Dodd-Frank Act also does not define the term. TILA section 128(a)(11) requires that the transaction-specific disclosures for closed-end consumer credit transactions disclose a “penalty” imposed upon prepayment in full of a closed-end transaction, without using the term “prepayment penalty.” 15 U.S.C. 1638(a)(11). Comment 18(k)(1)-1 clarifies that a “penalty” imposed upon prepayment in full is a charge assessed solely because of the prepayment of an obligation and includes, for example, “interest” charges for any period after prepayment in full is made and a minimum finance charge.

The Board's 2011 ATR Proposal proposed to implement the Dodd-Frank Act's prepayment penalty-related amendments to TILA for qualified mortgages by defining “prepayment penalty” for most closed-end, dwelling-secured transactions in new § 226.43(b)(10), and by cross-referencing proposed § 226.43(b)(10) in the proposed joint definition of points and fees for qualified and high-cost mortgages in § 226.32(b)(1)(v) and (vi). The definition of prepayment penalty proposed in the Board's 2011 ATR Proposal differed from the Board's prior proposals and existing guidance in the following respects: (1) Proposed § 226.43(b)(10) defined prepayment penalty with reference to a payment of “all or part of” the principal in a transaction covered by the provision, while § 1026.18(k) and associated commentary and the Board's 2009 Closed-End Proposal and 2010 Mortgage Proposal referred to payment “in full;” (2) the examples provided omitted reference to a minimum finance charge and loan guarantee fees; and (3) proposed § 226.43(b)(10) did not incorporate, and the Board's 2011 ATR Proposal did not otherwise address, the language in § 1026.18(k)(2) and associated commentary regarding disclosure of a rebate of a precomputed finance charge, or the language in § 1026.32(b)(6) and associated commentary concerning prepayment penalties for high-cost mortgages.

The Board proposal generally received support from industry commenters and consumer advocates for accurately implementing section 129C(c) by using a plain language definition of prepayment penalty. Many commenters, particularly consumer groups, supported a rule that eliminates or tightly restricts the availability of prepayment penalties. Some industry commenters, however, cautioned the Bureau against implementing an overbroad definition of prepayment penalty, citing primarily a concern over consumers' access to credit. At least one commenter argued that a prepayment penalty ban should be more narrowly focused on the subprime loan market, noting that the proposal affected prepayment penalties on a wider variety of products. Other industry commenters expressed a concern about the Board's approach to the monthly interest accrual amortization method, as discussed in more detail below as part of the discussion of comment 32(b)(6)-1.

The Bureau adopts the definition of prepayment penalty under § 1026.32(b)(6) largely as proposed by the Board in order to create a clear application of the term prepayment penalty that is consistent with the definitions proposed in the Bureau's 2012 TILA-RESPA Proposal (which itself draws from the definition adopted in the Bureau's 2013 HOEPA Final Rule). However, the Bureau adds to § 1026.32(b)(6) an explicit exclusion from the definition of prepayment penalty for a waived bona fide third-party charge that the creditor imposes if the consumer, sooner than 36 months after consummation, pays all of a covered transaction's principal before the date on which the principal is due. This addition is discussed in detail below. Consistent with TILA section 129(c)(1), existing § 1026.32(d)(6), and the Board's proposed § 226.43(b)(10) for qualified mortgages, § 1026.32(b)(6)(i) provides that, for a closed-end mortgage loan, a “prepayment penalty” means a charge imposed for paying all or part of the transaction's principal before the date on which the principal is due, though the Bureau has added a carve-out from this definition to accommodate the repayment of certain conditionally waived closing costs when the consumer prepays in full. The Bureau adopts this definition of prepayment penalty under § 1026.32(b)(6), rather than under § 1026.43(b)(10), to facilitate compliance for creditors across rulemakings. The definition of “prepayment penalty” under § 1026.32(b)(6) thus will apply to prepayment penalty restrictions, as applied under § 1026.43(g). Section 1026.32(b)(6) also contains requirements and guidance related to the Bureau's 2013 HOEPA Final Rule, such as a definition of prepayment penalty that applies to open-end credit.

The Board's 2011 ATR Proposal included as an example of a prepayment penalty a fee that the creditor waives unless the consumer prepays the covered transaction. Some industry commenters contended that such conditional fee waivers should be excluded from the definition of prepayment penalties. The commenters argued that creditors imposed conditional fee waivers not to increase profit, but to ensure compensation for fixed costs associated with originating the loan. At least one commenter directed the Bureau to a 1996 National Credit Union Administration opinion letter that concluded that a conditional waiver of closing costs by a credit union was a benefit to the consumer. Other comments characterized the conditional fee waiver as a “reimbursement,” rather than compensation.

The Bureau finds such comments persuasive, particularly with respect to a situation in which the creditor waives a bona fide third-party charge (or charges) on condition that the consumer reimburse the creditor for the cost of that charge if the consumer prepays the loan. In such situations, the Bureau recognizes that the creditor receives no profit from imposing or collecting such charges and the Bureau believes that treating such charges as a prepayment penalty might very well have the effect of reducing consumer choice without providing any commensurate consumer benefit. In an effort to provide a sensible way to permit a creditor to protect itself from losing money paid at closing to third parties on the consumer's behalf, prior to such time as the creditor can otherwise recoup such costs through the interest rate on the mortgage loan, while balancing consumer protection interests, the Bureau has concluded that such fees should be permissible for a limited time after consummation. The Bureau thus adopts § 1032(b)(6)(i) to clarify that the term prepayment penalty does not include a waived bona fide third-party charge imposed by the creditor if the consumer pays all of a covered transaction's principal before the date on which the principal is due sooner than 36 months after consummation. The Bureau concludes that limiting the duration of the possible charge to 36 months after consummation is consistent with TILA 129C(c)(3)(D), which prohibits any prepayment penalty three years after loan consummation, while accommodating the concerns discussed above. Moreover, § 1032(b)(6)(i) excludes from the definition of prepayment penalty only those charges that a creditor imposes to recoup waived bona-fide third party charges in such cases where the consumer prepays in full. Thus, for example, if one month after loan consummation, the consumer prepays $100 of principal earlier than it is due, where the total principal is $100,000, then any fee that the creditor imposes for such prepayment is a prepayment penalty under § 1032(b)(6)(i) and such a fee is restricted in accordance with § 1026.43(g).

The Bureau believes that § 1026.32(b)(6) accurately implements TILA section 129C(c), which significantly limits the applicability and duration of prepayment penalties. Some commenters argued that restrictions on prepayment penalties should be more narrowly focused on specific products or consumers, because not all consumers need protection from the pitfalls of prepayment penalties. The Bureau agrees that prepayment penalties are not always harmful to consumers and that, in some cases, allowing a creditor to charge a prepayment penalty may lead to increased consumer choice and access to credit. Congress recognized this balance by allowing a creditor to charge a prepayment penalty only in certain circumstances, such as requiring the loan to be a qualified mortgage, under TILA section 129C(c)(1)(A), and by limiting a creditor to charging a prepayment penalty to no more than three years following consummation, under TILA section 129C(c)(3)(D). Section 1026.32(b)(6) remains faithful to that balance, with the Bureau's minor clarification with respect to waived bona fide third party charges, as described above.

The Board's 2011 ATR Proposal included several other examples of a prepayment penalty under proposed § 226.43(b)(10)(i). For clarity, the Bureau incorporates these examples as comment 32(b)(6)-1.i and ii, and the Bureau is adding comment 32(b)(6)-1.iii and iv to provide additional clarity. Likewise, the Bureau is largely adopting the Board's proposed § 226.43(b)(10)(ii), an example of what is not a prepayment penalty, as comment 32(b)(6)-3.i, as well as adding comment 32(b)(6)-3.ii.

Comment 32(b)(6)-1.i through iv gives the following examples of prepayment penalties: (1) A charge determined by treating the loan balance as outstanding for a period of time after prepayment in full and applying the interest rate to such “balance,” even if the charge results from interest accrual amortization used for other payments in the transaction under the terms of the loan contract; (2) a fee, such as an origination or other loan closing cost, that is waived by the creditor on the condition that the consumer does not prepay the loan; (3) a minimum finance charge in a simple interest transaction; and (4) computing a refund of unearned interest by a method that is less favorable to the consumer than the actuarial method, as defined by section 933(d) of the Housing and Community Development Act of 1992, 15 U.S.C. 1615(d).

Post-payoff interest charges. The Board proposal included as an example of a prepayment penalty in proposed § 226.43(b)(10)(i)(A) a charge determined by the creditor or servicer treating the loan balance as outstanding for a period of time after prepayment in full. Some industry commenters expressed reservations about treating this monthly interest accrual amortization method as a prepayment penalty, arguing that such a rule might cause higher resale prices in the secondary mortgage market to account for cash flow uncertainty. Other commenters noted that this calculation method is currently used by FHA to compute interest on its loans (including loans currently in Ginnie Mae pools), or that such charges were not customarily considered a prepayment penalty. Some commenters expressed concern that the rule would disrupt FHA lending.

After careful consideration of the comments received, the Bureau concludes that going forward (e.g., for loans a creditor originates after the effective date), it is appropriate to designate higher interest charges for consumers based on accrual methods that treat a loan balance as outstanding for a period of time after prepayment in full as prepayment penalties under § 1026.32(b)(6) and comment 32(b)(6)-1.i. In such instances, the consumer submits a payment before it is due, but the creditor nonetheless charges interest on the portion of the principal that the creditor has already received. The Bureau believes that charging a consumer interest after the consumer has repaid the principal is the functional equivalent of a prepayment penalty. Comment 32(b)(6)-1.i further clarifies that “interest accrual amortization” refers to the method by which the amount of interest due for each period (e.g., month) in a transaction's term is determined and notes, for example, that “monthly interest accrual amortization” treats each payment as made on the scheduled, monthly due date even if it is actually paid early or late (until the expiration of any grace period). The proposed comment also provides an example where a prepayment penalty of $1,000 is imposed because a full month's interest of $3,000 is charged even though only $2,000 in interest was earned in the month during which the consumer prepaid.

With respect to FHA practices relating to monthly interest accrual amortization, the Bureau has consulted extensively with HUD in issuing this final rule as well as the 2013 HOEPA Final Rule. Based on these consultations, the Bureau understands that HUD must engage in rulemaking to end its practice of imposing interest charges on consumers for the balance of the month in which consumers prepay in full. The Bureau further understands that HUD requires approximately 24 months to complete its rulemaking process. Accordingly, in recognition of the important role that FHA-insured credit plays in the current mortgage market and to facilitate FHA creditors' ability to comply with this aspect of the 2013 ATR and HOEPA Final Rules, the Bureau is using its authority under TILA section 105(a) to provide for optional compliance until January 15, 2015 with § 1026.32(b)(6)(i) and the official interpretation of that provision in comment 32(b)(6)-1.i regarding monthly interest accrual amortization. Specifically, § 1026.32(b)(6)(i) provides that interest charged consistent with the monthly interest accrual amortization method is not a prepayment penalty for FHA loans consummated before January 21, 2015. FHA loans consummated on or after January 21, 2015 must comply with all aspects of the final rule. The Bureau is making this adjustment pursuant to its authority under TILA section 105(a), which provides that the Bureau's regulations may contain such additional requirements, classifications, differentiations, or other provisions, and may provide for such adjustments and exceptions for all or any class of transactions as in the Bureau's judgment are necessary or proper to effectuate the purposes of TILA, prevent circumvention or evasion thereof, or facilitate compliance therewith. 15 U.S.C. 1604(a). The purposes of TILA include the purposes that apply to 129C, to assure that consumers are offered and receive residential mortgage loans on terms that reasonably reflect their ability to repay the loan. See 15 U.S.C. 1639b(a)(2). The Bureau believes it is necessary and proper to make this adjustment to ensure that consumers receive loans on affordable terms and to facilitate compliance with TILA and its purposes while mitigating the risk of disruption to the market. For purposes of this rulemaking, the Bureau specifically notes that the inclusion of interest charged consistent with the monthly interest accrual amortization method in the definition of prepayment penalty for purposes of determining whether a transaction is in compliance with the requirements of § 1026.43(g) applies only to transactions consummated on or after January 10, 2014; for FHA loans, compliance with this aspect of the definition of prepayment penalties is optional for transactions consummated prior to January 21, 2015.

With regard to general concerns that loans subject to these interest accrual methods may be subject to higher prices on the secondary market, the Bureau is confident that the secondary market will be able to price the increased risk of prepayment, if any, that may occur as a result of the limits that will apply to monthly interest accrual amortization-related prepayment penalties. The secondary market already does so for various other types of prepayment risk on investor pools, such as the risk of refinancing or sale of the property.

Comment 32(b)(6)-1.ii further explains the 36 month carve-out for a waived bona fide third-party charge imposed by the creditor if the consumer pays all of a covered transaction's principal before the date on which the principal is due sooner than 36 months after consummation, as included in § 1026.32(b)(6)(i). The comment explains that if a creditor waives $3,000 in closing costs to cover bona fide third party charges but the terms of the loan agreement provide that the creditor may recoup $4,500, in part to recoup waived charges, then only $3,000 that the creditor may impose to cover the waived bona fide third party charges is considered not to be a prepayment penalty, while any additional $1,500 charge for prepayment is a prepayment penalty and subject to the restrictions under § 1026.43(g). This comment also demonstrates that the only amount excepted from the definition of prepayment penalty under § 1026.32(b)(6)(i) is the actual amount that the creditor pays to a third party for a waived, bona fide charge.

Minimum finance charges; unearned interest refunds. Although longstanding Regulation Z commentary has listed a minimum finance charge in a simple interest transaction as an example of a prepayment penalty, the Board proposed to omit that example from proposed § 226.43(b)(10) because the Board reasoned that such a charge typically is imposed with open-end, rather than closed-end, transactions. The Bureau did not receive substantial comment on this omission, but the Bureau has elected to continue using this example in comment 32(b)(6)-1.iii for consistency. Likewise, the Board did not propose to include the example of computing a refund of unearned interest by a method that is less favorable to the consumer than the actuarial method, but the Bureau is nonetheless using this example in comment 32(b)(6)-1.iv because similar language is found in longstanding Regulation Z commentary.

Examples of fees that are not prepayment penalties. The Board included in proposed § 226.43(b)(10)(ii) an example of a fee not considered a prepayment penalty. For the sake of clarity, the Bureau is moving this example into comment 32(b)(6)-2.i, rather than keep the example in the text of the regulation. The Bureau also is adding a second example in comment 32(b)(6)-2.ii.

Comment 32(b)(6)-2.i explains that fees imposed for preparing and providing documents when a loan is paid in full are not prepayment penalties when such fees are imposed whether or not the loan is prepaid or the consumer terminates the plan prior to the end of its term. Commenters did not provide substantial feedback on this example, which the Bureau has reworded slightly from the Board proposal to provide conformity and clarity.

The Board proposed omitting text from preexisting commentary on Regulation Z stating that a prepayment penalty did not include loan guarantee fees, noting that loan guarantee fees are not charges imposed for paying all or part of a loan's principal before the date on which the principal is due. The Bureau did not receive substantial comment on this omission. While the Bureau agrees with the Board's analysis, the Bureau nonetheless elects to include this example in comment 43(b)(6)-2.ii to clarify that loan guarantee fees continue to fall outside the definition of a prepayment penalty. Moreover, including this example of a fee that is not a prepayment penalty is consistent with the Bureau's efforts to streamline definitions and ease regulatory burden.

Construction-to-permanent financing. Some industry commenters advocated that, for construction-to-permanent loans, the Bureau should exclude from the definition of prepayment penalty charges levied by a creditor if a consumer does not convert the construction loan into a permanent loan with the same creditor within a specified time period. The Bureau believes that the concern expressed by these commenters that the cost of credit for these construction-to-permanent loans would increase if such charges were treated as prepayment penalties is misplaced primarily because in many cases, such charges are not, in fact, a prepayment penalty. A prepayment penalty is “a charge imposed for paying all or part of a covered transaction's principal before the date on which the principal is due.” First, the case where the creditor charges the consumer a fee for failing to convert a loan within a specified period after completing the repayment of a construction loan as scheduled is not a prepayment penalty; the fee is not assessed for an early payment of principal, but rather for the consumer's failure to take an action upon scheduled repayment of principal. Second, the case where a consumer does convert the construction loan to a permanent loan in a timely manner, but incurs a fee for converting the loan with another creditor, is also likely not prepayment penalty. While such cases depend highly on contractual wording, in the example above, the consumer is charged a fee not for his early payment of principal, but rather for his use of another creditor. Third, the case where the creditor charges the consumer a fee for converting the construction loan to a permanent loan earlier than specified by agreement, even with the same creditor, likely is a prepayment penalty. While this example is not the same as the hypothetical described by most commenters, who expressed concern if a consumer does not convert the construction loan into a permanent loan with the same creditor within a specified time period, this is an example of a prepayment penalty, as the creditor has imposed a charge for paying all or part of a covered transaction's principal before the date on which the principal was due. As the above examples demonstrate, whether a construction-to-permanent loan contains a prepayment penalty is fact-specific, and the Bureau has decided that adding a comment specifically addressing such loans would not be instructive. The Bureau sees no policy reason to generally exclude fees specific to construction-to-permanent loan from the definition of prepayment penalty and its statutory limits. The Bureau was not presented with any evidence that the risks inherent in construction-to-permanent loans could not be priced by creditors through alternative means, such as the examples described above, via interest rate, or charging closing costs. The Bureau also notes that, because of the scope of the rule, described in the section-by-section analysis of § 1026.43(a), as well as the prepayment penalty restrictions, described in the section-by-section analysis of § 1026.43(g), construction-to-permanent loans cannot be qualified mortgages, and thus under § 1026.43(g)(1)(ii)(B) cannot include a prepayment penalty. Construction-to-permanent loans are discussed in more detail in the section-by-section analysis of § 1026.43(a).

Open-end credit. The Bureau is concurrently adopting comments 32(b)(6)-3 and -4 to clarify its approach to prepayment penalties with respect to open-end credit. As the Board's 2011 ATR Proposal did not address open-end credit plans, the Bureau is not clarifying prepayment penalties with respect to open-end credit plans in this final rule. Instead, guidance is provided in comments 32(b)(6)-3 and -4, which the Bureau is adopting in the concurrent 2013 HOEPA Final Rule.

Section 1026.43 Minimum Standards for Transactions Secured by a Dwelling 43(a) Scope

Sections 1411, 1412 and 1414 of the Dodd-Frank Act add new TILA section 129C, which requires creditors to determine a consumer's ability to repay a “residential mortgage loan” and establishes new rules and prohibitions on prepayment penalties. Section 1401 of the Dodd-Frank Act adds new TILA section 103(cc), which defines “residential mortgage loan” to mean, with some exceptions, any consumer credit transaction secured by a mortgage, deed of trust, or other equivalent consensual security interest on “a dwelling or on residential real property that includes a dwelling.” TILA section 103(v) defines “dwelling” to mean a residential structure or mobile home which contains one- to four-family housing units, or individual units of condominiums or cooperatives. Thus, a “residential mortgage loan” is a dwelling-secured consumer credit transaction, regardless of whether the consumer credit transaction involves a home purchase, refinancing, home equity loan, first lien or subordinate lien, and regardless of whether the dwelling is a principal residence, second home, vacation home (other than a timeshare residence), a one- to four-unit residence, condominium, cooperative, mobile home, or manufactured home.

However, the Dodd-Frank Act specifically excludes from the term “residential mortgage loan” an open-end credit plan or an extension of credit secured by an interest in a timeshare plan, for purposes of the repayment ability and prepayment penalty provisions under TILA section 129C, among other provisions. See TILA section 103(cc)(5); see also TILA section 129C(i) (providing that timeshare transactions are not subject to TILA section 129C). Further, the repayment ability provisions of TILA section 129C(a) do not apply to reverse mortgages or temporary or “bridge” loans with a term of 12 months or less, including a loan to purchase a new dwelling where the consumer plans to sell another dwelling within 12 months. See TILA section 129C(a)(8). The repayment ability provisions of TILA section 129C(a) also do not apply to consumer credit transactions secured by vacant land. See TILA section 103(cc)(5) and 129C(a)(1).

TILA Section 103(cc) defines “residential mortgage loan” to mean a consumer credit transaction secured by a mortgage or equivalent consensual security interest “on a dwelling or on residential real property that includes a dwelling.” Under TILA and Regulation Z, the term “dwelling” means a residential structure with one to four units, whether or not the structure is attached to real property, and includes a condominium or cooperative unit, mobile home, and trailer, if used as a residence. See 15 U.S.C. 1602(v), § 1026.2(a)(19). To facilitate compliance by using consistent terminology throughout Regulation Z, the proposal used the term “dwelling,” as defined in § 1026.2(a)(19), and not the phrase “residential real property that includes a dwelling.” Proposed comment 43(a)-2 clarified that, for purposes of proposed § 226.43, the term “dwelling” would include any real property to which the residential structure is attached that also secures the covered transaction.

Proposed § 226.43(a) generally defined the scope of the ability-to-repay provisions to include any consumer credit transaction that is secured by a dwelling, other than home equity lines of credit, mortgage transactions secured by an interest in a timeshare plan, or for certain provisions reverse mortgages or temporary loans with a term of 12 months or less. Proposed comment 43(a)-1 clarified that proposed § 226.43 would not apply to an extension of credit primarily for a business, commercial, or agricultural purpose and cross-referenced the existing guidance on determining the primary purpose of an extension of credit in commentary on § 1026.3.

Numerous commenters requested additional exemptions from coverage beyond the statutory exemptions listed at proposed § 226.43(a)(1) through (3). The Bureau received requests for exemptions from the rule for seller-financed transactions, loans secured by non-primary residences, community development loans, downpayment assistance loans, loans eligible for purchase by GSEs, and housing stabilization refinances. The requested exemptions related to community development loans, downpayment assistance loans, and housing stabilization refinances are not being included in this final rule, but are addressed in the Bureau's proposed rule regarding amendments to the ability-to-repay requirements, published elsewhere in today's Federal Register. The requested exemptions that are not being included in the rule and are not being addressed in today's concurrent proposal are discussed immediately below.

The Bureau received numerous letters from individuals concerned that the rule would cover individual home sellers who finance the buyer's purchase, either through a loan or an installment sale. However, because the definition of “creditor” for mortgages generally covers only persons who extend credit secured by a dwelling more than five times in a calendar year, the overwhelming majority of individual seller-financed transactions will not be covered by the rule. Those creditors who self-finance six or more transactions in a calendar year, whether through loans or installment sales, will need to comply with the ability-to-repay provisions of § 1026.43, just as they must comply with other relevant provisions of Regulation Z.

An association of State bank regulators suggested that the scope of the ability-to-repay requirements be limited to owner-occupied primary residences, stating that ability to repay on vacation homes and investment properties should be left to an institution's business judgment. The Bureau believes it is not appropriate or necessary to exercise its exception authority to change the scope of the provision in this way for several reasons. First, as discussed in proposed comment 43(a)-1, loans that have a business purpose are not covered by TILA, and so would not be covered by the ability-to-repay provisions as proposed and adopted. Investment purpose loans are considered to be business purpose loans. Second, vacation home loans are consumer credit transactions that can have marked effects on a consumer's finances. If a consumer is unable to repay a mortgage on a vacation home, the consumer will likely suffer severe financial consequences and the spillover effects on property values and other consumers in the affected area can be substantial as well. Third, the Bureau understands that default rates on vacation homes are generally higher than those on primary residences, and an exemption could increase this disparity.

For the reasons discussed below, the general scope provision and the statutory exemptions in § 1026.43(a)(1) through (3)(ii) are adopted substantially as proposed, with minor changes as discussed in the relevant sections below, and the addition of § 1026.43(a)(3)(iii) to provide an exemption for the construction phase of a construction-to-permanent loan.

The general scope provision at § 1026.43(a) now includes language making clear that real property attached to a dwelling will be considered a part of the dwelling for purposes of compliance with § 1026.43. Although as discussed above similar language was included in the official commentary in the proposed rule, the Bureau believes this important legal requirement should be part of the regulatory text.

Comment 43(a)-1 now includes a reference to § 1026.20(a), which describes different types of changes to an existing loan that will not be treated as refinancings, to make clear that creditors may rely on that section in determining whether or not § 1026.43 will apply to a particular change to an existing loan.

43(a)(1)

The Board's proposal included an exemption from the scope of section 226.43 for “[a] home equity line of credit subject to § 226.5b,” which implemented the exclusion of HELOCs from coverage in the statutory definition of “residential mortgage loan.” Dodd-Frank Act section 1401. The Bureau received two comments asking that the HELOC exemption be reconsidered. The commenters stated that HELOCs had contributed to the crisis in the mortgage market and that failure to include them in the ability-to-repay rule's coverage would likely lead to more consumer abuse and systemic problems.

The Bureau notes that Congress specifically exempted open-end lines of credit from the ability-to-repay requirements, even though the Dodd-Frank Act extends other consumer protections to such loans, including the requirements for high-cost mortgages under HOEPA. The Bureau also notes that home equity lines of credit have consistently had lower delinquency rates than other forms of consumer credit. Furthermore, the requirements contained in the Dodd-Frank Act with respect to assessing a consumer's ability to repay a residential mortgage, and the regulations the Bureau is adopting thereunder, were crafted to apply to the underwriting of closed-end loans and are not necessarily transferrable to underwriting for an open-end line of credit secured by real estate. In light of these considerations, the Bureau does not believe there is sufficient justification to find it necessary or proper to use its adjustment and exception authority to expand the ability-to-repay provisions to HELOCs at this time. However, as discussed in detail below, the Bureau is adopting the Board's proposal to require creditors to consider and verify contemporaneous HELOCs in addition to other types of simultaneous loans for the purpose of complying with the ability-to-repay provisions. See the section-by-section analysis of § 1026.43(b)(12) below. In addition, the final rule includes the Board's proposed anti-evasion provision, which forbids the structuring of credit that does not meet the definition of open-end credit as an open-end plan in order to evade the requirements of this rule. See § 1026.43(h). Accordingly, § 1026.43(a)(1) is adopted as proposed, with the embedded citation updated. However, the Bureau intends to monitor the HELOC exemption through its supervision function and may revisit the issue as part of its broader review of the ability-to-repay rule under section 1022(d) of the Dodd-Frank Act, which requires the Bureau to publish an assessment of a significant rule or order not later than five years after its effective date.

43(a)(2)

The Bureau did not receive comments on the statutory timeshare exemption included in proposed § 226.43(a)(2). Accordingly, the Bureau is adopting § 1026.43(a)(2) as proposed.

43(a)(3)

43(a)(3)(i)

Proposed § 226.43(a)(3)(i) created an exemption from the ability-to-repay requirements in § 226.43(c) through (f) for reverse mortgages, as provided in the statute. The Bureau did not receive comments on this exemption. Accordingly, the Bureau is adopting § 1026.43(a)(3)(i) as proposed.

43(a)(3)(ii)

Proposed § 226.43(a)(3)(ii) provided an exemption from the ability-to-repay requirements in § 226.43(c) through (f) for “[a] temporary or `bridge' loan with a term of 12 months or less, such as a loan to finance the purchase of a new dwelling where the consumer plans to sell a current dwelling within 12 months or a loan to finance the initial construction of a dwelling.” Furthermore, proposed comment 43(a)-3 provided that, “[w]here a temporary or bridge loan is renewable, the loan term does not include any additional period of time that could result from a renewal provision.” The Board solicited comment on whether a decision to treat renewals in this manner would lead to evasion of the rule. The statute includes the one-year exemption implemented in the proposed rule but does not specifically address renewals. TILA section 129C(a)(8), 15 U.S.C. 1639c(a)(8).

Generally, commenters did not specifically address the proposal's request for comment on renewals of short-term financing; however, one industry commenter stated that the statutory one-year limitation would interfere with construction loans, which often require more than a year to complete. The Bureau understands that construction loans often go beyond a single year. Although the comment did not specify that disregarding potential renewals would alleviate this concern, the Bureau believes that disregarding renewals would facilitate compliance and prevent unwarranted restrictions on access to construction loans.

Commenters did not respond to the Board's query about whether or not disregarding renewals of transactions with one-year terms would lead to evasion of the rule. Upon further analysis, the Bureau believes that this concern does not warrant changing the proposed commentary. However, the Bureau intends to monitor the issue through its supervision function and to revisit the issue as part of its broader review of the ability-to-repay rule under section 1022(d) of the Dodd-Frank Act, which requires the Bureau to conduct an assessment of significant rules five years after they are adopted.

One industry trade association commented on the wording of the temporary financing exemption, suggesting that the inclusion of the two examples, bridge loans and construction loans, would create uncertainty as to whether the exemption would apply to temporary financing of other types. However, the Bureau believes further clarification is not required because the exemption applies to any temporary loan with a term of 12 months or less, and the examples are merely illustrative. The Bureau is aware of and provides clarifying examples of certain common loan products that are temporary or “bridge” loans. The commenter did not note other common types of temporary loan products. The Bureau further believes that the rule permits other types of temporary financing as long as the loan satisfies the requirements of the exemption.

Accordingly, § 1026.43(a)(3)(ii) and associated commentary are adopted substantially as proposed.

43(a)(3)(iii)

The Bureau also received comments requesting clarification on how the temporary financing exemption would apply to construction-to-permanent loans, i.e., construction financing that will be permanently financed by the same creditor. Typically, such loans have a short construction period, during which payments are made of interest only, followed by a fully amortizing permanent period, often an additional 30 years. Because of this hybrid form, the loans do not appear to qualify for the temporary financing exemption, nor would they be qualified mortgages because of the interest-only period and the fact that the entire loan term will often slightly exceed 30 years. However, such loans may have significant consumer benefits because they avoid the inconvenience and expense of a second closing, and also avoid the risk that permanent financing will be unavailable when the construction loan is due.

The Bureau notes that existing § 1026.17(c)(6)(ii) provides that construction-to-permanent loans may be disclosed as either a single transaction or as multiple transactions at the creditor's option. Consistent with that provision, the Bureau is using its adjustment and exception authority to allow the construction phase of a construction-to-permanent loan to be exempt from the ability-to-repay requirements as a temporary loan; however, the permanent phase of the loan is subject to § 1026.43. Because the permanent phase is subject to § 1026.43, it may be a qualified mortgage if it satisfies the appropriate requirements.

As amended by the Dodd-Frank Act, TILA section 105(a), 15 U.S.C. 1604(a), directs the Bureau to prescribe regulations to carry out the purposes of TILA, and provides that such regulations may contain additional requirements, classifications, differentiations, or other provisions, and may provide for such adjustments and exceptions for all or any class of transactions that the Bureau judges are necessary or proper to effectuate the purposes of TILA, to prevent circumvention or evasion thereof, or to facilitate compliance therewith. The main purpose of section 129C is articulated in section 129B(a)(2)—“to assure that consumers are offered and receive residential mortgage loans on terms that reasonably reflect their ability to repay the loans and that are not unfair, deceptive or abusive.” Creditors' ability to continue originating construction-to-permanent loans in a cost effective manner will help to ensure that consumers are offered and receive loans on terms that reasonably reflect their ability to repay. The construction-to-permanent product avoids the possibility of a consumer being unable to repay a construction loan, because the permanent financing is already part of the contract. Without the ability to treat the permanent financing as a qualified mortgage, and the construction phase as exempt, it is not clear how many creditors would continue to offer such loans, especially in the short term. In addition, consumers will benefit from the potentially lower costs associated with qualified mortgages. In addition to effectuating the purpose of ensuring ability to repay, this exemption will greatly facilitate compliance for creditors providing this product.

Proposed comment 43(a)(3)-1 provided that, where a temporary or “bridge” loan is renewable, the loan term does not include any additional period of time that could result from a renewal provision. The Bureau is adding comment 43(a)(3)-2 to make clear that if a construction-to-permanent loan is treated as multiple transactions in regard to compliance with the ability-to-repay requirements, and the initial one-year construction phase is renewable, the loan term of the construction phase does not include any additional period of time that could result from a renewal of that construction phase that is one year or less in duration. Comment 43(a)(3)-2 also makes clear that if the construction phase of a construction-to-permanent loan is treated as exempt, the permanent financing phase may be a qualified mortgage if it meets the appropriate requirements.

Accordingly, § 1026.43(a)(3)(iii) and comment 43(a)(3)-2 are added to this final rule.

43(b) Definitions

43(b)(1)

The definition of “covered transaction” restates the scope of the rule, discussed above, which implements the statutory term “residential mortgage loan” defined at TILA § 103(cc)(5). The Bureau did not receive any comments specifically on this provision and is adopting it as proposed in § 1026.43(b)(1). For clarity, the Bureau has added comment 43(b)(1)-1 explaining that the term “covered transaction” restates the scope of the rule as described in § 1026.43(a).

43(b)(2)

TILA section 129C(a)(3) requires that “[a] creditor shall determine the ability of the consumer to repay using a payment schedule that fully amortizes the loan over the term of the loan.” In implementing this provision, the proposed rule defined a “fully amortizing payment” as “a periodic payment of principal and interest that will fully repay the loan amount over the loan term.” The term “fully amortizing payment” is used in the general “payment calculation” provision in § 1026.43(c)(5)(i)(B), which requires the use of “[m]onthly, fully amortizing payments that are substantially equal.” The Bureau has determined that the definition of “fully amortizing payment” enables accurate implementation of the payment calculation process envisioned by the statute, and no comments focused on or questioned this definition. Accordingly, § 1026.43(b)(2) is adopted as proposed.

43(b)(3)

TILA section 129C(a)(6)(D) provides that, for purposes of making the repayment ability determination required under TILA section 129C(a), the creditor must calculate the monthly payment on the mortgage obligation based on several assumptions, including that the monthly payment be calculated using the fully indexed rate at the time of loan closing, without considering the introductory rate. See TILA section 129C(a)(6)(D)(iii). TILA section 129C(a)(7) defines the term “fully indexed rate” as “the index rate prevailing on a residential mortgage loan at the time the loan is made plus the margin that will apply after the expiration of any introductory interest rates.”

The term “fully indexed rate” appeared in proposed § 226.43(c)(5), which implemented TILA section 129C(a)(6)(D)(iii) and provided the payment calculation rules for covered transactions. The term also appeared in proposed § 226.43(d)(5), which provided special rules for creditors that refinance a consumer from a non-standard mortgage to a standard mortgage.

Proposed § 226.43(b)(3) defined the term “fully indexed rate” as “the interest rate calculated using the index or formula at the time of consummation and the maximum margin that can apply at any time during the loan term.” This proposed definition was consistent with the statutory language of TILA sections 129C(a)(6)(D)(iii) and 129C(a)(7), but revised certain text to provide clarity. First, for consistency with current Regulation Z and to facilitate compliance, the proposal replaced the phrases “at the time of the loan closing” in TILA section 129C(a)(6)(D)(iii) and “at the time the loan is made” in TILA section 129C(a)(7) with the phrase “at the time of consummation” for purposes of identifying the fully indexed rate. The Board interpreted these statutory phrases to have the same meaning as the phrase “at the time of consummation.” See current § 1026.2(a)(7), defining the term “consummation” for purposes of Regulation Z requirements as “the time that a consumer becomes contractually obligated on a credit transaction.”

In requiring that the fully indexed rate be determined using the specified index at consummation, the Board was concerned that the possible existence of loans that use more than one index could complicate this determination. Given the increasing relevance of market indices, the Board solicited comment on whether loan products currently exist that base the interest rate on a specific index at consummation, but then base subsequent rate adjustments on a different index, and whether further guidance addressing how to calculate the fully indexed rate for such loan products would be needed.

The proposed rule interpreted the statutory reference to the margin that will apply “after the expiration of any introductory interest rates” as a reference to the maximum margin that can apply “at any time during the loan term.” The Bureau agrees with this interpretation, because the statutory use of the plural “rates” modified by the all-inclusive term “any” clearly indicates not only that something more than the initial introductory rate is meant, but that “any” preliminary rate should be disregarded. In addition, the statutory term itself, “fully indexed rate,” appears to require such a reading. Referencing the entire loan term as the relevant period of time during which the creditor must identify the maximum margin that can occur under the loan makes the phrase “after the expiration of any introductory interest rates” unnecessary and allows for simplicity and consistency with new TILA section 103(bb), the high cost mortgage provision.

Because the proposal required that the creditor use the “maximum” margin that can apply when determining the fully indexed rate, the creditor would be required to take into account the largest margin that could apply under the terms of the legal obligation. The approach of using the maximum margin that can apply at any time during the loan term is consistent with the statutory language contained in TILA section 103(bb), as amended by section 1431 of the Dodd-Frank Act, which defines a high-cost mortgage. This statutory provision provides that, for purposes of the definition of a “high-cost mortgage” under HOEPA, for a mortgage with an interest rate that varies solely in accordance with an index, the annual percentage rate must be based on “the interest rate determined by adding the index rate in effect on the date of consummation of the transaction to the maximum margin permitted at any time during the loan agreement.” Furthermore, although the Board was not aware of any current loan products that possess more than one margin that may apply over the loan term, the Board proposed this clarification to address the possibility that creditors may create products that permit different margins to take effect at different points throughout the loan term. The proposal solicited comment on this approach.

The proposed definition of “fully indexed rate” was also generally consistent with the definition of “fully indexed rate” as used in the MDIA Interim Final Rule, and with the Federal banking agencies' use of the term “fully indexed rate” in the 2006 Nontraditional Mortgage Guidance and 2007 Subprime Mortgage Statement.

Proposed comment 43(b)(3)-1 noted that in some adjustable-rate transactions, creditors may set an initial interest rate that is not determined by the index or formula used to make later interest rate adjustments. This proposed comment explained that this initial rate charged to consumers will sometimes be lower than the rate would be if it were calculated using the index or formula at consummation (i.e., a “discounted rate”); in some cases, this initial rate may be higher (i.e., a “premium rate”). The proposed comment clarified that when determining the fully indexed rate where the initial interest rate is not determined using the index or formula for subsequent interest rate adjustments, the creditor must use the interest rate that would have applied had the creditor used such index or formula plus margin at the time of consummation. The proposed comment further clarified that this means, in determining the fully indexed rate, the creditor must not take into account any discounted or premium rate. (In addition, to facilitate compliance, this comment directed creditors to commentary that addresses payment calculations based on the greater of the fully indexed rate or “premium rate” for purposes of the repayment ability determination under proposed § 226.43(c)). See final rule § 1026.43(c)(5)(i)(A) and comment 43(c)(5)(i)-2.)

Proposed comment 43(b)(3)-1 differed from guidance on disclosure requirements in current comment 17(c)(1)-10.i, which provides that in cases where the initial interest rate is not calculated using the index or formula for later rate adjustments, the creditor should disclose a composite annual percentage rate that reflects both the initial rate and the fully indexed rate. The Board believed the different approach taken in proposed comment 43(b)(3)-1 was required by the statutory language and was appropriate in the present case where the purpose of the statute is to determine whether the consumer can repay the loan according to its terms, including any potential increases in required payments. TILA section 129B(a)(2), 15 U.S.C 1639b(a)(2).

Proposed comment 43(b)(3)-2 further clarified that if the contract provides for a delay in the implementation of changes in an index value or formula, the creditor need not use the index or formula in effect at consummation, and provides an illustrative example. This proposed comment was consistent with current guidance in Regulation Z regarding the use of the index value at the time of consummation where the contract provides for a delay. See comments 17(c)(1)-10.i and 18(s)(2)(iii)(C)-1, which address the fully indexed rate for purposes of disclosure requirements.

Proposed comment 43(b)(3)-3 explained that the creditor must determine the fully indexed rate without taking into account any periodic interest rate adjustment caps that may limit how quickly the fully indexed rate may be reached at any time during the loan term under the terms of the legal obligation. As the proposal noted, the guidance contained in proposed comment 43(b)(3)-3 differed from guidance contained in current comment 17(c)(1)-10.iii, which states that, when disclosing the annual percentage rate, creditors should give effect to periodic interest rate adjustment caps.

Nonetheless, the Board believed the approach in proposed comment 43(b)(3)-3 was consistent with, and required by, the statutory language that states that the fully indexed rate must be determined without considering any introductory rate and by using the margin that will apply after expiration of any introductory interest rates. See TILA section 129C(a)(6)(D)(iii) and (7). In addition, the Board noted that the proposed definition of fully indexed rate, and its use in the proposed payment calculation rules, was designed to assess whether the consumer has the ability to repay the loan according to its terms. TILA section 129B(a)(2), 15 U.S.C 1639b(a)(2). This purpose differs from the principal purpose of disclosure requirements, which is to help ensure that consumers avoid the uninformed use of credit. TILA section 102(a), 15 U.S.C. 1601(a). Furthermore, the guidance contained in proposed comment 43(b)(3)-3 was consistent with the Federal banking agencies' use of the term fully indexed rate in the 2006 Nontraditional Mortgage Guidance and 2007 Subprime Mortgage Statement.

Proposed comment 43(b)(3)-4 clarified that when determining the fully indexed rate, a creditor may choose, in its sole discretion, to take into account the lifetime maximum interest rate provided under the terms of the legal obligation. This comment explained, however, that where the creditor chooses to use the lifetime maximum interest rate, and the loan agreement provides a range for the maximum interest rate, the creditor must use the highest rate in that range as the maximum interest rate. In allowing creditors to use the lifetime maximum interest rate provided under the terms of the obligation, the Board was apparently interested in simplifying compliance and benefiting consumers by encouraging reasonable lifetime interest rate caps. In doing so, the Board was apparently reading its proposed definition of fully indexed rate to allow the maximum margin that can apply at any time during the loan term to refer to the maximum margin as determined at consummation. In other words, when the index value is determined at consummation, the maximum margin that can apply at any time during the loan term will be the difference between the lifetime interest rate cap and that index value. Consequently, adding the index value at consummation to that maximum margin, as required by the fully indexed rate definition, will yield the lifetime interest rate cap as the fully indexed rate.

Commenters generally did not focus specifically on the definition of “fully indexed rate” and associated commentary proposed by the Board, or provide examples of loans with more than one index or more than one margin. An organization representing state bank regulators supported the use of the maximum margin that can apply at any time during the loan term, suggesting that it would prevent evasion. (Some commenter groups did urge the Bureau to use its adjustment authority to require creditors to use a rate higher than the fully indexed rate in assessing a consumer's ability to repay; these comments are discussed below in the section-by-section analysis of § 1026.43(c)(5)(i)). The Bureau is adopting the rule and commentary largely as proposed, with some modifications for clarity. Specifically, the Bureau decided to include language in the definition that will make clear that the index used in determining the fully indexed rate is the index that will apply after the loan is recast, so that any index that might be used earlier in determining an initial or intermediate rate would not be used. This new language is included for clarification only, and does not change the intended meaning of the proposed definition.

In the proposed rule, the Board noted that the statutory construct of the payment calculation rules, and the requirement to calculate payments based on the fully indexed rate, apply to all loans that are subject to the ability-to-repay provisions, including loans that do not base the interest rate on an index and therefore, do not have a fully indexed rate. Specifically, the statute states that “[f]or purposes of making any determination under this subsection, a creditor shall calculate the monthly payment amount for principal and interest on any residential mortgage loan by assuming” several factors, including the fully indexed rate, as defined in the statute (emphasis added). See TILA section 129C(a)(6)(D). The statutory definition of “residential mortgage loan” includes loans with variable-rate features that are not based on an index or formula, such as step-rate mortgages. See TILA section 103(cc); see also proposed § 226.43(a), which addressed the proposal's scope, and proposed § 226.43(b)(1), which defined “covered transaction.” However, because step-rate mortgages do not have a fully indexed rate, it was unclear what interest rate the creditor should assume when calculating payment amounts for the purpose of determining the consumer's ability to repay the covered transaction.

As discussed above, the proposal interpreted the statutory requirement to use the “margin that can apply at any time after the expiration of any introductory interest rates” to mean that the creditor must use the “maximum margin that can apply at any time during the loan term” when determining the fully indexed rate. Accordingly, consistent with this approach, the proposal clarified in proposed comment 43(b)(3)-5 that where there is no fully indexed rate because the interest rate offered in the loan is not based on, and does not vary with, an index or formula, the creditor must use the maximum interest rate that may apply at any time during the loan term. Proposed comment 43(b)(3)-5 provided illustrative examples of how to determine the maximum interest rate for a step-rate and a fixed-rate mortgage.

The Board believed this approach was appropriate because the purpose of TILA section 129C is to require creditors to assess whether the consumer can repay the loan according to its terms, including any potential increases in required payments. TILA section 129B(a)(2), 15 U.S.C 1639b(a)(2). Requiring creditors to use the maximum interest rate would help to ensure that consumers could repay their loans. However, the Board was also concerned that by requiring creditors to use the maximum interest rate in a step-rate mortgage, the monthly payments used to determine the consumer's repayment ability might be overstated and potentially restrict credit availability. Therefore, the Board solicited comment on this approach, and whether authority under TILA sections 105(a) and 129B(e) should be used to provide an exception for step-rate mortgages, possibly requiring creditors to use the maximum interest rate that occurs in only the first 5 or 10 years, or some other appropriate time horizon.

The Bureau received few comments on the use of the maximum interest rate that may apply at any time during the loan term for step-rate mortgages. A consumer group and a regulatory reform group stated that this method was better and more protective of consumers than using a seven- or ten-year horizon. An organization representing state bank regulators suggested that the Bureau use a five-year horizon, provided that the loan has limits on later rate increases. An industry trade association suggested that the maximum rate only be applied to the balance remaining when that maximum rate is reached.

The Bureau believes that the proposal's method of using the maximum interest rate that may apply at any time during the loan term for step-rate mortgages is appropriate. This approach most closely approximates the statutorily required fully indexed rate because it employs the highest rate ascertainable at consummation, as does the fully indexed rate, and it applies that rate to the entire original principal of the loan, as the calculation in § 1026.43(c)(5)(i) does with the fully indexed rate. In addition, this method most effectively ensures the consumer's ability to repay the loan.

For the reasons stated above, § 1026.43(b)(3) is adopted substantially as proposed, with the clarification discussed above specifying that the index used in determining the fully indexed rate is the index that will apply after the loan is recast. Issues regarding the use of the fully indexed rate in the payment calculations required by § 1026.43(c)(5) are discussed in the section-by-section analysis of that section below.

43(b)(4)

The Dodd-Frank Act added TILA section 129C(a)(6)(D)(ii)(II), which provides that a creditor making a balloon-payment loan with an APR at or above certain thresholds must determine ability to repay “using the contract's repayment schedule.” The thresholds required by the statute are 1.5 or more percentage points above the average prime offer rate (APOR) for a comparable transaction for a first lien, and 3.5 or more percentage points above APOR for a subordinate lien. These thresholds are the same as those used in the Board's 2008 HOEPA Final Rule to designate a new category of “higher-priced mortgage loans” (HPMLs), which was amended by the Board's 2011 Jumbo Loans Escrows Final Rule to include a separate threshold for jumbo loans for purposes of certain escrows requirements. Implementing these thresholds for use with the payment underwriting determination for balloon-payment mortgages, the proposed rule defined a “higher-priced covered transaction” as one in which the annual percentage rate (APR) “exceeds the average prime offer rate (APOR) for a comparable transaction as of the date the interest rate is set by 1.5 or more percentage points for a first-lien covered transaction, or by 3.5 or more percentage points for a subordinate-lien covered transaction.” As explained further below and provided for in the statute, the designation of certain covered transactions as higher-priced affects the ability-to-repay determination for balloon-payment mortgages, and requires that those higher-priced transactions be analyzed using the loan contract's full repayment schedule, including the balloon payment. § 1026.43(c)(5)(ii)(A)(2).

Proposed comment 43(b)(4)-1 provided guidance on the term “average prime offer rate.” Proposed comment 43(b)(4)-2 stated that the table of average prime offer rates published by the Board would indicate how to identify the comparable transaction for a higher-priced covered transaction. Proposed comment 43(b)(4)-3 clarified that a transaction's annual percentage rate is compared to the average prime offer rate as of the date the transaction's interest rate is set (or “locked”) before consummation. This proposed comment also explained that sometimes a creditor sets the interest rate initially and then resets it at a different level before consummation, and clarified that in these cases, the creditor should use the last date the interest rate is set before consummation.

The Board explained in its proposed rule that it believed the ability-to-repay requirements for higher-priced balloon-payment loans was meant to apply to the subprime market, but that use of the annual percentage rate could lead to prime loans being exposed to this test. For this reason, the Board was concerned that the statutory formula for a higher-priced covered transaction might be over-inclusive. Accordingly, the Board solicited comment on whether the “transaction coverage rate” (TCR) should be used for this determination, instead of the annual percentage rate. 76 FR 27412. The TCR had previously been proposed in conjunction with a more inclusive version of the APR, in order to avoid having the more inclusive, hence higher, APRs trigger certain requirements unnecessarily. The TCR includes fewer charges, and the Board's 2011 Escrows Proposal proposed to use it in the threshold test for determining application of those requirements. 76 FR 11598, 11626-11627 (Mar. 2, 2011).

The only comment substantively discussing the possible substitution of the TCR for the APR was strongly opposed to the idea, stating that it would create unnecessary compliance difficulty and costs. The Bureau has determined that possible transition to a TCR standard will implicate several rules and is not appropriate at the present time. However, the issue will be considered further as part of the Bureau's TILA/RESPA rulemaking. See 77 FR 51116, 51126 (Aug. 23, 2012).

The Board also solicited comment on whether or not to provide a higher threshold for jumbo balloon-payment mortgages or for balloon-payment mortgages secured by a residence that is not the consumer's principal dwelling, e.g., a vacation home. 76 FR 27412. The Board requested this information due to its belief that higher interest rates charged for these loans might render them unavailable without the adjustment. The margin above APOR suggested for first-lien jumbo balloon-payment mortgages was 2.5 percentage points.

Two industry commenters supported the higher threshold for jumbo loans, arguing that the current thresholds would interfere with credit accessibility. One of these commenters also stated that the higher threshold should be available for all balloon-payment mortgages. No commenters discussed the non-principal-dwelling threshold.

Many other commenters objected strongly to the statutory requirement, implemented in the proposed rule, that the balloon payment be considered in applying the ability-to-repay requirements to higher-priced covered transaction balloon-payment mortgages. These industry commenters felt that the percentage point thresholds were too low, and that many loans currently being made would become unavailable. They did not, however, submit sufficient data to help the Bureau assess these claims. Other commenters, including several consumer protection advocacy organizations, argued that the higher-priced rule would be helpful in ensuring consumers' ability to repay their loans.

The Bureau has evaluated the proposed definition of “higher-priced covered transaction” not only in relation to its use in the payment determination for balloon-payment mortgages, but also in the light of its application in other provisions of the final rule. For example, as discussed below, the final rule varies the strength of the presumption of compliance for qualified mortgages. A qualified mortgage designated as a higher-priced covered transaction will be presumed to comply with the ability-to repay-provision at § 1026.43(c)(1), but will not qualify for the safe harbor provision. See § 1026.43(e)(1)(ii) and (i).

Specifically, the Bureau has considered whether to adopt a different threshold to define high price mortgage loans for jumbo loans than for other loans. The Bureau notes that the Board expressly addressed this issue in its 2008 HOEPA Final Rule and concluded not to do so. The Board explained that although prime jumbo loans have always had somewhat higher rates than prime conforming loans, the spread has been quite volatile. The Board concluded that it was sounder to err on the side of being over-inclusive than to set a higher threshold for jumbo loans and potentially fail to include subprime jumbo loans. The Bureau is persuaded by the Board's reasoning.

The Bureau recognizes that in the Dodd-Frank Act Congress, in requiring creditors to establish escrows accounts for certain transactions and in requiring appraisals for certain transactions based upon the interest rate of the transactions, did establish a separate threshold for jumbo loans. The Bureau is implementing that separate threshold in its 2013 Escrows Final Rule which is being issued contemporaneously with this final rule. However, the Bureau also notes that in the ability-to-repay provision of the Dodd-Frank Act, Congress mandated underwriting rules for balloon-payment mortgages which vary based upon the pricing of the loan, and in doing so Congress followed the thresholds adopted by the Board in its 2008 HOEPA Final Rule and did not add a separate threshold for jumbo loans. The fact that the Act uses the Board's criteria in the ability to repay context lends further support to the Bureau's decision to use those criteria as well in defining higher-priced loans under the final rule.

Accordingly, the Bureau is not providing for a higher threshold for jumbo or non-principal dwelling balloon-payment mortgages at this time. In regard to the possibility of a higher threshold for non-principal dwellings such as vacation homes, the Bureau understands that such products have historically been considered to be at higher risk of default than loans on principal dwellings. Therefore, any difference in rates is likely driven by the repayment risk associated with the product, and a rule meant to ensure a consumer's ability to repay the loan should not provide an exemption under these circumstances. And further, the Bureau did not receive and is not aware of any data supporting such an exemption.

The Bureau does not believe that these decisions regarding jumbo and non-principal-dwelling balloon-payment mortgages are likely to create any credit accessibility problems. In this final rule at § 1026.43(f), the Bureau is adopting a much wider area in which institutions that provide credit in rural or underserved areas may originate qualified mortgages that are balloon-payment loans than did the proposed rule. Because these are the areas in which balloon-payment loans are considered necessary to preserve access to credit, and higher-priced balloon-payment mortgages in these areas can meet the criteria for a qualified mortgage and thus will not have to include the balloon payment in the ability-to-repay evaluation, access to necessary balloon-payment mortgages will not be reduced.

Accordingly, § 1026.43(b)(4) is adopted as proposed. The associated commentary is amended with revisions to update information and citations.

43(b)(5)

The proposed rule defined “loan amount” as “the principal amount the consumer will borrow as reflected in the promissory note or loan contract.” This definition implemented the statutory language requiring that the monthly payment be calculated assuming that “the loan proceeds are fully disbursed on the date of consummation of the loan.” Dodd-Frank Act section 1411(a)(2), TILA section 129C(a)(6)(D)(i). The term “loan amount” was used in the proposed definition of “fully amortizing payment” in § 226.43(b)(2), which was then used in the general “payment calculation” at § 226.43(c)(5)(i)(B). The payment calculation required the use of payments that pay off the loan amount over the actual term of the loan.

The statute further requires that creditors assume that the loan amount is “fully disbursed on the date of consummation of the loan.” See TILA Section 129C(a)(6)(D)(i). The Board recognized that some loans do not disburse the entire loan amount to the consumer at consummation, but may, for example, provide for multiple disbursements up to an amount stated in the loan agreement. See current § 1026.17(c)(6), discussing multiple-advance loans and comment 17(c)(6)-2 and -3. In these cases, the loan amount, as reflected in the promissory note or loan contract, does not accurately reflect the amount disbursed at consummation. Thus, to reflect the statutory requirement that the creditor assume the loan amount is fully disbursed at consummation, the Board clarified that creditors must use the entire loan amount as reflected in the loan contract or promissory note, even where the loan amount is not fully disbursed at consummation. Proposed comment 43(b)(5)-1 provided an illustrative example and stated that generally, creditors should rely on § 1026.17(c)(6) and associated commentary regarding treatment of multiple-advance and construction loans that would be covered by the ability-to-repay requirements (i.e., loans with a term greater than 12 months). See § 1026.43(a)(3) discussing scope of coverage and term length.

The Board specifically solicited comment on whether further guidance was needed regarding determination of the loan amount for loans with multiple disbursements. The Bureau did not receive comments on the definition of “loan amount” or its application to loans with multiple disbursements. The Bureau believes that the loan amount for multiple disbursement loans that are covered transactions must be determined assuming that “the loan proceeds are fully disbursed on the date of consummation of the loan” as required by the statute and the rule, and explained in comment 43(b)(5)-1.

Accordingly, the Bureau is adopting § 1026.43(b)(5) and associated commentary as proposed.

43(b)(6)

The interchangeable phrases “loan term” and “term of the loan” appear in the ability-to-repay and qualified mortgage provisions of TILA, with no definition. See TILA section 129C(c)(3), 129C(a)(6)(D)(ii), 129C(b)(2)(A)(iv) and (v); 15 U.S.C. 1639c(c)(3), 1639c(a)(6)(D)(ii), 1639c(b)(2)(A)(iv) and (v). The proposed rule defined “loan term” as “the period of time to repay the obligation in full.” Proposed comment 43(b)(6)-1 clarified that the loan term is the period of time it takes to repay the loan amount in full, and provided an example. The term is used in § 1026.43(b)(2), the “fully amortizing payment” definition, which is then used in § 1026.43(c)(5)(i), the payment calculation general rule. It is also used in the qualified mortgage payment calculation at § 1026.43(e)(2)(iv). The Bureau did not receive any comments on this definition, and considers it to be an accurate and appropriate implementation of the statutory language. Accordingly, proposed § 1026.43(b)(6) is adopted as proposed.

43(b)(7)

The definition of “maximum loan amount” and the calculation for which it is used implement the requirements regarding negative amortization loans in new TILA section 129C(a)(6)(C) and (D). The statute requires that a creditor “take into consideration any balance increase that may accrue from any negative amortization provision.”

The “maximum loan amount” is defined in the proposed rule as including the loan balance and any amount that will be added to the balance as a result of negative amortization assuming the consumer makes only minimum payments and the maximum interest rate is reached at the earliest possible time. The “maximum loan amount” is used to determine a consumer's ability to repay for negative amortization loans under § 1026.43(c)(5)(ii)(C) by taking into account any loan balance increase that may occur as a result of negative amortization. The term “maximum loan amount” is also used for negative amortization loans in the “refinancing of non-standard mortgages” provision, at § 1026.43(d)(5)(i)(C)(3). The proposed rule included commentary on how to calculate the maximum loan amount, with examples. See comment 43(b)(7)-1 through -3.

The Bureau did not receive any comments on this definition and considers it to be an accurate and appropriate implementation of the statute. Accordingly, § 1026.43(b)(7) and associated commentary are adopted as proposed.

43(b)(8)

TILA section 129C(a)(1) and (3), as added by section 1411 of the Dodd-Frank Act, requires creditors to consider and verify mortgage-related obligations as part of the ability-to-repay determination “according to [the loan's] terms, and all applicable taxes, insurance (including mortgage guarantee insurance), and assessments.” TILA section 129C(a)(2) provides that consumers must have “a reasonable ability to repay the combined payments of all loans on the same dwelling according to the terms of those loans and all applicable taxes, insurance (including mortgage guarantee insurance), and assessments.” Although the Dodd-Frank Act did not establish or define a single, collective term, the foregoing requirements recite ongoing obligations that are substantially similar to the definition of “mortgage-related obligation” used elsewhere in Regulation Z. Section 1026.34(a)(4)(i), which was added by the 2008 HOEPA Final Rule, defines mortgage-related obligations as expected property taxes, premiums for mortgage-related insurance required by the creditor as set forth in the relevant escrow provisions of Regulation Z, and similar expenses. Comment 34(a)(4)(i)-1 clarifies that, for purposes of § 1026.34(a)(4)(i), similar expenses include homeowners association dues and condominium or cooperative fees. Section 1026.35(b)(3)(i), which addresses escrows, states that “premiums for mortgage-related insurance required by the creditor, [include] insurance against loss of or damage to property, or against liability arising out of the ownership or use of the property, or insurance protecting the creditor against the consumer's default or other credit loss.”

Under the Board's proposed § 226.43(b)(8), “mortgage-related obligations” was defined to mean property taxes; mortgage related insurance premiums required by the creditor as set forth in proposed § 226.45(b)(1); homeowners association, condominium, and cooperative fees; ground rent or leasehold payments; and special assessments. The Board's proposed definition was substantially similar to the definition under § 1026.34(a)(4)(i), with three clarifications. First, the proposed definition of mortgage-related obligations would have included a reference to ground rent or leasehold payments, which are payments made to the real property owner or leaseholder for use of the real property. Second, the proposed definition would have included a reference to “special assessments.” Proposed comment 43(b)(8)-1 would have clarified that special assessments include, for example, assessments that are imposed on the consumer at or before consummation, such as a one-time homeowners association fee that will not be paid by the consumer in full at or before consummation. Third, mortgage-related obligations would have referenced proposed § 226.45(b)(1), where the Board proposed to recodify the existing escrow requirement for higher-priced mortgage loans, to include mortgage-related insurance premiums required by the creditor, such as insurance against loss of or damage to property, or against liability arising out of the ownership or use of the property, or insurance protecting the creditor against the consumer's default or other credit loss. The Board solicited comment on how to address any issues that may arise in connection with homeowners association transfer fees and costs associated with loans for energy efficient improvements.

Proposed comment 43(b)(8)-1 would have clarified further that mortgage-related obligations include mortgage-related insurance premiums only if required by the creditor. This comment would have explained that the creditor need not include premiums for mortgage-related insurance that the creditor does not require, such as earthquake insurance or credit insurance, or fees for optional debt suspension and debt cancellation agreements. To facilitate compliance, this comment would have referred to commentary associated with proposed § 226.43(c)(2)(v), which sets forth the requirement to take into account any mortgage-related obligations for purposes of the repayment ability determination required under proposed § 226.43(c).

Industry commenters and consumer advocates generally supported the Board's proposed definition of mortgage-related obligations. One industry commenter opposed including community transfer fees, which are deed-based fees imposed upon the transfer of the property. This commenter was concerned that subjecting these fees to Federal law might affect existing contracts, deeds, and covenants related to these fees, which are subject to State and local regulation, as well as common law regarding the transfer of real property. The commenter also asked that special assessments not fall under the definition of mortgage-related obligations. The commenter recommended that, if special assessments are included, creditors be required to consider only current special assessments, not future special assessments. The commenter noted that, while common assessments should be included in the definition of mortgage-related obligations, the Bureau should provide guidance to creditors on the substance of questionnaires seeking information from third parties about mortgage-related obligations.

Certain consumer advocates suggested that voluntary insurance premiums be included in the definition of mortgage-related obligations. One consumer advocate explained that premiums such as these are technically voluntary, but many consumers believe them to be required, or have difficulty cancelling them if they choose to cancel them. Community advocates and several industry commenters also recommended that homeowners association dues, and similar charges, be included in the definition of mortgage-related obligations. They argued that such a requirement would further transparency in the mortgage loan origination process and would help ensure that consumers receive only credit they can reasonably expect to repay.

For the reasons discussed below, the Bureau concludes that property taxes, certain insurance premiums required by the creditor, obligations to community governance associations, such as cooperative, condominium, and homeowners associations, ground rent, and lease payments should be included in the definition of mortgage-related obligations. These obligations are incurred in connection with the mortgage loan transaction but are in addition to the obligation to repay principal and interest. Thus, the cost of these obligations should be considered with the obligation to repay principal and interest for purposes of determining a consumer's ability to repay. Further, the Bureau believes that the word `assessments' in TILA section 129C is most appropriately interpreted to refer to all obligations imposed on consumers in connection with ownership of the dwelling or real property, such as ground rent, lease payments, and, as discussed in detail below, obligations to community governance associations, whether denominated as association dues, special assessments, or otherwise. While the provision adopted by the Bureau is substantially similar to the provision proposed, the Bureau was persuaded by the comment letters that additional clarity and guidance is required. The Bureau is especially sensitive to the fact that many of the loans that will be subject to the ability-to-repay rules may be made by small institutions, which are often unable to devote substantial resources to analysis of regulatory compliance.

To address the concerns and feedback raised in the comment letters, the Bureau has revised § 1026.43(b)(8) and related commentary in two ways. First, the language of § 1026.43(b)(8) is being modified to add additional clarity. As adopted, § 1026.43(b)(8) refers to premiums and similar charges identified in § 1026.4(b)(5), (7), (8), or (10), if required by the creditor, instead of the proposed language, which referred to “mortgage-related insurance.” Second, the commentary is being significantly expanded to provide additional clarification and guidance.

As adopted, § 1026.43(b)(8) defines “mortgage-related obligations” to mean property taxes; premiums and similar charges identified in § 1026.4(b)(5), (7), (8), or (10) that are required by the creditor; fees and special assessments imposed by a condominium, cooperative, or homeowners association; ground rent; and leasehold payments. As proposed, comment 43(b)(8)-1 discussed all components of the proposed definition. To provide further clarity, the final rule splits the content of proposed comment 43(b)(8)-1 into four separate comments, each of which provides additional guidance. As adopted by the Bureau, comment 43(b)(8)-1 contains general guidance and a cross-reference to § 1026.43(c)(2)(v), which contains the requirement to take into account any mortgage-related obligations for purposes of determining a consumer's ability to repay.

The multitude of requests for additional guidance and clarification suggests that additional clarification of the meaning of “property tax” is needed. Comment 43(b)(8)-2 further clarifies that § 1026.43(b)(8) includes obligations that are functionally equivalent to property taxes, even if such obligations follow a different naming convention. For example, governments may establish independent districts with the authority to impose recurring levies on properties within the district to fund a special purpose, such as a local development bond district, water district, or other public purpose. These recurring levies may have a variety of names, such as taxes, assessments, or surcharges. Comment 43(b)(8)-2 clarifies that obligations such as these are property taxes based on the character of the obligation, as opposed to the name of the obligation, and therefore are mortgage-related obligations.

Most comments supported the inclusion of insurance premiums in the ability-to-repay determination. However, the Bureau believes that some modifications to the proposed “mortgage-related insurance premium” language are appropriate. The Bureau is persuaded that additional clarification and guidance is important, and the Bureau is especially sensitive to concerns related to regulatory complexity. The Bureau has determined that the proposed language should be clarified by revising the text to refer to the current definition of finance charge under § 1026.4. The components of the finance charge are long-standing parts of Regulation Z. Explicitly referring to existing language should facilitate compliance. Therefore, § 1026.43(b)(8) defines mortgage-related obligations to include all premiums or other charges related to protection against a consumer's default, credit loss, collateral loss, or similar loss as identified in § 1026.4(b)(5), (7), (8), or (10) except, as explained above, those premiums or charges that that are not required by the creditor. Comment 43(b)(8)-3 also contains illustrative examples of this definition. For example, if Federal law requires flood insurance to be obtained in connection with the mortgage loan, the flood insurance premium is a mortgage-related obligation for purposes of § 1026.43(b)(8).

Several commenters stated that insurance premiums and similar charges should be included in the determination even if the creditor does not require them in connection with the loan transaction. The Bureau has carefully considered these arguments, but has determined that insurance premiums and similar charges should not be considered mortgage-related obligations if such premiums and charges are not required by the creditor and instead have been voluntarily purchased by the consumer. The Bureau acknowledges that obligations such as these are usually paid from a consumer's monthly income and, in a sense, affect a consumer's ability to repay. But the consumer is free to cancel recurring obligations such as these at any time, provided they are truly voluntary. Thus, they are not “obligations” in the sense required by section 129C(a)(3) of TILA. The Bureau shares the concern raised by several commenters that unscrupulous creditors may mislead consumers into believing that these charges are not optional or cannot be cancelled. However, the Bureau does not believe that altering the ability-to-repay calculation for all is the appropriate method for combatting the harmful actions of a few. The Bureau believes that the better course of action is to exclude such premiums and charges from the definition of mortgage-related obligations only if they are truly voluntary, and is confident that violations of this requirement will be apparent in specific cases from the facts. Also, in the scenarios described by commenters where consumers are misled into believing that such charges are required, the premium or charge would not be voluntary for purposes of the definition of finance charge under § 1026.4(d), and would therefore be a mortgage-related obligation for the purposes of § 1026.43(b)(8). Therefore, comment 43(b)(8)-3 clarifies that insurance premiums and similar charges identified in § 1026.4(b)(5), (7), (8), or (10) that are not required by the creditor and that the consumer purchases voluntarily are not mortgage-related obligations for purposes of § 1026.43(b)(8). For example, if a creditor does not require earthquake insurance to be obtained in connection with the mortgage loan, but the consumer voluntarily chooses to purchase such insurance, the earthquake insurance premium is not a mortgage-related obligation for purposes of § 1026.43(b)(8). Or, if a creditor requires a minimum amount of coverage for homeowners' insurance and the consumer voluntarily chooses to purchase a more comprehensive amount of coverage, the portion of the premium allocated to the minimum coverage is a mortgage-related obligation for the purposes of § 1026.43(b)(8), while the portion of the premium allocated to the more comprehensive coverage voluntarily purchased by the consumer is not a mortgage-related obligation for the purposes of § 1026.43(b)(8). However, if the consumer purchases non-required insurance or similar coverage at consummation without having requested the specific non-required insurance or similar coverage and without having agreed to the premium or charge for the specific non-required insurance or similar coverage prior to consummation, the premium or charge is not voluntary for purposes of § 1026.43(b)(8) and is a mortgage-related obligation.

Several commenters supported the inclusion of mortgage insurance in the definition of mortgage-related obligations. The Bureau also has received several informal requests for guidance regarding the meaning of the term “mortgage insurance” in the context of certain disclosures required by Regulation Z. The Bureau has decided to clarify this issue with respect to the requirements of § 1026.43. Thus, comment 43(b)(8)-4 clarifies that § 1026.43(b)(8) includes all premiums or similar charges for coverage protecting the creditor against the consumer's default or other credit loss in the determination of mortgage-related obligations, whether denominated as mortgage insurance, guarantee insurance, or otherwise, as determined according to applicable State or Federal law. For example, monthly “private mortgage insurance” payments paid to a non-governmental entity, annual “guarantee fee” payments required by a Federal housing program, and a quarterly “mortgage insurance” payment paid to a State agency administering a housing program are all mortgage-related obligations for purposes of § 1026.43(b)(8). Comment 43(b)(8)-4 also clarifies that § 1026.43(b)(8) includes these charges in the definition of mortgage-related obligations if the creditor requires the consumer to pay them, even if the consumer is not legally obligated to pay the charges under the terms of the insurance program. Comment 43(b)(8)-4 also contains several other illustrative examples.

Several comment letters stressed the importance of including homeowners association dues and similar obligations in the determination of ability to repay. These letters noted that, during the subprime crisis, the failure to account for these obligations led to many consumers qualifying for mortgage loans that they could not actually afford. The Bureau agrees with these assessments. Recurring financial obligations payable to community governance associations, such as homeowners association dues, should be taken into consideration in determining whether a consumer has the ability to repay the obligation. While several comment letters identified practical problems with including obligations such as these in the calculation, these issues stemmed from difficulties that may arise in calculating, estimating, or verifying these obligations, rather than whether the obligations should be included in the ability-to-repay calculation. Based on this feedback, § 1026.43(b)(8) includes obligations to a homeowners association, condominium association, or condominium association in the determination of mortgage-related obligations. The Bureau has addressed the concerns related to difficulties in calculating, estimating, or verifying such obligations in the commentary to § 1026.43(c)(2)(v) and (c)(3).

One comment letter focused extensively on community transfer fees, which are deed-based fees imposed upon the transfer of the property. The Bureau recognizes that this topic is complex and is often the subject of special requirements imposed at the State and local level. However, the Bureau does not believe that the requirements of § 1026.43 implicate these complex issues. The narrow question is whether such obligations should be considered mortgage-related obligations for purposes of determining the consumer's ability to repay. The Bureau agrees with the argument, advanced by several commenters, that the entirety of the consumer's ongoing obligations should be included in the determination. A responsible determination of the consumer's ability to repay requires an accounting of such obligations, whether the purpose of the obligation is to satisfy the payment of a community transfer fee or traditional homeowners association dues. As with other obligations owed to condominium, cooperative, or homeowners associations discussed above, the Bureau believes that the practical problems with these obligations relate to when such obligations should be included in the determination of the consumer's ability to repay, rather than whether the obligations should be considered mortgage-related obligations. Therefore, the Bureau has addressed the concerns related to these obligations in the commentary to § 1026.43(c)(2)(v) and (c)(3).

In response to the request for feedback in the 2011 ATR Proposal, several commenters addressed the proposed treatment of special assessments. Unlike community transfer fees, which are generally identified in the deed or master community plan, creditors may encounter difficulty determining whether special assessments exist. However, as with similar charges discussed above, these concerns relate to determining the consumer's monthly payment for mortgage-related obligations, rather than whether these charges should be considered mortgage-related obligations. Special assessments may be significant and may affect the consumer's ability to repay a mortgage loan. Thus, the Bureau has concluded that special assessments should be included in the definition of mortgage-related obligations under § 1026.43(b)(8) and has addressed the concerns raised by commenters related to calculating, estimating, or verifying these obligations in the commentary to § 1026.43(c)(2)(v) and (c)(3).

New comment 43(b)(8)-5 explains that § 1026.43(b)(8) includes in the evaluation of mortgage-related obligations premiums and similar charges identified in § 1026.4(b)(5), (7), (8), or (10) that are required by the creditor. These premiums and similar charges are mortgage-related obligations regardless of whether the premium or similar charge is excluded from the finance charge pursuant to § 1026.4(d). For example, a premium for insurance against loss or damage to the property written in connection with the credit transaction is a premium identified in § 1026.4(b)(8). If this premium is required by the creditor, the premium is a mortgage-related obligation pursuant to § 1026.43(b)(8), regardless of whether the premium is excluded from the finance charge pursuant to § 1026.4(d)(2). Commenters did not request this guidance specifically, but the Bureau believes that this comment is needed to provide additional clarity.

43(b)(9)

TILA section 129C(b)(2)(C) generally defines “points and fees” for a qualified mortgage to have the same meaning as in TILA section 103(bb)(4), which defines points and fees for the purpose of determining whether a transaction exceeds the HOEPA points and fees threshold. Proposed § 226.43(b)(9) would have provided that “points and fees” has the same meaning as in § 226.32(b)(1). The Bureau adopts this provision as renumbered § 1026.43(b)(9).

43(b)(10)

Sections 1414, 1431, and 1432 of the Dodd-Frank Act amended TILA to restrict, and in many cases, prohibit a creditor from imposing prepayment penalties in dwelling-secured credit transactions. TILA does not, however, define the term “prepayment penalty.” In an effort to address comprehensively prepayment penalties in a fashion that eases compliance burden, as discussed above, the Bureau is defining prepayment penalty in § 1026.43(b)(10) by cross-referencing § 1026.32(b)(6). For a full discussion of the Bureau's approach to defining prepayment penalties, see § 1026.32(b)(6), its commentary, and the section-by-section analysis of those provisions above.

43(b)(11)

TILA in several instances uses the term “reset” to refer to the time at which the terms of a mortgage loan are adjusted, usually resulting in higher required payments. For example, TILA section 129C(a)(6)(E)(ii) states that a creditor that refinances a loan may, under certain conditions, “consider if the extension of new credit would prevent a likely default should the original mortgage reset and give such concerns a higher priority as an acceptable underwriting practice.” 15 U.S.C. 1639c(a)(6)(E)(ii). The legislative history further indicates that, for adjustable-rate mortgages with low, fixed introductory rates, Congress understood the term “reset” to mean the time at which low introductory rates convert to indexed rates, resulting in “significantly higher monthly payments for homeowners.”

Outreach conducted prior to issuance of the proposed rule indicated that the term “recast” is typically used in reference to the time at which fully amortizing payments are required for interest-only and negative amortization loans and that the term “reset” is more frequently used to indicate the time at which adjustable-rate mortgages with an introductory fixed rate convert to a variable rate. For simplicity and clarity, however, the Board proposed to use the term “recast” to cover the conversion to generally less favorable terms and higher payments not only for interest-only loans and negative amortization loans, but also for adjustable-rate mortgages.

Proposed § 226.43(b)(11) defined the term “recast,” which was used in two provisions of proposed § 226.43: (1) Proposed § 226.43(c)(5)(ii) regarding certain required payment calculations that creditors must consider in determining a consumer's ability to repay a covered transaction; and (2) proposed § 226.43(d) regarding payment calculations required for refinancings that are exempt from the ability-to-repay requirements in § 226.43(c).

Specifically, proposed § 226.43(b)(11) defined the term “recast” as follows: (1) For an adjustable-rate mortgage, as defined in § 1026.18(s)(7)(i), the expiration of the period during which payments based on the introductory interest rate are permitted under the terms of the legal obligation; (2) for an interest-only loan, as defined in § 1026.18(s)(7)(iv), the expiration of the period during which interest-only payments are permitted under the terms of the legal obligation; and (3) for a negative amortization loan, as defined in § 1026.18(s)(7)(v), the expiration of the period during which negatively amortizing payments are permitted under the terms of the legal obligation.

Proposed comment 43(b)(11)-1 explained that the date on which the “recast” occurs is the due date of the last monthly payment based on the introductory fixed rate, the last interest-only payment, or the last negatively amortizing payment, as applicable. Proposed comment 43(b)(11)-1 also provided an illustration showing how to determine the date of the recast.

Commenters did not focus specifically on the definition of “recast,” except that an association of State bank regulators agreed with the benefit of using a single term for the shift to higher payments for adjustable-rate, interest-only, and negative amortization loans.

The Bureau considers the proposed provision to be an accurate and appropriate implementation of the statute. Accordingly, the Bureau is adopting proposed § 226.43(b)(11) as proposed, in renumbered § 1026.43(b)(11).

43(b)(12)

New TILA section 129C(a)(2) provides that “if a creditor knows, or has reason to know, that 1 or more residential mortgage loans secured by the same dwelling will be made to the same consumer,” that creditor must make the ability-to-repay determination for “the combined payments of all loans on the same dwelling according to the terms of those loans and all applicable taxes, insurance (including mortgage guarantee insurance), and assessments.” This section, entitled “multiple loans,” follows the basic ability-to-repay requirements for a single loan, in new TILA section 129C(a)(1).

The proposed rule implemented the main requirement of the “multiple loans” provision by mandating in proposed § 226.43(c)(2)(iv) that a creditor, in making its ability-to-repay determination on the primary loan, take into account the payments on any “simultaneous loan” about which the creditor knows or has reason to know. “Simultaneous loan” was defined in proposed § 226.43(b)(12) as “another covered transaction or home equity line of credit subject to § 226.5b that will be secured by the same dwelling and made to the same consumer at or before consummation of the covered transaction.” Thus, although the statute referred only to closed-end “residential mortgage loans,” the Board proposed to expand the requirement to include consideration of simultaneous HELOCs. The proposed definition did not include pre-existing mortgage obligations, which would be considered as “current debt obligations” under § 1026.43(c)(2)(vi).

The Board chose to include HELOCs in the definition of “simultaneous loan” because it believed that new TILA section 129C(a)(2) was meant to help ensure that creditors account for the increased risk of consumer delinquency or default on the covered transaction where more than one loan secured by the same dwelling is originated concurrently. The Board believed that this increased risk would be present whether the other mortgage obligation was a closed-end credit obligation or a HELOC. For these reasons, and several others explained in detail below, the Board proposed to use its exception and adjustment authority under TILA section 105(a) to include HELOCs within the scope of new TILA section 129C(a)(2). 76 FR 27417-27418. Because one of the main reasons for including HELOCs was the likelihood of a consumer drawing on the credit line to provide the down payment in a purchase transaction, the Board solicited comment on whether this exception should be limited to purchase transactions.

TILA section 105(a), as amended by section 1100A of the Dodd-Frank Act, authorized the Board, and now the Bureau, to prescribe regulations to carry out the purposes of TILA and Regulation Z, to prevent circumvention or evasion, or to facilitate compliance. 15 U.S.C. 1604(a). The inclusion of HELOCs was further supported by the Board's authority under TILA section 129B(e) to condition terms, acts or practices relating to residential mortgage loans that the Board found necessary or proper to effectuate the purposes of TILA. 15 U.S.C. 1639b(e). One purpose of the statute is set forth in TILA section 129B(a)(2), which states that “[i]t is the purpose[] of * * * [S]ection 129C to assure that consumers are offered and receive residential mortgage loans on terms that reasonably reflect their ability to repay the loans.” 15 U.S.C. 1639b. For the reasons stated below, the Board believed that requiring creditors to consider simultaneous loans that are HELOCs for purposes of TILA section 129C(a)(2) would help to ensure that consumers are offered, and receive, loans on terms that reasonably reflect their ability to repay.

First, the Board proposed in § 226.43(c)(2)(vi) that the creditor must consider current debt obligations in determining a consumer's ability to repay a covered transaction. Consistent with current § 1026.34(a)(4), proposed § 226.43(c)(2)(vi) would not have distinguished between pre-existing closed-end and open-end mortgage obligations. The Board believed consistency required that it take the same approach when determining how to consider mortgage obligations that come into existence concurrently with a first-lien loan as would be taken for pre-existing mortgage obligations, whether the first-lien is a purchase or non-purchase transaction (i.e., refinancing). Including HELOCs in the proposed definition of “simultaneous loan” for purposes of TILA section 129C(a)(2) was also considered generally consistent with current comment 34(a)(4)-3, and the 2006 Nontraditional Mortgage Guidance regarding simultaneous second-lien loans.

Second, data indicate that where a subordinate loan is originated concurrently with a first-lien loan to provide some or all of the down payment (i.e., a “piggyback loan”), the default rate on the first-lien loan increases significantly, and in direct correlation to increasing combined loan-to-value ratios. The data does not distinguish between “piggyback loans” that are closed-end or open-end credit transactions, or between purchase and non-purchase transactions. However, empirical evidence demonstrates that approximately 60 percent of consumers who open a HELOC concurrently with a first-lien loan borrow against the line of credit at the time of origination, suggesting that in many cases the HELOC may be used to provide some, or all, of the down payment on the first-lien loan.

The Board recognized that consumers have varied reasons for originating a HELOC concurrently with the first-lien loan, for example, to reduce overall closing costs or for the convenience of having access to an available credit line in the future. However, the Board believed concerns relating to HELOCs originated concurrently for savings or convenience, and not to provide payment towards the first-lien home purchase loan, might be mitigated by the Board's proposal to require that a creditor consider the periodic payment on the simultaneous loan based on the actual amount drawn from the credit line by the consumer. See proposed § 226.43(c)(6)(ii), discussing payment calculation requirements for simultaneous loans that are HELOCs. Still, the Board recognized that in the case of a non-purchase transaction (e.g., a refinancing) a simultaneous loan that is a HELOC might be unlikely to be originated and drawn upon to provide payment towards the first-lien loan, except perhaps towards closing costs. Thus, the Board solicited comment on whether it should narrow the requirement to consider simultaneous loans that are HELOCs to apply only to purchase transactions.

Third, in developing this proposal Board staff conducted outreach with a variety of participants that consistently expressed the view that second-lien loans significantly impact a consumer's performance on the first-lien loan, and that many second-lien loans are HELOCs. One industry participant explained that the vast majority of “piggyback loans” it originated were HELOCs that were fully drawn at the time of origination and used to assist in the first-lien purchase transaction. Another outreach participant stated that HELOCs make up approximately 90 percent of its simultaneous loan book-of-business. Industry outreach participants generally indicated that it is a currently accepted underwriting practice to include HELOCs in the repayment ability assessment on the first-lien loan, and generally confirmed that the majority of simultaneous liens considered during the underwriting process are HELOCs. For these reasons, the Board proposed to use its authority under TILA sections 105(a) and 129B(e) to broaden the scope of TILA section 129C(a)(2), and accordingly proposed to define the term “simultaneous loan” to include HELOCs.

Proposed comment 43(b)(12)-1 clarified that the definition of “simultaneous loan” includes any loan that meets the definition, whether made by the same creditor or a third-party creditor, and provides an illustrative example of this principle.

Proposed comment 43(b)(12)-2 further clarified the meaning of the term “same consumer,” and explained that for purposes of the definition of “simultaneous loan,” the term “same consumer” would include any consumer, as that term is defined in § 1026.2(a)(11), that enters into a loan that is a covered transaction and also enters into another loan (e.g., a second-lien covered transaction or HELOC) secured by the same dwelling. This comment further explained that where two or more consumers enter into a legal obligation that is a covered transaction, but only one of them enters into another loan secured by the same dwelling, the “same consumer” includes the person that has entered into both legal obligations. The Board believed this comment would reflect statutory intent to include any loan that could impact the consumer's ability to repay the covered transaction according to its terms (i.e., to require the creditor to consider the combined payment obligations of the consumer(s) obligated to repay the covered transaction). See TILA § 129C(a)(2).

Both industry and consumer advocate commenters overwhelmingly supported inclusion of HELOCs as simultaneous loans, with only one industry commenter objecting. The objecting commenter stated that there was no persuasive policy argument for deviating from the statute, but did not provide any reason to believe that concurrent HELOCs are less relevant to an assessment of a consumer's ability to repay than concurrent closed-end second liens. As explained in the proposed rule, most industry participants are already considering HELOCs in the underwriting of senior-lien loans on the same property. 76 FR 27418.

For the reasons set forth by the Board and discussed above, the Bureau has determined that inclusion of HELOCs in the definition of simultaneous loans is an appropriate use of its TILA authority to make adjustments and additional requirements.

TILA section 105(a), as amended by section 1100A of the Dodd-Frank Act, authorizes the Bureau to prescribe regulations that may contain such additional requirements, classifications, differentiations, or other provisions, and may provide for such adjustments and exceptions for all or any class of transactions, as in the judgment of the Bureau are necessary or proper to effectuate the purposes of TILA, to prevent circumvention or evasion of TILA, or to facilitate compliance with TILA. 15 U.S.C. 1604(a). The Bureau finds that the inclusion of HELOCs is necessary and proper to effectuate the purposes of TILA. The inclusion of HELOCs is further supported by the Bureau's authority under TILA section 129B(e) to condition terms, acts or practices relating to residential mortgage loans that the Bureau finds necessary or proper to effectuate the purposes of TILA. 15 U.S.C. 1639b(e). TILA section 129B(a)(2) states that “[i]t is the purpose[] of * * * [S]ection 129C to assure that consumers are offered and receive residential mortgage loans on terms that reasonably reflect their ability to repay the loans.” 15 U.S.C. 1639b. Inclusion of HELOCs as simultaneous loans will help to carry out this purpose of TILA by helping to ensure that consumers receive loans on affordable terms, as further explained above.

Accordingly, the Bureau is adopting § 1026.43(b)(12) and associated commentary as proposed, with clarifying edits to ensure that simultaneous loans scheduled after consummation will be considered in determining ability to repay.

43(b)(13)

TILA section 129C(a)(1) requires that a creditor determine a consumer's repayment ability using “verified and documented information,” and TILA section 129C(a)(4) specifically requires the creditor to verify a consumer's income or assets relied on to determine repayment ability using a consumer's tax return or “third-party documents” that provide reasonably reliable evidence of the consumer's income or assets, as discussed in detail below in the section-by-section analysis of § 1026.43(c)(3) and (4). The Board proposed to define the term “third-party record” to mean: (1) A document or other record prepared or reviewed by a person other than the consumer, the creditor, any mortgage broker, as defined in § 1026.36(a)(2), or any agent of the creditor or mortgage broker; (2) a copy of a tax return filed with the Internal Revenue Service or a state taxing authority; (3) a record the creditor maintains for an account of the consumer held by the creditor; or (4) if the consumer is an employee of the creditor or the mortgage broker, a document or other record regarding the consumer's employment status or income. The Board explained that, in general, a creditor should refer to reasonably reliable records prepared by or reviewed by a third party to verify repayment ability under TILA section 129C(a), a principle consistent with verification requirements previously outlined under the Board's 2008 HOEPA Final Rule. See § 1026.34(a)(4)(ii).

Commenters generally supported the Board's broad definition of a third-party record as a reasonable definition that allows a creditor to use a wide variety of documents and sources, while ensuring that the consumer does not remain the sole source of information. Some consumer advocates, however, cautioned the Bureau against relying upon tax records to provide a basis for verifying income history, pursuant to amended TILA section 129C(a)(4)(A), to avoid penalizing consumers who may not have access to accurate tax records. The Bureau does not address comments with respect to consumers who may not maintain accurate tax records because the definition provided in 1026.43(b)(13) of third-party record merely ensures that a creditor may use any of a wide variety of documents, including tax records, as a method of income verification without mandating their use. Rather than rely solely on tax records, for example, a creditor might look to other third-party records for verification purposes, including the creditor's records regarding a consumer's savings account held by the creditor, which qualifies as a third-party record under § 1026.43(b)(13)(iii), or employment records for a consumer employed by the creditor, which qualifies as a third-party record under § 1026.43(b)(13)(iv).

The Board proposed comment 43(b)(13)-1 to clarify that third-party records would include records transmitted or viewed electronically, for example, a credit report prepared by a consumer reporting agency and transmitted or viewed electronically. The Bureau did not receive significant feedback on the proposed comment and is adopting the comment largely as proposed. The Bureau is clarifying that an electronic third-party record should be transmitted electronically, such as via email or if the creditor is able to click on a secure hyperlink to access a consumer's credit report. The Bureau is making this slight clarification to convey that mere viewing of a record, without the ability to capture or maintain the record, would likely be problematic with respect to record retention under § 1026.25(a) and (c). While it seems unlikely that an electronic record could be viewed without being transmitted as well, the Bureau is making this alteration to avoid any confusion.

The Bureau is adopting the remaining comments to 43(b)(13) largely as proposed by the Board. These comments did not elicit significant public feedback. Comment 43(b)(13)-1 assures creditors that a third-party record may be transmitted electronically. Comment 43(b)(13)-2 explains that a third-party record includes a form a creditor provides to a third party for providing information, even if the creditor completes parts of the form unrelated to the information sought. Thus, for example, a creditor may send a Webform, or mail a paper form, created by the creditor, to a consumer's current employer, on which the employer could check a box that indicates that the consumer works for the employer. The creditor may even elect to fill in the creditor's name, or other portions of the form, so long as those portions are unrelated to the information that the creditor seeks to verify, such as income or employment status.

Comment 43(b)(13)(i)-1 clarifies that a third-party record includes a document or other record prepared by the consumer, the creditor, the mortgage broker, or an agent of the creditor or mortgage broker, if the record is reviewed by a third party. For example, a profit-and-loss statement prepared by a self-employed consumer and reviewed by a third-party accountant is a third-party record under § 1026.43(b)(13)(i). The Bureau is including comment 43(b)(13)(i)-1 to explain how some first-party records, e.g., documents originally prepared by the consumer, may become third-party records by virtue of an appropriate, disinterested third-party's review or audit. It is the third party review, the Bureau believes, that provides reasonably reliable evidence of the underlying information in the document, just as if the document were originally prepared by the third party. Moreover, this clarification allows the creditor to consult a wider variety of documents in its determination of a consumer's ability to repay. Creditors should be cautioned not to assume, however, that merely because a document is a third-party record as defined by § 1026.43(b)(13), and the creditor uses the information provided by that document to make a determination as to whether the consumer will have a reasonable ability to repay the loan according to its terms, that the creditor has satisfied the requirements of this rule. The creditor also must make a reasonable and good faith determination at or before consummation that the consumer will have a reasonable ability, at the time of consummation, to repay the loan according to its terms. For a full discussion of the Bureau's approach to this determination, see § 1026.43(c)(1), its commentary, and the section-by-section analysis of those provisions below.

Finally, comment 43(b)(13)(iii)-1 clarifies that a third-party record includes a record that the creditor maintains for the consumer's account. Such examples might include records of a checking account, savings account, and retirement account that the consumer holds, or has held, with the creditor. Comment 43(b)(13)(iii)-1 also provides the example of a creditor's records for an account related to a consumer's outstanding obligations to the creditor, such as the creditor's records for a first-lien mortgage to a consumer who applies for a subordinate-lien home equity loan. This comment helps assure industry that such records are a legitimate basis for determining a consumer's ability to repay, and/or for verifying income and assets because it is unlikely to be in a creditor's interest to falsify such records for purposes of satisfying § 1026.43(b)(13), as falsifying records would violate the good faith requirement of § 1026.43(c)(1). In addition, this comment should help assure creditors that the rule does not inhibit a creditor's ability to “cross-sell” products to consumers, by avoiding placing the creditor at a disadvantage with respect to verifying a consumer's information by virtue of the creditor's existing relationship with the consumer.

43(c) Repayment Ability

As enacted by the Dodd-Frank Act, TILA section 129C(a)(1) provides that no creditor may make a residential mortgage loan unless the creditor makes a reasonable and good faith determination, based on verified and documented information, that, at the time the loan is consummated, the consumer has a reasonable ability to repay the loan according to its terms and all applicable taxes, insurance, and assessments. TILA section 129C(a)(2) extends the same requirement to a combination of multiple residential mortgage loans secured by the same dwelling where the creditor knows or has reason to know that such loans will be made to the same consumer. TILA sections 129C(a)(3) and (a)(4) specify factors that must be considered in determining a consumer's ability to repay and verification requirements for income and assets considered as part of that determination. Proposed § 226.43(c) would have implemented TILA section 129C(a)(1) through (4) in a manner substantially similar to the statute.

Proposed § 226.43(c)(1) would have implemented the requirement in TILA section 129C(a)(1) that creditors make a reasonable and good faith determination that a consumer will have a reasonable ability to repay the loan according to its terms. Proposed § 226.43(c)(2) would have required creditors to consider the following factors in making a determination of repayment ability, as required by TILA section 129C(a)(1) through (3): the consumer's current or reasonably expected income or assets (other than the property that secures the loan); the consumer's employment status, if the creditor relies on employment income; the consumer's monthly payment on the loan; the consumer's monthly payment on any simultaneous loan that the creditor knows or has reason to know will be made; the consumer's monthly payment for mortgage-related obligations; the consumer's current debt obligations; and the consumer's monthly debt-to-income ratio or residual income. Proposed § 226.43(c)(3) would have required that creditors verify the information they use in making an ability-to-repay determination using third-party records, as required by TILA section 129C(a)(1). Proposed § 226.43(c)(4) would have specified methods for verifying income and assets as required by TILA section 129C(a)(1) and (4). Proposed § 226.43(c)(5) and (6) would have specified how to calculate the monthly mortgage and simultaneous loan payments required to be considered under proposed § 226.43(c)(2). Proposed § 226.43(c)(7) would have specified how to calculate the monthly debt-to-income ratio or monthly residual income required to be considered under proposed § 226.43(c)(2). As discussed in detail below, the Bureau is adopting § 1026.43(c) substantially as proposed, with various modifications and clarifications.

Proposed comment 43(c)-1 would have indicated that creditors may look to widely accepted governmental or nongovernmental underwriting standards, such as the handbook on Mortgagee Credit Analysis for Mortgage Insurance on One- to Four-Unit Mortgage Loans issued by FHA, to evaluate a consumer's ability to repay. The proposed comment would have stated that creditors may look to such standards in determining, for example, whether to classify particular inflows, obligations, or property as “income,” “debt,” or “assets”; factors to consider in evaluating the income of a self-employed or seasonally employed consumer; or factors to consider in evaluating the credit history of a consumer who has obtained few or no extensions of traditional “credit” as defined in § 1026.2(a)(14). In the Supplemental Information regarding proposed comment 43(c)-1, the Board stated that the proposed rule and commentary were intended to provide flexibility in underwriting standards so that creditors could adapt their underwriting processes to a consumer's particular circumstances. The Board stated its belief that such flexibility is necessary because the rule covers such a wide variety of consumers and mortgage products.

Commenters generally supported giving creditors significant flexibility to develop and apply their own underwriting standards. However, commenters had concerns regarding the specific approach taken in proposed comment 43(c)-1. Commenters raised a number of questions about what kinds of underwriting standards might be considered widely accepted, such as whether a creditor's proprietary underwriting standards could ever be considered widely accepted. Commenters also were uncertain whether the proposed comment required creditors to adopt particular governmental underwriting standards in their entirety and requested clarification on that point. At least one commenter, an industry trade group, noted that FHA-insured loans constitute a small percentage of the mortgage market and questioned whether FHA underwriting standards therefore are widely accepted. This commenter also questioned whether it is appropriate to encourage creditors to apply FHA underwriting standards other than with respect to FHA-insured loans, as FHA programs are generally designed to make mortgage credit available in circumstances where private creditors are unwilling to extend such credit without a government guarantee. Finally, consumer group commenters asserted that underwriting standards do not accurately determine ability to repay merely because they are widely accepted and pointed to the widespread proliferation of lax underwriting standards that predated the recent financial crisis.

The Bureau believes that the Board did not intend to require creditors to use any particular governmental underwriting standards, including FHA standards, in their entirety or to prohibit creditors from using proprietary underwriting standards. The Bureau also does not believe that the Board intended to endorse lax underwriting standards on the basis that those standards may be prevalent in the mortgage market at a particular time. The Bureau therefore is adopting two new comments to provide greater clarity regarding the role of underwriting standards in ability-to-repay determinations and is not adopting proposed comment 43(c)-1.

The Bureau is concerned based on the comments received that referring creditors to widely accepted governmental and nongovernmental underwriting standards could lead to undesirable misinterpretations and confusion. The discussion of widely accepted standards in proposed comment 43(c)-1 could be misinterpreted to suggest that the underwriting standards of any single market participant with a large market share are widely accepted and therefore to be emulated. The widely accepted standard also could be misinterpreted to indicate that proprietary underwriting standards cannot yield reasonable, good faith determinations of a consumer's ability to repay because they are unique to a particular creditor and not employed throughout the mortgage market. Similarly, the widely accepted standard could be misinterpreted to encourage a creditor that lends in a limited geographic area or in a particular market niche to apply widely accepted underwriting standards that are inappropriate for that particular creditor's loans.

The Bureau also is concerned that evaluating underwriting standards based on whether they are widely accepted could have other undesirable consequences. In a market bubble or economic crisis, many creditors may change their underwriting standards in similar ways, leading to widely accepted underwriting standards becoming unreasonably lax or unreasonably tight. A regulatory directive to use underwriting standards that are widely accepted could exacerbate those effects. Also, referring creditors to widely accepted governmental and nongovernmental underwriting standards could hinder creditors' ability to respond to changing market and economic conditions and stifle market growth and positive innovation.

Finally, the Bureau is concerned that focusing on whether underwriting standards are widely accepted could distract creditors from focusing on their obligation under TILA section 129C and § 1026.43(c) to make ability-to-repay determinations that are reasonable and in good faith. The Bureau believes that a creditor's underwriting standards are an important factor in making reasonable and good faith ability-to-repay determinations. However, how those standards are applied to the individual facts and circumstances of a particular extension of credit is equally or more important.

In light of these issues, the Bureau is not adopting proposed comment 43(c)-1. Instead, the Bureau is adopting two new comments, comment 43(c)(1)-1 and comment 43(c)(2)-1. New comment 43(c)(1)-1 clarifies that creditors are permitted to develop and apply their own underwriting standards as long as those standards lead to ability-to-repay determinations that are reasonable and in good faith. New comment 43(c)(2)-1 clarifies that creditors are permitted to use their own definitions and other technical underwriting criteria and notes that underwriting guidelines issued by governmental entities such as the FHA are a source to which creditors may refer for guidance on definitions and technical underwriting criteria. These comments are discussed below in the section-by-section of § 1026.43(c)(1) and (2).

43(c)(1) General Requirement

Proposed § 226.43(c)(1) would have implemented TILA section 129C(a)(1) by providing that a creditor shall not make a loan that is a covered transaction unless the creditor makes a reasonable and good faith determination at or before consummation that the consumer will have a reasonable ability, at the time of consummation, to repay the loan according to its terms, including any mortgage-related obligations. Commenters generally agreed that creditors should not make loans to consumers unable to repay them and supported the requirement to consider ability to repay. Accordingly, § 1026.43(c)(1) is adopted substantially as proposed, with two technical and conforming changes.

As adopted, § 1026.43(c)(1) requires creditors to make a reasonable and good faith determination at or before consummation that the consumer will have a reasonable ability to repay the loan according to its terms. Section 1026.43(c)(1) as adopted omits the reference in the proposed rule to determining that a consumer has a reasonable ability “at the time of consummation” to repay the loan according to its terms. The Bureau believes this phrase is potentially misleading and does not accurately reflect the intent of either the Board or the Bureau. Mortgage loans are not required to be repaid at the time of consummation; instead, they are required to be repaid over months or years after consummation. Creditors are required to make a predictive judgment at the time of consummation that a consumer is likely to have the ability to repay a loan in the future. The Bureau believes that the rule more clearly reflects this requirement without the reference to ability “at the time of consummation” to repay the loan. The creditor's determination will necessarily be based on the consumer's circumstances at or before consummation and evidence, if any, that those circumstances are likely to change in the future. Section 1026.43(c)(1) as adopted also omits the reference in the proposed rule to mortgage-related obligations. The Bureau believes this reference is unnecessary because § 1026.43(c)(2) requires creditors to consider consumers' monthly payments for mortgage-related obligations and could create confusion because § 1026.43(c)(1) does not include references to other factors creditors must consider under § 1026.43(c)(2).

As noted above, the Bureau is adopting new comment 43(c)(1)-1, which provides guidance regarding, among other things, how the requirement to make a reasonable and good faith determination of ability to repay relates to a creditor's underwriting standards. New comment 43(c)(1)-1 replaces in part and responds to comments regarding proposed comment 43(c)-1, discussed above.

New comment 43(c)(1)-1 emphasizes that creditors are to be evaluated on whether they make a reasonable and good faith determination that a consumer will have a reasonable ability to repay as required by § 1026.43(c)(1). The comment acknowledges that § 1026.43(c) and the accompanying commentary describe certain requirements for making ability-to-repay determinations, but do not provide comprehensive underwriting standards to which creditors must adhere. As an example, new comment 43(c)(1)-1 notes that the rule and commentary do not specify how much income is needed to support a particular level of debt or how to weigh credit history against other factors.

The Bureau believes that a variety of underwriting standards can yield reasonable, good faith ability-to-repay determinations. New comment 43(c)(1)-1 explains that, so long as creditors consider the factors set forth in § 1026.43(c)(2) according to the requirements of § 1026.43(c), creditors are permitted to develop and apply their own proprietary underwriting standards and to make changes to those standards over time in response to empirical information and changing economic and other conditions. The Bureau believes this flexibility is necessary given the wide range of creditors, consumers, and mortgage products to which this rule applies. The Bureau also believes that there are no indicators in the statutory text or legislative history of the Dodd-Frank Act that Congress intended to replace proprietary underwriting standards with underwriting standards dictated by governmental or government-sponsored entities as part of the ability-to-repay requirements. The Bureau therefore believes that preserving this flexibility here is consistent with Congressional intent. The comment emphasizes that whether a particular ability-to-repay determination is reasonable and in good faith will depend not only on the underwriting standards adopted by the creditor, but on the facts and circumstances of an individual extension of credit and how the creditor's underwriting standards were applied to those facts and circumstances. The comment also states that a consumer's statement or attestation that the consumer has the ability to repay the loan is not indicative of whether the creditor's determination was reasonable and in good faith.

Concerns have been raised that creditors and others will have difficulty evaluating whether a particular ability-to-repay determination is reasonable and in good faith. Although the statute and the rule specifies certain factors that a creditor must consider in making such a determination, the Bureau does not believe that there is any litmus test that can be prescribed to determine whether a creditor, in considering those factors, arrived at a belief in the consumer's ability to repay which was both objectively reasonable and in subjective good faith. Nevertheless, new comment 43(c)(1)-1 lists considerations that may be relevant to whether a creditor who considered and verified the required factors in accordance with the rule arrived at an ability-to-repay determination that was reasonable and in good faith. The comment states that the following may be evidence that a creditor's ability-to-repay determination was reasonable and in good faith: (1) The consumer demonstrated actual ability to repay the loan by making timely payments, without modification or accommodation, for a significant period of time after consummation or, for an adjustable-rate, interest-only, or negative-amortization mortgage, for a significant period of time after recast; (2) the creditor used underwriting standards that have historically resulted in comparatively low rates of delinquency and default during adverse economic conditions; or (3) the creditor used underwriting standards based on empirically derived, demonstrably and statistically sound models.

In contrast, new comment 43(c)(1)-1 states that the following may be evidence that a creditor's ability-to-repay determination was not reasonable or in good faith: (1) The consumer defaulted on the loan a short time after consummation or, for an adjustable-rate, interest-only, or negative-amortization mortgage, a short time after recast; (2) the creditor used underwriting standards that have historically resulted in comparatively high levels of delinquency and default during adverse economic conditions; (3) the creditor applied underwriting standards inconsistently or used underwriting standards different from those used for similar loans without reasonable justification; (4) the creditor disregarded evidence that the underwriting standards it used are not effective at determining consumers' repayment ability; (5) the creditor consciously disregarded evidence that the consumer may have insufficient residual income to cover other recurring obligations and expenses, taking into account the consumer's assets other than the property securing the covered transaction, after paying his or her monthly payments for the covered transaction, any simultaneous loan, mortgage-related obligations and any current debt obligations; or (6) the creditor disregarded evidence that the consumer would have the ability to repay only if the consumer subsequently refinanced the loan or sold the property securing the loan.

New comment 43(c)(1)-1 states the Bureau's belief that all of these considerations may be relevant to whether a creditor's ability-to-repay determination was reasonable and in good faith. However, the comment also clarifies that these considerations are not requirements or prohibitions with which creditors must comply, nor are they elements of a claim that a consumer must prove to establish a violation of the ability-to-repay requirements. As an example, the comment clarifies that creditors are not required to validate their underwriting criteria using mathematical models.

New comment 43(c)(1)-1 also clarifies that these considerations are not absolute in their application; instead they exist on a continuum and may apply to varying degrees. As an example, the comment states that the longer a consumer successfully makes timely payments after consummation or recast the less likely it is that the creditor's determination of ability to repay was unreasonable or not in good faith.

Finally, new comment 43(c)(1)-1 clarifies that each of these considerations must be viewed in the context of all facts and circumstances relevant to a particular extension of credit. As an example, the comment states that in some cases inconsistent application of underwriting standards may indicate that a creditor is manipulating those standards to approve a loan despite a consumer's inability to repay. The creditor's ability-to-repay determination therefore may be unreasonable or in bad faith. However, in other cases inconsistently applied underwriting standards may be the result of, for example, inadequate training and may nonetheless yield a reasonable and good faith ability-to-repay determination in a particular case. Similarly, the comment states that although an early payment default on a mortgage will often be persuasive evidence that the creditor did not have a reasonable and good faith belief in the consumer's ability to repay (and such evidence may even be sufficient to establish a prima facie case of an ability-to-repay violation), a particular ability-to-repay determination may be reasonable and in good faith even though the consumer defaulted shortly after consummation if, for example, the consumer experienced a sudden and unexpected loss of income. In contrast, the comment states that an ability-to-repay determination may be unreasonable or not in good faith even though the consumer made timely payments for a significant period of time if, for example, the consumer was able to make those payments only by foregoing necessities such as food and heat.

The Board proposed comment 43(c)(1)-1 to clarify that a change in a consumer's circumstances after consummation of a loan, such as a significant reduction in income due to a job loss or a significant obligation arising from a major medical expense, that cannot reasonably be anticipated from the consumer's application or the records used to determine repayment ability, is not relevant to determining a creditor's compliance with the rule. The proposed comment would have further clarified that, if the application or records considered by the creditor at or before consummation indicate that there will be a change in the consumer's repayment ability after consummation, such as if a consumer's application states that the consumer plans to retire within 12 months without obtaining new employment or that the consumer will transition from full-time to part-time employment, the creditor must consider that information. Commenters generally supported proposed comment 43(c)(1)-1. Proposed comment 43(c)(1)-1 is adopted substantially as proposed and redesignated as comment 43(c)(1)-2.

The Board also proposed comment 43(c)(1)-2 to clarify that § 226.43(c)(1) does not require or permit the creditor to make inquiries or verifications prohibited by Regulation B, 12 CFR part 1002. Commenters generally supported proposed comment 43(c)(1)-2. Proposed comment 43(c)(1)-2 is adopted substantially as proposed and redesignated as comment 43(c)(1)-3.

43(c)(2) Basis for Determination

As discussed above, TILA section 129C(a)(1) generally requires a creditor to make a reasonable and good faith determination that a consumer has a reasonable ability to repay a loan and all applicable taxes, insurance, and assessments. TILA section 129C(a)(2) requires a creditor to include in that determination the cost of any other residential mortgage loans made to the same consumer and secured by the same dwelling. TILA section 129C(a)(3) enumerates several factors a creditor must consider in determining a consumer's ability to repay: credit history; current income; expected income; current obligations; debt-to-income ratio or residual income; employment status; and other financial resources other than equity in the property securing the loan.

Proposed § 226.43(c)(2) would have implemented the requirements under these sections of TILA that a creditor consider specified factors as part of a determination of a consumer's ability to repay. Proposed § 226.43(c)(2) would have required creditors to consider the following factors in making a determination of repayment ability, as required by TILA section 129C(a)(1) through (3): the consumer's current or reasonably expected income or assets, other than the dwelling that secures the loan; the consumer's employment status, if the creditor relies on employment income; the consumer's monthly payment on the loan; the consumer's monthly payment on any simultaneous loan that the creditor knows or has reason to know will be made; the consumer's monthly payment for mortgage-related obligations; the consumer's current debt obligations; the consumer's monthly debt-to-income ratio or residual income; and the consumer's credit history. As discussed in detail below, the Bureau is adopting § 1026.43(c)(2) substantially as proposed, with technical and conforming changes.

As indicated above, the Bureau also is adopting new comment 43(c)(2)-1. New comment 43(c)(2)-1 provides guidance regarding definitional and other technical underwriting issues related to the factors enumerated in § 1026.43(c)(2). New comment 43(c)(2)-1 replaces in part and responds to comments received regarding proposed comment 43(c)-1, as discussed above.

New comment 43(c)(2)-1 notes that § 1026.43(c)(2) sets forth factors creditors must consider when making the ability-to-repay determination required under § 1026.43(c)(1) and the accompanying commentary provides guidance regarding these factors. New comment 43(c)(2)-1 also notes that creditors must conform to these requirements and may rely on guidance provided in the commentary. New comment 43(c)(2)-1 also acknowledges that the rule and commentary do not provide comprehensive guidance on definitions and other technical underwriting criteria necessary for evaluating these factors in practice. The comment clarifies that, so long as a creditor complies with the provisions of § 1026.43(c), the creditor is permitted to use its own definitions and other technical underwriting criteria.

New comment 43(c)(2)-1 further provides that a creditor may, but is not required to, look to guidance issued by entities such as the FHA, VA, USDA, or Fannie Mae or Freddie Mac while operating under the conservatorship of the Federal Housing Finance Administration. New comment 43(c)(2)-1 gives several examples of instances where a creditor could refer to such guidance, such as: classifying particular inflows, obligations, and property as “income,” “debt,” or “assets”; determining what information to use when evaluating the income of a self-employed or seasonally employed consumer; or determining what information to use when evaluating the credit history of a consumer who has few or no extensions of traditional credit. The comment emphasizes that these examples are illustrative, and creditors are not required to conform to guidance issued by these or other such entities. The Bureau is aware that many creditors have, for example, existing underwriting definitions of “income” and “debt.” Creditors are not required to modify their existing definitions and other technical underwriting criteria to conform to guidance issued by such entities, and creditors' existing definitions and other technical underwriting criteria are not noncompliant merely because they differ from those used in such guidance.

Finally, new comment 43(c)(2)-1 emphasizes that a creditor must ensure that its underwriting criteria, as applied to the facts and circumstances of a particular extension of credit, result in a reasonable, good faith determination of a consumer's ability to repay. As an example, new comment 43(c)(2)-1 states that a definition used in underwriting that is reasonable in isolation may lead to ability-to-repay determinations that are unreasonable or not in good faith when considered in the context of a creditor's underwriting standards or when adopted or applied in bad faith. Similarly, an ability-to-repay determination is not unreasonable or in bad faith merely because the underwriting criteria used included a definition that was by itself unreasonable.

43(c)(2)(i)

TILA section 129C(a)(3) provides that, in making the repayment ability determination, a creditor must consider, among other factors, a consumer's current income, reasonably expected income, and “financial resources” other than the consumer's equity in the dwelling or real property that secures loan repayment. Furthermore, under TILA section 129C(a)(9), a creditor may consider the seasonality or irregularity of a consumer's income in determining repayment ability. The Board's proposal generally mirrored TILA section 129C(a)(3), but differed in two respects.

First, proposed § 226.43(c)(2)(i) used the term “assets” rather than “financial resources,” to conform with terminology used in other provisions under TILA section 129C(a) and Regulation Z. See, e.g., TILA section 129C(a)(4) (requiring that creditors consider a consumer's assets in determining repayment ability); § 1026.51(a) (requiring consideration of a consumer's assets in determining a consumer's ability to pay a credit extension under a credit card account). The Board explained that the terms “financial resources” and “assets” are synonymous as used in TILA section 129C(a), and elected to use the term “assets” throughout the proposal for consistency. The Bureau is adopting this interpretation as well, as part of its effort to streamline regulations and reduce compliance burden, and uses the term “assets” throughout Regulation Z.

Second, the Board's proposal provided that a creditor may not look to the value of the dwelling that secures the covered transaction, instead of providing that a creditor may not look to the consumer's equity in the dwelling, as provided in TILA section 129C(a). The Bureau received comments expressing concern that the Board had proposed dispensing with the term “equity.” These comments protested that the Board had assumed that congressional concern was over the foreclosure value of the home, rather than protecting all homeowners, including those who may have low home values. The commenters' concerns are likely misplaced, however, as the Board's language provides, if anything, broader protection for homeowners. TILA section 129C(a)(3) is intended to address the risk that a creditor will consider the amount that could be obtained through a foreclosure sale of the dwelling, which may exceed the amount of the consumer's equity in the dwelling. For example, the rule addresses the situation in which, several years after consummation, the value of a consumer's home has decreased significantly. The rule prohibits a creditor from considering, at or before consummation, any value associated with this home, even in the event that the “underwater” home is sold at foreclosure. The rule thus avoids the situation in which the creditor might assume that rising home values might make up the difference should the consumer be unable to make full mortgage payments, and therefore the rule is more protective of consumers because the rule forbids the creditor from considering any value associated with the dwelling whether the consumer's equity stake in the dwelling is large or small.

The Bureau is adopting the Board's proposal, providing that a creditor may not look to the value of the dwelling that secures the covered transaction, instead of providing that a creditor may not look to the consumer's equity in the dwelling, as provided in TILA section 129C(a). The Bureau is making this adjustment pursuant to its authority under TILA section 105(a), which provides that the Bureau's regulations may contain such additional requirements, classifications, differentiations, or other provisions, and may provide for such adjustments and exceptions for all or any class of transactions as in the Bureau's judgment are necessary or proper to effectuate the purposes of TILA, prevent circumvention or evasion thereof, or facilitate compliance therewith. 15 U.S.C. 1604(a). The purposes of TILA include the purposes that apply to 129C, to assure that consumers are offered and receive residential mortgage loans on terms that reasonably reflect their ability to repay the loan. See 15 U.S.C. 1639b(a)(2). As further explained above, the Bureau believes it is necessary and proper to make this adjustment to ensure that consumers receive loans on affordable terms and to facilitate compliance with TILA and its purposes.

The Board proposed comment 43(c)(2)(i)-1 to clarify that a creditor may base a determination of repayment ability on current or reasonably expected income from employment or other sources, assets other than the dwelling that secures the covered transaction, or both. The Bureau did not receive significant comment on the proposal and has adopted the Board's proposed comment. In congruence with the Bureau's adoption of the phrase “value of the dwelling” in § 1026.43(c)(2)(i), instead of the consumer's equity in the dwelling, as originally provided in TILA section 129C(a), comment 43(c)(2)(i)-1 likewise notes that the creditor may not consider the dwelling that secures the transaction as an asset in any respect. This comment is also consistent with comment 43(a)-2, which further clarifies that the term “dwelling” includes the value of the real property to which the dwelling is attached, if the real property also secures the covered transaction. Comment 43(c)(2)(i)-1 also provides examples of types of income the creditor may consider, including salary, wages, self-employment income, military or reserve duty income, tips, commissions, and retirement benefits; and examples of assets the creditor may consider, including funds in a savings or checking account, amounts vested in a retirement account, stocks, and bonds. The Bureau did not receive significant comment on the proposal and has adopted the Board's proposed comment. The Bureau notes that there may be assets other than those listed in comment 43(c)(2)(i)-1 that a creditor may consider; the Bureau does not intend for the list to be exhaustive, but merely illustrative.

The Board proposed comment 43(c)(2)(i)-2 to explain that, if a creditor bases its determination of repayment ability entirely or in part on a consumer's income, the creditor need consider only the income necessary to support a determination that the consumer can repay the covered transaction. The Bureau did not receive significant comment and has adopted the Board's comment largely as proposed. This comment clarifies that a creditor need not document and verify every aspect of the consumer's income, merely enough income to support the creditor's good faith determination. For example, if a consumer earns income from a full-time job and a part-time job and the creditor reasonably determines that the consumer's income from the full-time job is sufficient to repay the covered transaction, the creditor need not consider the consumer's income from the part-time job. Comment 43(c)(2)(i)-2 also cross-references comment 43(c)(4)-1 for clarity.

The Board proposed comment 43(c)(2)(i)-3 to clarify that the creditor may rely on the consumer's reasonably expected income either in addition to or instead of current income. This comment is similar to existing comment 34(a)(4)(ii)-2, which describes a similar income test for high-cost mortgages under § 1026.34(a)(4). This consistency should serve to reduce compliance burden for creditors. The Bureau did not receive significant comment on the proposal and is adopting the Board's comment as proposed. Comment 43(c)(2)(i)-3 further explains that, if a creditor relies on expected income, the expectation that the income will be available for repayment must be reasonable and verified with third-party records that provide reasonably reliable evidence of the consumer's expected income. Comment 43(c)(2)(i)-3 also gives examples of reasonably expected income, such as expected bonuses verified with documents demonstrating past bonuses or expected salary from a job verified with a written statement from an employer stating a specified salary. As the Board has previously stated, in some cases a covered transaction may have a likely payment increase that would not be affordable at the consumer's income at the time of consummation. A creditor may be able to verify a reasonable expectation of an increase in the consumer's income that will make the higher payment affordable to the consumer. See 73 FR 44522, 44544 (July 30, 2008).

TILA section 129C(a)(9) provides that a creditor may consider the seasonality or irregularity of a consumer's income in determining repayment ability. Accordingly, the Board proposed comment 43(c)(2)(i)-4 to clarify that a creditor reasonably may determine that a consumer can make periodic loan payments even if the consumer's income, such as self-employment or agricultural employment income, is seasonal or irregular. The Bureau received little comment on this proposal, although at least one consumer advocate expressed concern that creditors might interpret the rule to allow for a creditor to differentiate among types of income. Specifically, the commenter expressed concern that some creditors might differentiate types of income, for example salaried income as opposed to disability payments, and that these creditors might require the consumer to produce a letter stating that the disability income was guaranteed for a specified period. The Bureau understands these concerns, and cautions creditors not to overlook the requirements imposed by the Equal Credit Opportunity Act, implemented by the Bureau under Regulation B. See 15 U.S.C. 1601 et seq.; 12 CFR 1002.1 et seq. For example, 12 CFR 1002.6(b)(2) prohibits a creditor from taking into account whether an applicant's income derives from any public assistance program. The distinction here is that 43(c)(2)(i)-4 permits the creditor to consider the regularity of the consumer's income, but such consideration must be based on the consumer's income history, not based on the source of the income, as both a consumer's wages or a consumer's receipt of public assistance may or may not be irregular. The Bureau is adopting this comment largely as proposed, as the concerns discussed above are largely covered by Regulation B. Comment 43(c)(2)(i)-4 states that, for example, if the creditor determines that the income a consumer receives a few months each year from, for example, selling crops or from agricultural employment is sufficient to make monthly loan payments when divided equally across 12 months, then the creditor reasonably may determine that the consumer can repay the loan, even though the consumer may not receive income during certain months.

Finally, the Bureau is adding new comment 43(c)(2)(i)-5 to further clarify, in the case of joint applicants, the consumer's current or reasonably expected income or assets basis of the creditor's ability-to-repay determination. This comment is similar in approach to the Board's proposed comment 43(c)(4)-2, discussed below, however, proposed comment 43(c)(4)-2 discussed the verification of income in the case of joint applicants. The Bureau is adding comment 43(c)(2)(i)-5 to clarify the creditor's basis for making an ability-to-repay determination for joint applicants. Comment 43(c)(2)(i)-5 explains that when two or more consumers apply for an extension of credit as joint obligors with primary liability on an obligation, § 1026.43(c)(i) does not require the creditor to consider income or assets that are not needed to support the creditor's repayment ability determination. Thus, the comment explains that if the income or assets of one applicant are sufficient to support the creditor's repayment ability determination, then the creditor is not required to consider the income or assets of the other applicant.

43(c)(2)(ii)

TILA section 129C(a)(3) requires that a creditor consider a consumer's employment status in determining the consumer's repayment ability, among other requirements. The Board proposal implemented this requirement in proposed § 226.43(c)(2)(ii) and clarified that a creditor need consider a consumer's employment status only if the creditor relies on income from the consumer's employment in determining repayment ability. The Bureau did not receive significant comment on the Board's proposal and is adopting § 1026.43(c)(2)(ii) as proposed. The Bureau sees no purpose in requiring a creditor to consider a consumer's employment status in the case where the creditor need not consider the income from that employment in the creditor's reasonable and good faith determination that the consumer will have a reasonable ability to repay the loan according to its terms.

The Board proposed, and the Bureau is adopting, comment 43(c)(2)(ii)-1 to illustrate this point further. The comment states, for example, that if a creditor relies wholly on a consumer's investment income to determine the consumer's repayment ability, the creditor need not consider or verify the consumer's employment status. The proposed comment further clarifies that employment may be full-time, part-time, seasonal, irregular, military, or self-employment. Comment 43(c)(2)(ii)-1 is similar to comment 34(a)(4)-6, which discusses income, assets, and employment in determining repayment ability for high-cost mortgages.

In its proposal, the Board explained that a creditor generally must verify information relied on to determine repayment ability using reasonably reliable third-party records, but may verify employment status orally as long as the creditor prepares a record of the oral information. The Board proposed comment 43(c)(2)(ii)-2 to add that a creditor also may verify the employment status of military personnel using the electronic database maintained by the Department of Defense (DoD) to facilitate identification of consumers covered by credit protections provided pursuant to 10 U.S.C. 987, also known as the “Talent Amendment.” The Board solicited comment on whether creditors needed additional flexibility in verifying the employment status of military personnel, such as by verifying the employment status of a member of the military using a Leave and Earnings Statement. As this proposed comment was designed to provide clarification for creditors with respect to verifying a consumer's employment, this proposed comment is discussed in the section-by-section analysis of § 1026.43(c)(3) below.

43(c)(2)(iii)

Proposed § 226.43(c)(2)(iii) implemented the requirements under new TILA section 129C(a)(1) and (3), in part, by requiring that the creditor consider the consumer's monthly payment on the covered transaction, calculated in accordance with proposed § 226.43(c)(5), for purposes of determining the consumer's repayment ability. Proposed comment 43(c)(2)(iii)-1 clarified the regulatory language and made clear that mortgage-related obligations must also be considered.

The Bureau did not receive comments on this provision. Accordingly, the Bureau is adopting § 1026.43(c)(2)(iii) as proposed. Comment 43(c)(2)(iii)-1 has been edited to remove the reference to mortgage-related obligations as potentially confusing. The monthly payment for mortgage-related obligations must be considered under § 1026.43(c)(2)(v).

43(c)(2)(iv)

Proposed § 226.43(c)(2)(iv) implemented the requirements under new TILA section 129C(a)(2), in part, by requiring that the creditor consider “the consumer's monthly payment on any simultaneous loan that the creditor knows or has reason to know will be made, calculated in accordance with” proposed § 226.43(c)(6), for purposes of determining the consumer's repayment ability. As explained above in the section-by-section analysis of § 1026.43(b)(12), “simultaneous loan” is defined, in the proposed and final rules, to include HELOCs.

Proposed comment 43(c)(2)(iv)-1 clarified that for purposes of the repayment ability determination, a simultaneous loan includes any covered transaction or HELOC that will be made to the same consumer at or before consummation of the covered transaction and secured by the same dwelling that secures the covered transaction. This comment explained that a HELOC that is a simultaneous loan that the creditor knows or has reason to know about must be considered in determining a consumer's ability to repay the covered transaction, even though the HELOC is not a covered transaction subject to § 1026.43.

Proposed comment 43(c)(2)(iv)-3 clarified the scope of timing and the meaning of the phrase “at or before consummation” with respect to simultaneous loans that the creditor must consider for purposes of proposed § 226.43(c)(2)(iv). Proposed comment 43(c)(2)(iv)-4 provided guidance on the verification of simultaneous loans.

The Bureau received several industry comments on the requirement, in the regulation and the statute, that the creditor consider any simultaneous loan it “knows or has reason to know” will be made. The commenters felt that the standard was vague, and that it would be difficult for a creditor to understand when it “has reason to know” a simultaneous loan will be made.

The Board provided guidance on the “knows or has reason to know” standard in proposed comment 43(c)(2)(iv)-2. This comment provided that, in regard to “piggyback” second-lien loans, the creditor complies with the standard if it follows policies and procedures that are designed to determine whether at or before consummation that the same consumer has applied for another credit transaction secured by the same dwelling. The proposed comment provided an example in which the requested loan amount is less than the home purchase price, indicating that there is a down payment coming from a different funding source. The creditor's policies and procedures must require the consumer to state the source of the down payment, which must be verified. If the creditor determines that the source of the down payment is another extension of credit that will be made to the same consumer and secured by the same dwelling, the creditor knows or has reason to know of the simultaneous loan. Alternatively, if the creditor has verified information that the down payment source is the consumer's existing assets, the creditor would be under no further obligation to determine whether a simultaneous loan will be extended at or before consummation.

The Bureau believes that comment 43(c)(2)(iv)-2 provides clear guidance on the “knows or has reason to know” standard, with the addition of language clarifying that the creditor is not obligated to investigate beyond reasonable underwriting policies and procedures to determine whether a simultaneous loan will be extended at or before consummation of the covered transaction.

The Bureau considers the provision to be an accurate and appropriate implementation of the statute. Proposed § 226.43(c)(2)(iv) and associated commentary are adopted substantially as proposed, in renumbered § 1026.43(c)(2)(iv), with the addition of the language discussed above to comment 43(c)(2)(iv)-2 and other minor clarifying changes. Comment 43(c)(2)(iv)-3 now includes language making clear that if the consummation of the loan transaction is extended past the traditional closing, any simultaneous loan originated after that traditional closing may still be interpreted as having occurred “at” consummation. In addition, as discussed below, comment 43(c)(2)(iv)-4, Verification of simultaneous loans, has been grouped with other verification comments, in comment 43(c)(3)-4.

43(c)(2)(v)

As discussed above, TILA section 129C(a)(1) and (3) requires creditors to consider and verify mortgage-related obligations as part of the ability-to-repay determination “according to [the loan's] terms, and all applicable taxes, insurance (including mortgage guarantee insurance), and assessments.” Section 1026.34(a)(4), which was added by the 2008 HOEPA Final Rule, also requires creditors to consider mortgage-related obligations in assessing repayment ability. See the section-by-section analysis of § 1026.43(b)(8) for a discussion of the Bureau's interpretation of “mortgage-related obligations” and the definition adopted in the final rule.

The Board proposed to require creditors to consider the consumer's monthly payment for mortgage-related obligations as part of the repayment ability determination. Proposed comment 43(c)(2)(v)-1 explained that mortgage-related obligations must be included in the creditor's determination of repayment ability regardless of whether the amounts are included in the monthly payment or whether there is an escrow account established.

Proposed comment 43(c)(2)(v)-2 clarified that, in considering mortgage-related obligations that are not paid monthly, the creditor may look to widely accepted governmental or non-governmental standards to determine the pro rata monthly payment amount. The Board solicited comment on operational difficulties creditors may encounter when complying with this monthly requirement, and whether additional guidance was necessary.

Proposed comment 43(c)(2)(v)-3 explained that estimates of mortgage-related obligations should be based upon information known to the creditor at the time the creditor underwrites the mortgage obligation. This comment explained that information is known if it is “reasonably available” to the creditor at the time of underwriting the loan, and cross-referenced current comment 17(c)(2)(i)-1 for guidance regarding “reasonably available.” Proposed comment 43(c)(2)(v)-3 further clarified that, for purposes of determining repayment ability under proposed § 226.43(c), the creditor would not need to project potential changes.

Proposed comment 43(c)(2)(v)-4 stated that creditors must make the repayment ability determination required under proposed § 226.43(c) based on information verified from reasonably reliable records. This comment explained that guidance regarding verification of mortgage-related obligations could be found in proposed comments 43(c)(3)-1 and -2, which discuss verification using third-party records.

The Board solicited comment on any special concerns regarding the requirement to document certain mortgage-related obligations, for example, ground rent or leasehold payments, or special assessments. The Board also solicited comment on whether it should provide that the HUD-1 or -1A or a successor form could serve as verification of mortgage-related obligations reflected by the form, where a legal obligation exists to complete the form accurately.

Industry commenters and consumer advocates generally supported including consideration and verification of mortgage-related obligations in the ability-to-repay determination. Several industry commenters asked that the Bureau provide creditors more flexibility in considering and verifying mortgage-related obligations. They suggested that a reasonable and good faith determination be deemed sufficient, rather than use of all underwriting standards in any particular government or non-government handbook. Community banks asserted that flexible standards were necessary to meet their customers' needs. Some consumer advocates suggested that creditors be permitted to draw on only widely accepted standards that have been validated by experience or sanctioned by a government agency.

Some industry commenters asked for more guidance on how to calculate pro rata monthly payment amounts and estimated property taxes. One industry commenter asked that creditors be permitted to use pro rata monthly payment amounts for special assessments, not quarterly or yearly amounts. The commenter requested that estimates of common assessments be permitted. This commenter also recommended that creditors be permitted to verify the amount of common assessments with information provided by the consumer. One commenter noted that verification using HUD-1 forms should be permitted because there is a legal obligation to complete the HUD-1 accurately.

The Bureau is adopting the rule as proposed. For the reasons discussed below, the Bureau concludes that a creditor should consider the consumer's monthly payment for mortgage-related obligations in determining the consumer's ability to repay, pursuant to § 1026.43(c)(1). As commenters confirmed, obligations related to the mortgage may affect the consumer's ability to satisfy the obligation to make recurring payments of principal and interest. The Bureau also agrees with the argument raised by many commenters that the failure to account consistently for these obligations during the subprime crisis harmed many consumers. Thus, the Bureau has determined that it is appropriate to adopt § 1026.43(c)(2)(v) as proposed. However, the Bureau believes that additional guidance will facilitate compliance. As explained below, the Bureau has expanded on the proposed commentary language to provide additional clarity and illustrative examples.

The final version of comment 43(c)(2)(v)-1 is substantially similar to the language as proposed. As discussed under § 1026.43(b)(8) above, the Bureau is revising the language related to insurance premiums to provide additional clarity. The modifications to the language in proposed comment 43(c)(2)(v)-1 conform to the language adopted under § 1026.43(b)(8) and the related commentary. Furthermore, the final version of comment 43(c)(2)(v)-1 contains additional explanation regarding the determination of the consumer's monthly payment, and provides additional illustrative examples to clarify further the requirements of § 1026.43(c)(2)(v). For example, assume that a consumer will be required to pay mortgage insurance premiums, as defined by § 1026.43(b)(8), on a monthly, annual, or other basis after consummation. Section 1026.43(c)(2)(v) includes these recurring mortgage insurance payments in the evaluation of the consumer's monthly payment for mortgage-related obligations. However, if the consumer will incur a one-time fee or charge for mortgage insurance or similar purposes, such as an up-front mortgage insurance premium imposed at consummation, § 1026.43(c)(2)(v) does not include this up-front mortgage insurance premium in the evaluation of the consumer's monthly payment for mortgage-related obligations.

As discussed under § 1026.43(b)(8) above, several commenters discussed the importance of including homeowners association dues and similar obligations in the determination of ability to repay. These commenters argued, and the Bureau agrees, that recurring financial obligations payable to community governance associations, such as homeowners association dues, should be taken into consideration in determining whether a consumer has the ability to repay the obligation. The Bureau recognizes the practical problems that may arise with including obligations such as these in the evaluation of the consumer's monthly payment for mortgage-related obligations. Commenters identified issues stemming from difficulties which may arise in calculating, estimating, and verifying these obligations. Based on this feedback, the Bureau has determined that additional clarification is necessary. As adopted, comment 43(c)(2)(v)-2 clarifies that creditors need not include payments to community governance associations if such obligations are fully satisfied at or before consummation by the consumer. This comment further clarifies that § 1026.43(c)(2)(v) does not require the creditor to include these payments in the evaluation of the consumer's monthly payment for mortgage-related obligations if the consumer does not pay the fee directly at or before consummation, and instead finances the obligation. In these cases, the financed obligation will be included in the loan amount, and is therefore already included in the determination of ability to repay pursuant to § 1026.43(c)(2)(iii). However, if the consumer incurs the obligation and will satisfy the obligation with recurring payments after consummation, regardless of whether the obligation is escrowed, § 1026.43(c)(2)(v) requires the creditor to include the obligation in the evaluation of the consumer's monthly payment for mortgage-related obligations. The Bureau has also addressed the concerns raised by commenters related to calculating, estimating, and verifying these obligations in comments 43(c)(2)(v)-4 and -5 and 43(c)(3)-5, respectively.

As discussed under § 1026.43(b)(8) above, one comment letter focused extensively on community transfer fees. The Bureau agrees with the argument, advanced by several commenters, that the entirety of the consumer's ongoing obligations should be included in the determination. A responsible determination of the consumer's ability to repay requires an accounting of such obligations, whether the purpose of the obligation is to satisfy the payment of a community transfer fee or traditional homeowners association dues. An obligation that is not paid in full at or before consummation must be paid after consummation, which may affect the consumer's ability to repay ongoing obligations. Thus, comment 43(c)(2)(v)-2 clarifies that community transfer fees are included in the determination of the consumer's monthly payment for mortgage-related obligations if such fees are paid on a recurring basis after consummation. Additionally, the Bureau believes that a creditor is not required to include community transfer fees that are imposed on the seller, as many community transfer fees are, in the ability-to-repay calculation.

In response to the request for feedback in the proposed rule, several commenters addressed the proposed treatment of special assessments. Unlike community transfer fees, which are generally identified in the deed or master community plan, creditors may encounter difficulty determining whether special assessments exist. Special assessments are often imposed in response to some urgent or unexpected need. Consequently, neither the creditor nor the community governance association may be able to predict the frequency and magnitude of special assessments. However, this difficulty does not exist for special assessments that are known at the time of underwriting. Known special assessments, which the buyer must pay and which may be significant, may affect the consumer's ability to repay the obligation. Thus, comment 43(c)(2)(v)-3 clarifies that the creditor must include special assessments in the evaluation of the consumer's monthly payment for mortgage-related obligations if such fees are paid by the consumer on a recurring basis after consummation, regardless of whether an escrow is established for these fees. For example, if a homeowners association imposes a special assessment that the consumer will have to pay in full at or before consummation, § 1026.43(c)(2)(v) does not include the special assessment in the evaluation of the consumer's monthly payment for mortgage-related obligations. Section 1026.43(c)(2)(v) does not require a creditor to include special assessments in the evaluation of the consumer's monthly payment for mortgage-related obligations if the special assessments are imposed as a one-time charge. For example, if a homeowners association imposes a special assessment that the consumer will have to satisfy in one payment, § 1026.43(c)(2)(v) does not include this one-time special assessment in the evaluation of the consumer's monthly payment for mortgage-related obligations. However, if the consumer will pay the special assessment on a recurring basis after consummation, regardless of whether the consumer's payments for the special assessment are escrowed, § 1026.43(c)(2)(v) requires the creditor to include this recurring special assessment in the evaluation of the consumer's monthly payment for mortgage-related obligations. Comment 43(c)(2)(v)-3 also includes several other examples illustrating this requirement.

The Bureau agrees that clear and detailed guidance regarding determining pro rata monthly payments of mortgage-related obligations should be provided. However, the Bureau believes that it is important to strike a balance between providing clear guidance and providing creditors with the flexibility to serve the evolving mortgage market. The comments identified significant concerns with the use of “widely accepted governmental and non-governmental standards” for purposes of determining the pro rata monthly payment amount for mortgage-related obligations. While commenters generally stated that “widely accepted governmental standards” was an appropriate standard, others commented that “non-governmental standards” may not be sufficiently clear. The Bureau believes that “governmental standards” could be relied on to perform pro rata calculations of monthly mortgage related obligations because such standards provide detailed and comprehensive guidance and are frequently revised to adapt to the needs of the evolving residential finance market. However, the comments noted that “non-governmental standards” is not sufficiently descriptive to illustrate clearly how to calculate pro rata monthly payments. Additionally, the Bureau believes that clear guidance is also needed to address the possibility that a particular government program may not specifically describe how to calculate pro rata monthly payment amounts for mortgage-related obligations. Thus, the Bureau believes that it is appropriate to revise and further develop the concept of “widely accepted governmental and non-governmental standards.”

Based on this feedback, the Bureau has revised and expanded the comment clarifying how to calculate pro rata monthly mortgage obligations. As adopted, comment 43(c)(2)(v)-4 provides that, if the mortgage loan is originated pursuant to a governmental program, the creditor may determine the pro rata monthly amount of the mortgage-related obligation in accordance with the specific requirements of that program. If the mortgage loan is originated pursuant to a government program that does not contain specific standards for determining the pro rata monthly amount of the mortgage-related obligation, or if the mortgage loan is not originated pursuant to a government program, the creditor complies with § 1026.43(c)(2)(v) by dividing the total amount of a particular non-monthly mortgage-related obligation by no more than the number of months from the month that the non-monthly mortgage-related obligation last was due prior to consummation until the month that the non-monthly mortgage-related obligation next will be due after consummation. Comment 43(c)(2)(v)-4 also includes several examples which illustrate the conversion of non-monthly obligations into monthly, pro rata payments. For example, assume that a consumer applies for a mortgage loan on February 1st. Assume further that the subject property is located in a jurisdiction where property taxes are paid in arrears annually on the first day of October. The creditor complies with § 1026.43(c)(2)(v) by determining the annual property tax amount owed in the prior October, dividing the amount by 12, and using the resulting amount as the pro rata monthly property tax payment amount for the determination of the consumer's monthly payment for mortgage-related obligations. The creditor complies even if the consumer will likely owe more in the next year than the amount owed the prior October because the jurisdiction normally increases the property tax rate annually, provided that the creditor does not have knowledge of an increase in the property tax rate at the time of underwriting.

The Bureau is adopting comment 43(c)(2)(v)-5 in a form that is substantially similar to the version proposed. One industry commenter was especially concerned about estimating costs for community governance organizations, such as cooperative, condominium, or homeowners associations. This commenter noted that, because of industry concerns about TILA liability, many community governance organizations refuse to provide estimates of association expenses absent agreements disclaiming association liability. This commenter expressed concern that the ability-to-repay requirements would make community governance organizations less likely to provide estimates of association expenses, which would result in mortgage loan processing delays. The Bureau does not believe that the ability-to-repay requirements will lead to difficulties in exchanging information between creditors and associations because the ability-to-repay requirements generally apply only to creditors, as defined under § 1026.2(a)(17). However, the Bureau recognizes that consumers may be harmed if mortgage loan transactions are needlessly delayed by concerns arising from the ability-to-repay requirements. Thus, the Bureau has decided to address these concerns by adding several examples to comment 43(c)(2)(v)-5 illustrating the requirements of § 1026.43(c)(2)(v). For example, the creditor complies with § 1026.43(c)(2)(v) by relying on an estimate of mortgage-related obligations prepared by the homeowners association. In accordance with the guidance provided under comment 17(c)(2)(i)-1, the creditor need only exercise due diligence in determining mortgage-related obligations, and complies with § 1026.43(c)(2)(v) by relying on the representations of other reliable parties in preparing estimates. Or, assume that the homeowners association has imposed a special assessment on the seller, but the seller does not inform the creditor of the special assessment, the homeowners association does not include the special assessment in the estimate of expenses prepared for the creditor, and the creditor is unaware of the special assessment. The creditor complies with § 1026.43(c)(2)(v) if it does not include the special assessment in the determination of mortgage-related obligations. The creditor may rely on the representations of other reliable parties, in accordance with the guidance provided under comment 17(c)(2)(i)-1.

43(c)(2)(vi)

TILA section 129C(a)(1) and (3) requires creditors to consider “current obligations” as part of an ability-to-repay determination. Proposed § 226.43(c)(2)(vi) would have implemented the requirement under TILA section 129C(a)(1) and (3) by requiring creditors to consider current debt obligations. Proposed comment 43(c)(2)(vi)-1 would have specified that current debt obligations creditors must consider include, among other things, alimony and child support. The Bureau believes that it is reasonable to consider child support and alimony as “debts” given that the term “debt” is not defined in the statute. However, the Bureau understands that while alimony and child support are obligations, they may not be considered debt obligations unless and until they are not paid in a timely manner. Therefore, § 1026.43(c)(2)(vi) specifies that creditors must consider current debt obligations, alimony, and child support to clarify that alimony and child support are included whether or not they are paid in a timely manner.

Proposed comment 43(c)(2)(vi)-1 would have referred creditors to widely accepted governmental and non-governmental underwriting standards in determining how to define “current debt obligations.” The proposed comment would have given examples of current debt obligations, such as student loans, automobile loans, revolving debt, alimony, child support, and existing mortgages. The Board solicited comment on proposed comment 43(c)(2)(vi)-1 and on whether more specific guidance should be provided to creditors. Commenters generally supported giving creditors significant flexibility and did not encourage the Bureau to adopt more specific guidance. Because the Bureau believes that a wide range of criteria and guidelines for considering current debt obligations will contribute to reasonable, good faith ability-to-repay determinations, comment 43(c)(2)(vi)-1 as adopted preserves the flexible approach of the Board's proposed comment. The comment gives examples of current debt obligations but does not provide an exhaustive list. The comment therefore preserves substantial flexibility for creditors to develop their own underwriting guidelines regarding consideration of current debt obligations. Reference to widely accepted governmental and non-governmental underwriting standards has been omitted, as discussed above in the section-by-section analysis of § 1026.43(c).

The Board also solicited comment on whether additional guidance should be provided regarding consideration of debt obligations that are almost paid off. Commenters generally stated that creditors should be required to consider obligations that are almost paid off only if they affect repayment ability. The Bureau agrees that many different standards for considering obligations that are almost paid off could lead to reasonable, good faith ability-to-repay determinations. As adopted, comment 43(c)(2)(vi)-1 includes additional language clarifying that creditors have significant flexibility to consider current debt obligations in light of attendant facts and circumstances, including that an obligation is likely to be paid off soon after consummation. As an example, comment 43(c)(2)(vi)-1 states that a creditor may take into account that an existing mortgage is likely to be paid off soon after consummation because there is an existing contract for sale of the property that secures that mortgage.

The Board also solicited comment on whether additional guidance should be provided regarding consideration of debt obligations in forbearance or deferral. Several commenters, including both creditors and consumer advocates, supported requiring creditors to consider obligations in forbearance or deferral. At least one large creditor objected to requiring creditors to consider such obligations in all cases. The Bureau believes that many different standards for considering obligations in forbearance or deferral could lead to reasonable, good faith determinations of ability to repay. As adopted, comment 43(c)(2)(vi)-1 therefore includes additional language clarifying that creditors should consider whether debt obligations in forbearance or deferral at the time of underwriting are likely to affect a consumer's ability to repay based on the payment for which the consumer will be liable upon expiration of the forbearance or deferral period and other relevant facts and circumstances, such as when the forbearance or deferral period will expire.

Parts of proposed comment 43(c)(2)(vi)-1 and proposed comment 43(c)(2)(vi)-2 would have provided guidance on verification of current debt obligations. All guidance regarding verification has been moved to the commentary to § 1026.43(c)(3) and is discussed below in the section-by-section analysis of that provision.

The Board solicited comment on whether it should provide guidance on consideration of current debt obligations for joint applicants. Commenters generally did not comment on consideration of current debt obligations for joint applicants. One trade association commenter stated that joint applicants should be subject to the same standards as individual applicants. Because the Bureau believes that the current debt obligations of all joint applicants must be considered to reach a reasonable, good faith determination of ability to repay, the Bureau is adopting new comment 43(c)(2)(vi)-2. New comment 43(c)(2)(vi)-2 clarifies that when two or more consumers apply for credit as joint obligors, a creditor must consider the debt obligations of all such joint applicants. The comment also explains that creditors are not required to consider the debt obligations of a consumer acting merely as surety or guarantor. Finally, the comment clarifies that the requirements of § 1026.43(c)(2)(vi) do not affect various disclosure requirements.

43(c)(2)(vii)

TILA section 129C(a)(3) requires creditors to consider the consumer's monthly debt-to-income ratio or residual income the consumer will have after paying non-mortgage debt and mortgage-related obligations, as part of the ability-to-repay determination under TILA section 129C(a)(1). This provision is consistent with the 2008 HOEPA Final Rule, which grants a creditor in a high-cost or higher-priced mortgage loan a presumption of compliance with the requirement that the creditor assess repayment ability if, among other things, the creditor considers the consumer's debt-to-income ratio or residual income. See § 1026.34(a)(4)(iii)(C), (b)(1). Existing comment 34(a)(4)(iii)(C)-1 provides that creditors may look to widely accepted governmental and non-governmental underwriting standards in defining “income” and “debt” including, for example, those set forth in the FHA Handbook on Mortgage Credit Analysis for Mortgage Insurance on One-to-Four Unit Mortgage Loans.

Proposed § 226.43(c)(2)(vii) would have implemented TILA section 129C(a)(3) by requiring creditors, as part of the repayment ability determination, to consider the consumer's monthly debt-to-income ratio or residual income. Proposed comment 43(c)(2)(vii)-1 would have cross-referenced § 226.43(c)(7), regarding the definitions and calculations for the monthly debt-to-income and residual income. Consistent with the 2008 HOEPA Final Rule, the proposed rule would have provided creditors flexibility to determine whether to use a debt-to-income ratio or residual income metric in assessing the consumer's repayment ability. As the Board noted, if one of these metrics alone holds as much predictive power as the two together, then requiring creditors to use both metrics could reduce credit access without an offsetting increase in consumer protection. 76 FR 27390, 27424-25 (May 11, 2011), citing 73 FR 44550 (July 30, 2008). The proposed rule did not specifically address creditors' use of both metrics if such an approach would provide incremental predictive power of assessing a consumer's repayment ability. However, as discussed above in the section-by-section analysis of § 1026.43(c), the Board's proposed comment 43(c)-1 would have provided that, in evaluating the consumer's repayment ability under § 226.43(c), creditors may look to widely accepted governmental or non-governmental underwriting standards, such as the FHA Handbook on Mortgage Credit Analysis for Mortgage Insurance on One-to-Four Unit Mortgage Loans, consistent with existing comment 34(a)(4)(iii)(C)-1.

In response to the proposed rule, industry commenters and consumer advocates generally supported including consideration of the debt-to-income ratio or residual income in the ability-to-repay determination. Several industry commenters asked that the Bureau provide creditors more flexibility in considering and verifying the debt-to-income ratio or residual income. They suggested that a reasonable and good faith determination be deemed sufficient, rather than use of all underwriting standards in any particular government or non-government handbook. Community banks asserted that flexible standards are necessary to meet their customers' needs. Some consumer advocates suggested that creditors be permitted only to draw on widely accepted standards that have been validated by experience or sanctioned by a government agency. They argued that more specific standards would help ensure safe and sound underwriting criteria, higher compliance rates, and a larger number of performing loans.

Section 1026.43(c)(2)(vii) adopts the Board's proposal by requiring a creditor making the repayment determination under § 1026.43(c)(1) to consider the consumer's monthly debt-to-income ratio or residual income, in accordance with § 1026.43(c)(7). The Bureau believes that a flexible approach to evaluating a consumer's debt-to-income ratio or residual income is appropriate because stricter guidelines may limit access to credit and create fair lending problems. Broad guidelines will provide creditors necessary flexibility to serve the whole of the mortgage market effectively and responsibly. Accordingly, the final rule sets minimum underwriting standards while providing creditors with flexibility to use their own reasonable guidelines in making the repayment ability determination required by § 1026.43(c)(1). Moreover, and as in the 2008 HOEPA Final Rule, the approach would provide creditors flexibility to determine whether to use a debt-to-income ratio or residual income, or both, in assessing a consumer's repayment ability.

As discussed above in the section-by-section analysis of § 1026.43(c), the Bureau is not finalizing the Board's proposed comment 43(c)-1 regarding the use of widely accepted governmental or non-governmental underwriting standards in evaluating the consumer's repayment ability. Instead, for the reasons discussed above, comment 43(c)(2)-1 provides that the rule and commentary permit creditors to adopt reasonable standards for evaluating factors in underwriting a loan, such as whether to classify particular inflows or obligations as “income” or “debt,” and that, in evaluating a consumer's repayment ability, a creditor may look to governmental underwriting standards. See section-by-section analysis of § 1026.43(c)(2).

The Bureau believes a flexible approach to evaluating debt and income is appropriate in making the repayment ability determination under § 1026.43(c). However, for the reasons discussed below, the Bureau believes a quantitative standard for evaluating a consumer's debt-to-income ratio should apply to loans that are “qualified mortgages” that receive a safe harbor or presumption of compliance with the repayment ability determination under § 1026.43(c). For a discussion of the quantitative debt-to-income standard that applies to qualified mortgages pursuant to § 1026.43(e)(2) and the rationale for applying a quantitative standard in the qualified mortgage space, see the section-by-section analysis of § 1026.43(e)(2).

43(c)(2)(viii)

TILA section 129C(a)(1) and (3) requires creditors to consider credit history as part of the ability-to-repay determination. Proposed § 226.43(c)(2)(viii) would have implemented the requirement under TILA section 129C(a)(1) and (3) by adopting the statutory requirement that creditors consider credit history as part of an ability-to-repay determination. Proposed comment 43(c)(2)(viii)-1 would have referred creditors to widely accepted governmental and non-governmental underwriting standards to define credit history. The proposed comment would have given examples of factors creditors could consider, such as the number and age of credit lines, payment history, and any judgments, collections, or bankruptcies. The proposed comment also would have referred creditors to credit bureau reports or to nontraditional credit references such as rental payment history or public utility payments.

Commenters generally did not object to the proposed adoption of the statutory requirement to consider credit history as part of ability-to-repay determinations. Commenters generally supported giving creditors significant flexibility in how to consider credit history. Creditors also generally supported clarifying that creditors may look to nontraditional credit references such as rental payment history or public utility payments.

Section 1026.43(c)(2)(viii) is adopted as proposed. Comment 43(c)(2)(viii)-1 as adopted substantially maintains the proposed comment's flexible approach to consideration of credit history. Comment 43(c)(2)(viii)-1 notes that “credit history” may include factors such as the number and age of credit lines, payment history, and any judgments, collections, or bankruptcies. The comment clarifies that the rule does not require creditors to obtain or consider a consolidated credit score or prescribe a minimum credit score that creditors must apply. The comment further clarifies that the rule does not specify which aspects of credit history a creditor must consider or how various aspects of credit history could be weighed against each other or against other underwriting factors. The comment explains that some aspects of a consumer's credit history, whether positive or negative, may not be directly indicative of the consumer's ability to repay and that a creditor therefore may give various aspects of a consumer's credit history as much or as little weight as is appropriate to reach a reasonable, good faith determination of ability to repay. The Bureau believes that this flexible approach is appropriate because of the wide range of creditors, consumers, and loans to which the rule will apply. The Bureau believes that a wide range of approaches to considering credit history will contribute to reasonable, good faith ability-to-repay determinations. As in the proposal, the comment, as adopted, clarifies that creditors may look to non-traditional credit references such as rental payment history or public utility payments, but are not required to do so. Reference to widely accepted governmental and non-governmental underwriting standards has been omitted, as discussed in the section-by-section analysis of § 1026.43(c), above.

Portions of proposed comment 43(c)(2)(viii)-1 discussed verification of credit history. All guidance regarding verification has been moved to the commentary to § 1026.43(c)(3) and is discussed below in the section-by-section analysis of that provision.

Because the Bureau believes that the credit history of all joint applicants must be considered to reach a reasonable, good faith determination of joint applicants' ability to repay, and for conformity with the commentary to § 1026.43(c)(2)(vi) regarding consideration of current debt obligations for multiple applicants, the Bureau is adopting new comment 43(c)(2)(viii)-2 regarding multiple applicants. The comment clarifies that, when two or more consumers apply jointly for credit, the creditor is required by § 1026.43(c)(2)(viii) to consider the credit history of all joint applicants. New comment 43(c)(2)(viii)-2 also clarifies that creditors are not required to consider the credit history of a consumer who acts merely as a surety or guarantor. Finally, the comment clarifies that the requirements of § 1026.43(c)(2)(viii) do not affect various disclosure requirements.

43(c)(3) Verification Using Third-Party Records

TILA section 129C(a)(1) requires that a creditor make a reasonable and good faith determination, based on “verified and documented information,” that a consumer has a reasonable ability to repay the covered transaction. The Board's 2008 HOEPA Final Rule required that a creditor verify the consumer's income or assets relied on to determine repayment ability and the consumer's current obligations under § 1026.34(a)(4)(ii)(A) and (C). Thus, TILA section 129C(a)(1) differs from existing repayment ability rules by requiring a creditor to verify information relied on in considering the consumer's ability to repay according to the considerations required under TILA section 129C(a)(3), which are discussed above in the section-by-section analysis of § 1026.43(c)(2).

The Board's proposal would have implemented TILA section 129C(a)(1)'s general requirement to verify a consumer's repayment ability in proposed § 226.43(c)(3), which required that a creditor verify a consumer's repayment ability using reasonably reliable third-party records, with two exceptions. Under the first exception, proposed § 226.43(c)(3)(i) provided that a creditor may orally verify a consumer's employment status, if the creditor subsequently prepares a record of the oral employment status verification. Under the second exception, proposed § 226.43(c)(3)(ii) provided that, in cases where a creditor relies on a consumer's credit report to verify a consumer's current debt obligations and the consumer's application states a current debt obligation not shown in the consumer's credit report, the creditor need not independently verify the additional debt obligation, as reported. Proposed comment 43(c)(3)-1 clarified that records a creditor uses to verify a consumer's repayment ability under proposed § 226.43(c)(3) must be specific to the individual consumer. Records regarding, for example, average incomes in the consumer's geographic location or average incomes paid by the consumer's employer would not be specific to the individual consumer and are not sufficient.

Proposed comment 43(c)(3)-2 provided that a creditor may obtain third-party records from a third-party service provider, as long as the records are reasonably reliable and specific to the individual consumer. As stated in § 1026.43(c)(3), the standard for verification is that the creditor must use “reasonably reliable third-party records,” which is fulfilled for reasonably reliable documents, specific to the consumer, provided by a third-party service provider. Also, proposed comment 43(c)(3)-2 clarified that a creditor may obtain third-party records, for example, payroll statements, directly from the consumer, again as long as the records are reasonably reliable.

The Board also solicited comment on whether any documents or records prepared by the consumer and not reviewed by a third party appropriately could be considered in determining repayment ability, for example, because a particular record provides information not obtainable using third-party records. In particular, the Board solicited comment on methods currently used to ensure that documents prepared by self-employed consumers (such as a year-to-date profit and loss statement for the period after the period covered by the consumer's latest income tax return, or an operating income statement prepared by a consumer whose income includes rental income) are reasonably reliable for use in determining repayment ability.

Commenters generally supported the Board's proposal to implement the Dodd-Frank Act's verification requirements. Consumer groups generally found the proposal to be an accurate implementation of the statute and posited that the proposal would provide much-needed protection for consumers. Industry commenters generally also supported the proposal, noting that most underwriters already engaged in similarly sound underwriting practices. Some industry commenters noted that verifying a consumer's employment status imposes a burden upon the consumer's employer as well, however the Bureau has concluded that the oral verification provision provided by § 1026.43(c)(3)(ii), discussed below, alleviates such concerns.

The Bureau is adopting § 1026.43(c)(3) substantially as proposed, with certain clarifying changes which are described below. The final rule also adds new comment 43(c)(3)-3. In addition, for organizational purposes, the final rule generally adopts proposed comments 43(c)(2)(iv)-4, 43(c)(2)(v)-4, 43(c)(2)(vi)-1, 43(c)(2)(viii)-1, and 43(c)(2)(ii)-2 in renumbered comments 43(c)(3)-4 through -8 with revisions as discussed below. These changes and additions to § 1026.43(c)(3) and its commentary are discussed below.

First, the final rule adds a new § 1026.43(c)(3)(i), which provides that, for purposes of § 1026.43(c)(2)(i), a creditor must verify a consumer's income or assets in accordance with § 1026.43(c)(4). This is an exception to the general rule in § 1026.43(c)(3) that a creditor must verify the information that the creditor relies on in determining a consumer's repayment ability under § 1026.43(c)(2) using reasonably reliable third-party records. Because of this new provision, proposed § 226.43(c)(3)(i) and (ii) are adopted as proposed in § 1026.43(c)(3)(ii) and (iii), with minor technical revisions. In addition, the Bureau is adopting proposed comments 43(c)(3)-1 and -2 substantially as proposed with revisions to clarify that the guidance applies to both § 1026.43(c)(3) and (c)(4).

The Bureau is adding new comment 43(c)(3)-3 to clarify that a credit report generally is considered a reasonably reliable third-party record. The Board's proposed comment 43(c)(2)(vi)-2 stated, among other things, that a credit report is deemed a reasonably reliable third-party record under proposed § 226.43(c)(3). Commenters did not address that aspect of proposed comment 43(c)(2)(vi)-2. The Bureau believes credit reports are generally reasonably reliable third-party records for verification purposes. Comment 43(c)(3)-3 also explains that a creditor is not generally required to obtain additional reasonably reliable third-party records to verify information contained in a credit report, as the report itself is the means of verification. Likewise, comment 43(c)(3)-3 explains that if information is not included in the credit report, then the credit report cannot serve as a means of verifying that information. The comment further explains, however, that if the creditor may know or have reason to know that a credit report is not reasonably reliable, in whole or in part, then the creditor complies with § 1026.43(c)(3) by disregarding such inaccurate or disputed items or reports. The creditor may also, but is not required, to obtain other reasonably reliable third-party records to verify information with respect to which the credit report, or item therein, may be inaccurate. The Bureau believes that this guidance strikes the appropriate balance between acknowledging that in many cases, a credit report is a reasonably reliable third-party record for verification and documentation for many creditors, but also that a credit report may be subject to a fraud alert, extended alert, active duty alert, or similar alert identified in 15 U.S.C. 1681c-1, or may contain debt obligations listed on a credit report is subject to a statement of dispute pursuant to 15 U.S.C. 1681i(b). Accordingly, for the reasons discussed above, the Bureau is adopting new comment 43(c)(3)-3.

As noted above, the Bureau is adopting proposed comment 43(c)(2)(iv)-4 as comment 43(c)(3)-4 for organizational purposes. The Board proposed comment 43(c)(2)(iv)-4 to explain that although a creditor could use a credit report to verify current obligations, the report would not reflect a simultaneous loan that has not yet been consummated or has just recently been consummated. Proposed comment 43(c)(2)(iv)-2 clarified that if the creditor knows or has reason to know that there will be a simultaneous loan extended at or before consummation, then the creditor may verify the simultaneous loan by obtaining third-party verification from the third-party creditor of the simultaneous loan. The proposed comment provided, as an example, that the creditor may obtain a copy of the promissory note or other written verification from the third-party creditor in accordance with widely accepted governmental or non-governmental standards. In addition, proposed comment 43(c)(2)(iv)-2 cross-referenced comments 43(c)(3)-1 and -2, which discuss verification using third-party records. The Bureau generally did not receive comment with respect to this proposed comment; however, at least one commenter supported the example that a promissory note would serve as appropriate documentation for verifying a simultaneous loan. The Bureau is adopting proposed comment 43(c)(2)(iv)-4 as comment 43(c)(3)-4 with the following amendment. For consistency with other aspects of the rule, comment 43(c)(3)-4 does not include the Board's proposed reference to widely accepted governmental or non-governmental standards.

The Board proposed comment 43(c)(2)(v)-4, which stated that creditors must make the repayment ability determination required under proposed § 226.43(c) based on information verified from reasonably reliable records. The Board solicited comment on any special concerns regarding the requirement to document certain mortgage-related obligations, for example, ground rent or leasehold payments, or special assessments. The Board also solicited comment on whether it should provide that the HUD-1 or -1A or a successor form could serve as verification of mortgage-related obligations reflected by the form, where a legal obligation exists to complete the HUD-1 or -1A accurately. To provide additional clarity, the Bureau is moving guidance that discusses verification, including proposed comment 43(c)(2)(v)-4, as part of the section-by-section analysis of, and commentary to, § 1026.43(c)(3). Additional comments from the Board's proposal with respect to mortgage-related obligations are in the section-by-section analysis of § 1026.43(c)(2)(v), above.

Industry commenters and consumer advocates generally supported including consideration and verification of mortgage-related obligations in the ability-to-repay determination. Several industry commenters asked that the Bureau provide creditors more flexibility in considering and verifying mortgage-related obligations. Several consumer advocate commenters discussed the importance of verifying mortgage-related obligations based on reliable records, noting that inadequate, or non-existent, verification measures played a significant part in the subprime crisis. Industry commenters agreed that verification was appropriate, but these commenters also stressed the importance of clear and detailed guidance. Several commenters were concerned about the meaning of “reasonably reliable records” in the context of mortgage-related obligations. Some commenters asked the Bureau to designate certain items as reasonably reliable, such as taxes referenced in a title report, statements of common expenses provided by community associations, or items identified in the HUD-1 or HUD-1A.

The Bureau is adopting proposed comment 43(c)(2)(v)-4 as comment 43(c)(3)-5 with revision to provide further explanation of its approach to verifying mortgage-related obligations. While the reasonably reliable standard contains an element of subjectivity, the Bureau concludes that this flexibility is necessary. The Bureau believes that it is important to craft a regulation with the flexibility to accommodate an evolving mortgage market. The Bureau determines that the reasonably reliable standard is appropriate in this context given the nature of the items that are defined as mortgage-related obligations. Thus, comment 43(c)(3)-5 incorporates by reference comments 43(c)(3)-1 and -2. Mortgage-related obligations refer to a limited set of charges, such as property taxes and lease payments, which a creditor can generally verify from an independent or objective source. Thus, in the context of mortgage-related obligations this standard provides certainty while being sufficiently flexible to adapt as underwriting practices develop over time.

To address the concerns raised by several commenters, the Bureau is providing further clarification in 43(c)(3)-5 to provide detailed guidance and several examples illustrating these requirements. For example, comment 43(c)(3)-5 clarifies that records are reasonably reliable for purposes § 1026.43(c)(2)(v) if the information in the record was provided by a governmental organization, such as a taxing authority or local government. Comment 43(c)(3)-5 also explains that a creditor complies with § 1026.43(c)(2)(v) if it relies on, for example, homeowners association billing statements provided by the seller to verify other information in a record provided by an entity assessing charges, such as a homeowners association. Comment 43(c)(3)-5 further illustrates that records are reasonably reliable if the information in the record was obtained from a valid and legally executed contract, such as a ground rent agreement. Comment 43(c)(3)-5 also clarifies that other records may be reasonably reliable if the creditor can demonstrate that the source provided the information objectively.

The Board's proposal solicited comment regarding whether the HUD-1, or similar successor document, should be considered a reasonably reliable record. The Board noted, and commenters confirmed, that the HUD-1, HUD-1A, or successor form might be a reasonably reliable record because a legal obligation exists to complete the form accurately. Although the Bureau agrees with these considerations, the Bureau does not believe that a document provided in final form at consummation, such as the HUD-1, should be used for the purposes of determining ability to repay pursuant to § 1026.43(c)(2)(v). The Bureau expects the ability-to-repay determination to be conducted in advance of consummation. It therefore may be impractical for a creditor to rely on a document that is produced in final form at, or shortly before, consummation for verification purposes. The Bureau is also concerned that real estate transactions may be needlessly disrupted or delayed if creditors delay determining the consumer's ability to repay until the HUD-1, or similar successor document, is prepared. Given these concerns, and strictly as a matter of policy, the Bureau does not wish to encourage the use of the HUD-1, or similar successor document, for the purposes of determining a consumer's ability to repay, and the Bureau is not specifically designating the HUD-1 as a reasonably reliable record in either § 1026.43(c)(2)(v) or related commentary, such as comment 43(c)(3)-5. However, the Bureau acknowledges that the HUD-1, HUD-1A, or similar successor document may comply with § 1026.43(c)(3).

The Board proposed comment 43(c)(2)(vi)-1, which discussed both consideration and verification of current debt obligations. The Bureau discusses portions of proposed comment 43(c)(2)(vi)-1, regarding consideration of current debt obligations, in the section-by-section analysis of § 1026.43(c)(2)(vi). As noted above, for organizational purposes and to provide additional clarity, however, the Bureau is moving guidance that discusses verification, including portions of proposed comment 43(c)(2)(vi)-1, as part of the commentary to § 1026.43(c)(3). With respect to verification, proposed comment 43(c)(2)(vi)-1 stated that: (1) In determining how to verify current debt obligations, a creditor may look to widely accepted governmental and nongovernmental underwriting standards; and (2) a creditor may, for example, look to credit reports, student loan statements, automobile loan statements, credit card statements, alimony or child support court orders and existing mortgage statements. Commenters did not provide the Bureau with significant comment with respect to this proposal, although at least one large bank commenter specifically urged the Bureau to allow creditors to verify current debt obligations using a credit report. For the reasons discussed below, the Bureau is adopting, in relevant part, proposed comment 43(c)(2)(vi)-1 as comment 43(c)(3)-6. The Bureau believes that the proposed guidance regarding verification using statements and orders related to individual obligations could be misinterpreted as implying that credit reports are not sufficient verification of current debt obligations and that creditors must obtain statements and other documentation pertaining to each individual obligation. Comment 43(c)(3)-6 therefore explains that a creditor is not required to further verify the existence or amount of the obligation listed in a credit report, absent circumstances described in comment 43(c)(3)-3. The Bureau believes that a credit report is a reasonably reliable third-party record and is sufficient verification of current debt obligations in most cases. The Bureau also believes that this approach is reflected in the Board's proposal. For example, proposed comment 43(c)(2)(vi)-2 stated that a credit report is a reasonably reliable third-party record; and proposed § 1026.43(c)(3)(ii) indicated that a creditor could rely on a consumer's credit report to verify a consumer's current debt obligations. Unlike proposed comment 43(c)(2)(vi)-1, comment 43(c)(3)-6 does not include reference to widely accepted governmental and nongovernmental underwriting standards for consistency with the amendments in other parts of the rule. To understand the Bureau's approach to verification standards, see the section-by-section analysis, commentary, and regulation text of § 1026.43(c) and § 1026.43(c)(1) above. The Board proposed comment 43(c)(2)(viii)-1, which discussed both the consideration and verification of credit history. The Bureau discusses portions of proposed comment 43(c)(2)(viii)-1, those regarding consideration of credit history, in the section-by-section analysis of § 1026.43(c)(2)(viii). However, the Bureau is moving guidance on verification, including portions of proposed comment 43(c)(2)(viii)-1, to § 1026.43(c)(3) and its commentary. Regarding verification, proposed comment 43(c)(2)(viii)-1 stated that: (1) Creditors may look to widely accepted governmental and nongovernmental underwriting standards to determine how to verify credit history; and (2) a creditor may, for example, look to credit reports from credit bureaus, or other nontraditional credit references contained in third-party documents, such as rental payment history or public utility payments to verify credit history. Commenters did not object to the Board's proposed approach to verification of credit history. The Bureau is adopting this approach under comment 43(c)(3)-7 with the following exception. References to widely accepted governmental and nongovernmental underwriting standards have been removed, as discussed above in the section-by-section analysis of § 1026.43(c). Portions of proposed comment 43(c)(2)(viii)-1 regarding verification are otherwise adopted substantially as proposed in new comment 43(c)(3)-7.

The Board proposed comment 43(c)(2)(ii)-2 to clarify that a creditor also may verify the employment status of military personnel using the electronic database maintained by the DoD to facilitate identification of consumers covered by credit protections provided pursuant to 10 U.S.C. 987, also known as the “Talent Amendment.” The Board also sought additional comment as to whether creditors needed additional flexibility in verifying the employment status of military personnel, such as by verifying the employment status of a member of the military using a Leave and Earnings Statement.

Industry commenters requested that the Bureau provide additional flexibility for creditors to verify military employment. For example, some industry commenters noted that a Leave and Earnings Statement was concrete evidence of employment status and income for military personnel and other industry commenters stated that institutions that frequently work with military personnel have built their own expertise in determining the reliability of using the Leave and Earnings Statement. These commenters argued that using a Leave and Earnings Statement is as reliable a means of verifying the employment status of military personnel as using a payroll statement to verify that employment status of a civilian.

Accordingly, the Bureau is adopting proposed comment 43(c)(2)(ii)-2 as comment 43(c)(3)-8, for organizational purposes, with the following additional clarification. Comment 43(c)(3)-8 clarifies that a creditor may verify military employment by means of a military Leave and Earnings Statement. Therefore, comment 43(c)(3)-8 provides that a creditor may verify the employment status of military personnel by using either a military Leave and Earnings Statement or by using the electronic database maintained by the DoD.

The Board solicited comment on whether a creditor might appropriately verify a consumer's repayment ability using any documents or records prepared by the consumer and not reviewed by a third party, perhaps because a particular record might provide information not obtainable using third-party records. The Bureau did not receive sufficient indication that such records would qualify as reasonably reliable and has thus not added additional regulatory text or commentary to allow for the use of such records. However, a creditor using reasonable judgment nevertheless may determine that such information is useful in verifying a consumer's ability to repay. For example, the creditor may consider and verify a self-employed consumer's income from the consumer's 2013 income tax return, and the consumer then may offer an unaudited year-to-date profit and loss statement that reflects significantly lower expected income in 2014. The creditor might reasonably use the lower 2014 income figure as a more conservative method of underwriting. However, should the unverified 2014 income reflect significantly greater income than the income tax return showed for 2013, a creditor instead would verify this information in accordance with § 1026.43(c)(4).

43(c)(4) Verification of Income or Assets

TILA section 129C(a)(4) requires that a creditor verify amounts of income or assets that a creditor relied upon to determine repayment ability by reviewing the consumer's Internal Revenue Service (IRS) Form W-2, tax returns, payroll statements, financial institution records, or other third-party documents that provide reasonably reliable evidence of the consumer's income or assets. TILA section 129C(a)(4) further provides that, in order to safeguard against fraudulent reporting, any consideration of a consumer's income history must include the verification of income using either (1) IRS transcripts of tax returns; or (2) an alternative method that quickly and effectively verifies income documentation by a third-party, subject to rules prescribed by the Board, and now the Bureau. TILA section 129C(a)(4) is similar to existing § 1026.34(a)(4)(ii)(A), adopted by the Board's 2008 HOEPA Final Rule, although TILA section 129C(a)(4)(B) provides for the alternative methods of third-party income documentation (other than use of an IRS tax-return transcript) to be both “reasonably reliable” and to “quickly and effectively” verify a consumer's income. The Board proposed to implement TILA section 129C(a)(4)(B), adjusting the requirement to (1) require the creditor to use reasonably reliable third-party records, consistent with TILA section 129C(a)(4), rather than the “quickly and effectively” standard of TILA section 129C(a)(4)(B); and (2) provide examples of reasonably reliable records that a creditor can use to efficiently verify income, as well as assets. As discussed in the Board's proposal, the Board proposed these adjustments pursuant to its authority under TILA sections 105(a) and 129B(e). The Board believed that considering reasonably reliable records effectuates the purposes of TILA section 129C(a)(4), is an effective means of verifying a consumer's income, and helps ensure that consumers are offered and receive loans on terms that reasonably reflect their repayment ability.

Industry and consumer group commenters generally supported proposed § 226.43(c)(4) because the proposal would permit a creditor to use a wide variety of documented income and/or asset verification methods, while maintaining the appropriate goal of ensuring accurate verification procedures. Some commenters requested that the Bureau allow a creditor to underwrite a mortgage based on records maintained by a financial institution that show an ability to repay. Specifically, commenters raised concerns with respect to customers who may not have certain documents, such as IRS Form W-2, because of their employment or immigration status. The Bureau expects that § 1026.43(c)(4) provides that the answer to such concerns is self-explanatory; a creditor need not, by virtue of the requirements of § 1026.43(c)(4), require a consumer to produce an IRS Form W-2 in order to verify income. Some industry commenters argued that the Bureau should also permit creditors to verify information for certain applicants, such as the self-employed, by using non-third party reviewed documents, arguing it would reduce costs for consumers. The Bureau does not find such justification to be persuasive, as other widely available documents, such as financial institution records or tax records, could easily serve as means of verification without imposing significant cost to the consumer or creditor. See also the discussion of comment 43(b)(13)(i)-1, addressing third-party records.

Some industry commenters advocated, in addition, that creditors be allowed to employ broader, faster sources of income verification, such as internet-based tools that employ aggregate employer data, or be allowed to rely on statistically qualified models to estimate income or assets. The Bureau, however, believes that permitting creditors to use statistical models or aggregate data to verify income or assets would be contrary to the purposes of TILA section 129C(a)(4). Although the statute uses the words “quickly and effectively,” these words cannot be read in isolation, but should instead be read in context of the entirety of TILA section 129C(a)(4). As noted above, the Bureau believes that TILA section 129C(a)(4) is primarily intended to safeguard against fraudulent reporting and inaccurate underwriting, rather than accelerate the process of verifying a consumer's income, as the statute specifically notes that a creditor must verify a consumer's income history “[i]n order to safeguard against fraudulent reporting.” The Bureau further believes that permitting the use of aggregate data or non-individualized estimates would undermine the requirements to verify a consumer's income and to determine a consumer's ability to repay. Rather, the Bureau believes that the statute requires verification of the amount of income or assets relied upon using evidence of an individual's income or assets.

For substantially the same reasons stated in the Board's proposal, the Bureau is adopting proposed § 226.43(c)(4) and its accompanying commentary substantially as proposed in renumbered § 1026.43(c)(4), with revisions for clarity. Accordingly, the Bureau is implementing TILA section 129C(a)(4) in § 1026.43(c)(4), which provides that a creditor must verify the amounts of income or assets it relies on to determine a consumer's ability to repay a covered transaction using third-party records that provide reasonably reliable evidence of the consumer's income or assets. Section 1026.43(c)(4) further provides a list of illustrative examples of methods of verifying a consumer's income or asserts using reasonably reliable third-party records. Such examples include: (1) Copies of tax returns the consumer filed with the IRS or a State taxing authority; (2) IRS Form W-2s or similar IRS forms for reporting wages or tax withholding; (3) payroll statements, including military Leave and Earnings Statements; (4) financial institution records; (5) records from the consumer's employer or a third party that obtained consumer-specific income information from the consumer's employer; (6) records from a government agency stating the consumer's income from benefits or entitlements, such as a “proof of income” letter issued by the Social Security Administration; (7) check cashing receipts; and (8) receipts from a consumer's use of funds transfer services. The Bureau also believes that by providing such examples of acceptable records, the Bureau enables creditors to quickly and effectively verify a consumer's income, as provided in TILA section 129C(a)(4)(B).

Comment 43(c)(4)-1 clarifies that under § 1026.43(c)(4), a creditor need verify only the income or assets relied upon to determine the consumer's repayment ability. Comment 43(c)(4)-1 also provides an example where the creditor need not verify a consumer's annual bonus because the creditor relies on only the consumer's salary to determine the consumer's repayment ability. This comment also clarifies that comments 43(c)(3)-1 and -2, discussed above, are instructive with respect to income and asset verification.

Comment 43(c)(4)-2 clarifies that, if consumers jointly apply for a loan and each consumer lists his or her income or assets on the application, the creditor need verify only the income or assets the creditor relies on to determine repayment ability. Comment 43(c)(2)(i)-5, discussed above, may also be instructive in cases of multiple applicants.

Comment 43(c)(4)-3 provides that a creditor may verify a consumer's income using an IRS tax-return transcript that summarizes the information in the consumer's filed tax return, another record that provides reasonably reliable evidence of the consumer's income, or both. Comment 43(c)(4)-3 also clarifies that a creditor may obtain a copy of an IRS tax-return transcript or filed tax return from a service provider or from the consumer, and the creditor need not obtain the copy directly from the IRS or other taxing authority. For additional guidance, Comment 43(c)(4)-3 cross-references guidance on obtaining records in comment 43(c)(3)-2.

Finally, comment 43(c)(4)(vi)-1 states that an example of a record from a Federal, State, or local government agency stating the consumer's income from benefits or entitlements is a “proof of income letter” (also known as a “budget letter,” “benefits letter,” or “proof of award letter”) from the Social Security Administration.

As discussed above, the Bureau is adopting § 1026.43(c)(4) as enabling creditors to quickly and effectively verify a consumer's income, as provided in TILA section 129C(a)(4)(B). In addition, for substantially the same rationale as discussed in the Board's proposal, the Bureau is adopting § 1026.43(c)(4) using its authority under TILA section 105(a) to prescribe regulations to carry out the purposes of TILA. One of the purposes of TILA section 129C is to assure that consumers are offered and receive covered transactions on terms that reasonably reflect their ability to repay the loan. See TILA section 129B(a)(2). The Bureau believes that a creditor consulting reasonably reliable records is an effective means of verifying a consumer's income and helps ensure that consumers are offered and receive loans on terms that reasonably reflect their repayment ability. The Bureau further believes that TILA section 129C(a)(4) is intended to safeguard against fraudulent or inaccurate reporting, rather than to accelerate the creditor's ability to verify a consumer's income. Indeed, the Bureau believes that there is a risk that requiring a creditor to use quick methods to verify the consumer's income would undermine the effectiveness of the ability-to-repay requirements by sacrificing thoroughness for speed. The Bureau believes instead that requiring the use of reasonably reliable records effectuates the purposes of TILA section 129C(a)(4) without suggesting that creditors must obtain records or complete income verification within a specific period of time. The Bureau is adopting the examples of reasonably reliable records, proposed by the Board, that a creditor may use to efficiently verify income or assets, because the Bureau believes that it will facilitate compliance by providing clear guidance to creditors.

The Bureau notes that the Board proposal solicited comment on whether it should provide an affirmative defense for a creditor that can show that the amounts of the consumer's income or assets that the creditor relied upon in determining the consumer's repayment ability were not materially greater than the amounts the creditor could have verified using third-party records at or before consummation. Such an affirmative defense, while not specified under TILA, would be consistent with the Board's 2008 HOEPA Final Rule. See § 1026.34(a)(4)(ii)(B). Consumer group commenters generally opposed an affirmative defense, arguing that such an allowance would essentially gut the income and asset verification requirement provided by the rule. Other commenters noted that providing an affirmative defense might result in confusion, and possible litigation, over what the term “material” may mean, and that a rule permitting an affirmative defense would need to define materiality specifically, including from whose perspective materiality should be measured (i.e., the creditor's or the consumer's). Based on the comments received, the Bureau believes that an affirmative defense is not warranted. The Bureau believes that permitting an affirmative defense could result in circumvention of the § 1026.43(c)(4) verification requirement. For the reasons stated, the Bureau is not adopting an affirmative defense for a creditor that can show that the amounts of the consumer's income or assets that the creditor relied upon in determining the consumer's repayment ability were not materially greater than the amounts the creditor could have verified using third-party records at or before consummation.

43(c)(5) Payment Calculation

The Board proposed § 226.43(c)(5) to implement the payment calculation requirements of TILA section 129C(a), as enacted by section 1411 of the Dodd-Frank Act. TILA section 129C(a) contains the general requirement that a creditor determine the consumer's “ability to repay the loan, according to its terms, and all applicable taxes, insurance (including mortgage guarantee insurance), and assessments,” based on several considerations, including “a payment schedule that fully amortizes the loan over the term of the loan.” TILA section 129C(a)(1) and (3). The statutory requirement to consider mortgage-related obligations, as defined in § 1026.43(b)(8), is discussed above in the section-by-section analysis of § 1026.43(c)(2)(v).

TILA section 129C(a) requires, among other things, that a creditor make a determination that a consumer “has a reasonable ability to repay” a residential mortgage loan. TILA section 129C(a)(6)(D) provides the process for calculating the monthly payment amount “[f]or purposes of making any determination under this subsection,” i.e., subsection (a), for “any residential mortgage loan.” TILA section 129C(a)(6)(A) through (D) requires creditors to make uniform assumptions when calculating the payment obligation for purposes of determining the consumer's repayment ability for the covered transaction. Specifically, TILA section 129C(a)(6)(D)(i) through (iii) provides that, when calculating the payment obligation that will be used to determine whether the consumer can repay the covered transaction, the creditor must use a fully amortizing payment schedule and assume that: (1) The loan proceeds are fully disbursed on the date the loan is consummated; (2) the loan is repaid in substantially equal, monthly amortizing payments for principal and interest over the entire term of the loan with no balloon payment; and (3) the interest rate over the entire term of the loan is a fixed rate equal to the fully indexed rate at the time of the loan closing, without considering the introductory rate. The term “fully indexed rate” is defined in TILA section 129C(a)(7).

TILA section 129C(a)(6)(D)(ii)(I) and (II), however, provides two exceptions to the second assumption regarding “substantially equal, monthly payments over the entire term of the loan with no balloon payment” for loans that require “more rapid repayment (including balloon payment).” First, this statutory provision authorizes the Bureau to prescribe regulations for calculating the payment obligation for loans that require more rapid repayment (including balloon payment), and which have an annual percentage rate that does not exceed the threshold for higher-priced mortgage loans. TILA section 129C(a)(6)(D)(ii)(I). Second, for loans that “require more rapid repayment (including balloon payment),” and which exceed the higher-priced mortgage loan threshold, the statute requires that the creditor use the loan contract's repayment schedule. TILA section 129C(a)(6)(D)(ii)(II). The statute does not define the term “rapid repayment.”

The statute also provides three additional clarifications to the assumptions stated above for loans that contain certain features. First, for variable-rate loans that defer repayment of any principal or interest, TILA section 129C(a)(6)(A) states that for purposes of the repayment ability determination a creditor must use “a fully amortizing repayment schedule.” This provision generally reiterates the requirement provided under TILA section 129C(a)(3) to use a payment schedule that fully amortizes the loan. Second, for covered transactions that permit or require interest-only payments, the statute requires that the creditor determine the consumers' repayment ability using “the payment amount required to amortize the loan by its final maturity.” TILA section 129C(a)(6)(B). Third, for covered transactions with negative amortization, the statute requires the creditor to also take into account “any balance increase that may accrue from any negative amortization provision” when making the repayment ability determination. TILA section 129C(a)(6)(C). The statute does not define the terms “variable-rate,” “fully amortizing,” “interest-only,” or “negative amortization.” Proposed § 226.43(c)(5)(i) and (ii) implemented these statutory provisions, as discussed in further detail below.

TILA section 129C(a), as enacted by section 1411 of the Dodd-Frank Act, largely codifies many aspects of the repayment ability rule under § 1026.34(a)(4) from the Board's 2008 HOEPA Final Rule and extends such requirements to the entire mortgage market regardless of the loan's interest rate. Similarly to § 1026.34(a)(4), the statutory framework of TILA section 129C(a) focuses on prescribing the requirements that govern the underwriting process and extension of credit to consumers, rather than dictating which credit terms may or may not be permissible. However, there are differences between TILA section 129C(a) and the 2008 HOEPA Final Rule with respect to payment calculation requirements.

Current § 1026.34(a)(4) does not address how a creditor must calculate the payment obligation for a loan that cannot meet the presumption of compliance under § 1026.34(a)(4)(iii)(B). For example, § 1026.34(a)(4) does not specify how to calculate the periodic payment required for a negative amortization loan or balloon-payment mortgage with a term of less than seven years. In contrast, the Dodd-Frank Act lays out a specific framework for underwriting any loan subject to TILA section 129C(a). In taking this approach, the statutory requirements in TILA section 129C(a)(6)(D) addressing payment calculation requirements differ from § 1026.34(a)(4)(iii) in the following manner: (1) The statute generally premises repayment ability on monthly payment obligations calculated using the fully indexed rate, with no limit on the term of the loan that should be considered for such purpose; (2) the statute permits underwriting loans with balloon payments to differ depending on whether the loan's annual percentage rate exceeds the applicable loan pricing benchmark, or meets or falls below the applicable loan pricing benchmark; and (3) the statute expressly addresses underwriting requirements for loans with interest-only payments or negative amortization.

In 2006 and 2007 the Board and other Federal banking agencies addressed concerns regarding the increased risk to creditors and consumers presented by loans that permit consumers to defer repayment of principal and sometimes interest, and by adjustable-rate mortgages in the subprime market. The Interagency Supervisory Guidance stated that creditors should determine a consumer's repayment ability using a payment amount based on the fully indexed rate, assuming a fully amortizing schedule. In addition, the 2006 Nontraditional Mortgage Guidance addressed specific considerations for negative amortization and interest-only loans. State supervisors issued parallel statements to this guidance, which most states have adopted. TILA section 129C(a)(3) and (6) is generally consistent with this longstanding Interagency Supervisory Guidance and largely extends the guidance regarding payment calculation assumptions to all loan types covered under TILA section 129C(a), regardless of a loan's interest rate. The Board proposed § 226.43(c)(5) to implement the payment calculation requirements of TILA section 129C(a)(1), (3) and (6) for purposes of the repayment ability determination required under proposed § 226.43(c). Consistent with these statutory provisions, proposed § 226.43(c)(5) did not prohibit the creditor from offering certain credit terms or loan features, but rather focused on the calculation process the creditor would be required to use to determine whether the consumer could repay the loan according to its terms. Under the proposal, creditors generally would have been required to determine a consumer's ability to repay a covered transaction using the fully indexed rate or the introductory rate, whichever is greater, to calculate monthly, fully amortizing payments that are substantially equal, unless a special rule applies. See proposed § 226.43(c)(5)(i). For clarity and simplicity, proposed § 226.43(c)(5)(i) used the terms “fully amortizing payment” and “fully indexed rate,” which were defined separately under proposed § 226.43(b)(2) and (3), respectively, as discussed above. Proposed comment 43(c)(5)(i)-1 clarified that the general rule would apply whether the covered transaction is an adjustable-, step-, or fixed-rate mortgage, as those terms are defined in § 1026.18(s)(7)(i), (ii), and (iii), respectively.

Proposed § 226.43(c)(5)(ii)(A) through (C) created exceptions to the general rule and provided special rules for calculating the payment obligation for balloon-payment mortgages, interest-only loans or negative amortization loans, as follows:

Balloon-payment mortgages. Consistent with TILA section 129C(a)(6)(D)(ii)(I) and (II), for covered transactions with a balloon payment, proposed § 226.43(c)(5)(ii)(A) provided special rules that differed depending on the loan's rate. Proposed § 226.43(c)(5)(ii)(A)(1) stated that for covered transactions with a balloon payment that are not higher-priced covered transactions, the creditor must determine a consumer's ability to repay the loan using the maximum payment scheduled in the first five years after consummation. Proposed § 226.43(c)(5)(ii)(A)(2) further stated that for covered transactions with balloon payments that are higher priced covered transactions, the creditor must determine the consumer's ability to repay according to the loan's payment schedule, including any balloon payment. For clarity, proposed § 226.43(c)(5)(ii)(A) used the term “higher-priced covered transaction” to refer to a covered transaction that exceeds the applicable higher-priced mortgage loan coverage threshold. “Higher-priced covered transaction” is defined in § 1026.43(b)(4), discussed above. The term “balloon payment” has the same meaning as in current § 1026.18(s)(5)(i).

Interest-only loans. Consistent with TILA section 129C(a)(6)(B) and (D), proposed § 226.43(c)(5)(ii)(B) provided special rules for interest-only loans. Proposed § 226.43(c)(5)(ii)(B) required that the creditor determine the consumer's ability to repay the interest-only loan using (1) the fully indexed rate or the introductory rate, whichever is greater; and (2) substantially equal, monthly payments of principal and interest that will repay the loan amount over the term of the loan remaining as of the date the loan is recast. For clarity, proposed § 226.43(c)(5)(ii)(B) used the terms “loan amount” and “recast,” which are defined and discussed under § 1026.43(b)(5) and (11), respectively. The term “interest-only loan” has the same meaning as in current § 1026.18(s)(7)(iv).

Negative amortization loans. Consistent with TILA section 129C(a)(6)(C) and (D), proposed § 226.43(c)(5)(ii)(C) provided special rules for negative amortization loans. Proposed § 226.43(c)(5)(ii)(C) required that the creditor determine the consumer's ability to repay the negative amortization loan using (1) the fully indexed rate or the introductory rate, whichever is greater; and (2) substantially equal, monthly payments of principal and interest that will repay the maximum loan amount over the term of the loan remaining as of the date the loan is recast. Proposed comment 43(c)(5)(ii)(C)-1 clarified that for purposes of the rule, the creditor would first have to determine the maximum loan amount and the period of time that remains in the loan term after the loan is recast. For clarity, proposed § 226.43(c)(5)(ii)(C) used the terms “maximum loan amount” and “recast,” which are defined and discussed under § 1026.43(b)(7) and (11), respectively. The term “negative amortization loan” has the same meaning as in current § 1026.18(s)(7)(v) and comment 18(s)(7)-1.

43(c)(5)(i) General Rule

Proposed § 226.43(c)(5)(i) implemented the payment calculation requirements in TILA section 129C(a)(3), 129C(6)(D)(i) through (iii), and stated the general rule for calculating the payment obligation on a covered transaction for purposes of the ability-to-repay provisions. Consistent with the statute, proposed § 226.43(c)(5)(i) provided that unless an exception applies under proposed § 226.43(c)(5)(ii), a creditor must make the repayment ability determination required under proposed § 226.43(c)(2)(iii) by using the greater of the fully indexed rate or any introductory interest rate, and monthly, fully amortizing payments that are substantially equal. That is, under the proposed general rule the creditor would calculate the consumer's monthly payment amount based on the loan amount, and amortize that loan amount in substantially equal payments over the loan term, using the fully indexed rate.

Proposed comment 43(c)(5)(i)-1 explained that the payment calculation method set forth in proposed § 226.43(c)(5)(i) applied to any covered transaction that does not have a balloon payment or that is not an interest-only loan or negative amortization loan, whether it is a fixed-rate, adjustable-rate or step-rate mortgage. This comment further explained that the payment calculation method set forth in proposed § 226.43(c)(5)(ii) applied to any covered transaction that is a loan with a balloon payment, interest-only loan, or negative amortization loan. To facilitate compliance, this comment listed the defined terms used in proposed § 226.43(c)(5) and provided cross-references to their definitions.

The fully indexed rate or introductory rate, whichever is greater. Proposed § 226.43(c)(5)(i)(A) implemented the requirement in TILA section 129C(a)(6)(D)(iii) to use the fully indexed rate when calculating the monthly, fully amortizing payment for purposes of the repayment ability determination. Proposed § 226.43(c)(5)(i)(A) also provided that when creditors calculate the monthly, fully amortizing payment for adjustable-rate mortgages, they would have to use the introductory interest rate if it were greater than the fully indexed rate (i.e., a premium rate). In some adjustable-rate transactions, creditors may set an initial interest rate that is not determined by the index or formula used to make later interest rate adjustments. Sometimes this initial rate charged to consumers is lower than the rate would be if it were determined by using the index plus margin, or formula (i.e., the fully indexed rate). However, an initial rate that is a premium rate is higher than the rate based on the index or formula. Thus, requiring creditors to use only the fully indexed rate would result in creditors underwriting loans that have a “premium” introductory rate at a rate lower than the rate on which the consumer's initial payments would be based. The Board believed that requiring creditors to assess the consumer's ability to repay on the initial higher payments would better effectuate the statutory intent and purpose. Proposed comment 43(c)(5)(i)-2 provided guidance on using the greater of the premium or fully indexed rate.

Monthly, fully amortizing payments. For simplicity, proposed § 226.43(c)(5)(i) used the term “fully amortizing payment” to refer to the statutory requirements that a creditor use a payment schedule that repays the loan assuming that (1) the loan proceeds are fully disbursed on the date of consummation of the loan; and (2) the loan is repaid in amortizing payments for principal and interest over the entire term of the loan. See TILA sections 129C(a)(3) and (6)(D)(i) and (ii). As discussed above, § 1026.43(b)(2) defines “fully amortizing payment” to mean a periodic payment of principal and interest that will fully repay the loan amount over the loan term. The terms “loan amount” and “loan term” are defined in § 1026.43(b)(5) and (b)(6), respectively, and discussed above.

The statute also expressly requires that a creditor use “monthly amortizing payments” for purposes of the repayment ability determination. TILA section 129C(6)(D)(ii). The Board recognized that some loan agreements require consumers to make periodic payments with less frequency, for example quarterly or semi-annually. Proposed § 226.43(c)(5)(i)(B) did not dictate the frequency of payment under the terms of the loan agreement, but did require creditors to convert the payment schedule to monthly payments to determine the consumer's repayment ability. Proposed comment 43(c)(5)(i)-3 clarified that the general payment calculation rules do not prescribe the terms or loan features that a creditor may choose to offer or extend to a consumer, but establish the calculation method a creditor must use to determine the consumer's repayment ability for a covered transaction. This comment explained, by way of example, that the terms of the loan agreement may require that the consumer repay the loan in quarterly or bi-weekly scheduled payments, but for purposes of the repayment ability determination, the creditor must convert these scheduled payments to monthly payments in accordance with proposed § 226.43(c)(5)(i)(B). This comment also explained that the loan agreement may not require the consumer to make fully amortizing payments, but for purposes of the repayment ability determination the creditor must convert any non-amortizing payments to fully amortizing payments.

Substantially equal. Proposed comment 43(c)(5)(i)-4 provided additional guidance to creditors for determining whether monthly, fully amortizing payments are “substantially equal.” See TILA section 129C(a)(6)(D)(ii). This comment stated that creditors should disregard minor variations due to payment-schedule irregularities and odd periods, such as a long or short first or last payment period. The comment explained that monthly payments of principal and interest that repay the loan amount over the loan term need not be equal, but that the monthly payments should be substantially the same without significant variation in the monthly combined payments of both principal and interest. Proposed comment 43(c)(5)(i)-4 further explained that where, for example, no two monthly payments vary from each other by more than 1 percent (excluding odd periods, such as a long or short first or last payment period), such monthly payments would be considered substantially equal for purposes of the rule. The comment further provided that, in general, creditors should determine whether the monthly, fully amortizing payments are substantially equal based on guidance provided in current § 1026.17(c)(3) (discussing minor variations), and § 1026.17(c)(4)(i) through (iii) (discussing payment-schedule irregularities and measuring odd periods due to a long or short first period) and associated commentary. The proposal solicited comment on operational difficulties that arise by ensuring payment amounts meet the “substantially equal” condition. The proposal also solicited comment on whether a 1 percent variance is an appropriate tolerance threshold.

Examples of payment calculations. Proposed comment § 226.43(c)(5)(i)-5 provided illustrative examples of how to determine the consumer's repayment ability based on substantially equal, monthly, fully amortizing payments as required under proposed § 226.43(c)(5)(i) for a fixed-rate, adjustable-rate and step-rate mortgage.

The Board recognized that, although consistent with the statute, the proposed framework would require creditors to underwrite certain loans, such as hybrid ARMs with a discounted rate period of five or more years (e.g., 5/1, 7/1, and 10/1 ARMs) to a more stringent standard as compared to the underwriting standard set forth in proposed § 226.43(e)(2)(v) for qualified mortgages. The Board believed this approach was consistent with the statute's intent to ensure consumers can reasonably repay their loans, and that in both cases consumers' interests are properly protected. See TILA section 129B(a)(2), 15 U.S.C. 1639b(a)(2). To meet the definition of a qualified mortgage, a loan cannot have certain risky terms or features, such as provisions that permit deferral of principal or a term that exceeds 30 years; no similar restrictions apply to loans subject to the ability-to-repay standard. See proposed § 226.43(e)(2)(i) and (ii). As a result, the risk of potential payment shock is diminished significantly for qualified mortgages. For this reason, the Board believed that maintaining the potentially more lenient statutory underwriting standard for loans that satisfy the qualified mortgage criteria would help to ensure that responsible and affordable credit remains available to consumers. See TILA section 129B(a)(2), 15 U.S.C. 1639b(a)(2).

Loan amount or outstanding principal balance. As noted above, proposed § 226.43(c)(5)(i) was consistent with the statutory requirements regarding payment calculations for purposes of the repayment ability determination. The Board believed that the intent of these statutory requirements was to prevent creditors from assessing the consumer's repayment ability based on understated payment obligations, especially when risky features can be present on the loan. However, the Board was concerned that the statute, as implemented in proposed § 226.43(c)(5)(i), would require creditors to determine, in some cases, a consumer's repayment ability using overstated payment amounts because the creditor would have to assume that the consumer repays the loan amount in substantially equal payments based on the fully indexed rate, regardless of when the fully indexed rate could take effect under the terms of the loan. The Board was concerned that this approach might restrict credit availability, even where consumers were able to demonstrate that they can repay the payment obligation once the fully indexed rate takes effect.

For this reason, the proposal solicited comment on whether authority should be exercised under TILA sections 105(a) and 129B(e) to provide that the creditor may calculate the monthly payment using the fully indexed rate based on the outstanding principal balance as of the date the fully indexed rate takes effect under the loan's terms, instead of the loan amount at consummation.

Step-rate and adjustable-rate calculation s. Due to concerns regarding credit availability, the proposal also solicited comment on alternative means to calculate monthly payments for step-rate and adjustable-rate mortgages. The proposal asked for comment on whether or not the rule should require that creditors underwrite a step-rate or an adjustable-rate mortgage using the maximum interest rate in the first seven or ten years or some other appropriate time horizon that would reflect a significant introductory rate period. The section-by-section analysis of the “fully indexed rate” definition, at § 1026.43(b)(3) above, discusses this issue in regard to step-rate mortgages. For discussion of payment calculation methods for adjustable-rate mortgages, see below.

Safe harbor to facilitate compliance. The Board recognized that under its proposal, creditors would have to comply with multiple assumptions when calculating the particular payment for purposes of the repayment ability determination. The Board was concerned that the complexity of the proposed payment calculation requirements might increase the potential for unintentional errors to occur, making compliance difficult, especially for small creditors that might be unable to invest in advanced technology or software needed to ensure payment calculations are compliant. At the same time, the Board noted that the intent of the statutory framework and the proposal was to ensure consumers are offered and receive loans on terms that they can reasonably repay. Thus, the Board solicited comment on whether authority under TILA sections 105(a) and 129B(e) should be exercised to provide a safe harbor for creditors that use the largest scheduled payment that can occur during the loan term to determine the consumer's ability to repay, to facilitate compliance with the requirements under proposed § 226.43(c)(5)(i) and (ii).

Final Rule

The final rule requires creditors to underwrite the loan at the premium rate if greater than the fully indexed rate for purposes of the repayment ability determination using the authority under TILA section 105(a). 15 U.S.C. 1604(a). TILA section 105(a), as amended by section 1100A of the Dodd-Frank Act, provides that the Bureau's regulations may contain such additional requirements, classifications, differentiations, or other provisions, and may provide for such adjustments and exceptions for all or any class of transactions as in the Bureau's judgment are necessary or proper to effectuate the purposes of TILA, prevent circumvention or evasion thereof, or facilitate compliance therewith. 15 U.S.C. 1604(a). This approach is further supported by the authority under TILA section 129B(e) to condition terms, acts or practices relating to residential mortgage loans that the Bureau finds necessary and proper to ensure that responsible, affordable mortgage credit remains available to consumers in a manner consistent with and which effectuates the purposes of sections 129B and 129C, and which are in the interest of the consumer. 15 U.S.C. 1639b(e). The purposes of TILA include the purpose of TILA sections 129B and 129C, to assure that consumers are offered and receive residential mortgage loans on terms that reasonably reflect their ability to repay the loan, among other things. TILA section 129B(b), 15 U.S.C. 1639b. For the reasons discussed above, the Bureau believes that requiring creditors to underwrite the loan to the premium rate for purposes of the repayment ability determination is necessary and proper to ensure that consumers are offered, and receive, loans on terms that reasonably reflect their ability to repay, and to prevent circumvention or evasion. Without a requirement to consider payments based on a premium rate, a creditor could originate loans with introductory-period payments that consumers do not have the ability to repay. Therefore, this provision is also in the interest of consumers.

As discussed above, the Board solicited comment on whether payments for non-qualified mortgage ARMs should be calculated similarly to qualified mortgage ARMs, by using the maximum rate that will apply during a certain period, such as the first seven years or some other appropriate time horizon. Consumer and community groups were divided on this issue. Some supported use of the fully indexed rate, but one stated that underwriting ARMs based on the initial period of at least five years may be appropriate. Another suggested that for non-qualified mortgage ARMs the rule should require use of the maximum interest rate or interest rate cap, whichever is greater, to better protect against payment shock. A civil rights organization also advocated that ARMs that are not qualified mortgages should be underwritten to several points above the fully indexed rate. A combined comment from consumer advocacy organizations also supported non-qualified mortgage ARMs being underwritten more strictly, suggesting that because this is the market segment that will have the fewest controls, the predatory practices will migrate here, and there is significant danger of payment shock when using the fully indexed rate in a low-rate environment such as today's market. They suggested that the rule follow Fannie Mae's method, which requires underwriting that uses the fully indexed rate or the note rate plus 2 percent, whichever is greater, for ARMs with initial fixed periods of up to five years. In addition, one joint industry and consumer advocacy comment suggested adding 2 percent to the fully indexed rate in order to calculate the monthly payment amount.

Industry groups were strongly in favor of using a specific time period for underwriting, generally suggesting five years. One credit union association stated that use of the fully indexed rate is excessive and unnecessary, and will increase the cost of credit. Industry commenters stated that creditors generally consider only the fixed-rate period, and ARMs with fixed periods of at least five years are considered safe. One large bank stated that the calculation for ARMs, whether or not they are qualified mortgages, should be uniform to ease compliance.

The Bureau has determined that it will not use its exception and adjustment authority to change the statutory underwriting scheme for non-qualified mortgage ARMs. The statutory scheme clearly differentiates between the qualified mortgage and non-qualified mortgage underwriting strategies. The qualified mortgage underwriting rules ignore any adjustment in interest rate that may occur after the first five years; thus, for example, for an ARM with an initial adjustment period of seven years, the interest rate used for the qualified mortgage calculation will be the initial interest rate. In addition, the qualified mortgage rules, by using the “maximum interest rate,” take into account any adjustment in interest rate that can occur during the first five years, including adjustments attributable to changes in the index rate. In contrast, the non-qualified mortgage rules have an unlimited time horizon but do not take into account adjustments attributable to changes in the index rate.

Based on the its research and analysis, the Bureau notes that the data indicate that neither the fully indexed rate nor the maximum rate during a defined underwriting period produces consistent results with regard to ability-to-repay calculations. The Bureau finds that the underwriting outcomes under the two methods vary depending on a number of complex variables, such as the terms of the loan (e.g., the length of the initial adjustment period and interest rate caps) and the interest rate environment. In other words, for a particular loan, whether the monthly payment may be higher under a calculation that uses the fully indexed rate, as opposed to the maximum rate in the first five years, depends on a number of factors. Given the fact-specific nature of the payment calculation outcomes, the Bureau believes that overriding the statutory scheme would be inappropriate.

The Bureau also believes that adjusting the interest rate to be used for non-qualified mortgage ability-to-repay calculations to somewhere between the fully indexed rate specified in the statute and the maximum interest rate mandated for qualified mortgage underwriting; for example through an adjustment to the fully indexed rate of an additional 2 percent, would be inappropriate. The fully indexed rate had been in use since it was adopted by the Interagency Supervisory Guidance in 2006, and Congress was likely relying on that experience in crafting the statutory scheme. Adding to the fully indexed rate would potentially reduce the availability of credit. Such an adjustment also could result in a calculated interest rate and monthly payment that are higher than the interest rate and payment calculated for qualified mortgage underwriting, given that the qualified mortgage rules look only to potential adjustments during the first five years.

The Bureau recognizes that underwriting practices today often take into account potential adjustments in an ARM that can result from increases in the index rate. For example, Fannie Mae requires underwriting that uses the fully indexed rate or the note rate plus 2 percent, whichever is greater, for ARMs with initial fixed periods of up to five years. The Bureau notes that underwriters have the flexibility to adjust their practices in response to changing interest rate environments whereas the process an administrative agency like the Bureau must follow to amend a rule is more time consuming. The Bureau also notes that the creditor must make a reasonable determination that the consumer has the ability to repay the loan according to its terms. Therefore, in situations where there is a significant likelihood that the consumer will face an adjustment that will take the interest rate above the fully indexed rate, a creditor whose debt-to-income or residual income calculation indicates that a consumer cannot afford to absorb any such increase may not have a reasonable belief in the consumer's ability to repay the loan according to its terms. See comment 43(c)(1)-1.

Although the Bureau has determined to implement the statutory scheme as written and require use of the fully indexed rate for non-qualified mortgage ARMs, it will monitor this issue through its mandatory five-year review, and may make adjustments as necessary.

As discussed above, the Board also solicited comment on whether or not to allow the fully indexed rate to be applied to the balance projected to be remaining when the fully indexed rate goes into effect, instead of the full loan amount, and thus give a potentially more accurate figure for the maximum payment that would be required for purposes of determining ability to repay. A consumer group and a group advocating for financial reform supported this possibility, saying that allowing lenders to apply the fully indexed rate to the balance remaining when the rate changes, rather than the full loan amount, will encourage longer fixed-rate periods and safer lending, as well as preserve access to credit. An association representing credit unions also agreed with the possible amendment, stating that the new method would yield a more accurate measure of the maximum payment that could be owed.

The Bureau believes it is appropriate for the final rule to remain consistent with the statutory scheme. The Bureau believes that changing the calculation method, required by the statute, would not be an appropriate use of its exception and adjustment authority. The Bureau believes the potentially stricter underwriting method of calculating the monthly payment by applying the imputed (i.e., fully indexed) interest rate to the full loan amount for non-qualified mortgage ARMs, provides greater assurance of the ability to repay. In addition, payment calculation using the fully indexed rate can only approximate the consumer's payments after recast, since the index may have increased significantly by then. Accordingly, the Bureau believes that requiring the use of the full loan amount will reduce the potential inaccuracy of the ability-to-repay determination in such a situation.

In addition, the Board solicited comment on whether to provide a safe harbor for any creditor that underwrites using the “largest scheduled payment that can occur during the loan term.” To provide such a safe harbor the Bureau would have to employ its exception and adjustment authority because the use of the fully indexed rate calculation is required by TILA section 129C(a)(6)(D)(iii). Two industry commenters and an association of state bank regulators supported this exemption, but none of them provided a developed rationale for their support or included information useful in assessing the possible exemption. The Bureau does not believe that it would be appropriate at this time to alter the statutory scheme in this manner.

As discussed above, the Board also solicited comment on how to lessen any operational difficulties of ensuring that payment amounts meet the “substantially equal” condition, and whether or not allowing a one percent variance between payments provided an appropriate threshold. Only two commenters mentioned this issue. One industry commenter stated that the 1 percent threshold was appropriate, but an association of state bank regulators believed that a 5 percent threshold would work better. Because the 1 percent threshold appears to be sufficient to allow for payment variance and industry commenters did not express a need for a higher threshold, the Bureau does not believe that the provision should be amended.

For the reasons stated above, the Bureau is adopting § 1026.43(c)(5)(i) and associated commentary substantially as proposed, with minor clarifying revisions.

43(c)(5)(ii) Special Rules for Loans With a Balloon Payment, Interest-Only Loans, and Negative Amortization Loans

Proposed § 226.43(c)(5)(ii) created exceptions to the general rule under proposed § 226.43(c)(5)(i), and provided special rules in proposed § 226.43(c)(5)(ii)(A) through (C) for loans with a balloon payment, interest-only loans, and negative amortization loans, respectively, for purposes of the repayment ability determination required under proposed § 226.43(c)(2)(iii). In addition to TILA section 129C(a)(6)(D)(i) through (iii), proposed § 226.43(c)(5)(ii)(A) through (C) implemented TILA sections 129C(a)(6)(B) and (C), and TILA section 129C(a)(6)(D)(ii)(I) and (II). Each of these proposed special rules is discussed below.

43(c)(5)(ii)(A)

Implementing the different payment calculation methods in TILA section 129C(a)(6)(D)(ii), the Board proposed different rules for balloon-payment mortgages that are higher-priced covered transactions and those that are not, in § 1026.43(c)(5)(ii)(A)(1) and (2). Proposed comment 43(c)(5)(ii)(A)-1 provided guidance on applying these two methods. This guidance is adopted as proposed with minor changes for clarity and to update a citation. The language describing the calculation method for balloon-payment mortgages that are not higher-priced covered transactions has been changed to reflect the use of the first regular payment due date as the start of the relevant five-year period. Pursuant to the Bureau's rulewriting authority under TILA section 129C(a)(6)(D)(ii)(I), this change has been made to facilitate compliance through consistency with the amended underwriting method for qualified mortgages. See the section-by-section analysis of § 1026.43(e)(2)(iv)(A). As with the recast on five-year adjustable-rate qualified mortgages, the Bureau believes that consumers will benefit from having a balloon payment moved to at least five years after the first regular payment due date, rather than five years after consummation.

43(c)(5)(ii)(A)(1)

The statute provides an exception from the general payment calculation discussed above for loans that require “more rapid repayment (including balloon payment).” See TILA section 129C(a)(6)(D)(ii)(I) and (II). For balloon-payment loans that are not higher-priced covered transactions (as determined by using the margins above APOR in TILA section 129C(a)(6)(D)(ii)(I) and implemented at § 1026.43(b)(4)), the statute provides that the payment calculation will be determined by regulation. The Board proposed that a creditor be required to make the repayment determination under proposed § 226.43(c)(2)(iii) for “[t]he maximum payment scheduled during the first five years after consummation * * *”

The Board chose a five-year period in order to preserve access to affordable short-term credit, and because five years was considered an adequate period for a consumer's finances to improve sufficiently to afford a fully amortizing loan. The Board believed that balloon-payment loans of less than five years presented more risk of inability to repay. The Board also believed that the five-year period would facilitate compliance and create a level playing field because of its uniformity with the general qualified mortgage provision (see § 1026.43(e)), and balloon-payment qualified mortgage provision (see § 1026.43(f)). The Board solicited comment on whether the five-year horizon was appropriate. Proposed comment § 226.43(c)(5)(ii)(A)-2 provided further guidance to creditors on determining whether a balloon payment occurs in the first five years after consummation. Proposed comment 43(c)(5)(ii)(A)-3 addressed renewable balloon-payment loans. This comment discussed balloon-payment loans that are not higher-priced covered transactions which provide an unconditional obligation to renew a balloon-payment loan at the consumer's option or obligation to renew subject to conditions within the consumer's control. This comment clarified that for purposes of the repayment ability determination, the loan term does not include the period of time that could result from a renewal provision.

The Board recognized that proposed comment 43(c)(5)(ii)(A)-3 did not take the same approach as guidance contained in comment 17(c)(1)-11 regarding treatment of renewable balloon-payment loans for disclosure purposes, or with guidance contained in current comment 34(a)(4)(iv)-2 of the Board's 2008 HOEPA Final Rule. Although the proposal differed from current guidance in Regulation Z, the Board believed this approach was appropriate for several reasons. First, the ability-to-repay provisions in the Dodd-Frank Act do not address extending the term of a balloon-payment loan with an unconditional obligation to renew provision. Second, permitting short-term “prime” balloon-payment loans to benefit from the special payment calculation rule when a creditor includes an unconditional obligation to renew, but retains the right to increase the interest rate at the time of renewal, would create a significant loophole in the balloon payment rules. Such an approach could frustrate the objective to ensure consumers obtain mortgages on affordable terms for a reasonable period of time because the interest rate could escalate within a short period of time, increasing the potential risk of payment shock to the consumer. This is particularly the case where no limits exist on the interest rate that the creditor can choose to offer to the consumer at the time of renewal. See TILA Section 129B(a)(2), 15 U.S.C. 1639b(a)(2), and TILA Section 129C(b)(2)(A)(v). Moreover, the Board believed it would be speculative to posit the interest rate at the time of renewal for purposes of the repayment ability determination. Third, the guidance contained in comment 17(c)(1)-11 regarding treatment of renewable balloon-payment loans is meant to help ensure consumers are aware of their loan terms and avoid the uninformed use of credit, which differs from the stated purpose of this proposed provision, which was to help ensure that consumers receive loans on terms that reasonably reflect their repayment ability. TILA section 102(a), 15 U.S.C. 1601(a)(2), and TILA section 129B(a)(2), 15 U.S.C. 1639b(a)(2).

Proposed comment 43(c)(5)(ii)(A)-4 provided several illustrative examples of how to determine the maximum payment scheduled during the first five years after consummation for loans with a balloon payment that are not higher-priced covered transactions.

In regard to the proposed five-year underwriting period, some commenters suggested that the payment period considered should be increased to ten years, stating that balloon-payment loans were repeatedly used in an abusive manner during the years of heavy subprime lending. The combined consumer advocacy organizations' comment stated that the five-year underwriting might lead to an increase in five-year balloon-payment loans, which would be bad for sustainable lending. On the other hand, a trade association representing credit unions supported the five-year rule. One industry commenter objected to the whole balloon underwriting scheme, including the five-year rule, apparently preferring something less.

For the reasons discussed by the Board in the proposal, and described above, the Bureau has determined that five years is an appropriate time frame for determining the ability to repay on balloon-payment mortgages that are not higher-priced covered transactions. However, for the sake of uniformity and ease of compliance with the qualified mortgage calculation and ability-to-repay calculation for non-qualified mortgage adjustable-rate mortgages, the proposed provision has been changed to state that the five years will be measured from the date of the first regularly scheduled payment, rather than the date of consummation. The Bureau has made this determination pursuant to the authority granted by TILA section 129C(a)(6)(D)(ii)(I) to prescribe regulations for calculating payments to determine consumers' ability to repay balloon-payment mortgages that are not higher-cost covered transactions.

TILA section 129C(a)(6)(D)(ii)) refers to loans requiring “more rapid repayment (including balloon payment).” The Board solicited comment about whether this statutory language should be read as referring to loan types other than balloon-payment loans. The Bureau did not receive comments on this matter, and has determined that the rule language does not need to be amended to include other types of “rapid repayment” loans at this time.

The Board also solicited comment about balloon-payment loans that have an unconditional obligation to renew. The Board asked whether or not such loans should be allowed to comply with the ability-to-repay requirements using the total of the mandatory renewal terms, instead of just the first term. As discussed above, proposed comment 43(c)(5)(ii)(A)-3 made clear that this would not be allowed under the rule as proposed. The Board also solicited comment on any required conditions that the renewal obligation should have, if such an amendment were made. However, the Bureau did not receive comments on this matter, and the provision and staff comment are adopted as proposed. A creditor making any non-higher-priced balloon-payment mortgage of less than five years with a clear obligation to renew can avoid having the ability-to-repay determination applied to the balloon payment by including the renewal period in the loan term so that the balloon payment occurs after five years.

Accordingly, the Bureau is adopting § 1026.43(c)(5)(ii)(A)(1) and associated commentary substantially as proposed, with minor changes for clarification, as well as new language to reflect that the five-year underwriting period begins with the due date of the first payment, as discussed above. In addition, the Bureau has added a second example to comment 43(c)(5)(ii)(A)-2 to demonstrate the effect of the change to the beginning of the underwriting period.

43(c)(5)(ii)(A)(2)

Proposed § 226.43(c)(5)(ii)(A)(2) implemented TILA section 129C(a)(6)(D)(ii)(II) and provided that for a higher-priced covered transaction, the creditor must determine the consumer's ability to repay a loan with a balloon payment using the scheduled payments required under the terms of the loan, including any balloon payment. TILA section 129C(a)(6)(D)(ii)(II) states that for loans that require “more rapid repayment (including balloon payment),” and which exceed the loan pricing threshold set forth, the creditor must underwrite the loan using the “[loan] contract's repayment schedule.” For purposes of proposed § 226.43(c)(5)(i)(A), “higher-priced covered transaction” means a covered transaction with an annual percentage rate that exceeds the average prime offer rate for a comparable transaction as of the date the interest rate is set by 1.5 or more percentage points for a first-lien covered transaction, or by 3.5 or more percentage points for a subordinate-lien covered transaction. See § 1026.43(b)(4).

The proposed rule interpreted the statutory requirement that the creditor use the loan contract's payment schedule to mean that the creditor must use all scheduled payments under the terms of the loan needed to fully amortize the loan, consistent with the requirement under TILA section 129C(a)(3). Payment of the balloon, either at maturity or during any intermittent period, is necessary to fully amortize the loan, and so a consumer's ability to pay the balloon payment would need to be considered. Proposed comment 43(c)(5)(ii)(A)-5 provided an illustrative example of how to determine the consumer's repayment ability based on the loan contract's payment schedule, including any balloon payment. The proposed rule applied to “non-prime” loans with a balloon payment regardless of the length of the term or any contract provision that provides for an unconditional guarantee to renew.

In making this proposal, the Board expressed concern that this approach could lessen credit choice for non-prime consumers and solicited comment, with supporting data, on the impact of this approach for low-to-moderate income consumers. In addition, the Board asked for comment on whether or not a consumer's ability to refinance out of a balloon-payment loan should be considered in determining ability to repay.

Industry commenters who focused on this provision opposed applying the ability-to-repay determination to the entire payment schedule. Two trade associations representing small and mid-size banks strongly objected to including the balloon payment in the underwriting, and one stated that many of the loans its members currently make would fall into the higher-priced category, making these loans unavailable. However, the statutory scheme for including the balloon payment was supported by a state housing agency and the combined consumer protection advocacy organizations submitting joint comments.

None of the commenters submitted data supporting the importance of higher-priced balloon-payment mortgages for credit availability, or whether consideration of a consumer's ability to obtain refinancing would make the ability-to-repay determination less significant in this context. The Bureau notes that under § 1026.43(f) a balloon-payment mortgage that is a higher-priced covered transaction made by certain creditors in rural or underserved areas may also be a qualified mortgage and thus the creditor would not have to consider the consumer's ability to repay the balloon payment. Because this final rule adopts a wider definition of “rural or underserved area” than the Board proposed, potential credit accessibility concerns have been lessened. See the section-by-section analysis of § 1026.43(f), below.

The statute requires the consideration of the balloon payment for higher-priced covered transactions, and the Bureau does not believe that using its exception and adjustment authority would be appropriate for this issue. Accordingly, § 1026.43(c)(5)(ii)(A)(2) and associated commentary are adopted substantially as proposed, with minor changes for clarification.

43(c)(5)(ii)(B)

The Board's proposed § 226.43(c)(5)(ii)(B) implemented TILA section 129C(a)(6)(B), which requires that the creditor determine the consumer's repayment ability using “the payment amount required to amortize the loan by its final maturity.” For clarity, the proposed rule used the term “recast,” which is defined for interest-only loans as the expiration of the period during which interest-only payments are permitted under the terms of the legal obligation. See § 1026.43(b)(11). The statute does not define the term “interest-only.” For purposes of this rule, the terms “interest-only loan” and “interest-only” have the same meaning as in § 1026.18(s)(7)(iv).

For interest-only loans (i.e., loans that permit interest only payments for any part of the loan term), proposed § 226.43(c)(5)(ii)(B) provided that the creditor must determine the consumer's ability to repay the interest-only loan using (1) the fully indexed rate or any introductory rate, whichever is greater; and (2) substantially equal, monthly payments of principal and interest that will repay the loan amount over the term of the loan remaining as of the date the loan is recast. The proposed payment calculation rule for interest-only loans paralleled the general rule proposed in § 226.43(c)(5)(i), except that proposed § 226.43(c)(5)(ii)(B)(2) required a creditor to determine the consumer's ability to repay the loan amount over the term that remains after the loan is recast, rather than requiring the creditor to use fully amortizing payments, as defined under proposed § 226.43(b)(2).

The Board interpreted the statutory text in TILA section 129C(a)(6)(B) as requiring the creditor to determine the consumer's ability to repay an interest-only loan using the monthly principal and interest payment amount needed to repay the loan amount once the interest-only payment period expires, rather than using, for example, an understated monthly principal and interest payment that would amortize the loan over its entire term, similar to a 30-year fixed mortgage. The proposed rule would apply to all interest-only loans, regardless of the length of the interest-only period. The Board believed this approach most accurately assessed the consumer's ability to repay the loan once it begins to amortize; this is consistent with the approach taken for interest-only loans in the 2006 Nontraditional Mortgage Guidance.

Proposed comment 43(c)(5)(ii)(B)-1 provided guidance on the monthly payment calculation for interest-only loans, and clarified that the relevant term of the loan for calculating these payments is the period of time that remains after the loan is recast. This comment also explained that for a loan on which only interest and no principal has been paid, the loan amount will be the outstanding principal balance at the time of the recast.

Proposed comment 43(c)(5)(ii)(B)-2 provided illustrative examples for how to determine the consumer's repayment ability based on substantially equal monthly payments of principal and interest for interest-only loans.

Commenters did not focus on the calculation for interest-only loans. The Bureau considers the Board's interpretation and implementation of the statute to be accurate and appropriate. Accordingly, § 1026.43(c)(5)(ii)(B) and associated commentary are adopted as proposed.

43(c)(5)(ii)(C)

Proposed § 226.43(c)(5)(ii)(C) implemented the statutory requirement in TILA section 129C(a)(6)(C) that the creditor consider “any balance increase that may accrue from any negative amortization provision when making the repayment ability determination.” The statute does not define the term “negative amortization.”

For such loans, proposed § 226.43(c)(5)(ii)(C) provided that a creditor must determine the consumer's repayment ability using (1) the fully indexed rate or any introductory interest rate, whichever is greater; and (2) substantially equal, monthly payments of principal and interest that will repay the maximum loan amount over the term of the loan remaining as of the date the loan is recast. The proposed payment calculation rule for negative amortization loans paralleled the general rule in proposed § 226.43(c)(5)(i), except that proposed § 226.43(c)(5)(ii)(C)(2) required the creditor to use the monthly payment amount that repays the maximum loan amount over the term of the loan that remains after the loan is recast, rather than requiring the creditor to use fully amortizing payments, as defined under § 1026.43(b)(2). The proposed rule used the terms “maximum loan amount” and “recast,” which are defined and discussed at § 1026.43(b)(7) and (b)(11), respectively.

The Board proposed that the term “negative amortization loan” have the same meaning as set forth in § 226.18(s)(7)(v), which provided that the term “negative amortization loan” means a loan, other than a reverse mortgage subject to § 226.33, that provides for a minimum periodic payment that covers only a portion of the accrued interest, resulting in negative amortization. As defined, the term “negative amortization loan” does not cover other loan types that may have a negative amortization feature, but which do not permit the consumer multiple payment options, such as seasonal income loans. Accordingly, proposed § 226.43(c)(5)(ii)(C) covered only loan products that permit or require minimum periodic payments, such as payment-option loans and graduated payment mortgages with negative amortization. The Board believed that covering these types of loans in proposed § 226.43(c)(5)(ii)(C) was consistent with statutory intent to account for the negative equity that can occur when a consumer makes payments that defer some or all principal or interest for a period of time, and to address the impact that any potential payment shock might have on the consumer's ability to repay the loan. See TILA section 129C(a)(6)(C).

In contrast, in a transaction such as a seasonal loan that has a negative amortization feature, but which does not provide for minimum periodic payments that permit deferral of some or all principal, the consumer repays the loan with fully amortizing payments in accordance with the payment schedule. Accordingly, the same potential for payment shock due to accumulating negative amortization does not exist. These loans with a negative amortization feature are therefore not covered by the proposed term “negative amortization loan,” and would not be subject to the special payment calculation requirements for negative amortization loans at proposed § 226.43(c)(5)(ii)(C).

For purposes of determining the consumer's ability to repay a negative amortization loan under proposed § 226.43(c)(5)(ii)(C), creditors would be required to make a two-step payment calculation.

Step one: maximum loan amount. Proposed § 226.43(c)(5)(ii)(C) would have required that the creditor first determine the maximum loan amount and period of time that remains in the loan term after the loan is recast before determining the consumer's repayment ability on the loan. See comment 43(c)(5)(ii)(C)-1; s ee also proposed § 226.43(b)(11), which defined the term “recast” to mean the expiration of the period during which negatively amortizing payments are permitted under the terms of the legal obligation. Proposed comment 43(c)(5)(ii)(C)-2 further clarified that recast for a negative amortization loan occurs after the maximum loan amount is reached (i.e., the negative amortization cap) or the introductory minimum periodic payment period expires.

As discussed above, § 1026.43(b)(7) defines “maximum loan amount” as the loan amount plus any increase in principal balance that results from negative amortization, as defined in § 1026.18(s)(7)(v), based on the terms of the legal obligation. Under the proposal, creditors would make the following two assumptions when determining the maximum loan amount: (1) The consumer makes only the minimum periodic payments for the maximum possible time, until the consumer must begin making fully amortizing payments; and (2) the maximum interest rate is reached at the earliest possible time.

As discussed above under the proposed definition of “maximum loan amount,” the Board interpreted the statutory language in TILA section 129C(a)(6)(C) as requiring creditors to fully account for any potential increase in the loan amount that might result under the loan's terms where the consumer makes only the minimum periodic payments required. The Board believed the intent of this statutory provision was to help ensure that the creditor consider the consumer's capacity to absorb the increased payment amounts that would be needed to amortize the larger loan amount once the loan is recast. The Board recognized that the approach taken towards calculating the maximum loan amount requires creditors to assume a “worst-case scenario,” but believed this approach was consistent with statutory intent to take into account the greatest potential increase in the principal balance.

Moreover, the Board noted that calculating the maximum loan amount based on these assumptions is consistent with the approach in the 2010 MDIA Interim Final Rule, which addresses disclosure requirements for negative amortization loans, and also the 2006 Nontraditional Mortgage Guidance, which provides guidance to creditors regarding underwriting negative amortization loans.

Step two: payment calculation. Once the creditor knows the maximum loan amount and period of time that remains after the loan is recast, the proposed payment calculation rule for negative amortization loans would require the creditor to use the fully indexed rate or introductory rate, whichever is greater, to calculate the substantially equal, monthly payment amount that will repay the maximum loan amount over the term of the loan that remains as of the date the loan is recast. See proposed § 226.43(c)(5)(ii)(C)(1) and (2).

Proposed comment 43(c)(5)(ii)(C)-1 clarified that creditors must follow this two-step approach when determining the consumer's repayment ability on a negative amortization loan, and also provided cross-references to aid compliance. Proposed comment 43(c)(5)(ii)(C)-2 provided further guidance to creditors regarding the relevant term of the loan that must be used for purposes of the repayment ability determination. Proposed comment 43(c)(5)(ii)(C)-3 provided illustrative examples of how to determine the consumer's repayment ability based on substantially equal monthly payments of principal and interest as required under proposed § 226.43(c)(5)(ii)(C) for a negative amortization loan.

In discussing the ability-to-repay requirements for negative amortization loans, the Board noted the anomaly that a graduated payment mortgage may have a largest scheduled payment that is larger than the payment calculated under proposed § 226.43(c)(5)(ii)(C). The Board solicited comment on whether or not the largest scheduled payment should be used in determining ability to repay. The Bureau received one comment on this issue, from an association of State bank regulators, arguing that the rule should use the largest payment scheduled. However, the Bureau does not believe that a special rule for graduated payment mortgages, which would require an exception from the statute, is necessary to ensure ability to repay these loans. It is unlikely that the calculated payment will be very different from the largest scheduled payment, and introducing this added complexity to the rule is unnecessary. Also, the one comment favoring such a choice did not include sufficient data to support use of the exception and adjustment authority under TILA, and the Bureau is not aware any such data.

Final Rule

The Bureau did not receive comments on the proposed method for calculating payments for negative amortization loans. The Bureau believes that the method proposed by the Board implements the statutory provision accurately and appropriately. Accordingly, § 1026.43(c)(5)(ii)(C) and associated commentary are adopted substantially as proposed, with minor changes for clarification.

43(c)(6) Payment Calculation for Simultaneous Loans

43(c)(6)(i)

The Board's proposed rule provided that for purposes of determining a consumer's ability to repay a loan, “a creditor must consider a consumer's payment on a simultaneous loan that is—(i) a covered transaction, by following paragraphs (c)(5)(i) and (ii) of this section” (i.e., the payment calculation rules for the covered transaction itself).

Proposed comment 43(c)(6)-1 stated that in determining the consumer's repayment ability for a covered transaction, the creditor must include consideration of any simultaneous loan which it knows or has reason to know will be made at or before consummation of the covered transaction. Proposed comment 43(c)(6)-2 explained that for a simultaneous loan that is a covered transaction, as that term was defined in proposed § 226.43(b)(1), the creditor must determine a consumer's ability to repay the monthly payment obligation for a simultaneous loan as set forth in proposed § 226.43(c)(5), taking into account any mortgage-related obligations.

The Bureau did not receive comments on this specific language or the use of the covered transaction payment calculation for simultaneous loans. For discussion of other issues regarding simultaneous loans, see the section-by-section analysis of § 1026.43(b)(12), .43(c)(2)(iv) and .43(c)(6)(ii).

The Bureau considers the language of proposed§ 226.43(c)(6)(i) to be an accurate and appropriate implementation of the statute. Accordingly, the Bureau is adopting § 1026.43(c)(6)(i) and associated commentary substantially as proposed, with minor changes for clarity. The requirement to consider any mortgage-related obligations, presented in comment 43(c)(6)-2, is now also part of the regulatory text, at § 1026.43(c)(6).

43(c)(6)(ii)

For a simultaneous loan that is a HELOC, the consumer is generally not committed to using the entire credit line at consummation. The amount of funds drawn on a simultaneous HELOC may differ greatly depending, for example, on whether the HELOC is used as a “piggyback loan” to help towards payment on a home purchase transaction or if the HELOC is opened for convenience to be drawn down at a future time. In the proposed rule, the Board was concerned that requiring the creditor to underwrite a simultaneous HELOC assuming a full draw on the credit line might unduly restrict credit access, especially in connection with non-purchase transactions, because it would require creditors to assess the consumer's repayment ability using potentially overstated payment amounts. For this reason, the Board proposed under § 226.43(c)(6)(ii) that the creditor calculate the payment for the simultaneous HELOC based on the amount of funds to be drawn by the consumer at consummation of the covered transaction. The Board solicited comment on whether this approach was appropriate.

Proposed comment 43(c)(6)-3 clarified that for a simultaneous loan that is a HELOC, the creditor must consider the periodic payment required under the terms of the plan when assessing the consumer's ability to repay the covered transaction secured by the same dwelling as the simultaneous loan. This comment explained that under proposed § 226.43(c)(6)(ii), the creditor must determine the periodic payment required under the terms of the plan by considering the actual amount of credit to be drawn by the consumer at or before consummation of the covered transaction. This comment clarified that the amount to be drawn is the amount requested by the consumer; when the amount requested will be disbursed, or actual receipt of funds, is not determinative.

Several industry commenters objected that it is difficult to know the actual amount drawn on a HELOC if it is held by another lender. One commenter suggested finding another way to do this calculation, such as by adding 1 percent of the full HELOC line to the overall monthly payment. Two banking trade associations said that the full line of credit should be considered, and if the consumer does not qualify, the line of credit can be reduced in order to qualify safely. One bank stated that creditors regulated by Federal banking agencies are bound by the interagency “Credit Risk Guidance for Home Equity Lending” (2005) to consider the full line of credit, and this will create an uneven playing field.

Other industry commenters supported use of the actual amount drawn at consummation. Both Freddie Mac and Fannie Mae stated that the Board's proposal for considering the actual amount drawn at closing was consistent with their underwriting standards. In addition, an association representing one state's credit unions stated that requiring consideration of a 100 percent draw would be onerous and inaccurate. It also asked that we make clear that the creditor does not have to recalculate a consumer's ability to repay if the amount drawn changes at consummation.

The Bureau believes that requiring consideration of 100 percent of a home equity line of credit would unnecessarily restrict credit availability for consumers. Available but unaccessed credit is not considered in determining ability to repay a mortgage when the consumer has other types of credit lines, such as credit cards. Although HELOCs are secured by the consumer's dwelling, and thus differ from other types of available but unaccessed credit, this difference does not seem determinative. Any potential dwelling-secured home equity line of credit that a creditor might grant to a consumer could simply be requested by the consumer immediately following consummation of the covered transaction. The fact that the potential credit line has been identified and enumerated prior to the transaction, rather than after, does not seem significant compared to the fact that the consumer has chosen not to access that credit, and will not be making payments on it. As with the rest of the ability-to-repay requirements, creditors should apply appropriate underwriting procedures, and are not restricted to the legally mandated minimum required by this rule, as long as they satisfy that minimum.

The requirements of the 2005 “Credit Risk Guidance for Home Equity Lending” do not change the Bureau's view of this issue. The Guidance covers home equity lending itself, not consideration of HELOCs as simultaneous loans when determining ability to repay for senior non-HELOCs. The requirement to consider the entire home equity line of credit controls only a bank's granting of that line of credit. For this reason, the Bureau does not believe that banks following this guidance will be disadvantaged. In addition, the Bureau will not be implementing the suggested alternative of adding 1 percent to the calculated monthly payment on the covered transaction. The Bureau is not aware of any data supporting the accuracy of such an approach.

In regard to the comments concerning difficulty in determining the amount of the draw and the monthly HELOC payment, the Bureau as discussed above in the section-by-section analysis of § 1026.43(c)(2)(iv) has added language to comment 43(c)(2)(iv)-4 providing more specific guidance in applying the knows or has reason to know standard. In addition, language has been added to comment 43(c)(6)-3, regarding payment calculations for simultaneous HELOCs, making clear that a creditor does not need to reconsider ability to repay if the consumer unexpectedly draws more money than planned at closing from a HELOC issued by a different creditor. In addition, the regulation language has been clarified to state that the creditor must use the amount of credit “to be” drawn at consummation, making clear that a violation does not occur if the creditor did not know or have reason to know that a different amount would be drawn.

The Board also solicited comment on whether or not a safe harbor should be given to those creditors who consider the full HELOC credit line. However, commenters did not focus on this possibility. The Bureau believes that although a creditor may choose to underwrite using the full credit line as a means of considering ability to repay in relation to the actual draw, a safe harbor is not warranted. Because the full credit line should always be equal to or greater than the actual draw, appropriate use of the full credit line in underwriting will constitute appropriate compliance without a safe harbor.

In addition to the amount of a HELOC that needs to be considered in determining ability to repay, the Board also solicited comment on whether the treatment of HELOCs as simultaneous loans should be limited to purchase transactions. The Board suggested that concerns regarding “piggyback loans” were not as acute with non-purchase transactions.

Consumer and public interest groups opposed limiting the consideration of HELOCs to purchase transactions. Several consumer advocacy groups suggested that if only purchase transactions were covered, the abuses would migrate to the unregulated space. Some commenters said they did not see a reason to exclude the cost of a simultaneous loan when it is extended as part of a refinance. Industry commenters did not focus much on this issue, but an association representing credit unions supported limiting consideration to purchase transactions in order to reduce regulatory burden on credit unions and streamline the refinancing process.

The Bureau believes that requiring consideration of HELOCs as simultaneous loans is appropriate in both purchase and non-purchase transactions. In both situations the HELOC is a lien on the consumer's dwelling with a cost that affects the viability of the covered transaction loan. The Bureau recognizes that a simultaneous HELOC in connection with a refinancing is more likely to be a convenience than one issued simultaneously with a purchase transaction, which will often cover down payment, transaction costs or other major expenses. However, the final rule accommodates this difference by allowing the creditor to base its ability-to-repay determination on the actual draw. The Bureau did not receive and is not aware of any information or data that justifies excluding actual draws on simultaneous HELOCs in connection with refinances from this rule.

For the reasons stated above, the Bureau considers the language of proposed§ 226.43(c)(6)(ii) to be an accurate and appropriate implementation of the statute. Accordingly, the Bureau is adopting § 1026.43(c)(6)(ii) and associated commentary as proposed, with minor changes for clarity.

43(c)(7) Monthly Debt-to-Income Ratio or Residual Income

As discussed above, TILA section 129C(a)(3) requires creditors to consider the debt-to-income ratio or residual income the consumer will have after paying non-mortgage debt and mortgage-related obligations, as part of the ability-to-repay determination under TILA section 129C(a)(1). The Board's proposal would have implemented this requirement in § 226.43(c)(2)(vii). The Board proposed definitions and calculations for the monthly debt-to-income ratio and residual income in § 226.43(c)(7).

With respect to the definitions, proposed § 226.43(c)(7)(i)(A) would have defined the total monthly debt obligations as the sum of: the payment on the covered transaction, as required to be calculated by proposed § 226.43(c)(2)(iii) and (c)(5); the monthly payment on any simultaneous loans, as required to be calculated by proposed § 226.43(c)(2)(iv) and (c)(6); the monthly payment amount of any mortgage-related obligations, as required to be considered by proposed § 226.43(c)(2)(v); and the monthly payment amount of any current debt obligations, as required to be considered by proposed § 226.43(c)(2)(vi). Proposed § 1026.43(c)(7)(i)(B) would have defined the total monthly income as the sum of the consumer's current or reasonably expected income, including any income from assets, as required to be considered by proposed § 226.43(c)(2)(i) and (c)(4).

With respect to the calculations, proposed § 226.43(c)(7)(ii)(A) would have required the creditor to consider the consumer's monthly debt-to-income ratio by taking the ratio of the consumer's total monthly debt obligations to total monthly income. Proposed § 226.43(c)(7)(ii)(B) would have required the creditor to consider the consumer's residual income by subtracting the consumer's total monthly debt obligations from the total monthly income. The Board solicited comment on whether consideration of residual income should account for loan amount, region of the country, and family size, and on whether creditors should be required to include Federal and State taxes in the consumer's obligations to calculate the residual income.

Proposed comment 43(c)(7)-1 would have stated that a creditor must calculate the consumer's total monthly debt obligations and total monthly income in accordance with the requirements in proposed § 226.43(c)(7). The proposed comment would have explained that creditors may look to widely accepted governmental and non-governmental underwriting standards to determine the appropriate thresholds for the debt-to-income ratio or residual income.

Proposed comment 43(c)(7)-2 would have clarified that if a creditor considers both the consumer's debt-to-income ratio and residual income, the creditor may base its determination of ability to repay on either the consumer's debt-to-income ratio or residual income, even if the determination would differ with the basis used. In the section-by-section analysis of proposed § 226.43(c)(7), the Board explained that it did not wish to create an incentive for creditors to consider and verify as few factors as possible in the repayment ability determination.

Proposed comment 43(c)(7)-3 would have provided that creditors may consider compensating factors to mitigate a higher debt-to-income ratio or lower residual income. The proposed comment would have provided that the creditor may, for example, consider the consumer's assets other than the dwelling securing the covered transaction or the consumer's residual income as a compensating factor for a higher debt-to-income ratio. The proposed comment also would have provided that, in determining whether and in what manner to consider compensating factors, creditors may look to widely accepted governmental and non-governmental underwriting standards. The Board solicited comment on whether it should provide more guidance on what factors creditors may consider, and on how creditors may include compensating factors in the repayment ability determination.

In addition, the Board solicited comment on two issues related to the use of automated underwriting systems. The Board solicited comment on providing a safe harbor for creditors relying on automated underwriting systems that use monthly debt-to-income ratios, if the system developer certifies that the system's use of monthly debt-to-income ratios in determining repayment ability is empirically derived and statistically sound. The Board also solicited comment on other methods to facilitate creditor reliance on automated underwriting systems, while ensuring that creditors can demonstrate compliance with the rule.

As discussed above in the section-by-section analysis of § 1026.43(c)(2)(vii), industry commenters and consumer advocates largely supported including consideration of the monthly debt-to-income ratio or residual income in the ability-to-repay determination and generally favored a flexible approach to consideration of those factors. In response to the Board's proposal, some consumer advocates asked that the Bureau conduct research on the debt-to-income ratio and residual income. They requested a standard that reflects the relationship between the debt-to-income ratio and residual income. One industry commenter recommended that the Bureau adopt the VA calculation of residual income. Another industry commenter suggested that the Bureau adopt the same definitions of the debt-to-income ratio and residual income as for qualified residential mortgages, to reduce compliance burdens and the possibility of errors. One industry commenter asked that consideration of residual income be permitted to vary with family size and geographic location. The commenter suggested that the residual income calculation account for Federal and State taxes. Several consumer advocates suggested that the Bureau review the VA residual income guidelines and update the cost of living tiers. They affirmed that all regularly scheduled debt payments should be included in the residual income calculation. They noted that residual income should be sufficient to cover basic living necessities, including food, utilities, clothing, transportation, and known health care expenses.

One industry commenter asked that the Bureau provide guidance on and additional examples of compensating factors, for example, situations where a consumer has many assets but a low income or high debt-to-income ratio. The commenter suggested that the Bureau clarify that the list of examples was not exclusive. Consumer advocates recommended that the Bureau not permit extensions of credit based on a good credit history or involving a high loan-to-value ratio if the debt-to-income ratio or residual income does not reflect an ability to repay. These commenters argued that credit scores and down payments reflect past behavior and incentives to make down payments, not ability to repay.

The Bureau is largely adopting § 1026.43(c)(7) as proposed, with certain clarifying changes to the commentary. Specifically, comment 43(c)(7)-1 clarifies that § 1026.43(c) does not prescribe a specific debt-to-income ratio with which creditors must comply. For the reasons discussed above in the section-by-section analysis of § 1026.43(c), the Bureau is not finalizing the portion of proposed comment 43(c)(7)-1 which would have provided that the creditor may look to widely accepted governmental and non-governmental underwriting standards to determine the appropriate threshold for the monthly debt-to-income ratio or the monthly residual income. Instead, comment 43(c)(7)-1 provides that an appropriate threshold for a consumer's monthly debt-to-income ratio or monthly residual income is for the creditor to determine in making a reasonable and good faith determination of a consumer's repayment ability.

Comment 43(c)(7)-2 clarifies guidance regarding use of both monthly debt-to-income and monthly residual income by providing that if a creditor considers the consumer's monthly debt-to-income ratio, the creditor may also consider the consumer's residual income as further validation of the assessment made using the consumer's monthly debt-to-income ratio. The Bureau is not finalizing proposed comment 43(c)(7)-2, which would have provided that if a creditor considers both the consumer's monthly debt-to-income ratio and residual income, the creditor may base the ability-to-repay determination on either metric, even if the ability-to-repay determination would differ with the basis used. The Bureau believes the final guidance better reflects how the two standards work together in practice, but the change is not intended to alter the rule.

Comment 43(c)(7)-3 also clarifies guidance regarding the use of compensating factors in assessing a consumer's ability to repay by providing that, for example, the creditor may reasonably and in good faith determine that an individual consumer has the ability to repay despite a higher monthly debt-to-income ratio or lower residual income in light of the consumer's assets other than the dwelling securing the covered transaction, such as a savings account. The creditor may also reasonably and in good faith determine that a consumer has the ability to repay despite a higher debt-to-income ratio in light of the consumer's residual income. The Bureau believes that not permitting use of compensating factors may reduce access to credit in some cases, even if the consumer could afford the mortgage. The Bureau does not believe, however, that the rule should provide an extensive list of compensating factors that the creditor may consider in assessing repayment ability. Instead, creditors should make reasonable and good faith determinations of the consumer's repayment ability in light of the facts and circumstances. This approach to compensating factors is consistent with the final rule's flexible approach to the requirement that creditors make a reasonable and good faith of a consumer's repayment ability throughout § 1026.43(c).

The Bureau will consider conducting a future study on the debt-to-income ratio and residual income. Except for one small creditor and the VA, the Bureau is not aware of any creditors that routinely use residual income in underwriting, other than as a compensating factor. The VA underwrites its loans to veterans based on a residual income table developed in 1997. The Bureau understands that the table shows the residual income desired for the consumer based on the loan amount, region of the country, and family size, but does not account for differences in housing or living costs within regions (for instance rural Vermont versus New York City). The Bureau also understands that the residual income is calculated by deducting obligations, including Federal and State taxes, from effective income. However, at this time, the Bureau is unable to conduct a detailed review of the VA residual income guidelines, which would include an analysis of whether those guidelines are predictive of repayment ability, to determine if those standards should be incorporated, in whole or in part, into the ability-to-repay analysis that applies to the entire residential mortgage market. Further, the Bureau believes that providing broad standards for the definition and calculation of residual income will help preserve flexibility if creditors wish to develop and refine more nuanced residual income standards in the future. The Bureau accordingly does not find it necessary or appropriate to specify a detailed methodology in the final rule for consideration of residual income.

The final rule also does not provide a safe harbor for creditors relying on automated underwriting systems that use monthly debt-to-income ratios. The Bureau understands that creditors routinely rely on automated underwriting systems, many of which are proprietary and thus lack transparency to the individual creditors using the systems. Such systems may decide, for example, whether the debt-to-income ratio and compensating factors are appropriate, but may not disclose to the individual creditors using such systems which compensating factors were used for loan approval. However, the Bureau does not believe a safe harbor is necessary in light of the flexibility the final rule provides to creditors in assessing a consumer's repayment ability, including consideration of monthly debt-to-income ratios. See comments 43(c)(1)-1 and 43(c)(2)-1.

Finally, the Bureau notes the contrast between the flexible approach to considering and calculating debt-to-income in § 1026.43(c)(2)(vii) and (7) and the specific standards for evaluating debt-to-income for purposes of determining whether a covered transaction is a qualified mortgage under § 1026.43(e)(2). For the reasons discussed below in the section-by-section analysis of § 1026.43(e)(2), the Bureau believes a specific, quantitative standard for evaluating a consumer's debt-to-income ratio is appropriate in determining whether a loan receives either a safe harbor or presumption of compliance with the repayment ability requirements of § 1026.43(c)(1) pursuant to § 1026.43(e)(2). However, the ability-to-repay requirements in § 1026.43(c) will apply to the whole of the mortgage market and therefore require flexibility to permit creditors to assess repayment ability while ensuring continued access to responsible, affordable mortgage credit. Accordingly, the final rule sets minimum underwriting standards while providing creditors with flexibility to use their own quantitative standards in making the repayment ability determination required by § 1026.43(c)(1).

43(d) Refinancing of Non-Standard Mortgages

Two provisions of section 1411 of the Dodd-Frank Act address the refinancing of existing mortgage loans under the ability-to-repay requirements. As provided in the Dodd-Frank Act, TILA section 129C(a)(5) provides that certain Federal agencies may create an exemption from the income verification requirements in TILA section 129C(a)(4) if certain conditions are met. 15 U.S.C. 1639c(a)(5). In addition, TILA section 129C(a)(6)(E) provides certain special ability-to-repay requirements to encourage applications to refinance existing “hybrid loans” into a “standard loans” with the same creditor, where the consumer has not been delinquent on any payments on the existing loan and the monthly payments would be reduced under the refinanced loan. The statute allows creditors to give special weight to the consumer's good standing and to consider whether the refinancing would prevent a likely default, as well as other potentially favorable treatment to the consumer. However, it does not expressly exempt applications for such “payment shock refinancings” from TILA's general ability-to-repay requirements or define “hybrid” or “standard loans.” 15 U.S.C. 1639c(a)(6)(E).

The Board noted in its proposal that it reviewed the Dodd-Frank Act's legislative history, consulted with consumer advocates and representatives of both industry and the GSEs, and examined underwriting rules and guidelines for the refinance programs of private creditors, GSEs and Federal agencies, as well as for the Home Affordable Modification Program (HAMP). The Board noted that it also considered TILA section 129C(a)(5), which permits Federal agencies to adopt rules exempting refinancings from certain of the ability-to-repay requirements in TILA section 129C(a).

In proposing § 226.43(d) to implement TILA section 129C(a)(6)(E), the Board interpreted the statute as being intended to afford greater flexibility to creditors of certain home mortgage refinancings when complying with the general ability-to-repay provisions in TILA section 129C(a). Consistent with this reading of the statute, the proposal would have provided an exemption from certain criteria required to be considered as part of the general repayment ability determination under TILA section 129C(a). Specifically, the Board's proposal would have permitted creditors to evaluate qualifying applications without having to verify the consumer's income and assets as prescribed in the general ability-to-repay requirements, provided that a number of additional conditions were met. In addition, the proposal would have permitted a creditor to calculate the monthly payment used for determining the consumer's ability to repay the new loan based on assumptions that would typically result in a lower monthly payment than those required to be used under the general ability-to-repay requirements. The proposal also clarified the conditions that must be met in a home mortgage refinancing in order for this greater flexibility to apply.

The Board noted that TILA section 129C(a)(6)(E)(ii) permits creditors to give prevention of a “likely default should the original mortgage reset a higher priority as an acceptable underwriting practice.” 15 U.S.C. 1639c(a)(6)(E)(ii). The Board interpreted this provision to mean that certain ability-to-repay criteria under TILA section 129C(a) should not apply to refinances that meet the requisite conditions. TILA section 129C(a) specifically prescribes the requirements that creditors must meet to satisfy the obligation to determine a consumer's ability to repay a mortgage loan. The Board concluded that the term “underwriting practice” could reasonably be interpreted to refer to the underwriting rules prescribed in earlier portions of TILA section 129C(a); namely, those concerning the general ability-to-repay underwriting requirements.

The Board also structured its proposal to provide for flexibility in underwriting that is characteristic of so-called “streamlined refinances,” which are offered by creditors to existing customers without having to go through a full underwriting process appropriate for a new origination. The Board noted that section 1411 of the Dodd-Frank Act specifically authorizes streamlined refinancings of loans made, guaranteed, or insured by Federal agencies, and concluded that TILA section 129C(a)(6)(E) is most reasonably interpreted as being designed to address the remaining market for streamlined refinancings; namely, those offered under programs of private creditors and the GSEs. The Board stated that in its understanding typical streamlined refinance programs do not require documentation of income and assets, although a verbal verification of employment may be required. The Board further noted that TILA section 129C(a)(6)(E) includes three central elements of typical streamlined refinance programs, in that it requires that the creditor be the same for the existing and new mortgage loan obligation, that the consumer have a positive payment history on the existing mortgage loan obligation, and that the payment on the new refinancing be lower than on the existing mortgage loan obligation.

One difference the Board noted between the statute and typical streamlined refinance programs is that the statute targets consumers facing “likely default” if the existing mortgage “reset[s].” The Board indicated that, by contrast, streamlined refinance programs may not be limited to consumers at risk in this way. For example, streamlined refinancing programs may assist consumers who are not facing potential default but who simply wish to take advantage of lower rates despite a drop in their home value or wish to switch from a less stable variable-rate product to a fixed-rate product. The Board noted parallels between TILA's new refinancing provisions and the focus of HAMP, a government program specifically aimed at providing modifications for consumers at risk of “imminent default,” or in default or foreclosure. However, the Board noted that underwriting criteria for a HAMP modification are considerably more stringent than for a typical streamlined refinance.

On balance, the Board interpreted the statutory language as being modeled on the underwriting standards of typical streamlined refinance programs rather than the tighter standards of HAMP. The Board concluded that Congress intended to facilitate opportunities to refinance loans on which payments could become significantly higher and thus unaffordable. The Board cautioned that applying underwriting standards that are too stringent could impede refinances that Congress intended to encourage. In particular, the statutory language permitting creditors to give “likely default” a “higher priority as an acceptable underwriting practice” indicates that flexibility in these special refinances should be permitted. In addition, underwriting standards that go significantly beyond those used in existing streamlined refinance programs could create a risk that these programs would be unable to meet the TILA ability-to-repay requirements; thus, an important refinancing resource for at-risk consumers would be compromised and the overall mortgage market potentially disrupted at a vulnerable time.

The Board noted, however, that consumers at risk of default when higher payments are required might present greater credit risks to the institutions holding their loans when those loans are refinanced without verifying the consumer's income and assets. Accordingly, the Board's proposal would have imposed some requirements that are more stringent than those of typical streamlined refinance programs as a prerequisite to the refinancing provision under proposed § 226.43(d). For example, the proposal would have permitted a consumer to have had only one delinquency of more than 30 days in the 24 months immediately preceding the consumer's application for a refinance. By contrast, the Board indicated that streamlined refinance programs of which it is aware tend to consider the consumer's payment history for only the last 12 months. In addition, the proposal would have defined the type of loan into which a consumer may refinance under TILA's new refinancing provisions to include several characteristics designed to ensure that those loans are stable and affordable. These include a requirement that the interest rate be fixed for the first five years after consummation and that the points and fees be capped at three percent of the total loan amount, subject to a limited exemption for smaller loans.

43(d)(1) Definitions

In the Board's proposal, § 226.43(d)(1) established the scope of paragraph (d) and set forth the conditions under which the special refinancing provisions applied, while proposed § 226.43(d)(2) addressed the definitions for “non-standard mortgage,” “standard mortgage,” and “refinancing.” The Bureau believes that paragraph (d) should begin with the relevant definitions, before proceeding to the scope and conditions of the special refinancing provisions. The rule finalized by the Bureau is accordingly reordered. The following discussion details the definitions adopted in § 1026.43(d)(1), which were proposed by the Board under § 226.43(d)(2).

Proposed § 226.43(d)(2) defined the terms “non-standard mortgage” and “standard mortgage.” As noted earlier, the statute does not define the terms “hybrid loan” and “standard loan” used in the special refinancing provisions of TILA section 129C(a)(6)(E). Therefore, the Board proposed definitions it believed to be consistent with the policy objective underlying these special provisions: Facilitating the refinancing of home mortgages on which consumers risk a likely default due to impending payment shock into more stable and affordable products.

43(d)(1)(i) Non-Standard Mortgage

As noted above, the statute does not define the terms “hybrid loan” and “standard loan” used in TILA section 129C(a)(6)(E). The Board proposed definitions it believed to be consistent with Congress's objectives. Proposed § 226.43(d)(2)(i) substituted the term “non-standard mortgage” for the statutory term “hybrid loan” and would have defined non-standard mortgage as any “covered transaction,” as defined in proposed § 226.43(b)(1), that is:

• An adjustable-rate mortgage, as defined in § 226.18(s)(7)(i), with an introductory fixed interest rate for a period of one year or longer;

• An interest-only loan, as defined in § 226.18(s)(7)(iv); or

• A negative amortization loan, as defined in § 226.18(s)(7)(v).

Proposed comment 43(d)(2)(i)(A)-1 explained the application of the definition of non-standard mortgage to an adjustable-rate mortgage with an introductory fixed interest rate for one or more years. This proposed comment clarified that, for example, a covered transaction with a fixed introductory rate for the first two, three or five years that then converts to a variable rate for the remaining 28, 27 or 25 years, respectively, is a non-standard mortgage. By contrast, a covered transaction with an introductory rate for six months that then converts to a variable rate for the remaining 29 and 1/2 years is not a non-standard mortgage.

The Board articulated several rationales for its proposed definition of a non-standard mortgage. First, the Board noted that the legislative history of the Dodd-Frank Act describes “hybrid” mortgages as mortgages with a “blend” of fixed-rate and adjustable-rate characteristics—generally loans with an initial fixed period and adjustment periods, such as “2/23s and 3/27s.” The Board also stated that the legislative history indicates that Congress was concerned about consumers being trapped in mortgages likely to result in payments that would suddenly become significantly higher—often referred to as “payment shock”—because their home values had dropped, thereby “making refinancing difficult.”

The Board interpreted Congress' concern about consumers being at risk due to payment shock as supporting an interpretation of the term “hybrid loan” to encompass both loans that are “hybrid” in that they start with a fixed interest rate and convert to a variable rate, but also loans that are “hybrid” in that consumers can make payments that do not pay down principal for a period of time that then convert to higher payments covering all or a portion of principal. By defining “non-standard mortgage” in this way, the proposal was intended to increase refinancing options for a wide range of at-risk consumers while conforming to the statutory language and legislative intent.

The proposed definition of “non-standard mortgage” would not have included adjustable-rate mortgages whose rate is fixed for an initial period of less than one year. In those instances, the Board posited that a consumer may not face “payment shock” because the consumer has paid the fixed rate for such a short period of time. The Board also expressed concern that allowing streamlined refinancings under this provision where the interest rate is fixed for less than one year could result in “loan flipping.” A creditor, for example, could make a covered transaction and then only a few months later refinance that loan under proposed § 226.43(d) to take advantage of the exemption from certain ability-to-repay requirements while still profiting from the refinancing fees.

The Board expressed concern that under its proposed definition, a consumer could refinance out of a relatively stable product, such as an adjustable-rate mortgage with a fixed interest rate for a period of 10 years, which then adjusts to a variable rate for the remaining loan term, and that it was unclear whether TILA section 129C(a)(6)(E) was intended to cover this type of product. The Board solicited comment on whether adjustable-rate mortgages with an initial fixed rate should be considered non-standard mortgages regardless of how long the initial fixed rate applies, or if the proposed initial fixed-rate period of at least one year should otherwise be revised.

The proposed definition of non-standard mortgage also did not include balloon-payment mortgages. The Board noted that balloon-payment mortgages are not clearly “hybrid” products, given that the monthly payments on a balloon-payment mortgage do not necessarily increase or change from the time of consummation; rather, the entire outstanding principal balance becomes due on a particular, predetermined date. The Board stated that consumers of balloon-payment mortgages typically expect that the entire loan balance will be due at once at a certain point in time and are generally aware well in advance that they will need to repay the loan or refinance.

The Board solicited comment on whether to use its legal authority to include balloon-payment mortgages in the definition of non-standard mortgage for purposes of the special refinancing provisions of TILA section 129C(a)(6)(E). The Board also requested comment generally on the appropriateness of the proposed definition of non-standard mortgage.

Commenters on this aspect of the proposal generally urged the Bureau to expand in various ways the proposed definition of non-standard mortgage and either supported or did not address the proposed definition's inclusion of adjustable-rate mortgages, interest-only loans, or negative amortization loans. One consumer group commented that it supported the Board's proposed definition of non-standard mortgage. Other consumer group commenters stated that the Bureau should use its exemption and adjustment authority under TILA to include balloon-payment loans within the scope of proposed § 226.43(d). In addition, one industry commenter stated that creditors should have flexibility to refinance a performing balloon-payment loan within the six months preceding, or three months following, a balloon payment date without regard to the ability-to-pay requirements. In contrast, one industry commenter stated that balloon-payment loans should not be included in the definition of non-standard mortgage, because consumers are generally well aware of the balloon payment feature in a loan, which is clearly explained to customers. This industry commenter further stated that during the life of a balloon-payment loan, its customers often make regular payments that reduce the principal balance and that balloon-payment loans do not make it more likely that a consumer will default.

While the Bureau agrees that many consumers may need to seek a refinancing when a balloon loan payment comes due, given the approach that the Bureau has taken to implementing the payment shock refinancing provision in § 1026.43(d), the Bureau is declining to expand the definition of non-standard mortgage to include balloon-payment mortgages. As discussed in more detail in the supplementary information to § 1026.43(d)(3), as adopted § 1026.43(d) provides a broad exemption to all of the general ability-to-repay requirements set forth in § 1026.43(c) when a non-standard mortgage is refinanced into a standard mortgage provided that certain conditions are met. The point of this exemption is to enable creditors, without going through full underwriting, to offer consumers who are facing increased monthly payments due to the recast of a loan a new loan with lower monthly payments. Thus, a key element of the exemption is that the monthly payment on the standard mortgage be materially lower than the monthly payment for the non-standard mortgage. As discussed in the section-by-section analysis of § 1026.43(d)(1) below, the Bureau is adopting a safe harbor for reductions of 10 percent. Balloon payments pose a different kind of risk to consumers, one that arises not from the monthly payments (which often tend to be low) but from the balloon payment due when the entire remaining balance becomes due. The provisions of § 1026.43(d)(1) are not meant to address this type of risk. Accordingly, the Bureau declines to expand the definition of non-standard mortgage to include balloon-payment loans. The Bureau believes, however, that where a consumer is performing under a balloon-payment mortgage and is offered a new loan of a type that would qualify as a standard loan with monthly payments at or below the payments of the balloon-payment mortgage, creditors will have little difficulty in satisfying the ability-to-repay requirements.

Consumer group commenters and one GSE commenter argued that the definition of non-standard mortgage should accommodate GSE-held loans. These commenters stated that these loans should receive the same income verification exemption as Federal agency streamlined refinancing programs. These commenters noted that while the GSEs are held in conservatorship by the Federal government, GSE-held loans should be treated the same as FHA for purposes of streamlined refinance programs, which are ultimately about reducing the risk to the taxpayer by avoiding default by consumers who could receive lower-cost mortgage loans. Consumer group commenters further urged that GSE streamlined refinance programs should be subject to standards at least as stringent as those for the FHA streamlined refinance program.

In addition, one of the GSEs questioned the policy justification for the differences between sections 129C(a)(5) and 129C(a)(6)(E) of TILA. TILA section 129C(a)(5), which applies to certain government loans, permits Federal agencies to exempt certain refinancings from the income and asset verification requirement without regard to the original mortgage product, in contrast to TILA section 129C(a)(6)(E), which as discussed above applies only when the original loan is a “hybrid” loan. This commenter noted that consumers with certain types of mortgage loans, such as fixed-rate and balloon-payment loans, may have to go through a more costly and cumbersome process to refinance their mortgages than consumers with government loans.

The Bureau declines to adopt regulations implementing TILA section 129C(a)(5). The Bureau notes that TILA section 129C(a)(5) expressly confers authority on certain Federal agencies (i.e., HUD, VA, USDA, and RHS) to exempt from the income verification requirement refinancings of certain loans made, guaranteed, or insured by such Federal agencies. The scope of TILA section 129C(a)(5) is limited to such Federal agencies or government-guaranteed or -insured loans. The Bureau also declines to expand the scope of § 1026.43(d) to include GSE refinancings that do not otherwise fall within the scope of § 1026.43(d). While accommodation for GSE-held mortgage loans that are not non-standard mortgages under § 1026.43(d) may be appropriate, the Bureau wishes to obtain additional information in connection with GSE refinancings and has requested feedback in a proposed rule published elsewhere in today's Federal Register. However, the Bureau notes that to the extent a loan held by the GSEs (or a loan made, guaranteed or insured by the Federal agencies above) qualifies as a non-standard mortgage under § 1026.43(d)(1)(i) and the other conditions in § 1026.43(d) are met, the refinancing provisions of general applicability in § 1026.43(d) would be available for refinancing a GSE-held loan.

Industry commenters and one industry trade association commented that special ability-to-repay requirements should be available for all rate-and-term refinancings, regardless of whether the refinancings are insured or guaranteed by the Federal government or involve a non-standard mortgage. One industry trade association stated that such special ability-to-repay requirements should incorporate similar standards to those established for certain government loans in TILA section 129C(a)(5), including a requirement that the consumer not be 30 or more days delinquent. For such loans, this trade association stated that other requirements under TILA section 129C(a)(6)(E) regarding payment history should not be imposed, because the consumer is already obligated to pay the debt and the note holder in many cases will already bear the credit risk. Other commenters stated that because a rate-and-term refinancing would offer the consumer a better rate (except in the case of adjustable rate mortgages), there is no reason to deny the creditor the ability to improve its credit risk and to offer the consumer better financing. Several industry commenters and one GSE noted that streamlined refinancing programs are an important resource for consumers seeking to refinance into a lower monthly payment mortgage even when the underlying mortgage loan is not a non-standard mortgage, and urged the Bureau to considering modifying proposed § 226.43(d) to include conventional loans where the party making or purchasing the new loan already owns the credit risk.

The Bureau declines to expand the scope of § 1026.43(d) to include rate-and-term refinancings when the underlying mortgage is not a non-standard mortgage, as defined in § 1026.43(d)(1)(i). The Bureau believes that the statute clearly limits the refinancing provision in TILA section 129(C)(6)(E) to circumstances where the loan being refinanced is a “hybrid loan” and where the refinancing could “prevent a likely default.” The Bureau agrees with the Board that TILA section 129C(a)(6)(E) is intended to address concerns about loans involving possible payment shock. Where a consumer has proven capable of making payments, is about to experience payment shock, is at risk of default, and is refinancing to a mortgage with a lower monthly payment and with product terms that do not pose any increased risk, the Bureau believes that the benefits of the refinancing outweigh the consumer protections afforded by the ability-to-repay requirements. Absent these exigent circumstances, the Bureau believes that creditors should determine that the consumer has the ability to repay the mortgage loan. The Bureau does not believe that a consumer who receives an initial lower monthly payment from a rate-and-term refinancing actually receives a benefit if the consumer cannot reasonably be expected to repay the loan. Also, the Bureau notes that some of the scenarios identified by commenters, such as offering a consumer a better rate with a rate-and-term refinancing where the creditor bears the credit risk, would be exempt from the ability-to-repay requirements. A refinancing that results in a reduction in the APR with a corresponding change in the payment schedule and meets the other conditions in § 1026.20(a) is not a “refinancing” for purposes of § 1026.43, and therefore is not subject to the ability-to-repay requirements. As with other terms used in TILA section 129C, the Bureau believes that this interpretation is necessary to achieve Congress's intent.

Several other industry commenters urged the Bureau to broaden the definition of non-standard mortgage to include refinancings extended pursuant to the Home Affordable Refinance Program (HARP) and similar programs. One such commenter indicated that under HARP, a loan can only be refinanced if the consumer is not in default, the new payment is fully amortizing, and both the original and new loans comply with agency requirements. This commenter stated that HARP permits consumers who would not otherwise be able to refinance due to a high loan-to-value ratio or other reasons to refinance into another loan, providing a consumer benefit. The commenter indicated that HARP loans do not meet all of the proposed ability-to-repay requirements and that the Bureau should use its authority to provide that HARP and other similar programs are exempt from the ability-to-repay requirements, as they promote credit availability and increasing stability in the housing market. The Bureau acknowledges that HARP refinancings and the payment shock refinancings addressed under TILA section 129C(a)(6)(E) are both intended to assist consumers harmed by the financial crisis. Although both types of refinancings are motivated by similar goals, the Bureau does not believe that expanding § 1026.43(d) to include all HARP refinancings is consistent with TILA section 129C(a)(6)(E) because HARP refinancings are not predicated on the occurrence of payment shock and a consumer's likely default. For example, a consumer with a mortgage loan that will not recast and who is not at risk of default may qualify for a HARP refinancing if the consumer's loan-to-value ratio exceeds 80 percent. The Bureau strongly believes that § 1026.43(d) should be limited to instances where a consumer is facing payment shock and likely default.

While not limited to the prevention of payment shock and default, the Bureau acknowledges that extensions of credit made pursuant to programs such as HARP are intended to assist consumers harmed by the financial crisis. Furthermore, these programs employ complex underwriting requirements to determine a consumer's ability to repay. Thus, it may be appropriate to modify the ability-to-repay requirements to accommodate such programs. However, an appropriate balance between helping affected consumers and ensuring that these consumers are offered and receive residential mortgage loans on terms that reasonably reflect consumers' ability to repay must be found. To determine how to strike this balance, the Bureau wishes to obtain additional information in connection with these programs and has requested feedback in a proposed rule published elsewhere in today's Federal Register.

Accordingly, the definition of “non-standard mortgage” is adopted as proposed, renumbered as § 1026.43(d)(1)(i). In addition, comment 43(d)(2)(i)(A)-1 also is adopted as proposed, renumbered as 43(d)(1)(i)(A)-1.

43(d)(1)(ii) Standard Mortgage

Proposed § 226.43(d)(2)(ii) would have substituted the term “standard mortgage” for the statutory term “standard loan” and defined this term to mean a covered transaction that has the following five characteristics:

• First, the regular periodic payments may not: (1) Cause the principal balance to increase; (2) allow the consumer to defer repayment of principal; or (3) result in a balloon payment.
• Second, the total points and fees payable in connection with the transaction may not exceed three percent of the total loan amount, with exceptions for smaller loans specified in proposed § 226.43(e)(3).
• Third, the loan term may not exceed 40 years.
• Fourth, the interest rate must be fixed for the first five years after consummation.
• Fifth, the proceeds from the loan may be used solely to pay—(1) the outstanding principal balance on the non-standard mortgage; and (2) closing or settlement charges required to be disclosed under RESPA.

Proposed limitations on regular periodic payments. Proposed § 226.43(d)(2)(ii)(A) would have required that a standard mortgage provide for regular periodic payments that do not result in negative amortization, deferral of principal repayment, or a balloon payment. Proposed comment 43(d)(2)(ii)(A)-1 clarified that “regular periodic payments” are payments that do not result in an increase of the principal balance (negative amortization) or allow the consumer to defer repayment of principal. The proposed comment explained that the requirement for “regular periodic payments” means that the contractual terms of the standard mortgage must obligate the consumer to make payments of principal and interest on a monthly or other periodic basis that will repay the loan amount over the loan term. Proposed comment 43(d)(2)(ii)(A)-1 further explained that, with the exception of payments resulting from any interest rate changes after consummation in an adjustable-rate or step-rate mortgage, the periodic payments must be substantially equal, with a cross-reference to proposed comment 43(c)(5)(i)-3 regarding the meaning of “substantially equal.” In addition, the comment clarified that “regular periodic payments” do not include a single-payment transaction and cross-referenced similar commentary on the meaning of “regular periodic payments” under proposed comment 43(e)(2)(i)-1. Proposed comment 43(d)(2)(ii)(A)-1 also cross-referenced proposed comment 43(e)(2)(i)-2 to explain the prohibition on payments that “allow the consumer to defer repayment of principal.”

One consumer group commenter stated that it supported the exclusion of negative amortization, interest-only payments, and balloon payments from the definition of standard mortgage. In addition, several other consumer groups commented in support of the Board's proposal to exclude balloon-payment loans from the definition of standard mortgage. These commenters stated that balloon-payment products, even with self-executing renewal, should not be permitted to take advantage of an exemption from the general underwriting standards in § 1026.43(c). Consumer groups expressed concern that, in cases where the consumer does not have assets sufficient to make the balloon payment, balloon-payment loans will necessarily require another refinance or will lead to a default. The Bureau agrees with the concerns expressed by such commenters and believes that it is appropriate to require that balloon-payment loans be underwritten in accordance with the general ability-to-repay standard, rather than under the payment shock refinancing provision in § 1026.43(d). Accordingly, the Bureau is not expanding the definition of standard mortgage to include balloon-payment mortgages.

The Bureau received no other comment on this proposed definition. Accordingly, the Bureau is adopting the definition of standard mortgage as proposed, renumbered as § 1026.43(d)(1)(ii)(A). Similarly, the Bureau received no comment on proposed comment 43(d)(2)(ii)(A)-1, which is adopted as proposed and renumbered as 43(d)(1)(ii)(A)-1.

Proposed three percent cap on points and fees. Proposed § 226.43(d)(2)(ii)(B) would have prohibited creditors from charging points and fees on the mortgage loan of more than three percent of the total loan amount, with certain exceptions for small loans. Specifically, proposed § 226.43(d)(2)(ii)(B) cross-referenced the points and fees provisions under proposed § 226.43(e)(3), thereby applying the points and fees limitations for a “qualified mortgage” to a standard mortgage. The points and fees limitation for a “qualified mortgage” and the relevant exception for small loans are discussed in detail in the section-by-section analysis of § 1026.43(e)(3) below.

The Board noted several reasons for the proposed limitation on the points and fees that may be charged on a standard mortgage. First, the limitation was intended to prevent creditors from undermining the provision's purpose—placing at-risk consumers into more affordable loans—by charging excessive points and fees for the refinance. Second, the points and fees limitation was intended to ensure that consumers attain a net benefit in refinancing their non-standard mortgage. The higher a consumer's up-front costs to refinance a home mortgage, the longer it will take for the consumer to recoup those costs through lower payments on the new mortgage. By limiting the amount of points and fees that can be charged in a refinance covered by proposed § 226.43(d), the provision increases the likelihood that the consumer will hold the loan long enough to recoup those costs. Third, the proposed limitation was intended to be consistent with the provisions set forth in TILA section 129C(a)(5) regarding certain refinancings under Federal agency programs.

The Board requested comment on the proposal to apply the same limit on the points and fees that may be charged for a “qualified mortgage” under § 226.43(e) to the points and fees that may be charged on a “standard mortgage” under § 226.43(d). The Bureau received no comments on this proposed points and fees threshold, which is adopted as proposed, renumbered as § 1026.43(d)(1)(ii)(B). See the section-by-section analysis of § 1026.43(e)(3) below for more specific information regarding the limitations applicable to “points and fees” for qualified mortgages and refinancings under § 1026.43(d).

Proposed loan term of no more than 40 years. Proposed § 226.43(d)(2)(ii)(C) would have provided that, to qualify as a standard mortgage under proposed § 226.43(d), a covered transaction may not have a loan term of more than 40 years. The Board stated that this condition was intended to ensure that creditors and consumers have sufficient options to refinance a 30-year loan, for example, which is unaffordable for the consumer in the near term, into a loan with lower, more affordable payments over a longer term. This flexibility may be especially important in higher cost areas where loan amounts on average exceed loan amounts in other areas.

The Board noted that loans with longer terms may cost more over time, but indicated that it was reluctant to foreclose options for consumers for whom the lower payment of a 40-year loan might make the difference between defaulting and not defaulting. The Board also noted that prevalent streamlined refinance programs permit loan terms of up to 40 years and expressed concern about disrupting the current mortgage market at a vulnerable time. The Board specifically requested comment on the proposed condition to allow a standard mortgage to have a loan term of up to 40 years. The Bureau received no comment on this proposed condition, which is adopted as proposed, renumbered as § 1026.43(d)(1)(ii)(C).

Proposed requirement that the interest rate be fixed for the first five years. Proposed § 226.43(d)(2)(ii)(D) would have required that a standard mortgage have a fixed interest rate for the first five years after consummation. Proposed comment 43(d)(2)(ii)(D)-1 provided an illustrative example. The proposed comment also cross-referenced proposed comment 43(e)(2)(iv)-3.iii for guidance regarding step-rate mortgages.

The Board articulated several reasons for requiring a minimum five-year fixed-rate period for standard mortgages. First, the Board noted that a fixed rate for five years is consistent with TILA section 129C(b)(2)(A)(v), which requires the creditor to underwrite a qualified mortgage based on the maximum interest rate that may apply during the first five years. The Board indicated that Congress intended both qualified mortgages and standard mortgages to be stable loan products, and therefore that the required five-year fixed-rate period for qualified mortgages would also be an appropriate benchmark for standard mortgages. The Board further stated that the safeguard of a fixed rate for five years after consummation would help to ensure that consumers refinance into products that are stable for a substantial period of time. In particular, a fixed payment for five years after consummation would constitute a significant improvement in the circumstances of a consumer who may have defaulted absent the refinance. The Board specifically noted that the proposal would permit so-called “5/1 ARMs,” where the interest rate is fixed for the first five years, after which time the rate becomes variable, to be standard mortgages.

The Board requested comment on the proposal defining a standard mortgage as a mortgage loan with an interest rate that is fixed for at least the first five years after consummation, including on whether the rate should be required to be fixed for a shorter or longer period and data to support any alternative time period. One consumer group commenter stated that the use of adjustable-rate mortgages should be limited in the definition of standard mortgage. This commenter stated that adjustable-rate mortgage loans contributed to the subprime lending expansion and the financial crisis that followed. In particular, this commenter expressed concern that adjustable-rate mortgage loans were utilized in loan-flipping schemes that trapped consumers in unaffordable loans, forcing such consumers to refinance into less affordable mortgage loans. This commenter indicated that standard mortgages should be limited to fixed and step-rate loans and, in low or moderate interest rate environments, adjustable-rate mortgages with a 5-year or longer-term fixed period. However, this commenter urged the Bureau to consider permitting shorter-term adjustable-rate mortgages to be standard mortgages in high interest rate environments because in such circumstance, an adjustable-rate mortgage could potentially reduce the consumer's monthly payments at recast, which may outweigh the risks of increased payments for some consumers.

The Bureau is adopting the requirement that a standard mortgage have a fixed interest rate for the first five years after consummation as proposed, renumbered as § 1026.43(d)(1)(ii)(D). The Bureau agrees with the Board that the intent of TILA section 129C(a)(6)(E) appears to be to facilitate refinances of riskier mortgages into more stable loan products, and accordingly, believes that a standard mortgage should provide for a significant period of time during which payments will be predictable, based on a fixed rate or step rates that are set at the time of consummation. The Bureau believes that five years is an appropriate standard in part because it is consistent with the statutory requirement for a qualified mortgage under section 129C(b)(2)(A)(v). The Bureau believes that predictability for consumers is best effectuated by a single rule that applies in all interest rate environments, rather than a rule that depends on the interest rate environment in effect at the time of the refinancing. Further, given that § 1026.43(d) provides an exemption from the general ability-to-repay requirements in § 1026.43(c), the Bureau believes that it is important that a refinancing conducted in accordance with § 1026.43(d) result in a stable loan product and predictable payments for a significant period of time.

In addition, the Board solicited comment on whether a balloon-payment mortgage of at least five years should be considered a standard mortgage under the refinancing provisions of proposed § 226.43(d). The Board noted that in some circumstances, a balloon-payment mortgage with a fixed, monthly payment for five years might benefit a consumer who otherwise would have defaulted. The Board further noted that a five-year balloon-payment mortgage may not be appreciably less risky for the consumer than a “5/1 ARM,” which is permitted under the proposal, depending on the terms of the rate adjustment scheduled to occur in year five.

As discussed above, several consumer groups stated that balloon products, even with self-executing renewal, should not be permitted to take advantage of an exemption from the general underwriting standards in § 1026.43(c). Consumer groups expressed concern that, in cases where the consumer does not have assets sufficient to make the balloon payment, balloon-payment mortgages will necessarily require another refinance or will lead to a default. For the reasons discussed in the supplementary information to § 1026.43(d)(1)(ii)(A) above, the Bureau is not expanding the definition of “standard mortgage” to include balloon-payment mortgages.

Proposed requirement that loan proceeds be used for limited purposes. Proposed § 226.43(d)(2)(ii)(E) would have restricted the use of the proceeds of a standard mortgage to two purposes:

• To pay off the outstanding principal balance on the non-standard mortgage; and
• To pay closing or settlement charges required to be disclosed under the Real Estate Settlement Procedures Act, 12 U.S.C. 2601 et seq., which includes amounts required to be deposited in an escrow account at or before consummation.

Proposed comment 43(d)(2)(ii)(E)-1 clarified that if the proceeds of a covered transaction are used for other purposes, such as to pay off other liens or to provide additional cash to the consumer for discretionary spending, the transaction does not meet the definition of a “standard mortgage.”

The Board expressed concern that permitting the consumers to lose additional equity in their homes under the proposed refinancing provisions could undermine the financial stability of those consumers, thus contravening the purposes of TILA section 129C(a)(6)(E). The Board requested comment, however, on whether some de minimis amount of cash to the consumer should be permitted, either because this allowance would be operationally necessary to cover transaction costs or for other reasons, such as to reimburse a consumer for closing costs that were over-estimated but financed.

The Bureau received only one comment on this aspect of the proposal. An association of State bank regulators agreed that the rule should generally restrict the use of the proceeds of the standard mortgage to paying off the outstanding balance on the non-standard mortgage or to pay closing or settlement costs. However, they urged the Bureau to provide an exemption that would permit loan proceeds to be used to pay for known home repair needs and suggested that any such exemption require the consumer to provide verified estimates in advance in order to ensure that loan proceeds are used only for required home repairs.

The Bureau is adopting the limitation on the use of loan proceeds as proposed, renumbered as § 1026.43(d)(1)(ii)(E). The Bureau declines to permit the proceeds of a refinancing conducted in accordance with § 1026.43(d) to be used for home repair purposes, for several reasons. First, the Bureau believes that such an exemption would be inconsistent with the statutory purposes of TILA section 129C(a)(6)(E), which is intended to permit refinancings on the basis of less stringent underwriting in the narrow circumstances where a consumer's non-standard mortgage is about to recast and lead to a likely default by the consumer. The Bureau believes that permitting a consumer to utilize home equity for home repairs in connection with a refinancing conducted pursuant to § 1026.43(d) could further compromise the financial position of consumers who are already in a risky financial position. The Bureau believes that it would be more appropriate, where home repairs are needed, for a creditor to perform the underwriting required to advance any credit required in connection with those repairs. In addition, the Bureau believes that such an exemption could be subject to manipulation by fraudulent home contractors, by the creditor, and even by a consumer. It would be difficult, even with a requirement that the consumer provide verified estimates, to ensure that amounts being disbursed for home repairs actually are needed, and in fact used, for that purpose.

43(d)(1)(iii)

Proposed § 226.43(d)(2)(iii) would have defined the term “refinancing” to have the same meaning as in § 1026.20(a). Section 1026.20(a) defines the term “refinancing” generally to mean a transaction in which an existing obligation is “satisfied and replaced by a new obligation undertaken by the same consumer.” Official commentary explains that “[w]hether a refinancing has occurred is determined by reference to whether the original obligation has been satisfied or extinguished and replaced by a new obligation, based on the parties' contract and applicable law.” See comment 20(a)-1. However, the following are not considered “refinancings” for purposes of § 1026.20(a): (1) A renewal of a payment obligation with no change in the original terms; and (2) a reduction in the annual percentage rate with a corresponding change in the payment schedule. See § 1026.20(a)(1) and (a)(2), and comment 20(a)-2.

The Board requested comment on whether the proposed meaning of “refinancing” should be expanded to include a broader range of transactions or otherwise should be defined differently or explained more fully than proposed. The Bureau received no comments on this proposed definition. Accordingly, the Bureau is adopting the definition of refinancing as proposed, renumbered as § 1026.43(d)(1)(iii).

43(d)(2) Scope

In the Board's proposal, § 226.43(d)(2) addressed the definitions for “non-standard mortgage,” “standard mortgage,” and “refinancing,” while proposed § 226.43(d)(1) established the scope of paragraph (d) and set forth the conditions under which the special refinancing provisions applied. The Bureau believes that paragraph (d) should begin with the relevant definitions, before proceeding to the scope and conditions of the special refinancing provisions. The rule finalized by the Bureau is accordingly reordered. The following discussion details the provisions adopted in § 1026.43(d)(2), which were proposed by the Board under § 226.43(d)(1).

Proposed § 226.43(d)(1) would have defined the scope of the refinancing provisions under proposed § 226.43(d). Specifically, proposed § 226.43(d) applied when a non-standard mortgage is refinanced into a standard mortgage and the following conditions are met—

• The creditor of the standard mortgage is the current holder of the existing non-standard mortgage or the servicer acting on behalf of the current holder.
• The monthly payment for the standard mortgage is significantly lower than the monthly payment for the non-standard mortgage, as calculated under proposed § 226.43(d)(5).
• The creditor receives the consumer's written application for the standard mortgage before the non-standard mortgage is “recast.”
• The consumer has made no more than one payment more than 30 days late on the non-standard mortgage during the 24 months immediately preceding the creditor's receipt of the consumer's written application for the standard mortgage.
• The consumer has made no payments more than 30 days late during the six months immediately preceding the creditor's receipt of the consumer's written application for the standard mortgage.

Proposed comment 43(d)(1)-1 clarified that the requirements for a “written application,” a term that appears in § 226.43(d)(1)(iii), (d)(1)(iv) and (d)(1)(v), discussed in detail below, are found in comment 19(a)(1)(i)-3. Comment 19(a)(1)(i)-3 states that creditors may rely on the Real Estate Settlement Procedures Act (RESPA) and Regulation X (including any interpretations issued by HUD) in deciding whether a “written application” has been received. This comment further states that, in general, Regulation X defines “application” to mean the submission of a borrower's financial information in anticipation of a credit decision relating to a federally related mortgage loan. See 12 CFR 1024.2(b). Comment 19(a)(1)(i)-3 clarifies that an application is received when it reaches the creditor in any of the ways applications are normally transmitted, such as by mail, hand delivery, or through an intermediary agent or broker. The comment further clarifies that, if an application reaches the creditor through an intermediary agent or broker, the application is received when it reaches the creditor, rather than when it reaches the agent or broker. Comment 19(a)(1)(i)-3 also cross-references comment 19(b)-3 for guidance in determining whether or not the transaction involves an intermediary agent or broker. The Bureau received no comments on this proposed comment, which is adopted as proposed, renumbered as 43(d)(2)-1.

43(d)(2)(i)

Proposed § 226.43(d)(1)(i) would have required that the creditor for the new mortgage loan also be either the current holder of the existing non-standard mortgage or the servicer acting on behalf of the current holder. This provision was intended to implement the requirement in TILA section 129C(a)(6)(E) that the existing loan must be refinanced by “the creditor into a standard loan to be made by the same creditor.”

The Board interpreted the statutory phrase “same creditor” to mean that the creditor refinancing the loan must have an existing relationship with the consumer. The Board explained that the existing relationship is important because the creditor must be able to easily access the consumer's payment history and potentially other information about the consumer in lieu of documenting the consumer's income and assets. The Board also noted that this statutory provision is intended to ensure that the creditor of the refinancing has an interest in placing the consumer into a new loan that is affordable and beneficial. The proposal would have permitted the creditor of the refinanced loan to be the holder, or servicer acting on behalf of the holder, of the existing mortgage. The Board further explained that the existing servicer may be the entity conducting the refinance, particularly for refinances held by GSEs. By also permitting the creditor on the refinanced loan to be the servicer acting on behalf of the holder of the existing mortgage, the proposal was intended to apply to a loan that has been sold to a GSE, refinanced by the existing servicer, and continues to be held by the same GSE. The Board solicited comment on whether the proposed rule could be structured differently to better ensure that the creditor retains an interest in the performance of the new loan and whether additional guidance is needed.

Several commenters urged the Bureau to impose a specific period following a refinancing under § 226.43(d) during which the creditor must remain the current holder of the loan. Consumer group commenters suggested that to be eligible for the non-standard mortgage refinancing the creditor should be required to maintain full interest in the refinanced loan for a minimum of 12 months. These commenters expressed concern that the lack of such a retention requirement would permit creditors to refinance loans that are likely to fail without performing the robust underwriting that would otherwise be required for a new loan. If such loans were to be immediately sold to a third party, consumer groups indicated that it could invite abuse by creditors with an incentive to sell riskier loans without providing full value to the consumer. An association of State bank regulators urged the Bureau to adopt a two-year holding period during which the creditor must remain the current holder of the loan.

One industry commenter indicated that the Bureau should broaden the scope to permit a subservicer of the loan to be the creditor with respect to the standard loan. Another industry commenter stated that the scope should be expanded to allow a creditor to refinance a non-standard mortgage that it did not originate or is not servicing. This commenter indicated that due to the volume of requests for refinancing received by some creditors, consumers may benefit from more timely refinancing if a third-party creditor is eligible to use non-standard refinancing provisions.

The Bureau is adopting this requirement as proposed, renumbered as § 1026.43(d)(2)(i). As discussed in more detail below, as adopted § 1026.43(d) provides a broad exemption to all of the ability-to-repay requirements set forth in § 1026.43(c) when a non-standard mortgage is refinanced into a standard mortgage provided that certain conditions are met. Section 1026.43(d)(2)(i) is adopted pursuant to the Bureau's authority under section 105(a) of TILA. The Bureau finds that this adjustment is necessary to effectuate the purposes of TILA by ensuring that consumers are offered and receive residential mortgage loans on terms that reasonably reflect their ability to repay, while ensuring that consumers at risk of default due to payment shock are able to obtain responsible, affordable refinancing credit from the current holder of the consumer's mortgage loan, or the servicer acting on behalf of the current holder. To prevent unscrupulous creditors from using § 1026.43(d) to engage in loan-flipping, and to ensure that this exemption is available only in those cases where consumer benefit is the most likely, the Bureau believes that it is important that the creditor of the standard loan be the holder of, or the servicer acting on behalf of the holder of, the non-standard loan. In such cases, the Bureau agrees with the Board that the creditor has a better incentive to refinance the consumer into a more stable and affordable loan. Therefore, the Bureau declines to extend the scope of § 1026.43(d) to cover cases in which the creditor of the non-standard loan is not the current holder of the nonstandard loan or servicer acting on behalf of that holder.

The Bureau believes that the combination of this restriction and the other protections contained in § 1026.43(d) is sufficient to prevent unscrupulous creditors from engaging in loan-flipping. Therefore, the Bureau does not believe that it is necessary to impose a specified period during which the creditor of the standard mortgage must remain the holder of the loan. As discussed in the section-by-section analysis of § 1026.43(d)(2)(vi) below, the Bureau has conditioned use of § 1026.43(d), for non-standard loans consummated after the effective date of this final rule, on the non-standard loan having been made in accordance with the ability-to-repay requirements in § 1026.43(c), including consideration of the eight factors listed in § 1026.43(c)(2). The Bureau believes that this will help to ensure that creditors cannot use the refinancing provisions of § 1026.43(d) to systematically make and divest riskier mortgages, or to cure substandard underwriting on a non-standard mortgage by refinancing the consumer into a loan with a lower, but still unaffordable, payment. TILA section 130(k)(1) provides that consumers may assert as a defense to foreclosure by way of recoupment or setoff violations of TILA section 129C(a) (of which TILA section 129C(a)(6)(E) comprises a subpart). 15 U.S.C. 1640(k)(1). This defense to foreclosure applies against assignees of the loan in addition to the original creditor. Therefore, given that the non-standard loan having been originated in accordance with § 1026.43(c) is a condition for using the refinancing provision in § 1026.43(d), a consumer may assert violations of § 1026.43(c) on the original non-standard loan as a defense to foreclosure for the standard loan made under § 1026.43(d), even if that standard loan is subsequently sold by the creditor.

In addition to believing that imposition of a holding period is unnecessary, the Bureau has concerns that imposition of a holding period also could create adverse consequences for the safety and soundness of financial institutions. In some circumstances, a creditor may need for safety and soundness reasons to sell a portion of its portfolio, which may include a residential mortgage loan that was made in accordance with § 1026.43(d). However, such a creditor may not know at the time of the refinancing that it ultimately will need to sell the loan, and may even intend to remain the holder the loan for a longer period of time at the time of consummation. The Bureau has concerns about the burden imposed on issuers by a holding period in such circumstances where the creditor does not or cannot know at the time of the refinance under § 1026.43(d) that the loan will need to be sold within the next 12 months.

43(d)(2)(ii)

Proposed § 226.43(d)(1)(ii) would have required that the monthly payment on the new mortgage loan be “materially lower” than the monthly payment for the existing mortgage loan. This proposed provision would have implemented the requirement in TILA section 129C(a)(6)(E) that there be “a reduction in monthly payment on the existing hybrid loan” in order for the special provisions to apply to a refinancing. Proposed comment 43(d)(1)(ii)-1 provided that the monthly payment for the new loan must be “materially lower” than the monthly payment for an existing non-standard mortgage and clarifies that the payments that must be compared must be calculated according to proposed § 226.43(d)(5). The proposed comment also clarified that whether the new loan payment is “materially lower” than the non-standard mortgage payment depends on the facts and circumstances, but that, in all cases, a payment reduction of 10 percent or greater would meet the “materially lower” standard.

Consumer groups and an association of State bank regulators supported the adoption of a 10 percent safe harbor for the “materially lower” standard. In contrast, industry commenters opposed the requirement that payment on the standard mortgage be “materially lower” than the payment on the non-standard mortgage. These commenters urged the Bureau not to adopt the 10 percent safe harbor proposed by the Board and stated that the 10 percent safe harbor would become the de facto rule if adopted. These commenters expressed concerns that the “materially lower” standard would unduly restrict access to credit for many consumers and suggested that the Bureau instead adopt a standard that would permit more consumers to qualify for the non-standard refinancing provisions. Several commenters indicated that the Bureau should adopt a five percent safe harbor rather than the proposed ten percent. One industry commenter recommended that the Bureau permit reductions of a minimum dollar amount to satisfy the rule, particularly in cases where the monthly payment is already low. Finally, one industry commenter asked the Bureau to provide guidance regarding the meaning of “materially lower” when the reduction in payment is less than 10 percent.

The Bureau is adopting as proposed the requirement that the payment on the standard mortgage be “materially lower” than the non-standard mortgage and the safe harbor for a 10 percent or greater reduction, renumbered as § 1026.43(d)(2)(ii) and comment 43(d)(2)(ii)-1. The Bureau agrees with the Board that it would be inconsistent with the statutory purpose to permit the required reduction to be merely de minimis. In such cases, the consumer likely would not obtain a meaningful benefit that would help to prevent default. As discussed in the section-by-section analysis below, § 1026.43(d)(3) exempts refinancings from the ability-to-repay requirements in § 1026.43(c), provided that certain conditions are met. Given that § 1026.43(d) provides a broad exemption to the ability-to-repay requirements, the Bureau believes that it is important that the reduction in payment provide significant value to the consumer and increase the likelihood that the refinancing will improve the consumer's ability to repay the loan. Accordingly, the Bureau is adopting the 10 percent safe harbor as proposed. The Bureau declines to adopt a dollar amount safe harbor because the appropriate dollar amount would depend on a number of factors, including the amount of the loan and monthly payment, but notes that reductions of less than 10 percent could nonetheless meet the “materially lower” standard depending on the relevant facts and circumstances.

43(d)(2)(iii)

Proposed § 226.43(d)(1)(iii) would have required that the creditor for the refinancing receive the consumer's written application for the refinancing before the existing non-standard mortgage is “recast.” As discussed in the section-by-section analysis of § 1026.43(b)(11) above, the proposal defined the term “recast” to mean, for an adjustable-rate mortgage, the expiration of the period during which payments based on the introductory fixed rate are permitted; for an interest-only loan, the expiration of the period during which the interest-only payments are permitted; and, for a negative amortization loan, the expiration of the period during which negatively amortizing payments are permitted.

The Board explained that the proposal was intended to implement TILA section 129C(a)(6)(E)(ii), which permits creditors of certain refinances to “consider if the extension of new credit would prevent a likely default should the original mortgage reset.” This statutory language implies that the special refinancing provisions apply only where the original mortgage has not yet “reset.” Accordingly, the Board concluded that Congress's concern likely was prevention of default in the event of a “reset,” not loss mitigation on a mortgage for which a default on the “reset” payment has already occurred.

However, in recognition of the fact that a consumer may not realize that a loan will be recast until the recast occurs and that the consumer could not refinance the loan under proposed § 226.43(d), the Board also requested comment on whether it would be appropriate to use legal authority to make adjustments to TILA to permit refinancings after a loan is recast.

Consumer groups urged the Bureau to expand the scope of the non-standard refinancing provisions to apply to applications filed after the initial recast of a non-standard loan has occurred. These commenters stated that the intent of the proposal is to avoid “likely default” and indicated that for some consumers, notification that the consumer's interest rate has adjusted and their payment has increased may be their first notice that their payment has gone up and increased their likelihood of default. One consumer group commenter stated that these consumers may be better credit risks than those consumers whose loans have not yet recast and they would clearly benefit from a materially lower monthly payment.

Several industry commenters similarly urged the Bureau to modify the provisions to apply to applications for refinancings received after recast of the non-standard loan. One of these commenters stated that the timing of the application is irrelevant to the consumer's ability to repay or the consumer's need to refinance. One industry commenter stated that processing an application and assessing a consumer's ability to repay a new loan may require additional time well before the recast date. This commenter urged the Bureau to expand the scope of the non-standard refinancing provisions to include refinancings after a loan is recast that are in the best interests of consumers.

For the reasons discussed below, the Bureau is adopting § 1026.43(d)(2)(iii), which provides that § 1026.43(d) applies to the refinancing of a non-standard mortgage into a standard mortgage when the creditor receives the consumer's written application for the standard mortgage no later than two months after the non-standard mortgage has recast, provided certain other conditions are met. The Bureau believes that the best reading of TILA section 129C(a)(6)(E) is that it is intended to facilitate refinancings for consumers at risk of default due to the “payment shock” that may occur upon the recast of the consumer's loan to a higher rate or fully amortizing payments. The Bureau acknowledges that the statutory language contemplates that such recast has not yet occurred. However, the Bureau does not believe that Congress intended to provide relief for consumers facing imminent “payment shock” based on how promptly the consumer filed, or how quickly the creditor processed, an application for a refinancing. For example, the periodic rate on a mortgage loan may recast on July 1st, but the higher payment reflecting the recast interest rate would not be due until August 1st. In this example, a consumer may not experience payment shock until a month after the consumer's rate recasts. Additionally, it may take a significant amount of time for a consumer to provide the creditor with all of the information required by the creditor, thereby triggering the receipt of an application for purposes of the ability-to-repay requirements. The Bureau does not believe that Congress intended the special treatment afforded by TILA section 129C(a)(6)(E) to hinge on paperwork delays such as these. The Bureau agrees with the arguments raised by commenters and believes that the purposes of TILA are best effectuated by permitting consumers to submit applications for refinancings for a short period of time after recast occurs. The Bureau has determined that permitting a consumer to apply for a refinancing within two months of the date of recast strikes the appropriate balance between the language of the statute and the practical considerations involved with submitting an application for a refinancing in response to payment shock. Pursuant to its authority under TILA section 105(a), the Bureau finds that modifying § 1026.43(d) to apply to extensions of credit where the creditor receives the consumer's written application for the standard mortgage no later than two months after the non-standard mortgage has recast ensures that consumers are offered and receive residential mortgage loans on terms that reasonably reflect their ability to repay while ensuring that responsible, affordable mortgage credit remains available to consumers at risk of default due to higher payments resulting from the recast.

43(d)(2)(iv)

Proposed § 226.43(d)(1)(iv) would have required that, during the 24 months immediately preceding the creditor's receipt of the consumer's written application for the standard mortgage, the consumer has made no more than one payment on the non-standard mortgage more than 30 days late. Proposed comment 43(d)(1)(iv)-1 provided an illustrative example. Together with proposed § 226.43(d)(1)(v), proposed § 226.43(d)(1)(iv) would have implemented the portion of TILA section 129C(a)(6)(E) that requires that the consumer not have been “delinquent on any payment on the existing hybrid loan.”

Although TILA section 129C(a)(6)(E) contains a statutory prohibition on “any” delinquencies on the existing non-standard (“hybrid”) mortgage, the Board interpreted its proposal as consistent with the statute in addition to being consistent with the consumer protection purpose of TILA and current industry practices. In addition, the Board noted its authority under TILA sections 105(a) and 129B(e)—which has since transferred to the Bureau—to adjust provisions of TILA and condition practices “to assure that consumers are offered and receive residential mortgage loan on terms that reasonably reflect their ability to repay the loans and that are understandable and not unfair, deceptive, or abusive.” 15 U.S.C. 1604(a); 15 U.S.C. 1639b(e); TILA section 129B(a)(2), 15 U.S.C. 1639b(a)(2).

The Board provided several reasons for proposing to require a look-back period for payment history of 24 months, rather than a 12-month period. First, the Board noted that consumers at risk of default when higher payments are required might present greater credit risks to the institutions holding their loans, even if the institutions refinance those loans. Second, the Board noted views expressed during outreach by GSE and creditor representatives that consumers with positive payment histories tend to be less likely than other consumers to become obligated on a new loan for which they cannot afford the monthly payments. The Board solicited comment on the proposal to require that the consumer have only one delinquency during the 24 months prior to applying for a refinancing, particularly on whether a longer or shorter look-back period should be required.

In addition, under the proposal, late payments of 30 days or fewer on the existing, non-standard mortgage would not disqualify a consumer from refinancing the non-standard mortgage under the streamlined refinance provisions of proposed § 226.43(d). The Board stated that allowing delinquencies of 30 or fewer days is consistent with the statutory prohibition on “any” delinquency for several reasons. First, the Board noted that delinquencies of this length may occur for many reasons outside of the consumer's control, such as mailing delays, miscommunication about where the payment should be sent, or payment crediting errors. Second, many creditors incorporate a late fee “grace period” into their payment arrangements, which permits consumers to make their monthly payments for a certain number of days after the contractual due date without incurring a late fee. Accordingly, the Board noted that the statute should not be read to prohibit consumers from obtaining needed refinances due to payments that are late but within a late fee grace period. Finally, the Board indicated that the predominant streamlined refinance programs of which it is aware uniformly measure whether a consumer has a positive payment history based on whether the consumer has made any payments late by 30 days (or, as in the proposal, more than 30 days).

Proposed comment 43(d)(1)(iv)-2 would have clarified that whether a payment is more than 30 days late depends on the contractual due date not accounting for any grace period and provided an illustrative example. The Board indicated that using the contractual due date for determining whether a payment has been made more than 30 days after the due date would facilitate compliance and enforcement by providing clarity. Whereas late fee “grace periods” are often not stated in writing, the contractual due date is unambiguous. Finally, the Board stated that using the contractual due date for determining whether a loan payment is made on time is consistent with standard home mortgage loan contracts. The Board requested comment on whether the delinquencies that creditors are required to consider under § 226.43(d)(1) should be late payments of more than 30 days as proposed, 30 days or more, or some other time period.

Consumer groups supported the Board's proposal to identify late payments as late payments of more than 30 days. However, they stated that the requirement that consumers not have more than one delinquency in the past 24 months to qualify for a refinance under § 1026.43(d) was overly stringent and that the appropriate standard would be no delinquencies in the past 12 months.

Several industry commenters similarly urged the Bureau to adopt a 12-month period rather than the proposed 24-month period in which a consumer may have one late payment. These commenters stated that permitting only one 30-day late payment in the past 24 months is too restrictive and would require a creditor to overlook a recent history of timely payments. In addition, one industry commenter stated that the standard for defining a late payment should be late payments of more than 60 days.

The Bureau is adopting this provision generally as proposed, renumbered as § 1026.43(d)(2)(iv), with one substantive change. The Bureau is adopting a 12-month look-back period rather than the 24-month period proposed by the Board. The Bureau believes that reviewing a consumer's payment history over the last 12 months would be more appropriate than a 24-month period, and agrees that a 24-month period may unduly restrict consumer access to the § 1026.43(d) refinancing provisions. The Bureau believes that the requirement that a consumer's account have no more than one 30-day late payment in the past 12 months will best effectuate the purposes of TILA by ensuring that only those consumers with positive payment histories are eligible for the non-standard refinancing provisions under § 1026.43(d). Section 1026.43(d)(2)(iv) is adopted pursuant to the Bureau's authority under section 105(a) of TILA. The Bureau finds that this adjustment is necessary and proper to effectuate the purposes of TILA by ensuring that consumers are offered and receive residential mortgage loans on terms that reasonably reflect their ability to repay, while ensuring that consumers at risk of default due to payment shock are able to obtain responsible, affordable refinancing credit.

The Bureau also is adopting comments 43(d)(1)(iv)-1 and 43(d)(1)(iv)-2 generally as proposed, with conforming amendments to reflect the 12-month look-back period in § 1026.43(d)(2)(iv), and renumbered as 43(d)(2)(iv)-1 and 43(d)(2)(iv)-2. The Bureau has made several technical amendments to the example in comment 43(d)(2)(iv)-1 for clarity. As proposed, the examples in the comment referred to dates prior to the effective date of this rule; the Bureau has updated the dates in the examples so that they will occur after this rule becomes effective.

43(d)(2)(v)

Proposed § 226.43(d)(1)(v) would have required that the consumer have made no payments on the non-standard mortgage more than 30 days late during the six months immediately preceding the creditor's receipt of the consumer's written application for the standard mortgage. This provision complemented proposed § 226.43(d)(1)(iv), discussed above, in implementing the portion of TILA section 129C(a)(6)(E) that requires that the consumer not have been “delinquent on any payment on the existing hybrid loan.” Taken together with proposed § 226.43(d)(1)(iv), the Board believed that this is a reasonable interpretation of the prohibition on “any” delinquencies on the non-standard mortgage and is supported by the Board's authority under TILA sections 105(a) and 129B(e)—which has transferred to the Bureau—to adjust provisions of TILA and condition practices “to assure that consumers are offered and receive residential mortgage loans on terms that reasonably reflect their ability to repay the loans and that are understandable and not unfair, deceptive, or abusive.” 15 U.S.C. 1604(a); TILA section 129B(a)(2), 15 U.S.C. 1639b(a)(2).

The Board stated that a six-month “clean” payment record indicates a reasonable level of financial stability on the part of the consumer applying for a refinancing. In addition, the Board noted that participants in its outreach indicated that a prohibition on delinquencies of more than 30 days for the six months prior to application for the refinancing was generally consistent with common industry practice and would not be unduly disruptive to existing streamlined refinance programs with well-performing loans.

Proposed comment 43(d)(1)(v)-1 provided an illustrative example of the proposed rule and clarified that if the number of months between consummation of the non-standard mortgage and the consumer's application for the standard mortgage is six or fewer, the consumer may not have made any payment more than 30 days late on the non-standard mortgage. The comment cross-referenced proposed comments 43(d)(1)-2 and 43(d)(1)(iv)-2 for an explanation of “written application” and how to determine the payment due date, respectively.

One industry commenter stated that the prohibition on late payments in the past six months should be amended to provide flexibility when the late payment was due to extenuating circumstances. The Bureau declines to adopt a rule providing an adjustment for extenuating circumstances, for several reasons. First, the existence or absence of extenuating circumstances is a fact-specific question and it would be difficult to distinguish by regulation between extenuating circumstances that reflect an ongoing risk with regard to the consumer's ability to repay the loan versus extenuating circumstances that present less risk. In addition, an adjustment for extenuating circumstances appears to be inconsistent with the purposes of TILA section 129C(a)(6)(E), which contemplates that the consumer “has not been delinquent on any payment on the existing hybrid loan,” without distinguishing between payments that are delinquent due to extenuating circumstances or otherwise. Furthermore, by defining a late payment as more than 30 days late, the Bureau believes that many extenuating circumstances, for example a payment made three weeks late due to mail delivery issues, will not preclude use of § 1026.43(d).

Accordingly, the Bureau is adopting this provision as proposed, renumbered as § 1026.43(d)(2)(v). Similarly, the Bureau is adopting comment 43(d)(1)(v)-1 generally as proposed, with several technical amendments for clarity and renumbered as 43(d)(2)(v)-1. As proposed, the examples in the comment referred to dates prior to the effective date of this rule; the Bureau has updated the dates in the examples so that they will occur after this rule becomes effective. Pursuant to its authority under TILA section 105(a), the Bureau finds that requiring that the consumer have made no payments on the non-standard mortgage more than 30 days late during the six months immediately preceding the creditor's receipt of the consumer's written application for the standard mortgage ensures that consumers are offered and receive residential mortgage loans on terms that reasonably reflect their ability to repay while ensuring that responsible, affordable mortgage credit remains available to consumers at risk of default due to higher payments resulting from the recast.

43(d)(2)(vi)

For the reasons discussed in the section-by-section analysis of § 1026.43(d)(3), the Bureau is adopting a new § 1026.43(d)(2)(vi) that generally conditions use of § 1026.43(d) on the existing non-standard mortgage having been made in accordance with § 1026.43(c), provided that the existing non-standard mortgage loan was consummated on or after January 10, 2014. For the reasons discussed in the section-by-section analysis of § 1026.43(d)(3), the Bureau believes that this provision is necessary and proper to prevent use of § 1026.43(d)'s streamlined refinance provision to circumvent or “cure” violations of the ability-to-repay requirements in § 1026.43(c). Section 1026.43(d)(2)(vi) is adopted pursuant to the Bureau's authority under TILA section 105(a). The Bureau finds that this adjustment is necessary to effectuate the purposes of TILA by ensuring that consumers are offered and receive residential mortgage loans on terms that reasonably reflect their ability to repay, while ensuring that consumers at risk of default due to payment shock are able to obtain responsible and affordable refinancing credit. Furthermore, the Bureau believes that this adjustment is necessary to prevent unscrupulous creditors from using § 1026.43(d) to engage in loan-flipping or other practices that are harmful to consumers, thereby circumventing the requirements of TILA.

43(d)(3) Exemption From Repayment Ability Requirements

Under specific conditions, proposed § 226.43(d)(3) would have exempted a creditor in a refinancing from two of the ability-to-repay requirements under proposed § 226.43(c). First, the proposal provided that a creditor is not required to comply with the income and asset verification requirements of proposed § 226.43(c)(2)(i) and (c)(4). Second, the proposal provided that the creditor is not required to comply with the payment calculation requirements of proposed § 226.43(c)(2)(iii) and (c)(5); the creditor may instead use payment calculations prescribed in proposed § 226.43(d)(5)(ii).

For these exemptions to apply, proposed § 226.43(d)(3)(i)(A) would have required that all of the conditions in proposed § 226.43(d)(1)(i) through (v) be met. In addition, proposed § 226.43(d)(3)(i)(B) would have required that the creditor consider whether the standard mortgage will prevent a likely default by the consumer on the non-standard mortgage when the non-standard mortgage is recast. This proposed provision implemented TILA section 129C(a)(6)(E)(ii), which permits a creditor to “consider if the extension of new credit would prevent a likely default should the original mortgage reset and give such concerns a higher priority as an acceptable underwriting practice.” As clarified in proposed comment 43(d)(3)(i)-1, the Board interpreted TILA section 129(a)(6)(E)(ii) to require a creditor to consider whether: (1) The consumer is likely to default on the existing mortgage once new, higher payments are required; and (2) the new mortgage will prevent the consumer's default. The Board solicited comment regarding whether these proposed provisions were appropriate, and also specifically solicited comment on whether exemptions from the ability-to-repay requirements, other than those proposed, were appropriate.

Several commenters expressly supported this proposed provision. An association of State bank supervisors stated that refinancing designed to put a consumer in a higher-quality standard mortgage before the existing lower-quality mortgage recasts should be given greater deference and further stated that it is sound policy to encourage refinancing where it protects both the economic interest of the creditor and the financial health of the consumer. Consumer groups commented that limited and careful exemption from income verification, provided that protections are in place, can help consumers and communities, while preventing reckless and abusive lending on the basis of little or no documentation. Civil rights organizations also stated that the streamlined refinance option would provide much-needed relief for consumers with loans that are not sustainable in the long term but who are not yet in default. These commenters also stated that minority consumers have been targeted in the past for unsustainable loans and that this provision could help to prevent further foreclosures and economic loss in minority communities, as well as for homeowners in general.

Other consumer group commenters stated that an exemption to the income verification requirement for refinancing into standard mortgages is problematic. One commenter stated that, because the refinance would be executed by the same creditor that made the original hybrid loan, income verification would not be difficult. This commenter urged the Bureau to encourage income documentation when implementing the Dodd-Frank Act.

Several industry commenters urged the Bureau to provide additional relief for refinancings made in accordance with proposed § 226.43(d), either by permitting the standard loan to be classified as a qualified mortgage or by providing exemptions from other of the proposed ability-to-repay requirements. One industry commenter stated that in addition to the proposed exemption for the verification of income and assets, refinancings conducted in accordance with § 226.43(d) also should be exempt from the requirements to consider the consumer's debt-to-income ratio or residual income, if the consumer is still employed and has not incurred significant additional debt obligations prior to the refinance. This commenter stated that overly rigid standards could significantly reduce the number of consumers who qualify for this exemption. Similarly, one industry trade association urged the Bureau to exempt refinancings from the requirement to consider the consumer's debt obligations, debt-to-income ratio, and employment. This commenter stated that the proposed requirement to consider these additional underwriting factors was seemingly in conflict with the purpose of proposed § 226.43(d) and would preclude consumers from taking advantage of beneficial and less costly refinancing opportunities. In addition, several industry commenters and one industry trade association commented that standard mortgages made in accordance with § 226.43(d) should be treated as qualified mortgages.

The Bureau agrees with the concerns raised by commenters that the proposed exemptions were drawn too narrowly. The Bureau believes that TILA section 129C(a)(6)(E) is intended to create incentives for creditors to refinance loans in circumstances where consumers have non-standard loans on which they are currently able to make payments but on which they are likely to be unable to make the payments after recast and therefore default on the loan. Accordingly, the Bureau believes that in order to create incentives for creditors to use the non-standard refinancing provision, TILA section 129C(a)(6)(E) must be intended to provide at least a limited exemption from the general ability-to-repay determination as adopted in § 1026.43(c). Otherwise, creditors may have little incentive to provide consumers at risk of default with refinancings that result in “materially lower” payments. The Bureau believes, however, that in implementing TILA section 129C(a)(6)(E) it is important to balance the creation of additional flexibility and incentives for creditors to refinance non-standard mortgages into standard mortgages against the likelihood of benefit to the consumer.

The Bureau notes that under the final rule as adopted, the availability of the non-standard refinancing provision contains several conditions that are intended to benefit the consumer. First, the special ability-to-repay requirements in § 1026.43(d) are available only if the conditions in § 1026.43(d)(2) are met. These conditions include limiting the scope of § 1026.43(d) to refinancings of non-standard mortgages into standard mortgages, which generally are more stable products with reduced risk of payment shock. The definition of standard mortgage in § 1026.43(d)(1)(ii) includes a number of limitations that are intended to ensure that creditors may only use the provisions in § 1026.43(d) to offer a consumer a product with safer features. For example, as discussed in the section-by-section analysis of § 1026.43(d)(1)(ii) a standard mortgage may not include negative amortization, an interest-only feature, or a balloon payment; in addition, the term of the standard mortgage may not exceed 40 years, the interest rate must be fixed for at least the first five years, the loan is subject to a limitation on the points and fees that may be charged, and there are limitations on the use of proceeds from the refinancing. Furthermore, § 1026.43(d)(2)(ii) requires that the monthly payment on the standard mortgage be materially lower than the monthly payment for the non-standard mortgage and, as discussed above, the Bureau is adopting a 10 percent safe harbor for what constitutes a “material” reduction.

The Bureau has concerns that, as proposed by the Board, an exemption only from the requirement to consider and verify the consumer's income or assets may create insufficient incentives for creditors to make refinancings to assist consumers at risk of default. For example, the proposal would have required creditors to comply with the requirement in § 1026.43(c)(2)(vii) to consider the consumer's debt-to-income ratio or residual income. Accordingly, notwithstanding an exemption from income or asset verification, the proposal would have required consideration of income, as well as consideration of all of the other underwriting criteria set forth in § 1026.43(c)(2).

The Bureau believes that in light of the safeguards imposed by other portions of § 1026.43(d), as discussed above, it is appropriate to provide an exemption to all of the ability-to-repay requirements under § 1026.43(c) for a refinance conducted in accordance with § 1026.43(d). The Bureau believes that a broad exemption from the general ability-to-repay determination is appropriate in order to create incentives for creditors to quickly and efficiently refinance consumers whose non-standard mortgages are about to recast, thus rendering them likely to default, into more affordable, more stable mortgage loans. The Bureau is aware that some consumers may nonetheless default on a standard mortgage made in accordance with § 1026.43(d), but those consumers likely would have defaulted had the non-standard mortgage remained in place. For others, the material reduction in payment required under § 1026.43(d)(2) and the more stable product type following refinancing may be sufficient to enable consumers to avoid default. The Bureau believes that a refinancing conducted in accordance with § 1026.43(d) will generally improve a consumer's chances of avoiding default. Section 1026.43(d)(3) is adopted pursuant to the Bureau's authority under TILA section 105(a). The Bureau finds that this adjustment is necessary to effectuate the purposes of TILA by ensuring that consumers are offered and receive residential mortgage loans on terms that reasonably reflect their ability to repay, while ensuring that consumers at risk of default due to payment shock are able to obtain responsible and affordable refinancing credit.

However, to prevent evasion or circumvention of the ability-to-repay requirements in § 1026.43(c), the Bureau is imposing one additional condition on the use of § 1026.43(d). Specifically, new § 1026.43(d)(2)(vi) conditions the use of § 1026.43(d), for non-standard mortgages consummated on or after the effective date of this rule, on the non-standard mortgage having been made in accordance with § 1026.43(c). The Bureau has concerns that absent § 1026.43(d)(2)(vi), a creditor might attempt to use a refinancing conducted in accordance with § 1026.43(d) to “cure” substandard underwriting of the prior non-standard mortgage. For example, without § 1026.43(d)(2)(vi), if a creditor discovered that it had made an error in consideration of the underwriting factors under § 1026.43(c)(2) for a non-standard mortgage, the creditor might consider conducting a refinancing under § 1026.43(d), in order to argue that the consumer may no longer raise as a defense to foreclosure the underwriting of the original non-standard mortgage. The Bureau believes that conditioning the use of § 1026.43(d) on the earlier loan having been made in accordance with § 1026.43(c) will better effectuate the purposes of TILA by ensuring that consumers are offered and receive residential mortgage loans on terms that reasonably reflect their ability to repay while preventing unscrupulous creditors from evading the ability-to-repay requirements.

New § 1026.43(d)(2)(vi) applies only to non-standard mortgages consummated on or after the effective date of this rule. For non-standard loans consummated before the effective date of this final rule, a refinancing under § 1026.43(d) would not be subject to this condition. The Bureau believes that non-standard mortgages made prior to the effective date, to which the ability-to-repay requirements in § 1026.43(c) did not apply, may present an increased risk of default when they are about to recast, so that facilitating refinancing into more stable mortgages may be particularly important even if the consumer could not qualify for a new loan under traditional ability-to-repay requirements. The Bureau believes that, on balance, given the conditions that apply to refinances under § 1026.43(d), refinances of these loans are more likely to benefit consumers than to harm consumers, notwithstanding the inapplicability of § 1026.43(d)(2)(vi). In addition, the concern about a creditor using § 1026.43(d) to “cure” prior violations of § 1026.43(c) does not apply to loans made before the effective date of this rule, as such loans were not required to be made in accordance with § 1026.43.

Proposed condition that the consumer will likely default. Proposed comment 43(d)(3)(i)-2 would have clarified that, in considering whether the consumer's default on the non-standard mortgage is “likely,” the creditor may look to widely accepted governmental and non-governmental standards for analyzing a consumer's likelihood of default. The proposal was not intended, however, to constrain servicers and other relevant parties from using other methods to determine a consumer's likelihood of default, including those tailored specifically to that servicer. As discussed in the supplementary information to the proposal, the Board considered certain government refinancing programs as well as feedback from outreach participants, each of which suggested that there may be legitimate differences in servicer assessments of a consumer's likelihood of default. The Board noted that it considered an “imminent default” standard but heard from consumer advocates that “imminent default” may be a standard that is too high for the refinancing provisions in TILA section 129C(a)(6)(E) and could prevent many consumers from obtaining a refinancing to avoid payment shock. Accordingly, the Board's proposal used the exact statutory wording—“likely default”—in implementing the provision permitting a creditor to prioritize prevention of default in underwriting a refinancing. The Board solicited comment on the proposal to use the term “likely default” in implementing TILA section 129C(a)(6)(E)(ii) and on whether additional guidance is needed on how to meet the requirement that a creditor must reasonably and in good faith determine that a standard mortgage will prevent a likely default should the non-standard mortgage be recast.

Two industry trade associations urged the Bureau to remove proposed § 226.43(d)(3)(i)(B) as a condition to the availability of the non-standard refinancing provisions. One of these commenters noted that a creditor would have to underwrite a consumer's income and assets to determine whether the consumer would likely default, which would defeat the purpose of the proposed provision. Several industry commenters also indicated that the “likelihood of default” standard is vague and accordingly subjects creditors to potential liability for waiving certain ability-to-repay requirements, and questioned the extent to which creditors would utilize the streamline refinance option in light of this potential liability. One such commenter urged the Bureau to eliminate this requirement or, in the alternative, to provide additional guidance regarding when a consumer is “likely to go into default.”

An association of State bank supervisors stated that there can be no quantifiable standard for the definition of “likely default.” These commenters further stated that institutions must use sound judgment and regulators must provide responsible oversight to ensure that abuses are not occurring through the refinancing exemption set forth in § 1026.43(d).

The Bureau is adopting the provision as proposed, renumbered as § 1026.43(d)(3)(i)(B), and is also adopting comments 43(d)(3)(i)-1 and 43(d)(3)(i)-2 as proposed. The Bureau believes that eliminating the requirement that a creditor consider whether the extension of new credit would prevent a likely default would be inconsistent with TILA section 129C(a)(6)(E), which expressly includes language regarding consideration by the creditor of “[whether] the extension of new credit would prevent a likely default should the original mortgage reset.” At the same time, the Bureau agrees with the association of State bank supervisors that it would be difficult to impose by regulation a single standard for what constitutes a likely default. Accordingly, the Bureau is adopting the flexible approach proposed by the Board, which would permit but not require creditors to look to widely-accepted standards for analyzing a consumer's likelihood of default. The Bureau does not believe that this flexible approach requires a creditor to consider the consumer's income and assets if, for example, statistical evidence indicates that consumers who experience a payment shock of the type that the consumer is about to experience have a high incidence of defaulting following the payment shock.

Proposed payment calculation for repayment ability determination. Proposed comment 43(d)(3)(ii)-1 would have explained that, if the conditions in proposed § 226.43(d)(1) are met, the creditor may satisfy the payment calculation requirements for determining a consumer's ability to repay the new loan by applying the calculation prescribed under proposed § 226.43(d)(5)(ii), rather than the calculation prescribed under proposed § 226.43(c)(2)(iii) and (c)(5). As discussed in the section-by-section analysis above, as adopted § 1026.43(d)(3) provides an exemption from the requirements of § 1026.43(c) if certain conditions are met. Accordingly, while the creditor is required to determine whether there is a material reduction in payment consistent with § 1026.43(d)(2)(ii) by using the payment calculations prescribed in § 1026.43(d)(5), the creditor is not required to use these same payment calculations for purposes of § 1026.43(c). Accordingly, the Bureau is withdrawing proposed comment 43(d)(3)(ii)-1 as unnecessary.

43(d)(4) Offer of Rate Discounts and Other Favorable Terms

Proposed § 226.43(d)(4) would have provided that a creditor making a loan under the special refinancing provisions of § 226.43(d) may offer to the consumer the same or better rate discounts and other terms that the creditor offers to any new consumer, consistent with the creditor's documented underwriting practices and to the extent not prohibited by applicable State or Federal law. This aspect of the proposal was intended to implement TILA section 129C(a)(6)(E)(iii), which permits creditors of refinancings subject to special ability-to-repay requirements in TILA section 129C(a)(6)(E) to “offer rate discounts and other favorable terms” to the consumer “that would be available to new customers with high credit ratings based on such underwriting practice.”

The Bureau received no comments on this provision, which is adopted as proposed and renumbered as § 1026.43(d)(4). The Bureau is concerned that the phrase “consistent with the creditor's underwriting practice” could be misinterpreted to refer to the underwriting requirements in § 1026.43(c). As this final rule provides an exemption under § 1026.43(d) for all of the requirements in § 1026.43(c), subject to the other conditions discussed above, the Bureau believes that additional clarification is needed to address this potential misinterpretation. Thus, the Bureau is adopting comment 43(d)(4)-1, which clarifies that in connection with a refinancing made pursuant to § 1026.43(d), § 1026.43(d)(4) requires a creditor offering a consumer rate discounts and terms that are the same as, or better than, the rate discounts and terms offered to new consumers to make such an offer consistent with the creditor's documented underwriting practices. Section 1026.43(d)(4) does not require a creditor making a refinancing pursuant to § 1026.43(d) to comply with the underwriting requirements of § 1026.43(c). Rather, § 1026.43(d)(4) requires creditors providing such discounts to do so consistent with documented policies related to loan pricing, loan term qualifications, or other similar underwriting practices. For example, assume that a creditor is providing a consumer with a refinancing made pursuant to § 1026.43(d) and that this creditor has a documented practice of offering rate discounts to consumers with credit scores above a certain threshold. Assume further that the consumer receiving the refinancing has a credit score below this threshold, and therefore would not normally qualify for the rate discount available to consumers with high credit scores. This creditor complies with § 1026.43(d)(4) by offering the consumer the discounted rate in connection with the refinancing made pursuant to § 1026.43(d), even if the consumer would not normally qualify for that discounted rate, provided that the offer of the discounted rate is not prohibited by applicable State or Federal law. However, § 1026.43(d)(4) does not require a creditor to offer a consumer such a discounted rate.

43(d)(5) Payment Calculations

Proposed § 226.43(d)(5) would have prescribed the payment calculations for determining whether the consumer's monthly payment for a standard mortgage will be “materially lower” than the monthly payment for the non-standard mortgage. Proposed § 226.43(d)(5) thus was intended to complement proposed § 226.43(d)(1)(ii) in implementing TILA section 129C(a)(6)(E), which requires a “reduction” in the monthly payment for the existing non-standard (“hybrid”) mortgage when refinanced into a standard mortgage.

43(d)(5)(i) Non-Standard mortgage

Proposed § 226.43(d)(5)(i) would have required that the monthly payment for a non-standard mortgage be based on substantially equal, monthly, fully amortizing payments of principal and interest that would result once the mortgage is recast. The Board stated that comparing the payment on the standard mortgage to the payment amount on which the consumer likely would have defaulted (i.e., the payment resulting on the existing non-standard mortgage once the introductory terms cease and a higher payment results) would promote needed refinances consistent with Congress's intent.

The Board noted that the payment that the consumer is currently making on the existing non-standard mortgage may be an inappropriately low payment to compare to the standard mortgage payment. The existing payments may be interest-only or negatively amortizing; these temporarily lower payment amounts would be difficult for creditors to “reduce” with a refinanced loan that has a comparable term length and principal amount. Indeed, the payment on a new loan with a fixed-rate rate and fully-amortizing payment, as is required for the payment calculation of a standard mortgage under the proposal, for example, is likely to be higher than the interest-only or negative amortization payment. As a result, few refinancings would yield a lower monthly payment, so many consumers could not receive the benefits of refinancing into a more stable loan product.

Accordingly, the proposal would have required a creditor to calculate the monthly payment for a non-standard mortgage using—

• The fully indexed rate as of a reasonable period of time before or after the date on which the creditor receives the consumer's written application for the standard mortgage;
• The term of the loan remaining as of the date of the recast, assuming all scheduled payments have been made up to the recast date and the payment due on the recast date is made and credited as of that date; and
• A remaining loan amount that is—

○ For an adjustable-rate mortgage, the outstanding principal balance as of the date the mortgage is recast, assuming all scheduled payments have been made up to the recast date and the payment due on the recast date is made and credited as of that date;

○ For an interest-only loan, the loan amount, assuming all scheduled payments have been made up to the recast date and the payment due on the recast date is made and credited as of that date;

○ For a negative amortization loan, the maximum loan amount.

Proposed comment 43(d)(5)(i)-1 would have explained that, to determine whether the monthly periodic payment for a standard mortgage is materially lower than the monthly periodic payment for the non-standard mortgage under proposed § 226.43(d)(1)(ii), the creditor must consider the monthly payment for the non-standard mortgage that will result after the loan is recast, assuming substantially equal payments of principal and interest that amortize the remaining loan amount over the remaining term as of the date the mortgage is recast. The proposed comment noted that guidance regarding the meaning of “substantially equal” and “recast” is provided in comment 43(c)(5)(i)-4 and § 226.43(b)(11), respectively.

Proposed comment 43(d)(5)(i)-2 would have explained that the term “fully indexed rate” used for calculating the payment for a non-standard mortgage is generally defined in proposed § 226.43(b)(3) and associated commentary. The proposed comment explained an important difference between the “fully indexed rate” as defined in proposed § 226.43(b)(3), however, and the meaning of “fully indexed rate” in § 226.43(d)(5)(i). Specifically, under proposed § 226.43(b)(3), the fully indexed rate is calculated at the time of consummation. Under proposed § 226.43(d)(5)(i), the fully indexed rate would be calculated within a reasonable period of time before or after the date on which the creditor receives the consumer's written application for the standard mortgage. Comment 43(d)(5)(i)-2 clarified that 30 days would generally be considered a “reasonable period of time.”

Proposed comment 43(d)(5)(i)-3 would have clarified that the term “written application” is explained in comment 19(a)(1)(i)-3. Comment 19(a)(1)(i)-3 states that creditors may rely on RESPA and Regulation X (including any interpretations issued by HUD) in deciding whether a “written application” has been received. In general, Regulation X defines “application” to mean the submission of a borrower's financial information in anticipation of a credit decision relating to a federally related mortgage loan. See 12 CFR 1024.2(b). As explained in comment 19(a)(1)(i)-3, an application is received when it reaches the creditor in any of the ways applications are normally transmitted, such as by mail, hand delivery, or through an intermediary agent or broker. If an application reaches the creditor through an intermediary agent or broker, the application is received when it reaches the creditor, rather than when it reaches the agent or broker. This proposed comment also cross-referenced comment 19(b)-3 for guidance in determining whether the transaction involves an intermediary agent or broker.

Proposed payment calculation for an adjustable-rate mortgage with an introductory fixed rate. Proposed comments 43(d)(5)(i)-4 and -5 would have clarified the payment calculation for an adjustable-rate mortgage with an introductory fixed rate under proposed § 226.43(d)(5)(i). Proposed comment 43(d)(5)(i)-4 clarified that the monthly periodic payment for an adjustable-rate mortgage with an introductory fixed interest rate for a period of one or more years must be calculated based on several assumptions. First, the payment must be based on the outstanding principal balance as of the date on which the mortgage is recast, assuming all scheduled payments have been made up to that date and the last payment due under those terms is made and credited on that date. Second, the payment calculation must be based on substantially equal monthly payments of principal and interest that will fully repay the outstanding principal balance over the term of the loan remaining as of the date the loan is recast. Third, the payment must be based on the fully indexed rate, as defined in § 226.43(b)(3), as of the date of the written application for the standard mortgage. The proposed comment set forth an illustrative example. Proposed comment 43(d)(5)(i)-5 would have provided a second illustrative example of the payment calculation for an adjustable-rate mortgage with an introductory fixed rate.

Proposed payment calculation for an interest-only loan. Proposed comments 43(d)(5)(i)-6 and -7 would have explained the payment calculation for an interest-only loan under proposed § 226.43(d)(5)(i). Proposed comment 43(d)(5)(i)-6 would have clarified that the monthly periodic payment for an interest-only loan must be calculated based on several assumptions. First, the payment must be based on the loan amount, as defined in § 226.43(b)(5), assuming all scheduled payments are made under the terms of the legal obligation in effect before the mortgage is recast. The comment provides an example of a mortgage with a 30-year loan term for which the first 24 months of payments are interest-only. The comment then explains that, if the 24th payment is due on September 1, 2013, the creditor must calculate the outstanding principal balance as of September 1, 2013, assuming that all 24 payments under the interest-only payment terms have been made and credited.

Second, the payment calculation must be based on substantially equal monthly payments of principal and interest that will fully repay the loan amount over the term of the loan remaining as of the date the loan is recast. Thus, in the example above, the creditor must assume a loan term of 28 years (336 payments). Third, the payment must be based on the fully indexed rate as of the date of the written application for the standard mortgage.

Proposed comment 43(d)(5)(i)-7 would have provided an illustration of the payment calculation for an interest-only loan. The example assumes a loan in an amount of $200,000 that has a 30-year loan term. The loan agreement provides for a fixed interest rate of 7 percent, and permits interest-only payments for the first two years, after which time amortizing payments of principal and interest are required. Second, the example states that the non-standard mortgage is consummated on February 15, 2011, and the first monthly payment is due on April 1, 2011. The loan is recast on the due date of the 24th monthly payment, which is March 1, 2013. Finally, the example assumes that on March 15, 2012, the creditor receives the consumer's written application for a refinancing, after the consumer has made 12 monthly on-time payments.

Proposed comment 43(d)(5)(i)-7 would have further explained that, to calculate the non-standard mortgage payment that must be compared to the standard mortgage payment, the creditor must use—

• The loan amount, which is the outstanding principal balance as of March 1, 2013, assuming all scheduled interest-only payments have been made and credited up to that date. In this example, the loan amount is $200,000.
• An interest rate of 7 percent, which is the interest rate in effect at the time of consummation of this fixed-rate non-standard mortgage.
• The remaining loan term as of March 1, 2013, the date of the recast, which is 28 years.

The comment concluded by stating that, based on the assumptions above, the monthly payment for the non-standard mortgage for purposes of determining whether the standard mortgage monthly payment is lower than the non-standard mortgage monthly payment is $1,359. This is the substantially equal, monthly payment of principal and interest required to repay the loan amount at the fully indexed rate over the remaining term.

Proposed payment calculation for a negative amortization loan. Proposed comments 43(d)(5)(i)-8 and -9 would have explained the payment calculation for a negative amortization loan under proposed § 226.43(d)(5)(i)(C). Proposed comment 43(d)(5)(i)-8 would have clarified that the monthly periodic payment for a negative amortization loan must be calculated based on several assumptions. First, the calculation must be based on the maximum loan amount. The comment further stated that examples of how to calculate the maximum loan amount are provided in proposed comment 43(b)(7)-3.

Second, the payment calculation must be based on substantially equal monthly payments of principal and interest that will fully repay the maximum loan amount over the term of the loan remaining as of the date the loan is recast. For example, the comment states, if the loan term is 30 years and the loan is recast on the due date of the 60th monthly payment, the creditor must assume a loan term of 25 years. Third, the payment must be based on the fully indexed rate as of the date of the written application for the standard mortgage.

Proposed comment 43(d)(5)(i)-9 would have provided an illustration of the payment calculation for a negative amortization loan. The example assumes a loan in an amount of $200,000 that has a 30-year loan term. The loan agreement provides that the consumer can make minimum monthly payments that cover only part of the interest accrued each month until the date on which the principal balance increases to the negative amortization cap of 115 percent of the loan amount, or for the first five years of monthly payments, whichever occurs first. The loan is an adjustable-rate mortgage that adjusts monthly according to a specified index plus a margin of 3.5 percent.

The example also assumed that the non-standard mortgage is consummated on February 15, 2011, and the first monthly payment is due on April 1, 2011. Further, the example assumes that, based on the calculation of the maximum loan amount required under § 226.43(b)(7) and associated commentary, the negative amortization cap of 115 percent is reached on July 1, 2013, the due date of the 28th monthly payment. Finally, the example assumes that on March 15, 2012, the creditor receives the consumer's written application for a refinancing, after the consumer has made 12 monthly on-time payments. On this date, the index value is 4.5 percent.

Proposed comment 43(d)(5)(i)-9 then stated that, to calculate the non-standard mortgage payment that must be compared to the standard mortgage payment under proposed § 226.43(d)(1)(ii), the creditor must use—

• The maximum loan amount of $229,243 as of July 1, 2013.
• The fully indexed rate of 8 percent, which is the index value of 4.5 percent as of March 15, 2012 (the date on which the creditor receives the application for a refinancing) plus the margin of 3.5 percent.
• The remaining loan term as of July 1, 2013, the date of the recast, which is 27 years and 8 months (332 monthly payments).

The comment concluded by stating that, based on the assumptions above, the monthly payment for the non-standard mortgage for purposes of determining whether the standard mortgage monthly payment is lower than the non-standard mortgage monthly payment is $1,717. This is the substantially equal, monthly payment of principal and interest required to repay the maximum loan amount at the fully indexed rate over the remaining term.

The Board requested comment on the proposed payment calculation for a non-standard mortgage and on the appropriateness and usefulness of the proposed payment calculation examples.

The Bureau received no specific comment on the payment calculations for non-standard mortgages set forth in proposed § 226.43(d)(5)(i) and its associated commentary. Accordingly, the provision that is being adopted is substantially similar to the version proposed, renumbered as § 1026.43(d)(5)(i). The Bureau also is adopting the associated commentary generally as proposed. The Bureau has made several technical amendments to the examples in comments 43(d)(5)(i)-4, -5, -6, -7, and -9 for clarity. As proposed, the examples in the comment referred to dates prior to the effective date of this rule; the Bureau has updated the dates in the examples so that they will occur after this rule becomes effective.

The Bureau believes that it is necessary to clarify the provisions related to payment calculations for interest-only loans and negative amortization loans. The provisions adopted clarify that the payment calculation required by § 1026.43(d)(5)(i) must be based on the outstanding principal balance, rather than the original amount of credit extended. Accordingly, as adopted § 1026.43(d)(5)(i)(C)(2) requires the remaining loan amount for an interest-only loan to be based on the outstanding principal balance as of the date of the recast, assuming all scheduled payments have been made up to the recast date and the payment due on the recast date is made and credited as of that date. Similarly, § 1026.43(d)(5)(i)(C)(3) requires the remaining loan amount for a negative amortization loan to be based on the maximum loan amount, determined after adjusting for the outstanding principal balance. The Bureau has made technical amendments to the example in comments 43(d)(5)(i)-6, -7, -8, and -9 to conform to this clarification.

Additionally, the Bureau has added new comment 43(d)(5)(i)-10 to add an additional illustration of the payment calculation for a negative amortization loan. As adopted, comment 43(d)(5)(i)-10 provides an illustrative example, clarifying that, pursuant to the example and assumptions included in the example, to calculate the non-standard mortgage payment on a negative amortization loan for which the consumer has made more than the minimum required payment that must be compared to the standard mortgage payment under § 1026.43(d)(1)(i), the creditor must use the maximum loan amount of $229,219 as of March 1, 2019, the fully indexed rate of 8 percent, which is the index value of 4.5 percent as of March 15, 2012 (the date on which the creditor receives the application for a refinancing) plus the margin of 3.5 percent, and the remaining loan term as of March 1, 2019, the date of the recast, which is 25 years (300 monthly payments). The comment further explains that, based on these assumptions, the monthly payment for the non-standard mortgage for purposes of determining whether the standard mortgage monthly payment is lower than the non-standard mortgage monthly payment is $1,769. This is the substantially equal, monthly payment of principal and interest required to repay the maximum loan amount at the fully indexed rate over the remaining term. The Bureau finds that comment 43(d)(5)(i)-10, which is adopted pursuant to the Bureau's authority under section 105(a) of TILA, is necessary to facilitate compliance with TILA.

43(d)(5)(ii) Standard Mortgage

Proposed § 226.43(d)(5)(ii) would have prescribed the required calculation for the monthly payment on a standard mortgage that must be compared to the monthly payment on a non-standard mortgage under proposed § 226.43(d)(1)(ii). The same payment calculation must also be used by creditors of refinances under proposed § 226.43(d) in determining whether the consumer has a reasonable ability to repay the standard mortgage, as would have been required under proposed § 226.43(c)(2)(ii).

Specifically, the monthly payment for a standard mortgage must be based on substantially equal, monthly, fully amortizing payments using the maximum interest rate that may apply to the standard mortgage within the first five years after consummation. Proposed comment 43(d)(5)(ii)-1 would have clarified that the meaning of “fully amortizing payment” is defined in § 226.43(b)(2), and that guidance regarding the meaning of “substantially equal” may be found in proposed comment 43(c)(5)(i)-4. Proposed comment 43(d)(5)(ii)-1 also explained that, for a mortgage with a single, fixed rate for the first five years, the maximum rate that will apply during the first five years after consummation will be the rate at consummation. For a step-rate mortgage, however, which is a type of fixed-rate mortgage, the rate that must be used is the highest rate that will apply during the first five years after consummation. For example, if the rate for the first two years is 4 percent, the rate for the second two years is 5 percent, and the rate for the next two years is 6 percent, the rate that must be used is 6 percent.

Proposed comment 43(d)(5)(ii)-2 would have provided an illustration of the payment calculation for a standard mortgage. The example assumes a loan in an amount of $200,000 with a 30-year loan term. The loan agreement provides for an interest rate of 6 percent that is fixed for an initial period of five years, after which time the interest rate will adjust annually based on a specified index plus a margin of 3 percent, subject to a 2 percent annual interest rate adjustment cap. The comment states that, based on the above assumptions, the creditor must determine whether the standard mortgage payment is materially lower than the non-standard mortgage payment based on a standard mortgage payment of $1,199. This is the substantially equal, monthly payment of principal and interest required to repay $200,000 over 30 years at an interest rate of 6 percent.

The Bureau received no specific comment on the payment calculations for standard mortgages set forth in proposed § 226.43(d)(5)(ii) and its associated commentary. Accordingly, this provisions is adopted as proposed, renumbered as § 1026.43(d)(5)(ii). The Bureau also is adopting the associated commentary generally as proposed, with several technical amendments for clarity.

43(e) Qualified Mortgages

Background

As discussed above, TILA section 129C(a)(1) prohibits a creditor from making a residential mortgage loan unless the creditor makes a reasonable and good faith determination, at or before consummation, based on verified and documented information, that at the time of consummation the consumer has a reasonable ability to repay the loan. TILA section 129C(a)(1) through (4) and (6) through (9) requires creditors specifically to consider and verify various factors relating to the consumer's income and other assets, debts and other obligations, and credit history. However, the ability-to-repay provisions do not directly restrict features, term, or costs of the loan.

TILA section 129C(b), in contrast, provides that loans that meet certain requirements shall be deemed “qualified mortgages,” which are entitled to a presumption of compliance with the ability-to-repay requirements. The section sets forth a number of qualified mortgage requirements which focus mainly on prohibiting certain risky features and practices (such as negative amortization and interest-only periods or underwriting a loan without verifying the consumer's income) and on generally limiting points and fees in excess of 3 percent of the total loan amount. The only underwriting provisions in the statutory definition of qualified mortgage are a requirement that “income and financial resources relied upon to qualify the [borrowers] be verified and documented” and a further requirement that underwriting be based upon a fully amortizing schedule using the maximum rate permitted during the first five years of the loan. TILA section 129C(b)(2)(A)(iii) through (v). However, TILA section 129C(b)(2)(A)(vi) authorizes the Bureau to adopt “guidelines or regulations * * * relating to ratios of total monthly debt to monthly income or alternative measures of ability to pay * * * .” And TILA section 129C(b)(3)(B)(i) further authorizes the Bureau to revise, add to, or subtract from the criteria that define a qualified mortgage upon a finding that the changes are necessary or proper to ensure that responsible, affordable mortgage credit remains available to consumers in a manner consistent with the purposes of TILA section 129C, necessary and appropriate to effectuate the purposes of TILA sections 129C and 129B, to prevent circumvention or evasion thereof, or to facilitate compliance with TILA sections 129C and 129B.

The qualified mortgage requirements are critical to implementation of various parts of the Dodd-Frank Act. For example, several consumer protection requirements in title XIV of the Dodd-Frank Act treat qualified mortgages differently than non-qualified mortgages or key off elements of the qualified mortgage definition. In addition, the requirements concerning retention of risk by parties involved in the securitization process under title IX of the Dodd-Frank Act provide special treatment for “qualified residential mortgages,” which under section 15G of the Securities Exchange Act of 1934, as amended by section 941(b) of the Dodd-Frank Act, “shall be no broader than the term `qualified mortgage,'” as defined by TILA section 129C(b) and the Bureau's implementing regulations. 15 U.S.C. 780-11(e)(4).

For present purposes, however, the definition of a qualified mortgage is perhaps most significant because of its implications for ability-to-repay claims. TILA section 129C(b)(1) provides that “[a]ny creditor with respect to any residential mortgage loan, and any assignee of such loan subject to liability under this title, may presume that the loan has met the [ability-to-repay] requirements of subsection (a), if the loan is a qualified mortgage.” But the statute does not describe the strength of the presumption or what if anything could be used to rebut it. As discussed further below, there are legal and policy arguments that support interpreting the presumption as either rebuttable or conclusive.

Determining the definition and scope of protection afforded to qualified mortgages is the area of this rulemaking which has engendered perhaps the greatest interest and comment. Although TILA section 129C(a)(1) requires only that a creditor make a “reasonable and good faith determination” of the consumer's “reasonable ability to repay” a residential mortgage, considerable concern has arisen about the actual and perceived litigation and liability risk to creditors and assignees under the statute. Commenters tended to focus heavily on the choice between a presumption that is rebuttable and one that is conclusive as a means of mitigating that risk, although the criteria that define a qualified mortgage are also important because a creditor would have to prove status as a qualified mortgage in order to invoke any (rebuttable or conclusive) presumption of compliance.

In assessing the potential impacts of the statute, it is important to note that regulations issued after the mortgage crisis but prior to the enactment of the Dodd-Frank Act have already imposed ability-to-repay requirements for high-cost and higher-priced mortgages and created a presumption of compliance for such mortgages if the creditor satisfied certain underwriting and verification requirements. Specifically, under provisions of the Board's 2008 HOEPA Final Rule that took effect in October 2009, creditors are prohibited from extending high-cost or higher-priced mortgage loans without regard to the consumer's ability to repay. See § 1026.34(a)(4). The rules provide a presumption of compliance with those ability-to-repay requirements if the creditor follows certain optional procedures regarding underwriting the loan payment, assessing the debt-to-income (DTI) ratio or residual income, and limiting the features of the loan, in addition to following certain procedures mandated for all creditors. See § 1026.34(a)(4)(iii) and (iv) and comment 34(a)(4)(iii)-1. However, the 2008 HOEPA Final Rule makes clear that even if the creditor follows these criteria, the presumption of compliance is rebuttable. See comment 34(a)(4)(iii)-1. The consumer can still overcome that presumption by showing that, despite following the required and optional procedures, the creditor nonetheless disregarded the consumer's ability to repay the loan. For example, the consumer could present evidence that although the creditor assessed the consumer's debt-to-income ratio or residual income, the debt-to-income ratio was very high or the residual income was very low. This evidence may be sufficient to overcome the presumption of compliance and demonstrate that the creditor extended credit without regard to the consumer's ability to repay the loan.

The Dodd-Frank Act extends a requirement to assess consumers' ability to repay to the full mortgage market, and establishes a presumption using a different set of criteria that focus more on product features than underwriting practices. Further, the statute establishes similar but slightly different remedies than are available under the existing requirements. Section 1416 of the Dodd-Frank Act amended TILA section 130(a) to provide that a consumer who brings a timely action against a creditor for a violation the ability-to-repay requirements may be able to recover special statutory damages equal to the sum of all finance charges and fees paid by the consumer. The statute of limitations is three years from the date of the occurrence of the violation. Moreover, as amended by section 1413 of the Dodd-Frank Act, TILA section 130(k) provides that when a creditor, assignee, or other holder initiates a foreclosure action, a consumer may assert a violation of the ability-to-repay requirements as a matter of defense by recoupment or setoff. There is no time limit on the use of this defense, but the amount of recoupment or setoff is limited with respect to the special statutory damages to no more than three years of finance charges and fees. This limit on setoff is more restrictive than under the existing regulations, but also expressly applies to assignees.

In light of the statutory ambiguities, complex policy considerations, and concerns about litigation risk, the Board's proposal mapped out two alternatives at the opposite ends of a spectrum for defining a qualified mortgage and the protection afforded to such mortgages. At one end, the Board's Alternative 1 would have defined qualified mortgage only to include the mandated statutory elements listed in TILA section 129C(b)(2), most of which, as noted above, relate to product features and not to the underwriting decision or process itself. This alternative would have provided creditors with a safe harbor to establish compliance with the general repayment ability requirement in proposed § 226.43(c)(1). As the Board recognized, this would provide strong incentives for creditors to make qualified mortgages in order to minimize litigation risk and compliance burden under general ability-to-repay requirements, but might prevent consumers from seeking redress for failure to assess their ability to repay. In Alternative 2, the Board proposed a definition of qualified mortgage which incorporated both the statutory product feature restrictions and additional underwriting elements drawn from the general ability-to-repay requirements, as well as seeking comment on whether to establish a specific debt-to-income requirement. Alternative 2 also specified that consumers could rebut the presumption of compliance by demonstrating that a creditor did not adequately determine the consumers' ability to repay the loan. As the Board recognized, this would better ensure that creditors fully evaluate consumers' ability to repay qualified mortgages and preserve consumers' rights to seek redress. However, the Board expressed concern that Alternative 2 would provide little incentive to make qualified mortgages in the first place, given that the requirements may be challenging to satisfy and the strength of protection afforded would be minimal.

Overview of Final Rule

As noted above and discussed in greater detail in the section-by-section analysis below, the Dodd-Frank Act accords the Bureau significant discretion in defining the scope of, and legal protections afforded to, a qualified mortgage. In developing the rules for qualified mortgages, the Bureau has carefully considered numerous factors, including the Board's proposal to implement TILA section 129C(b), comments and ex parte communications, current regulations and the current state of the mortgage market, and the implications of the qualified mortgage rule on other parts of the Dodd-Frank Act. The Bureau is acutely aware of the problematic practices that gave rise to the financial crisis and sees the ability-to-pay requirement as an important bulwark to prevent a recurrence of those practices by establishing a floor for safe underwriting. At the same time, the Bureau is equally aware of the anxiety in the mortgage market today concerning the continued slow pace of recovery and the confluence of multiple major regulatory and capital initiatives. Although every industry representative that has communicated with the Bureau acknowledges the importance of assessing a consumer's ability to repay before extending a mortgage to the consumer—and no creditor claims to do otherwise—there is nonetheless a widespread fear about the litigation risks associated with the Dodd-Frank Act ability-to-repay requirements. Even community banks, deeply ingrained within their local communities and committed to a relationship lending model, have expressed to the Bureau their fear of litigation. In crafting the rules to implement the qualified mortgage provision, the Bureau has sought to balance creating new protections for consumers and new responsibilities for creditors with preserving consumers' access to credit and allowing for appropriate lending and innovation.

The Bureau recognizes both the need for certainty in the short term and the risk that actions taken by the Bureau in order to provide such certainty could, over time, defeat the prophylactic aims of the statute or impede recovery in various parts of the market. For instance, in defining the criteria for a qualified mortgage, the Bureau is called upon to identify a class of mortgages which can be presumed to be affordable. The boundaries must be clearly drawn so that consumers, creditors, and secondary market investors can all proceed with reasonable assurance as to whether a particular loan constitutes a qualified mortgage. Yet the Bureau believes that it is not possible by rule to define every instance in which a mortgage is affordable, and the Bureau fears that an overly broad definition of qualified mortgage could stigmatize non-qualified mortgages or leave insufficient liquidity for such loans. If the definition of qualified mortgage is so broad as to deter creditors from making non-qualified mortgages altogether, the regulation would curtail access to responsible credit for consumers and turn the Bureau's definition of a qualified mortgage into a straitjacket setting the outer boundary of credit availability. The Bureau does not believe such a result would be consistent with congressional intent or in the best interests of consumers or the market.

The Bureau is thus attuned to the problems of the past, the pressures that exist today, and the ways in which the market might return in the future. As a result, the Bureau has worked to establish guideposts in the final rule to make sure that the market's return is healthy and sustainable for the long-term. Within that framework, the Bureau is defining qualified mortgages to strike a clear and calibrated balance as follows:

First, the final rule provides meaningful protections for consumers while providing clarity to creditors about what they must do if they seek to invoke the qualified mortgage presumption of compliance. Accordingly, the qualified mortgage criteria include not only the minimum elements required by the statute—including prohibitions on risky loan features, a cap on points and fees, and special underwriting rules for adjustable-rate mortgages—but additional underwriting features to ensure that creditors do in fact evaluate individual consumers' ability to repay the qualified mortgages. The qualified mortgage criteria thus incorporate key elements of the verification requirements under the ability-to-repay standard and strengthen the consumer protections established by the ability-to-repay requirements.

In particular, the final rule provides a bright-line threshold for the consumer's total debt-to-income ratio, so that under a qualified mortgage, the consumer's total monthly debt payments cannot exceed 43 percent of the consumer's total monthly income. The bright-line threshold for debt-to-income serves multiple purposes. First, it protects consumer interests because debt-to-income ratios are a common and important tool for evaluating consumers' ability to repay their loans over time, and the 43 percent threshold has been utilized by the Federal Housing Administration (FHA) for many years as its general boundary for defining affordability. Relative to other benchmarks that are used in the market (such as GSE guidelines) that have a benchmark of 36 percent, before consideration of compensating factors, this threshold is a relatively liberal one which allows ample room for consumers to qualify for an affordable mortgage. Second, it provides a well-established and well-understood rule that will provide certainty for creditors and help to minimize the potential for disputes and costly litigation over whether a mortgage is a qualified mortgage. Third, it allows room for a vibrant market for non-qualified mortgages over time. The Bureau recognizes that there will be many instances in which individual consumers can afford an even higher debt-to-income ratio based on their particular circumstances, although the Bureau believes that such loans are better evaluated on an individual basis under the ability-to-repay criteria rather than with a blanket presumption. The Bureau also believes that there are a sufficient number of potential borrowers who can afford a mortgage that would bring their debt-to-income ratio above 43 percent that responsible creditors will continue to make such loans as they become more comfortable with the new regulatory framework. To preserve access to credit during the transition period, the Bureau has also adopted temporary measures as discussed further below.

The second major feature of the final rule is the provision of carefully calibrated presumptions of compliance afforded to different types of qualified mortgages. Following the approach developed by the Board in the existing ability-to-repay rules to distinguish between prime and subprime loans, the final rule distinguishes between two types of qualified mortgages based on the mortgage's Annual Percentage Rate (APR) relative to the Average Prime Offer Rate (APOR). For loans that exceed APOR by a specified amount—loans denominated as “higher-priced mortgage loans”—the final rule provides a rebuttable presumption. In other words, the creditor is presumed to have satisfied the ability-to-repay requirements, but a consumer may rebut that presumption under carefully defined circumstances. For all other loans, i.e., loans that are not “higher-priced,” the final rule provides a conclusive presumption that the creditor has satisfied the ability-to-repay requirements once the creditor proves that it has in fact made a qualified mortgage. In other words, the final rule provides a safe harbor from ability-to-repay challenges for the least risky type of qualified mortgages, while providing room to rebut the presumption for qualified mortgages whose pricing is indicative of a higher level of risk. The Bureau believes that this calibration will further encourage creditors to extend credit responsibly and provide certainty that promotes access to credit.

The Bureau believes that loans that fall within the rebuttable presumption category will be loans made to consumers who are more likely to be vulnerable so that, even if the loans satisfy the criteria for a qualified mortgage, those consumers should be provided the opportunity to prove that, in an individual case, the creditor did not have a reasonable belief that the loan would be affordable for that consumer. Under a qualified mortgage with a safe harbor, most of the loans within this category will be the loans made to prime borrowers who pose fewer risks. Furthermore, considering the difference in historical performance levels between prime and subprime loans, the Bureau believes that it is reasonable to presume conclusively that a creditor who has verified a consumer's debt and income, determined in accordance with specified standards that the consumer has a debt-to-income ratio that does not exceed 43 percent, and made a prime mortgage with the product features required for a qualified mortgage has satisfied its obligation to assess the consumer's ability to repay. This approach will provide significant certainty to creditors operating in the prime market. The approach will also create lesser but still important protection for creditors in the subprime market who follow the qualified mortgage rules, while preserving consumer remedies and creating strong incentives for more responsible lending in the part of the market in which the most abuses occurred prior to the financial crisis.

Third, the final rule provides a temporary special rule for certain qualified mortgages to provide a transition period to help ensure that sustainable credit will return in all parts of the market over time. The temporary special rule expands the definition of a qualified mortgage to include any loan that is eligible to be purchased, guaranteed, or insured by various Federal agencies or by the GSEs while they are operating under conservatorship. This temporary provision preserves access to credit in today's market by permitting a loan that does not satisfy the 43 percent debt-to-income ratio threshold to nonetheless be a qualified mortgage based upon an underwriting determination made pursuant to guidelines created by the GSEs while in conservatorship or one of the Federal agencies. This temporary provision will sunset in a maximum of seven years. As with loans that satisfy the 43 percent debt-to-income ratio threshold, qualified mortgages under this temporary rule will receive either a rebuttable or conclusive presumption of compliance depending upon the pricing of the loan relative to APOR. The Bureau believes this provision will provide sufficient consumer protection while providing adequate time for creditors to adjust to the new requirements of the final rule as well as to changes in other regulatory, capital, and economic conditions.

A detailed description of the qualified mortgage definition is set forth below. Section 1026.43(e)(1) provides the presumption of compliance provided to qualified mortgages. Section 1026.43(e)(2) provides the criteria for a qualified mortgage under the general definition, including the restrictions on certain product features, verification requirements, and a specified debt-to-income ratio threshold. Section 1026.43(e)(3) provides the limits on points and fees for qualified mortgages, including the limits for smaller loan amounts. Section 1026.43(e)(4) provides the temporary special rule for qualified mortgages. Lastly, § 1026.43(f) implements a statutory exemption permitting certain balloon-payment loans by creditors operating predominantly in rural or underserved areas to be qualified mortgages.

43(e)(1) Safe Harbor and Presumption of Compliance

As discussed above, the Dodd-Frank Act provides a presumption of compliance with the ability-to-repay requirements for qualified mortgages, but the statute is not clear as to whether that presumption is intended to be conclusive so as to create a safe harbor that cuts off litigation or a rebuttable presumption of compliance with the ability-to-repay requirements. The title of section 1412 refers to both a “safe harbor and rebuttable presumption,” and as discussed below there are references to both safe harbors and presumptions in other provisions of the statute. As the Board's proposal discussed, an analysis of the statutory construction and policy implications demonstrates that there are sound reasons for adopting either interpretation. See 76 FR 27390, 27452-55 (May 11, 2011).

Several aspects of the statutory structure favor a safe harbor interpretation. First, TILA section 129C(b)(1) states that a creditor or assignee may presume that a loan has “met the requirements of subsection (a), if the loan is a qualified mortgage.” TILA section 129C(a) contains the general ability-to repay requirement, and also a set of specific underwriting criteria that must be considered by a creditor in assessing the consumer's repayment ability. Rather than stating that the presumption of compliance applies only to TILA section 129C(a)(1) for the general ability-to-repay requirements, it appears Congress intended creditors who make qualified mortgages to be presumed to comply with both the ability-to-repay requirements and all of the specific underwriting criteria. Second, TILA section 129C(b)(2) does not define a qualified mortgage as requiring compliance with all of the underwriting criteria of the general ability-to-repay standard. Therefore, unlike the approach found in the 2008 HOEPA Final Rule, it appears that meeting the criteria for a qualified mortgage is an alternative way of establishing compliance with all of the ability-to-repay requirements, which could suggest that meeting the qualified mortgage criteria conclusively satisfies these requirements. In other words, given that a qualified mortgage satisfies the ability-to-repay requirements, one could assume that meeting the qualified mortgage definition conclusively establishes compliance with those requirements.

In addition, TILA section 129C(b)(3)(B), which provides the Bureau authority to revise, add to, or subtract from the qualified mortgage criteria upon making certain findings, is titled “Revision of Safe Harbor Criteria.” Further, in section 1421 of the Dodd-Frank Act, Congress instructed the Government Accountability Office to issue a study on the effect “on the mortgage market for mortgages that are not within the safe harbor provided in the amendments made by this subtitle.”

Certain policy considerations also favor a safe harbor. Treating a qualified mortgage as a safe harbor provides greater legal certainty for creditors and secondary market participants than a rebuttable presumption of compliance. Increased legal certainty may benefit consumers if as a result creditors are encouraged to make loans that satisfy the qualified mortgage criteria, as such loans cannot have certain risky features and have a cap on upfront costs. Furthermore, increased certainty may result in loans with a lower cost than would be charged in a world of legal uncertainty. Thus, a safe harbor may also allow creditors to provide consumers additional or more affordable access to credit by reducing their expected total litigation costs.

On the other hand, there are also several aspects of the statutory structure that favor interpreting qualified mortgage as creating a rebuttable presumption of compliance. With respect to statutory construction, TILA section 129C(b)(1) states that a creditor or assignee “may presume” that a loan has met the repayment ability requirement if the loan is a qualified mortgage. As the Board's proposal notes, this could suggest that originating a qualified mortgage provides a presumption of compliance with the repayment ability requirements, which the consumer can rebut with evidence that the creditor did not, in fact, make a good faith and reasonable determination of the consumer's ability to repay the loan. Similarly, in the smaller loans provisions in TILA section 129C(b)(2)(D), Congress instructed the Bureau to adjust the points and fees cap for qualified mortgages “to permit lenders that extend smaller loans to meet the requirements of the presumption of compliance” in TILA section 129C(b)(1). As noted above, the 2008 HOEPA Final Rule also contains a rebuttable presumption of compliance with respect to the ability-to-repay requirements that currently apply to high-cost and higher-priced mortgages.

The legislative history of the Dodd-Frank Act may also favor interpreting “qualified mortgage” as a rebuttable presumption of compliance. As described in a joint comment letter from several consumer advocacy groups, a prior version of Dodd-Frank Act title XIV from 2007 contemplated a dual track for liability in litigation: a rebuttable presumption for creditors and a safe harbor for secondary market participants. That draft legislation would have provided that creditors, assignees, and securitizers could presume compliance with the ability-to-repay provision if the loan met certain requirements. However, the presumption of compliance would have been rebuttable only against the creditor, effectively creating a safe harbor for assignees and securitizers. The caption “safe harbor and rebuttable presumption” appears to have originated from the 2007 version of the legislation. The 2009 version of the legislation did not contain this dual track approach. Instead, the language simply stated that creditors, assignees, and securitizers “may presume” that qualified mortgages satisfied ability-to-repay requirements, without specifying the nature of the presumption. The committee report of the 2009 bill described the provision as establishing a “limited safe harbor” for qualified mortgages, while also stating that “the presumption can be rebutted.” This suggests that Congress contemplated that qualified mortgages would receive a rebuttable presumption of compliance with the ability-to-repay provisions, notwithstanding Congress's use of the term “safe harbor” in the heading of section 129C(b) and elsewhere in the statute and legislative history.

There are also policy reasons that favor interpreting “qualified mortgage” as a rebuttable presumption of compliance. The ultimate aim of the statutory provisions is to assure that, before making a mortgage loan, the creditor makes a determination of the consumer's ability to repay. No matter how many elements the Bureau might add to the definition of qualified mortgage, it still would not be possible to define a class of loans which ensured that every consumer within the class could necessarily afford a particular loan. In light of this, interpreting the statute to provide a safe harbor that precludes a consumer from challenging the creditor's determination of repayment ability seems to raise tensions with the requirement to determine repayment ability. In contrast, interpreting a qualified mortgage as providing a rebuttable presumption of compliance would better ensure that creditors consider each consumer's ability to repay the loan rather than only satisfying the qualified mortgage criteria.

The Board's Proposal

As described above, in light of the statutory ambiguity and competing policy considerations, the Board proposed two alternative definitions for a qualified mortgage, which generally represent two ends of the spectrum of possible definitions. Alternative 1 would have applied only the specific requirements listed for qualified mortgages in TILA section 129C(b)(2), and would have provided creditors with a safe harbor to establish compliance with the general repayment ability requirement in proposed § 226.43(c)(1). Alternative 2 would have required a qualified mortgage to satisfy the specific requirements listed in the TILA section 129C(b)(2), as well as additional requirements taken from the general ability-to-repay standard in proposed § 226.43(c)(2) through (7). Alternative 2 would have provided a rebuttable presumption of compliance with the ability-to-repay requirements. Although the Board specifically proposed two alternative qualified mortgage definitions, it also sought comment on other approaches by soliciting comment on other alternative definitions. The Board also specifically solicited comment on what criteria should be included in the definition of a qualified mortgage to ensure that the definition provides an incentive to creditors to make qualified mortgages, while also ensuring that consumers have the ability to repay those loans. In particular, the Board sought comment on whether the qualified mortgage definition should require consideration of a consumer's debt-to-income ratio or residual income, including whether and how to include a quantitative standard for the debt-to-income ratio or residual income for the qualified mortgage definition.

Comments

Generally, numerous industry and other commenters, including some members of Congress, supported a legal safe harbor while consumer groups and other commenters, including an association of State bank regulators, supported a rebuttable presumption. However, as described below, commenters did not necessarily support the two alternative proposals specifically as drafted by the Board. For instance, a significant number of industry commenters advocated incorporating the general ability-to-repay requirements into the qualified mortgage definition, while providing a safe harbor for those loans that met the enhanced standards. And a coalition of industry and consumer advocates presented a proposal to the Bureau that would have provided a tiered approach to defining a qualified mortgage. Under the first tier, if the consumer's back-end debt-to-income (total debt-to-income) ratio is 43 percent or less, the loan would be a qualified mortgage, and no other tests would be required. Under the second tier, if the consumer's total debt-to-income ratio is more than 43 percent, the creditor would apply a series of tests related to the consumer's front-end debt-to-income ratio (housing debt-to-income), stability of income and past payment history, availability of reserves, and residual income to determine if a loan is a qualified mortgage.

Comments in favor of safe harbor. Industry commenters strongly supported a legal safe harbor from liability for qualified mortgages. These commenters believe that a broad safe harbor with clear, bright lines would provide certainty and clarity for creditors and assignees. Generally, industry commenters argued that a safe harbor is needed in order: (i) To ensure creditors make loans, (ii) to ensure the availability of and access to affordable credit without increasing the costs of borrowing; (iii) to promote certainty and saleability in the secondary market, and (iv) to contain litigation risk and costs for creditors and assignees.

Generally, although acknowledging ambiguities in the statutory language, industry commenters argued that the statute's intent and legislative history indicate that qualified mortgages are meant to be a legal safe harbor, in lieu of the ability-to-repay standards. Industry commenters argued that a safe harbor would best ensure safe, well-documented, and properly underwritten loans without limiting the availability of credit or increasing the costs of credit to consumers. Many industry commenters asserted that a legal safe harbor from liability would ensure access to affordable credit. Other industry commenters argued that a safe harbor ultimately benefits consumers with increased access to credit, reduced loan fees and interest rates, and less-risky loan features. In contrast, various industry commenters contended that a rebuttable presumption would not provide enough certainty for creditors and the secondary market. Commenters argued that if creditors cannot easily ascertain whether a loan satisfies the ability-to-repay requirements, creditors will either not make loans or will pass the cost of uncertain legal risk to consumers, which in turn would increase the cost of borrowing.

Numerous industry commenters argued for a legal safe harbor because of the liabilities of an ability-to-repay violation and the costs associated with ability-to-repay litigation. Generally, commenters argued that a rebuttable presumption for qualified mortgages would invite more extensive litigation than necessary that will result in greater costs being borne by all consumers. Commenters emphasized the relatively severe penalties for ability-to-repay violations under the Dodd-Frank Act, including enhanced damages, an extended three-year statute of limitations, a recoupment or set-off provision as a defense to foreclosure, and new enforcement authorities by State attorneys general. In addition, assignee liabilities are amplified because of the recoupment and set-off provision in TILA section 130(k). Commenters asserted that the increased costs associated with litigation could make compliance too costly for smaller creditors, which would reduce competition and credit availability from the market. In particular, community bank trade association commenters argued that the Bureau should adopt a safe harbor for qualified mortgage loans and include bright-line requirements to protect community banks from litigation and ease the compliance burden. Ultimately, community bank trade association commenters stated that few, if any, banks would risk providing a mortgage that only has a rebuttable presumption attached.

Industry commenters generally believed that a rebuttable presumption would increase the incidence of litigation because any consumer who defaults on a loan would be likely to sue for recoupment in foreclosure. Commenters were also concerned about frivolous challenges in court as well as heightened scrutiny by regulators. In particular, a credit union association commenter supported a safe harbor because of concerns that a rebuttable presumption would cause credit unions to be faced with significant amounts of frivolous foreclosure defense litigation in the future. In addition to increased incidence of litigation, industry commenters and other interested parties argued that the estimated costs of litigation under a rebuttable presumption would be overly burdensome for creditors and assignees. Some commenters and interested parties presented estimates of the litigation costs associated with claims alleging a violation of the ability-to-repay requirements. For example, one industry trade association commenter estimated that the attorney's fees for a claim involving a qualified mortgage under a safe harbor would cost $30,000, compared to $50,000 for a claim under a rebuttable presumption. That commenter provided a separate estimation from a law firm that the attorneys' fees to the creditor will be approximately $26,000 in cases where the matter is disposed of on a motion to dismiss, whereas the fees for the cost of a full trial could reach $155,000. That commenter asserted that safe harbor claims are more likely to be dismissed on a motion to dismiss than the rebuttable presumption.

An industry commenter and other interested parties argued that the estimated costs to creditors associated with litigation and penalties for an ability-to-repay violation could be substantial and provided illustrations of costs under the proposal, noting potential cost estimates of the possible statutory damages and attorney's fees. For example, the total estimated costs and damages ranged between approximately $70,000 and $110,000 depending on various assumptions, such as the interest rate on a loan or whether the presumption of compliance is conclusive or rebuttable.

Industry commenters also generally argued that a safe harbor would promote access to credit because creditors would be more willing to extend credit where they receive protections under the statutory scheme. One industry trade association commenter cited the 2008 HOEPA Final Rule, which provided a rebuttable presumption of compliance with the requirement to consider a consumer's repayment ability upon meeting certain criteria, as causing a significant drop in higher-priced mortgage loan originations, and suggested that access to general mortgage credit would be similarly restricted if the final rule adopts a rebuttable presumption for the market as a whole. A large bank commenter similarly noted the lack of lending in the higher-priced mortgage space since the 2008 HOEPA Final Rule took effect.

In addition to the liquidity constraints for non-qualified mortgages, commenters argued that the liability and damages from a potential ability-to-repay TILA violation would be a disincentive for a majority of creditors to make non-qualified mortgage loans. Further, some commenters suggested that creditors could face reputational risk from making non-qualified mortgage loans because consumers would view them as “inferior” to qualified mortgages. Other commenters argued that reducing the protections afforded to qualified mortgages could cause creditors to act more conservatively and restrict credit or result in the denial of credit at a higher rate and increase the cost of credit. Many commenters argued that the most serious effects and impacts on the availability and cost of credit would be for minority, low- to moderate-income, and first-time borrowers. Therefore, industry commenters believed that a bright-line safe harbor would provide the strongest incentive for creditors to provide sustainable mortgage credit to the widest array of qualified consumers. Furthermore, one industry trade association commenter argued that not providing strong incentives for creditors would diminish the possibility of recovery of the housing market and the nation's economy.

Industry commenters also expressed concerns regarding secondary market considerations and assignee liability. Commenters urged the Bureau to consider commercial litigation costs associated with the contractually required repurchase (“put-back”) of loans sold on the secondary market where there is litigation over those loans, as well as the risk of extended foreclosure timelines because of ongoing ability-to-repay litigation. Industry commenters asserted that a safe harbor is critical to promote saleability of loans in the secondary market. In particular, they stated that clarity and certainty provided by a safe harbor would promote efficiencies in the secondary market because investors in securitized residential mortgage loans (mortgage backed securities, or MBS) could be more certain that they are not purchasing compliance risk along with their investments. Commenters asserted that without a safe harbor, the resulting uncertainty would eliminate the efficiencies provided by secondary sale or securitization of loans. By extension, commenters claimed that the cost of borrowing for consumers would ultimately increase. Large bank commenters stated that although they might originate non-qualified mortgage loans, the number would be relatively small and held in portfolio because they believe it is unlikely that non-qualified mortgage loans will be saleable in the secondary market. Generally, industry commenters asserted that creditors, regardless of size, would be unwilling to risk exposure outside the qualified mortgage space. One large bank commenter stated that the 2008 HOEPA Final Rule did not create a defense to foreclosure against assignees for the life of the loan, as does the Dodd-Frank Act's ability-to-repay provisions. Accordingly, industry commenters strongly supported broad coverage of qualified mortgages, as noted above.

Commenters asserted that the secondary market will demand a “safe harbor” for quality assurance and risk avoidance. If the regulatory framework does not provide a safe harbor, commenters asserted that investors would require creditors to agree to additional, strict representations and warranties when assigning loans. Contracts between loan originators and secondary market purchasers often require originators to repurchase loans should a loan perform poorly, and these commenters expect that future contracts will include provisions related to the ability-to-repay rule. Commenters assert that the risks and costs associated with additional potential put-backs to the creditor would increase liability and risk to creditors, which would ultimately increase the cost of credit to consumers. Furthermore, commenters contended that if the rule is too onerous in its application to the secondary market, then the secondary market participants may purchase fewer loans or increase pricing to account for the additional risk, such as is now the case for high-cost mortgages.

Commenters noted that the risks associated with assignee liability are heightened by any vagueness in standards in the rule. One secondary market purchaser commenter argued that a rebuttable presumption would present challenges because purchasers (or assignees) are not part of the origination process. It is not feasible for purchasers to evaluate all of the considerations that went into an underwriting decision, so they must rely on the creditor's representations that the loan was originated in compliance with applicable laws and the purchaser's requirements. However, assignees may have to defend a creditor's underwriting decision at any time during the life of the loan because there is no statute of limitations on raising the failure to make an ability-to-repay determination as a defense to foreclosure. The commenters argued that defending these cases would be difficult and costly, and that such burdens would be reduced by safe harbor protections.

Comments in favor of rebuttable presumption of compliance. Consumer group commenters generally urged the Bureau to adopt a rebuttable presumption for qualified mortgages. Commenters argued that Congress intended a rebuttable presumption, not a safe harbor. In particular, commenters contended that the Dodd-Frank Act's legislative history and statutory text strongly support a rebuttable presumption. Commenters noted that the statute is designed to strike a fair balance between market incentives and market discipline, as well as a balance between consumers' legal rights and excessive exposure to litigation risk for creditors. Commenters asserted that the purpose of the qualified mortgage designation is to foster sustainable lending products and practices built upon sound product design and sensible underwriting. To that end, a rebuttable presumption would accomplish the goal of encouraging creditors to originate loans that meet the qualified mortgage definition while assuring consumers of significantly greater protection from abusive or ineffective underwriting than if a safe harbor were adopted. Consumer group commenters contended that qualified mortgages can earn and deserve the trust of both consumers and investors only if they carry the assurance that they are soundly designed and properly underwritten. Many consumer group commenters asserted that a rebuttable presumption would provide better protections for consumers as well as improving safeguards against widespread risky lending while helping ensure that there would be no shortcuts on common sense underwriting. They argued that a legal safe harbor could invite abusive lending because consumers will have no legal recourse. Several commenters also asserted that no qualified mortgage definition could cover all contingencies in which such abuses could occur.

Some commenters argued that a legal safe harbor would leave consumers unprotected against abuses, such as those associated with simultaneous liens or from inadequate consideration of employment and income. An association of State bank regulators favored a rebuttable presumption because, although a rebuttable presumption provides less legal protection than a safe harbor, a rebuttable presumption encourages institutions to consider repayment factors that are part of a sound underwriting process. That commenter contended that a creditor should not be granted blanket protection from a foreclosure defense of an ability-to-repay violation if the creditor failed to consider and verify such crucial information as a consumer's employment status and credit history, for example. On this point, the rebuttable presumption proposed by the Board would require creditors to make individualized determinations that the consumer has the ability to repay the loan based on all of the underwriting factors listed in the general ability-to-repay standard.

Consumer group commenters observed that a rebuttable presumption would better ensure that creditors actually consider a consumer's ability to repay the loan. Consumer group commenters also asserted that the goals of safe, sound, sustainable mortgage lending and a balanced system of accountability are best served by a rebuttable presumption because consumers should be able to put evidence before a court that the creditor's consideration and verification of the consumer's ability to repay the loan was unreasonable or in bad faith. To that end, a rebuttable presumption would allow the consumer to assert that, despite complying with the criteria for a qualified mortgage and the ability-to-repay standard, the creditor did not make a reasonable and good faith determination of the consumer's ability to repay the loan. Without this accountability, commenters argued that the Dodd-Frank Act's effectiveness would be undermined.

Ultimately, consumer group commenters believed that a rebuttable presumption would not exacerbate current issues with credit access and availability, but would instead allow room for honest, efficient competition and affordable credit. Consumer group commenters generally contended that the fear of litigation and estimated costs and risks associated with ability-to-repay violations are overstated and based on misunderstanding of the extent of exposure to TILA liability. Consumer group commenters and some ex parte communications asserted that the potential incidence of litigation is relatively small, and therefore liability cost and risk are minimal for any given mortgage creditor. For example, consumer group commenters asserted that there are significant practical limitations to consumers bringing an ability-to-repay claim, suggesting that few distressed homeowners would be able to obtain legal representation often necessary to mount a successful rebuttal in litigation. Consumer groups provided percentages of borrowers in foreclosure who are represented by lawyers, noting the difficulty of bringing a TILA violation claim, and addressed estimates of litigation costs, such as attorneys' fees. Consumer groups provided estimates of the number of cases in foreclosure and the percentage of cases that involve TILA claims, such as a claim of rescission.

Furthermore, consumer group commenters argued that the three-year cap on enhanced damages (equal to the sum if all finance charges and fees paid by the consumer within three years of consummation) for violation of the ability-to-repay requirements limits litigation risk significantly. Commenters contended that, as a general rule, a court is more likely to find that the ability-to-repay determination at consummation was not reasonable and in good faith the earlier in the process a default occurs, and at that point the amount of interest paid by a consumer (a component of enhanced damages) will be relatively small. Commenters argued that the longer it takes a consumer to default, the harder the burden it will be for the consumer to show that the default was reasonably predictable at consummation and was caused by improper underwriting rather than a subsequent income or expense shock; moreover, even if the consumer can surmount that burden, the amount of damages is still capped at three years' worth of paid interest. In addition, consumer group commenters contended that the penalties to which creditors could be subject on a finding of failure to meet the ability-to-repay requirements would not be so injurious or even so likely to be applied in all but the most egregious situations as to impose any meaningful risk upon creditors.

Moreover, many consumer group commenters observed that creditors that comply with the rules and ensure that their loan originators are using sound, well documented and verified underwriting will be adequately protected by a rebuttable presumption.

Final Rule

As described above, the presumption afforded to qualified mortgages in the final rule balances consumers' ability to invoke the protections of the Dodd-Frank Act scheme with the need to create sufficient certainty to promote access to credit in all parts of the market. Specifically, the final rule provides a safe harbor with the ability-to-repay requirements for loans that meet the qualified mortgage criteria and pose the least risk, while providing a rebuttable presumption for “higher-priced” mortgage loans, defined as having an APR that exceeds APOR by 1.5 percentage points for first liens and 3.5 percentage points for second liens. The final rule also specifically defines the grounds on which the presumption accorded to more expensive qualified mortgages can be rebutted. In issuing this final rule, the Bureau has drawn on the experiences from the current ability-to-repay provisions that apply to higher-priced mortgages, described above. Based on the difference in historical performance levels between prime and subprime loans, the Bureau believes that this approach will provide significant certainty to creditors while preserving consumer remedies and creating strong incentives for more responsible lending in the part of the market in which the most abuses occurred prior to the financial crisis.

In issuing this final rule, the Bureau carefully considered the comments received and the interpretive and policy considerations for providing qualified mortgages either a safe harbor or rebuttable presumption of compliance with the repayment ability requirements. For the reasons set forth by the Board and discussed above, the Bureau finds that the statutory language is ambiguous and does not mandate a particular approach. In adopting the final rule, the Bureau accordingly focused on which interpretation would best promote the various policy goals of the statute, taking into account the Bureau's authority, among other things, to make adjustments and exceptions necessary or proper to effectuate the purposes of TILA, as amended by the Dodd-Frank Act.

Discouraging unsafe underwriting. As described in part II above, the ability-to-repay provisions of the Dodd-Frank Act were codified in response to lax lending terms and practices in the mid-2000's, which led to increased foreclosures, particularly for subprime borrowers. The statutory underwriting requirements for a qualified mortgage—for example, the requirement that loans be underwritten on a fully amortized basis using the maximum interest rate during the first five years and not a teaser rate, and the requirement to consider and verify a consumer's income or assets—will help prevent a return to such lax lending. So, too, will the requirement that a consumer's debt-to-income ratio (including mortgage-related obligations and obligations on simultaneous second liens) not exceed 43 percent, as discussed further below.

Notwithstanding these requirements, however, the Bureau recognizes that it is not possible to define by a bright-line rule a class of mortgages as to which it will always be the case that each individual consumer has the ability to repay his or her loan. That is especially true with respect to subprime loans. In many cases, the pricing of a subprime loan is the result of loan level price adjustments established by the secondary market and calibrated to default risk. Furthermore, the subprime segment of the market is comprised of borrowers who tend to be less sophisticated and who have fewer options available to them, and thus are more susceptible to being victimized by predatory lending practices. The historical performance of subprime loans bears all this out. The Bureau concludes, therefore, that for subprime loans there is reason to impose heightened standards to protect consumers and otherwise promote the policies of the statute. Accordingly, the Bureau believes that it is important to afford consumers the opportunity to rebut the presumption of compliance that applies to qualified mortgages with regard to higher-priced mortgages by showing that, in fact, the creditor did not have a good faith and reasonable belief in the consumer's reasonable ability to repay the loan at the time the loan was made.

These same considerations lead to the opposite result with respect to prime loans which satisfy the requirements for a qualified mortgage. The fact that a consumer receives a prime rate is itself indicative of the absence of any indicia that would warrant a loan level price adjustment, and thus is suggestive of the consumer's ability to repay. Historically, prime rate loans have performed significantly better than subprime rate loans and the prime segment of the market has been subject to fewer abuses. Moreover, requiring creditors to prove that they have satisfied the qualified mortgage requirements in order to invoke the presumption of compliance will itself ensure that the loans in question do not contain certain risky features and are underwritten with careful attention to consumers' debt-to-income ratios. Accordingly, the Bureau believes that where a loan is not a higher-priced covered transaction and meets both the product and underwriting requirements for a qualified mortgage, there are sufficient grounds for concluding that the creditor had a reasonable and good faith belief in the consumer's ability to repay to warrant a safe harbor.

This approach carefully balances the likelihood of consumers needing redress with the potential benefits to both consumers and industry of reducing uncertainty concerning the new regime. To the extent that the rule reduces litigation risk concerns for prime qualified mortgages, consumers in the prime market may benefit from enhanced competition (although, as discussed below, the Bureau believes litigation costs will be small and manageable for almost all creditors). In particular, the Bureau believes that larger creditors may expand correspondent lending relationships with smaller banks with respect to prime qualified mortgages. Larger creditors may also relax currently restrictive credit overlays (creditor-created underwriting requirements that go beyond GSE or agency guidelines), thereby increasing access to credit.

Scope of rebuttable presumption. In light of the heightened protections for subprime loans, the final rule also carefully defines the grounds on which the presumption that applies to higher-priced qualified mortgages can be rebutted. The Bureau believes that this feature is critical to ensuring that creditors have sufficient incentives to provide higher-priced qualified mortgages to consumers. Given the historical record of abuses in the subprime market, the Bureau believes it is particularly important to ensure that consumers are able to access qualified mortgages in light of their product feature restrictions and other protections.

Specifically, the final rule defines the standard by which a consumer may rebut the presumption of compliance afforded to higher-priced qualified mortgages, and provides an example of how a consumer may rebut the presumption. As described below, the final rule provides that consumers may rebut the presumption with regard to a higher-priced covered transaction by showing that, at the time the loan was originated, the consumer's income and debt obligations left insufficient residual income or assets to meet living expenses. The analysis would consider the consumer's monthly payments on the loan, mortgage-related obligations, and any simultaneous loans of which the creditor was aware, as well as any recurring, material living expenses of which the creditor was aware.

The Bureau believes the rebuttal standard in the final rule appropriately balances the consumer protection and access to credit considerations described above. This standard is consistent with the standard in the 2008 HOEPA Final Rule, and is specified as the exclusive means of rebutting the presumption. Commentary to the existing rule provides as an example of how its presumption may be rebutted that the consumer could show “a very high debt-to-income ratio and a very limited residual income.” Under the definition of qualified mortgage that the Bureau is adopting, however, the creditor generally is not entitled to a presumption if the debt-to-income ratio is “very high.” As a result, the Bureau is focusing the standard for rebutting the presumption in the final rule on whether, despite meeting a debt-to-income test, the consumer nonetheless had insufficient residual income to cover the consumer's living expenses. The Bureau believes this standard is sufficiently broad to provide consumers a reasonable opportunity to demonstrate that the creditor did not have a good faith and reasonable belief in the consumer's repayment ability, despite meeting the prerequisites of a qualified mortgage. At the same time, the Bureau believes the rebuttal standard in the final rule is sufficiently clear to provide certainty to creditors, investors, and regulators about the standards by which the presumption can successfully be challenged in cases where creditors have correctly followed the qualified mortgage requirements.

Several commenters raised concerns about the use of oral evidence to impeach the information contained in the loan file. For example, a consumer may seek to show that a loan does not meet the requirements of a qualified mortgage by relying on information provided orally to the creditor or loan originator to establish that the debt-to-income ratio was miscalculated. Alternatively, a consumer may seek to show that the creditor should have known, based upon facts disclosed orally to the creditor or loan originator, that the consumer had insufficient residual income to be able to afford the mortgage. The final rule does not preclude the use of such oral evidence in ability-to-repay cases. The Bureau believes that courts will determine the weight to be given to such evidence on a case-by-case basis. To exclude such evidence across the board would invite abuses in which consumers could be misled or coerced by an unscrupulous loan originator into keeping certain facts out of the written record.

Litigation risks and access to credit. In light of the continuing and widespread concern about litigation risk under the Dodd-Frank Act regime, the Bureau, in the course of developing the framework described above, carefully analyzed the impacts of potential litigation on non-qualified mortgages, any qualified mortgages with a rebuttable presumption, and any qualified mortgages with a safe harbor. The Bureau also considered secondary market dynamics, including the potential impacts on creditors from loans that the secondary market “puts back” on the originators because of ability-to-repay litigation. The Bureau's analysis is described in detail in the section 1022(b)(2) analysis under part VII; the results of that analysis helped to shape the calibrated approach that the Bureau is adopting in the final rule and suggest that the mortgage market will be able to absorb litigation risks under the rule without jeopardizing access to credit.

Specifically, as discussed in the section 1022(b)(2) analysis under part VII, the Bureau believes that even without the benefit of any presumption of compliance, the actual increase in costs from the litigation risk associated with ability-to-pay requirements would be quite modest. This is a function of the relatively small number of potential claims, the relatively small size of those claims, and the relatively low likelihood of claims being filed and successfully prosecuted. The Bureau notes that litigation likely would arise only when a consumer in fact was unable to repay the loan (i.e. was seriously delinquent or had defaulted), and even then only if the consumer elects to assert a claim and is able to secure a lawyer to provide representation; the consumer can prevail only upon proving that the creditor lacked a reasonable and good faith belief in the consumer's ability to repay at consummation or failed to consider the statutory factors in arriving at that belief.

The rebuttable presumption of compliance being afforded to qualified mortgages that are higher-priced reduces the litigation risk, and hence the potential transaction costs, still further. As described above, the Bureau has crafted the presumption of compliance being afforded to subprime loans so that it is not materially different than the presumption that exists today under the 2008 HOEPA Final Rule. Indeed, the Bureau is defining with more particularity the requirements for rebutting this presumption. No evidence has been presented to the Bureau to suggest that the presumption under the 2008 HOEPA Final Rule has led to significant litigation or to any distortions in the market for higher-priced mortgages. As noted above, commenters noted the lack of lending in the higher-priced mortgage space since the 2008 HOEPA Final Rule took effect, but the Bureau is unaware of evidence suggesting the low lending levels are the result of the Board's rule, as compared to the general state of the economy, uncertainty over multiple regulatory and capital initiatives, and other factors.

Relative to the Dodd-Frank Act, the Bureau notes that the existing regime already provides for attorneys' fees and the same remedies against creditors in affirmative cases, and actually provides for greater remedies against creditors in foreclosure defense situations. Nevertheless, the incidence of claims under the existing ability-to-repay rules for high-cost and higher-priced loans and analogous State laws is relatively low. The Bureau's analysis shows that cost estimates remain modest for both loans that are not qualified mortgages and loans that are qualified mortgages with a rebuttable presumption of compliance, and even more so for qualified mortgages with a safe harbor.

The Bureau recognizes, of course, that under the Dodd-Frank Act ability-to-repay provisions, a consumer can assert a claim against an assignee as a “defense by recoupment or set off” in a foreclosure action. There is no time limit on the use of this defense, but the consumer cannot recover as special statutory damages more than three years of finance charges and fees. To the extent this leads to increased litigation potential with respect to qualified mortgages as to which the presumption of compliance is rebuttable, this may cause creditors to take greater care when underwriting these riskier products to avoid potential put-back risk from investors. The Bureau believes that this is precisely what Congress intended—to create incentives for creditors to engage in sound underwriting and for secondary market investors to monitor the quality of the loans they buy—and that these incentives are particularly warranted with respect to the subprime market.

At the same time, the Bureau does not believe that the potential assignee liability with respect to higher-priced qualified mortgages will preclude such loans from being sold on the secondary market. Specifically, in analyzing impacts on the secondary market the Bureau notes that investors are purchasing higher-priced mortgage loans that are subject to the existing ability-to-repay requirements and presumption of compliance and that the GSEs have already incorporated into their contracts with creditors a representation and warranty designed to provide investor protection in the event of an ability-to-repay violation. The Bureau agrees with industry and secondary market participant commenters that investors will likely require creditors to agree to similar representations and warranties when assigning or selling loans under the new rule because secondary market participants will not want to be held accountable for ability-to-repay compliance which investors will view as the responsibility of the creditor. For prime loans, this may represent an incremental risk of put-back to creditors, given that such loans are not subject to the current regime, but those loans are being provided a safe harbor if they are qualified mortgages. For subprime (higher risk) loans it is not clear that there is any incremental risk beyond that which exists today under the Board's rule. There are also some administrative costs associated with such “put-backs” (e.g., costs associated with the process of putting back loans from the issuer or insurer or servicer on behalf of the securitization trust to the creditor as a result of the ability-to-repay claims), but those costs are unlikely to be material for qualified mortgages subject to the rebuttable presumption and will not affect either the pricing of the loans or the availability of a secondary market for these loans.

In sum, the Bureau has crafted the calibrated presumptions to ensure that these litigation and secondary market impacts do not jeopardize access to credit. With regard to subprime loans, there is some possibility that creditors who are less sophisticated or less able to bear any litigation risk may elect to refrain from engaging in subprime lending, but as discussed below, the Bureau believes that there are sufficient creditors with the capabilities of making responsible subprime loans so as to avoid significant adverse impact on credit availability in that market.

Specific provisions. For the reasons discussed above, in § 1026.43(e)(1), the Bureau is providing a safe harbor and rebuttable presumption with the ability-to-repay requirements for loans that meet the definition of a qualified mortgage. As explained in comment 43(e)(1)-1, § 1026.43(c) requires a creditor to make a reasonable and good faith determination at or before consummation that a consumer will be able to repay a covered transaction. Section 1026.43(e)(1)(i) and (ii) provide a safe harbor and rebuttable presumption of compliance, respectively, with the repayment ability requirements of § 1026.43(c) for creditors and assignees of covered transactions that satisfy the requirements of a qualified mortgage under § 1026.43(e)(2), (e)(4), or (f).

Section 1026.43(e)(1)(i) provides a safe harbor for qualified mortgages that are not higher-priced covered transactions, by stating that a creditor or assignee of a qualified mortgage as defined in § 1026.43(e)(2), (e)(4), or (f) that is not a higher-priced covered transaction, as defined in § 1026.43(b)(4), complies with the repayment ability requirements of § 1026.43(c). Comment 43(e)(1)(i)-1 clarifies that, to qualify for the safe harbor in § 1026.43(e)(1)(i), a covered transaction must meet the requirements of a qualified mortgage in § 1026.43(e)(2), (e)(4), or (f) and must not be a higher-priced covered transaction, as defined in § 1026.43(b)(4).

For qualified mortgages that are higher-priced covered transactions, § 1026.43(e)(1)(ii)(A) provides a rebuttable presumption of compliance with the repayment ability requirements. That section provides that a creditor or assignee of a qualified mortgage as defined in § 1026.43(e)(2), (e)(4), or (f) that is a higher-priced covered transaction, as defined § 1026.43(b)(4), is presumed to comply with the repayment ability requirements of § 1026.43(c). Section 1026.43(e)(1)(ii)(B) provides that to rebut the presumption of compliance, it must be proven that, despite meeting the requirements of § 1026.43(e)(2), (e)(4), or (f), the creditor did not make a reasonable and good faith determination of the consumer's repayment ability at the time of consummation, by showing that the consumer's income, debt obligations, alimony, child support, and the consumer's monthly payment (including mortgage-related obligations) on the covered transaction and on any simultaneous loans of which the creditor was aware at consummation would leave the consumer with insufficient residual income or assets other than the value of the dwelling (including any real property attached to the dwelling) that secures the loan with which to meet living expenses, including any recurring and material non-debt obligations of which the creditor was aware at the time of consummation.

Comment 43(e)(1)(ii)-1 clarifies that a creditor or assignee of a qualified mortgage under § 1026.43(e)(2), (e)(4), or (f) that is a higher-priced covered transaction is presumed to comply with the repayment ability requirements of § 1026.43(c). To rebut the presumption, it must be proven that, despite meeting the standards for a qualified mortgage (including either the debt-to-income standard in § 1026.43(e)(2)(vi) or the standards of one of the entities specified in § 1026.43(e)(4)(ii)), the creditor did not have a reasonable and good faith belief in the consumer's repayment ability. To rebut the presumption, it must be proven that, despite meeting the standards for a qualified mortgage (including either the debt-to-income standard in § 1026.43(e)(2)(vi) or the standards of one of the entities specified in § 1026.43(e)(4)(ii)), the creditor did not have a reasonable and good faith belief in the consumer's repayment ability. Specifically, it must be proven that, at the time of consummation, based on the information available to the creditor, the consumer's income, debt obligations, alimony, child support, and the consumer's monthly payment (including mortgage-related obligations) on the covered transaction and on any simultaneous loans of which the creditor was aware at consummation would leave the consumer with insufficient residual income or assets other than the value of the dwelling (including any real property attached to the dwelling) that secures the loan with which to meet living expenses, including any recurring and material non-debt obligations of which the creditor was aware at the time of consummation, and that the creditor thereby did not make a reasonable and good faith determination of the consumer's repayment ability. The comment also provides, by way of example, that a consumer may rebut the presumption with evidence demonstrating that the consumer's residual income was insufficient to meet living expenses, such as food, clothing, gasoline, and health care, including the payment of recurring medical expenses of which the creditor was aware at the time of consummation, and after taking into account the consumer's assets other than the value of the dwelling securing the loan, such as a savings account. In addition, the longer the period of time that the consumer has demonstrated actual ability to repay the loan by making timely payments, without modification or accommodation, after consummation or, for an adjustable-rate mortgage, after recast, the less likely the consumer will be able to rebut the presumption based on insufficient residual income and prove that, at the time the loan was made, the creditor failed to make a reasonable and good faith determination that the consumer had the reasonable ability to repay the loan.

As noted above, the Bureau believes that the statutory language regarding whether qualified mortgages receive either a safe harbor or rebuttable presumption of compliance is ambiguous, and does not plainly mandate one approach over the other. Furthermore, the Bureau has the authority to tailor the strength of the presumption of compliance based on the characteristics associated with the different types of qualified mortgages. Accordingly, the Bureau interprets TILA section 129C(b)(1) to create a rebuttable presumption, but exercises its adjustment authority under TILA section 105(a) to limit the ability to rebut the presumption in two ways, because an open-ended rebuttable presumption would unduly restrict access to credit without a corresponding benefit to consumers.

First, the Bureau uses its adjustment authority under section 105(a) to limit the ability to rebut the presumption to insufficient residual income or assets other than the dwelling that secures the transaction because the Bureau believes exercise of this authority is necessary and proper to facilitate compliance with and to effectuate a purpose of section 129 and TILA. The Bureau believes this approach, while preserving consumer remedies, provides clear standards to creditors and courts regarding the basis upon which the presumption of compliance that applies to higher-priced covered transactions may be rebutted, thereby enhancing creditor certainty and encouraging lending in the higher-priced mortgage market. The Bureau finds this approach is necessary and proper to ensure that consumers are offered and receive residential mortgage loans on terms that reasonably reflect their ability to repay the loans, a purpose of section 129 and TILA.

Second, with respect to prime loans (loans with an APR that does not exceed APOR by 1.5 percentage points for first liens and 3.5 percentage points for second liens), the Bureau also uses its adjustment authority under TILA section 105(a) to provide a conclusive presumption (e.g., a safe harbor). Under the conclusive presumption, if a prime loan satisfies the criteria for being a qualified mortgage, the loan will be deemed to satisfy section 129C's ability-to-repay criteria and will not be subject to rebuttal based on residual income or otherwise. The Bureau finds that this approach balances the competing consumer protection and access to credit considerations described above. As discussed above, the Bureau will not extend the safe harbor to higher-priced loans because that approach would provide insufficient protection to consumers in loans with higher interest rates who may require greater protection than consumers in prime rate loans. On the other hand, an approach that provided a rebuttable presumption of compliance for all qualified mortgages (including prime loans which historically have a low default rate) could lead creditors to make fewer mortgage loans to certain consumers, which could restrict access to credit (or unduly raise the cost of credit) without a corresponding benefit to consumers. The Bureau finds that this adjustment providing a safe harbor for prime loans is necessary and proper to facilitate compliance with and to effectuate the purposes of section 129C and TILA, including to assure that consumers are offered and receive residential mortgage loans on terms that reasonably reflect their ability to repay the loans.

43(e)(2) Qualified Mortgage Defined—General

As discussed above, TILA section 129C(b)(2) defines the requirements for qualified mortgages to limit certain loan terms and features. The statute generally prohibits a qualified mortgage from permitting an increase of the principal balance on the loan (negative amortization), interest-only payments, balloon payments (except for certain balloon-payment qualified mortgages pursuant to TILA section 129C(b)(2)(E)), a term greater than 30 years, or points and fees that exceed a specified threshold.

In addition, the statute incorporates limited underwriting criteria that overlap with some elements of the general ability-to-repay standard. Specifically, the statutory definition of qualified mortgage requires the creditor to (1) verify and document the income and financial resources relied upon to qualify the obligors on the loan; and (2) underwrite the loan based on a fully amortizing payment schedule and the maximum interest rate during the first five years, taking into account all applicable taxes, insurance, and assessments. As noted above, these requirements appear to be focused primarily on ensuring that certain mortgage products—no-documentation loans and loans underwritten based only on a consumer's ability to make payments during short introductory periods with low “teaser” interest rates—are not eligible to be qualified mortgages.

In addition to these limited underwriting criteria, the statute also authorizes the Bureau to establish additional criteria relating to ratios of total monthly debt to monthly income or alternative measures of ability to pay regular expenses after payment of total monthly debt, taking into account the income levels of the consumer and other factors the Bureau determines relevant and consistent with the purposes described in TILA section 129C(b)(3)(B)(i). To the extent the Bureau incorporates a debt-to-income or residual income requirement into the qualified mortgage definition, several additional elements of the general ability-to-repay standard would effectively also be incorporated into the qualified mortgage definition, since debt-to-income and residual income analyses by their nature require assessment of income, debt (including simultaneous loans), and mortgage-related obligations. As discussed above, the Board proposed two alternatives to implement the qualified mortgage elements. Both alternatives under the Board's proposal would have incorporated the statutory elements of a qualified mortgage (e.g., product feature and loan term restrictions, limits on points and fees, payment calculation requirements, and the requirement to consider and verify the consumer's income or assets). However, Alternative 2 also included the additional factors in the general ability-to-repay standard.

Comments

Qualified mortgage definition. As an initial matter, the majority of commenters generally favored defining qualified mortgages to reach a broad portion of the overall market and to provide clarity with regard to the required elements. Commenters agreed that clarity promotes the benefits of creditors lending with confidence and consumers receiving loans that comply with the basic requirements of an affordable loan. In addition, commenters generally agreed that a qualified mortgage should be broad, encompassing the vast majority of the existing mortgage market. Numerous commenters indicated that creditors believed that the difference between the legal protections afforded (or risks associated with) qualified mortgages and non-qualified mortgages would result in very little lending outside of qualified mortgages. Commenters asserted that a narrowly defined qualified mortgage would leave loans outside the legal protections of qualified mortgages and would result in constrained credit or increased cost of credit.

As discussed in the section-by-section analysis of § 1026.43(e)(1), commenters did not necessarily support the two alternatives specifically as proposed by the Board, but suggested variations on the definition of qualified mortgage that contain some or all of the Board's proposed criteria, or additional criteria not specifically included in either of the Board's proposed alternatives. For example, as described below, a coalition of industry and consumer advocates suggested a tiered approach to defining qualified mortgage, based primarily on meeting a specific back-end debt-to-income requirement, with alternative means of satisfying the qualified mortgage definition (such as housing debt-to-income, reserves, and residual income) if the back-end debt-to-income test is not satisfied. Similarly, one industry commenter suggested using a weighted approach to defining qualified mortgage, which would weight some underwriting factors more heavily than others and permit a significant factor in one area to compensate for a weak or missing factor in another area.

Consumer group commenters and some industry commenters generally supported excluding from the definition of qualified mortgage certain risky loan features which result in “payment shock,” such as negative amortization or interest-only features. Consumer group commenters also supported limiting qualified mortgages to a 30-year term, as required by statute. Consumer group commenters and one industry trade association strongly supported requiring creditors to consider and verify the all the ability-to-repay requirements. These commenters contended that the ability-to-repay requirements represent prudent mortgage underwriting techniques and are essential to sustainable lending. To that point, these commenters argued that qualified mortgage loans should represent the best underwritten and most fully documented loans, which would justify some form of protection from future liability. In addition, several consumer group commenters suggested adding a further requirement that when assessing the consumer's income and determining whether the consumer will be able to meet the monthly payments, a creditor must also take into account other recurring but non-debt related expenses. These commenters argued that many consumers, and especially low- and moderate-income consumers, face significant monthly recurring expenses, such as medical care or prescriptions and child care expense needed to enable the borrower or co-borrower to work outside the home. These commenters further argued that even where the percentage of disposable income in such situations seems reasonable, the nominal amounts left to low- and moderate-income consumers may be insufficient to enable such households to reasonably meet all their obligations. While one consumer group commenter specifically supported the inclusion of a consumer's credit history as an appropriate factor for a creditor to consider and verify when underwriting a loan, several commenters argued that the consumer's credit history should be not included in the ability-to-repay requirements because, although credit history may be relevant in prudent underwriting, it involves a multitude of factors that need to be taken into consideration. In addition, one association of State bank regulators also favored consideration of the repayment factors that are part of a sound underwriting process.

As noted above, some industry commenters also generally supported including the underwriting requirements as proposed in Alternative 2, with some adjustments, so long as the resulting qualified mortgage was entitled to a safe harbor. These commenters stated that most creditors today are already complying with the full ability-to-repay underwriting standards, and strong standards will help them resist competitive forces to lower underwriting standards in the future. Other industry commenters argued that the qualified mortgage criteria should not exclude specific loan products because the result will be that such products will be unavailable in the market.

Some commenters generally supported aligning the definition of qualified mortgage with the definition proposed by several Federal agencies to define “qualified residential mortgages” (QRM) for purposes of the risk retention requirements in title IX of the Dodd-Frank Act. For example, one commenter suggested that the required payment calculation for qualified mortgages be consistent with the QRM proposed requirement that the payment calculation be based on the maximum rate in the first five years after the first full payment required. An association of reverse mortgage lenders requested that a “qualified” reverse mortgage be defined to ensure that the Federal agencies finalizing the QRM rule are able to make a proprietary reverse mortgage a QRM, which would be exempt from the risk retention requirements. Lastly, numerous consumer group commenters argued that high-cost mortgages be excluded from being a qualified mortgage.

Quantitative standards. Some industry commenters supported including quantitative standards for such variables as debt-to-income ratios and credit score with compensating factors in the qualified mortgage definition. These commenters contended that quantitative standards provide certainty and would help ensure creditworthy consumers have access to qualified mortgage loans. One consumer group commenter argued that, without specific quantitative standards, bank examiners and assignees would have no benchmarks against which to measure a creditor's compliance or safety and soundness. One industry commenter favored quantitative standards such as a maximum back-end debt-to-income ratio because that would provide sufficient certainty to creditors and investors. One consumer group commenter supported including quantitative standards for the debt-to-income ratio because, without this, every loan would be open to debate as to whether the consumer had the ability to repay at the time of loan consummation.

As described further below, certain commenters and interested parties requested that the Bureau adopt a specific debt-to-income ratio requirement for qualified mortgages. For example, some suggested that if a consumer's total debt-to-income ratio is below a specified threshold, the mortgage loan should satisfy the qualified mortgage requirements, assuming other relevant conditions are met. In addition to a debt-to-income requirement, some commenters and interested parties suggested that the Bureau should include within the definition of a “qualified mortgage” loans with a debt-to-income ratio above a certain threshold if the consumer has a certain amount of assets, such as money in a savings or similar account, or a certain amount of residual income.

Some industry commenters advocated against including quantitative standards for such variables as debt-to-income ratios and residual income. Those commenters argued that underwriting a loan involves weighing a variety of factors, and creditors and investors should be allowed to exercise discretion and weigh risks for each individual loan. To that point, one industry trade group commenter argued that community banks, for example, generally have conservative requirements for a consumer's debt-to-income ratio, especially for loans that are held in portfolio by the bank, and consider many factors when underwriting mortgage loans, such as payment history, liquid reserves, and other assets. Because several factors are considered and evaluated in the underwriting process, this commenter asserted that community banks can be flexible when underwriting mortgage loans and provide arrangements for certain consumers that fall outside of the normal debt-to-income ratio for a certain loan. This commenter contended that strict quantitative standards would inhibit community banks' relationship lending and ability to use their sound judgment in the lending process. Some commenters contended that requiring specific quantitative standards could restrict credit access and availability for consumers.

Generally, industry commenters and some consumer group commenters believed compensating factors are beneficial in underwriting and should be permitted. These commenters generally believe compensating factors should be incorporated into the qualified mortgage criteria, such as in circumstances when a specified debt-to-income ratio threshold was exceeded. In their view, lending is an individualized decision and compensating factors can, for example, mitigate a consumer's high debt-to-income ratio or low residual income. One industry trade group commenter argued that the inclusion of compensating factors would allow for a broader underwriting approach and should include family history, repayment history, potential income growth, and inter-family transactions. One association of State bank regulators suggested that the rule provide guidance on mitigating factors for creditors to consider when operating outside of standard parameters. For example, creditors lending outside of typical debt-to-income standards can rely upon other assets or the fact that a consumer has a high income. Other industry commenters argued that the rule should provide for enough flexibility to allow for common-sense underwriting and avoid rigid limits or formulas that would exclude consumers on the basis of one or a few underwriting factors.

Another commenter stated that the rule should not set thresholds or limits on repayment ability factors. Instead, the rule should allow the creditor to consider the required factors and be held to a good faith standard. Such a rule permits individualized determinations to be made based on each consumer, local markets, and the risk tolerance of each creditor.

Final Rule

Section 1026.43(e)(2) of the final rule contains the general qualified mortgage definition. As set forth below, the final rule defines qualified mortgages under § 1026.43(e)(2) as loans that satisfy all of the qualified mortgage criteria required by the statute (including underwriting to the maximum interest rate during the first five years of the loan and consideration and verification of the consumer's income or assets), for which the creditor considers and verifies the consumer's current debt obligations, alimony, and child support, and that have a total (“back-end”) monthly debt-to-income ratio of no greater than 43 percent, following the standards for “debt” and “income” set forth in appendix Q.

While the general definition of qualified mortgage in § 1026.43(e)(2) contains all of the statutory qualified mortgage elements, it does not separately incorporate all of the general ability-to-repay underwriting requirements that would have been part of the qualified mortgage definition under the Board's proposed Alternative 2. In particular, the definition of qualified mortgage in § 1026.43(e)(2) does not specifically require consideration of the consumer's employment status, monthly payment on the covered transaction (other than the requirement to underwrite the loan to the maximum rate in the first five years), monthly payment on any simultaneous loans, or the consumer's credit history, which are part of the general ability-to-repay analysis under § 1026.43(c)(2). Instead, most of these requirements are incorporated into the standards for determining “debt” and “income” pursuant to § 1026.43(e)(2)(vi)(A) and (B), to which the creditor must look to determine if the loan meets the 43 percent debt-to-income ratio threshold as required in § 1026.43(e)(2)(vi). In particular, that calculation will require the creditor to verify, among other things, the consumer's employment status (to determine current or expected income) and the monthly payment on the covered transaction (including mortgage-related obligations) and on any simultaneous loans that the creditor knows or has reason to know will be made. In addition, although consideration and verification of a consumer's credit history is not specifically incorporated into the qualified mortgage definition, creditors must verify a consumer's debt obligations using reasonably reliable third-party records, which may include use of a credit report or records that evidence nontraditional credit references. See section-by-section analysis of § 1026.43(e)(2)(v) and (c)(3).

The final rule adopts this approach because the Bureau believes that the statute is fundamentally about assuring that the mortgage credit consumers receive is affordable. Qualified mortgages are intended to be mortgages as to which it can be presumed that the creditor made a reasonable determination of the consumer's ability to repay. Such a presumption would not be reasonable—indeed would be imprudent—if a creditor made a mortgage loan without considering and verifying core aspects of the consumer's individual financial picture, such as income or assets and debt. Incorporating these ability-to-repay underwriting requirements into the qualified mortgage definition thus ensures that creditors assess the consumer's repayment ability for a qualified mortgage using robust and appropriate underwriting procedures. The specific requirements for a qualified mortgage under § 1026.43(e)(2) are described below.

The Bureau notes that the final rule does not define a “qualified” reverse mortgage. As described above, TILA section 129C(a)(8) excludes reverse mortgages from the repayment ability requirements. See section-by-section analysis of § 1026.43(a)(3)(i). However, TILA section 129C(b)(2)(ix) provides that the term “qualified mortgage” may include a “residential mortgage loan” that is “a reverse mortgage which meets the standards for a qualified mortgage, as set by the Bureau in rules that are consistent with the purposes of this subsection.” The Board's proposal did not include reverse mortgages in the definition of a “qualified mortgage.” Because reverse mortgages are exempt from the ability-to-repay requirements, the effects of defining a reverse mortgage as a “qualified mortgage” would be, for example, to allow for certain otherwise banned prepayment penalties and permit reverse mortgages to be QRMs under the Dodd-Frank Act's risk retention rules. The Bureau believes that the first effect is contrary to the purposes of the statute. With respect to the QRM rulemaking, the Bureau will continue to coordinate with the Federal agencies finalizing the QRM rulemaking to determine the appropriate treatment of reverse mortgages.

43(e)(2)(i)

TILA section 129C(b)(2)(A)(i) states that the regular periodic payments of a qualified mortgage may not result in an increase of the principal balance or allow the consumer to defer repayment of principal (except for certain balloon-payment loans made by creditors operating predominantly in rural or underserved areas, discussed below in the section-by-section analysis of § 1026.43(f)). TILA section 129C(b)(2)(A)(ii) states that the terms of a qualified mortgage may not include a balloon payment (subject to an exception for creditors operating predominantly in rural or underserved areas). The statute defines “balloon payment” as “a scheduled payment that is more than twice as large as the average of earlier scheduled payments.” TILA section 129C(b)(2)(A)(ii).

The Board's proposed § 226.43(e)(2)(i) would have implemented TILA sections 129C(b)(2)(A)(i) and (ii). First, the proposed provision would have required that a qualified mortgage provide for regular periodic payments. Second, proposed § 226.43(e)(2)(i) would have provided that the regular periodic payments may not (1) result in an increase of the principal balance; (2) allow the consumer to defer repayment of principal, except as provided in proposed § 226.43(f); or (3) result in a balloon payment, as defined in proposed § 226.18(s)(5)(i), except as provided in proposed § 226.43(f).

Proposed comment 43(e)(2)(i)-1 would have explained that, as a consequence of the foregoing requirements, a qualified mortgage must require the consumer to make payments of principal and interest, on a monthly or other periodic basis, that will fully repay the loan amount over the loan term. These periodic payments must be substantially equal except for the effect that any interest rate change after consummation has on the payment amount in the case of an adjustable-rate or step-rate mortgage. The proposed comment would have also provided that, because proposed § 226.43(e)(2)(i) would have required that a qualified mortgage provide for regular, periodic payments, a single-payment transaction may not be a qualified mortgage. This comment would have clarified a potential evasion of the regulation, as a creditor otherwise could structure a transaction with a single payment due at maturity that technically would not be a balloon payment as defined in proposed § 226.18(s)(5)(i) because it is not more than two times a regular periodic payment.

Proposed comment 43(e)(2)(i)-2 would have provided additional guidance on the requirement in proposed § 226.43(e)(2)(i)(B) that a qualified mortgage may not allow the consumer to defer repayment of principal. The comment would have clarified that, in addition to interest-only terms, deferred principal repayment also occurs if the payment is applied to both accrued interest and principal but the consumer makes periodic payments that are less than the amount that would be required under a payment schedule that has substantially equal payments that fully repay the loan amount over the loan term. Graduated payment mortgages, for example, allow deferral of principal repayment in this manner and therefore may not be qualified mortgages.

As noted above, the Dodd-Frank Act defines “balloon payment” as “a scheduled payment that is more than twice as large as the average of earlier scheduled payments.” However, proposed § 226.43(e)(2)(i)(C) would have cross-referenced Regulation Z's existing definition of “balloon payment” in § 226.18(s)(5)(i), which provides that a balloon payment is “a payment that is more than two times a regular periodic payment.” The Board noted that this definition is substantially similar to the statutory one, except that it uses as its benchmark any regular periodic payment rather than the average of earlier scheduled payments. The Board explained that the difference in wording between the statutory definition and the existing regulatory definition does not yield a significant difference in what constitutes a “balloon payment” in the qualified mortgage context. Specifically, the Board stated its belief that because a qualified mortgage generally must provide for substantially equal, fully amortizing payments of principal and interest, a payment that is greater than twice any one of a loan's regular periodic payments also generally will be greater than twice the average of its earlier scheduled payments.

Accordingly, to facilitate compliance, the Board proposed to cross-reference the existing definition of “balloon payment.” The Board proposed this adjustment to the statutory definition pursuant to its authority under TILA section 105(a) to make such adjustments for all or any class of transactions as in the judgment of the Board are necessary or proper to facilitate compliance with TILA. The Board stated that this approach is further supported by its authority under TILA section 129B(e) to condition terms, acts or practices relating to residential mortgage loans that the Board finds necessary or proper to facilitate compliance.

Finally, in the preamble to the Board's proposal, the Board noted that some balloon-payment loans are renewable at maturity and that such loans might appropriately be eligible to be qualified mortgages, provided the terms for renewal eliminate the risk of the consumer facing a large, unaffordable payment obligation, which underlies the rationale for generally excluding balloon-payment loans from the definition of qualified mortgages. If the consumer is protected by the terms of the transaction from that risk, the Board stated that such a transaction might appropriately be treated as though it effectively is not a balloon-payment loan even if it is technically structured as one. The Board solicited comment on whether it should include an exception providing that, notwithstanding proposed § 226.43(e)(2)(i)(C), a qualified mortgage may provide for a balloon payment if the creditor is unconditionally obligated to renew the loan at the consumer's option (or is obligated to renew subject to conditions within the consumer's control). The Board sought comment on how such an exception should be structured to ensure that the large-payment risk ordinarily accompanying a balloon-payment loan is fully eliminated by the renewal terms and on how such an exception might be structured to avoid the potential for circumvention.

As discussed above, commenters generally supported excluding from the definition of qualified mortgage certain risky loan features which result in “payment shock,” such as negative amortization or interest-only features. Commenters generally recognized such features as significant contributors to the recent housing crisis. Industry commenters noted that such restrictions are objective criteria which creditors can conclusively demonstrate were met at the time of origination. However, one mortgage company asserted that such limitations should not apply in loss mitigation transactions, such as loan modifications and extensions, or to loan assumptions. That commenter noted that while negative amortization is not common in most loan modification programs, the feature can be used at times to help consumers work through default situations. The commenter also noted that deferral of payments, including principal payments, and balloon payment structures are commonly used to relieve payment default burdens. One bank commenter argued that the rule should permit qualified mortgages to have balloon payment features if the creditor is unconditionally obligated to renew the loan at the consumer's option, or is obligated to renew subject to conditions within the consumer's control.

For the reasons discussed in the proposed rule, the Bureau is adopting § 226.43(e)(2)(i) as proposed in renumbered § 1026.43(e)(2)(i), with certain clarifying changes. In particular, in addition to the proposed language, section 1026.43(e)(2)(i) specifies that a qualified mortgage is a covered transaction that provides for regular periodic payments that are substantially equal, “except for the effect that any interest rate change after consummation has on the payment in the case of an adjustable-rate or step-rate mortgage.” This language appeared in the commentary to § 226.43(e)(2)(i) in the proposed rule, but to provide clarity, the Bureau is adopting this language in the text of § 1026.43(e)(2)(i) in the final rule.

Notably, the Bureau is adopting in § 1026.43(e)(2)(i) the proposed cross-reference to the existing Regulation Z definition of balloon payment. Like the Board, the Bureau finds that the statutory definition and the existing regulatory definition do not yield a significant difference in what constitutes a “balloon payment” in the qualified mortgage context. Accordingly, the Bureau makes this adjustment pursuant to its authority under TILA section 105(a) because the Bureau believes that affording creditors a single definition of balloon payment within Regulation Z is necessary and proper to facilitate compliance with and effectuate the purposes of TILA.

In addition, like the proposal, the final rule does not provide exceptions from the prohibition on qualified mortgages providing for balloon payments, other than the exception for creditors operating predominantly in rural or underserved areas, described below in the section-by-section analysis of § 1026.43(f). The Bureau believes that it is appropriate to implement the rule consistent with statutory intent, which specifies only a narrow exception from this general rule for creditors operating predominantly in rural or underserved areas rather than a broader exception to the general prohibition on qualified mortgages containing balloon payment features. With respect to renewable balloon-payment loans, the Bureau does not believe that the risk that a consumer will face a significant payment shock from the balloon feature can be fully eliminated, and that a rule that attempts to provide such special treatment for renewable balloon-payment loans would be subject to abuse.

43(e)(2)(ii)

TILA section 129C(b)(2)(A)(viii) requires that a qualified mortgage must not provide for a loan term that exceeds 30 years, “except as such term may be extended under paragraph (3), such as in high-cost areas.” As discussed above, TILA section 129C(b)(3)(B)(i) authorizes the Bureau to revise, add to, or subtract from the qualified mortgage criteria if the Bureau makes certain findings, including that such revision is necessary or proper to ensure that responsible, affordable mortgage credit remains available to consumers in a manner consistent with the purposes of TILA section 129C(b) or necessary and appropriate to effectuate the purposes of section 129C.

Proposed § 226.43(e)(2)(ii) would have implemented the 30-year maximum loan term requirement in the statute without exception. The preamble to the proposed rule explains that, based on available information, the Board believed that mortgage loans with terms greater than 30 years are rare and, when made, generally are for the convenience of consumers who could qualify for a loan with a 30-year term but prefer to spread out their payments further. Therefore, the Board believed such an exception is generally unnecessary. The Board solicited comment on whether there are any “high-cost areas” in which loan terms in excess of 30 years are necessary to ensure that responsible, affordable credit is available and, if so, how they should be identified for purposes of such an exception. The Board also sought comment on whether any other exceptions would be appropriate, consistent with the Board's authority in TILA section 129C(b)(3)(B)(i).

As noted above, commenters generally supported the 30-year term limitation. One commenter suggested the final rule should clarify that a loan term that is slightly longer than 30 years because of the due date of the first regular payment nevertheless meets the 30-year term requirement. One trade association commenter suggested that creditors be provided flexibility to originate 40-year loans in order to accommodate consumers in regions of the country where housing prices are especially high, but did not provide any information regarding the historic performance of 40-year loans or discuss how the Bureau should define high-cost areas in a way that avoids abuse. An association of State bank regulators also suggested that the rule permit loan terms beyond 30 years in high-cost areas and suggested that those areas could be determined based on housing price indices. That commenter, two large industry trade associations, and one mortgage company commenter argued that the 30-year term limitation should not apply to loan modifications that provide a consumer with a loan with a lower monthly payment than he or she may otherwise face. One such commenter noted that, as a general matter, the rule should clarify that modifications of existing loans should not be subject to the same ability-to-repay requirements to avoid depriving consumers of beneficial modifications.

For the reasons discussed in the proposed rule, the Bureau is generally adopting § 226.43(e)(2)(ii) as proposed in renumbered § 1026.43(e)(2)(ii). In response to commenter concern, the final rule clarifies in comment 43(e)(2)(ii)-1 that the 30-year term limitation in § 1026.43(e)(2)(ii) is applied without regard to any interim period between consummation and the beginning of the first full unit period of the repayment schedule. Consistent with the Board's analysis, the final rule does not provide exceptions to the 30-year loan limitation. Like the Board, the Bureau is unaware of a basis upon which to conclude that an exception to the 30-year loan term limitation for qualified mortgages in high-cost areas is appropriate. In particular, the Bureau believes that loans with terms greater than 30 years are rare and that, when made, generally are for the convenience of consumers who could qualify for a loan with a 30-year term.

The final rule also does not provide additional guidance on the 30-year loan term limitation in the context of loan modifications. The Bureau understands that private creditors may offer loan modifications to consumers at risk of default or foreclosure, and that such modifications may extend the duration of the loan beyond the initial term. If such modification results in the satisfaction and replacement of the original obligation, the loan would be a refinance under current § 1026.20(a), and therefore the new transaction must comply with the ability-to-repay requirements of § 1026.43(c) or satisfy the criteria for a qualified mortgage, independent of any ability-to-repay analysis or the qualified mortgage status of the initial transaction. However, if the transaction does not meet the criteria in 1026.20(a), which determines a refinancing—generally resulting in the satisfaction and replacement of the original obligation—the loan would not be a refinance under § 1026.20(a), and would instead be an extension of the original loan. In such a case, compliance with the ability-to-repay provision, including a loan's qualified mortgage status, would be determined as of the date of consummation of the initial transaction, regardless of a later modification.

43(e)(2)(iii)

TILA section 129C(b)(2)(A)(vii) defines a qualified mortgage as a loan for which, among other things, the total points and fees payable in connection with the loan do not exceed three percent of the total loan amount. TILA section 129C(b)(2)(D) requires the Bureau to prescribe rules adjusting this threshold to “permit lenders that extend smaller loans to meet the requirements of the presumption of compliance.” The statute further requires the Bureau, in prescribing such rules, to “consider the potential impact of such rules on rural areas and other areas where home values are lower.”

Proposed § 226.43(e)(2)(iii) would have implemented these provisions by providing that a qualified mortgage is a loan for which the total points and fees payable in connection with the loan do not exceed the amounts specified under proposed § 226.43(e)(3). As discussed in detail in the section-by-section analysis of § 1026.43(e)(3), the Board proposed two alternatives for calculating the allowable points and fees for a qualified mortgage: One approach would have consisted of five “tiers” of loan sizes and corresponding limits on points and fees, while the other approach would have consisted of three “tiers” of points and fees based on a formula yielding a greater allowable percentage of the total loan amount to be charged in points and fees for each dollar increase in loan size. Additionally, proposed § 226.43(b)(9) would have defined “points and fees” to have the same meaning as in proposed § 226.32(b)(1).

For the reasons discussed in the proposed rule, the Bureau is generally adopting § 226.43(e)(2)(iii) as proposed in renumbered § 1026.43(e)(2)(iii). For a discussion of the final rule's approach to calculating allowable points and fees for a qualified mortgage, see the section-by-section analysis of § 1026.43(e)(3). For a discussion of the definition of points and fees, see the section-by-section analysis of § 1026.32(b)(1).

As noted above, several consumer group commenters requested that high-cost mortgages be prohibited from receiving qualified mortgage status. Those commenters noted that high-cost mortgages have been singled out by Congress as deserving of special regulatory treatment because of their potential to be abusive to consumers. They argue that it would seem incongruous for any high-cost mortgage to be given a presumption of compliance with the ability-to-repay rule. However, the final rule does not prohibit a high-cost mortgage from being a qualified mortgage. Under the Dodd-Frank Act, a mortgage loan is a high-cost mortgage when (1) the annual percentage rate exceeds APOR by more than 6.5 percentage points for first-liens or 8.5 percentage points for subordinate-liens; (2) points and fees exceed 5 percent, generally; or (3) when prepayment penalties may be imposed more than three years after consummation or exceed 2 percent of the amount prepaid. Neither the Board's 2011 ATR-QM Proposal nor the Bureau's 2012 HOEPA Proposal would have prohibited loans that are high-cost mortgages as a result of a high interest rate from receiving qualified mortgage status.

As a general matter, the ability-to-repay requirements in this final rule apply to most closed-end mortgage loans, including closed-end high-cost mortgages. Notwithstanding the Dodd-Frank Act's creation of a new ability-to-repay regime for mortgage loans, Congress did not modify an existing prohibition in TILA section 129(h) against originating a high-cost mortgage without regard to a consumer's repayment ability (HOEPA ability-to-repay). Thus, under TILA (as amended by the Dodd-Frank Act), closed-end high-cost mortgages are subject both to the general ability-to-repay provisions and to HOEPA's ability-to-repay requirement. As implemented in existing § 1026.34(a)(4), the HOEPA ability-to-repay rules contain a rebuttable presumption of compliance if the creditor takes certain steps that are generally less rigorous than the Dodd-Frank Act's ability-to-repay requirements, as implemented in this rule. For this reason, and as explained further in that rulemaking, the Bureau's 2013 HOEPA Final Rule provides that a creditor complies with the high-cost mortgage ability-to-repay requirement by complying with the general ability-to-repay provision, as implemented by this final rule.

The final rule does not prohibit high-cost mortgages from being qualified mortgages for several reasons. First, the Dodd-Frank Act does not prohibit high-cost mortgages from receiving qualified mortgage status. While the statute imposes a points and fees limit on qualified mortgages (3 percent, generally) that effectively prohibits loans that trigger the high-cost mortgage points and fee threshold from receiving qualified mortgage status, it does not impose an annual percentage rate limit on qualified mortgages. Therefore, nothing in the statute prohibits a creditor from making a loan with a very high interest rate such that the loan is a high-cost mortgage while still meeting the criteria for a qualified mortgage.

In addition, the final rule does not prohibit high-cost mortgages from being qualified mortgages because the Bureau believes that, for loans that meet the qualified mortgage definition, there is reason to presume, subject to rebuttal, that the creditor had a reasonable and good faith belief in the consumer's ability to repay notwithstanding the high interest rate. High-cost mortgages will be less likely to meet qualified mortgage criteria because the higher interest rate will generate higher monthly payments and thus require higher income to satisfy the debt-to-income test for a qualified mortgage. But where that test is satisfied—that is, where the consumer has an acceptable debt-to-income ratio calculated in accordance with qualified mortgage underwriting rules—there is no logical reason to exclude the loan from the definition of a qualified mortgage.

Allowing a high-cost mortgage to be a qualified mortgage can benefit consumers. The Bureau anticipates that, in the small loan market, creditors may sometimes exceed high-cost mortgage thresholds due to the unique structure of their business. The Bureau believes it would be in the interest of consumers to afford qualified mortgage status to loans meeting the qualified mortgage criteria so as to remove any incremental impediment that the general ability-to-repay provisions would impose on making such loans. The Bureau also believes this approach could provide an incentive to creditors making high-cost mortgages to satisfy the qualified mortgage requirements, which would provide additional consumer protections, such as restricting interest-only payments and limiting loan terms to 30 years, which are not requirements under HOEPA.

Furthermore, allowing high-cost mortgage loans to be qualified mortgages would not impact the various impediments to making high-cost mortgage loans, including enhanced disclosure and counseling requirements and the enhanced liability for HOEPA violations. Thus, there would remain strong disincentives to making high-cost mortgages. The Bureau does not believe that allowing high-cost mortgages to be qualified mortgages would incent creditors who would not otherwise make high-cost mortgages to start making them.

43(e)(2)(iv)

TILA section 129C(b)(2)(A)(iv) and (v) provides as a condition to meeting the definition of a qualified mortgage, in addition to other criteria, that the underwriting process for a fixed-rate or adjustable-rate loan be based on “a payment schedule that fully amortizes the loan over the loan term and takes into account all applicable taxes, insurance, and assessments.” The statute further states that for an adjustable-rate loan, the underwriting must be based on “the maximum rate permitted under the loan during the first 5 years.” See TILA section 129C(b)(2)(A)(v). The statute does not define the terms “fixed rate,” “adjustable-rate,” or “loan term,” and provides no additional assumptions regarding how to calculate the payment obligation.

These statutory requirements differ from the payment calculation requirements set forth in existing § 1026.34(a)(4)(iii), which provides a presumption of compliance with the repayment ability requirements for higher-priced mortgage loans, where the creditor underwrites the loan using the largest payment of principal and interest scheduled in the first seven years following consummation. The existing presumption of compliance under § 1026.34(a)(4)(iii) is available for all high-cost and higher-priced mortgage loans, except for loans with negative amortization or balloon-payment mortgages with a term less than seven years. In contrast, TILA section 129C(b)(2)(A) requires the creditor to underwrite the loan based on the maximum payment during the first five years, and does not extend the scope of qualified mortgages to any loan that contains certain risky features or a loan term exceeding 30 years. Loans with a balloon-payment feature would not meet the definition of a qualified mortgage regardless of term length, unless made by a creditor that satisfies the conditions in § 1026.43(f).

The Board proposed to implement the underwriting requirements of TILA section 129C(b)(2)(A)(iv) and (v), for purposes of determining whether a loan meets the definition of a qualified mortgage, in proposed § 226.43(e)(2)(iv). Under the proposal, creditors would have been required to underwrite a loan that is a fixed-, adjustable-, or step-rate mortgage using a periodic payment of principal and interest based on the maximum interest rate permitted during the first five years after consummation. The terms “adjustable-rate mortgage,” “step-rate mortgage,” and “fixed-rate mortgage” would have had the meaning as in current § 1026.18(s)(7)(i) through (iii), respectively.

Specifically, proposed § 226.43(e)(2)(iv) would have provided that meeting the definition of a qualified mortgage is contingent, in part, on creditors meeting the following underwriting requirements:

(1) Proposed § 226.43(e)(2)(iv) would have required that the creditor take into account any mortgage-related obligations when underwriting the consumer's loan;

(2) Proposed § 226.43(e)(2)(iv)(A) would have required the creditor to use the maximum interest rate that may apply during the first five years after consummation; and

(3) Proposed § 226.43(e)(2)(iv)(B) would have required that the periodic payments of principal and interest repay either the outstanding principal balance over the remaining term of the loan as of the date the interest rate adjusts to the maximum interest rate that can occur during the first five years after consummation, or the loan amount over the loan term.

These three underwriting conditions under proposed § 226.43(e)(2)(iv), and the approach to these criteria adopted in the final rule, are discussed below.

Proposed § 226.43(e)(2)(iv) would have implemented TILA section 129C(b)(2)(A)(iv) and (v), in part, and provided that, to be a qualified mortgage under proposed § 1026.43(e)(2), the creditor must underwrite the loan taking into account any mortgage-related obligations. Proposed comment 43(e)(2)(iv)-6 would have provided cross-references to proposed § 226.43(b)(8) and associated commentary. The Board proposed to use the term “mortgage-related obligations” in place of “all applicable taxes, insurance (including mortgage guarantee insurance), and assessments.” Proposed § 226.43(b)(8) would have defined the term “mortgage-related obligations” to mean property taxes; mortgage-related insurance premiums required by the creditor as set forth in proposed § 226.45(b)(1); homeowners association, condominium, and cooperative fees; ground rent or leasehold payments; and special assessments.

Commenters generally supported the inclusion of mortgage-related obligations in the underwriting requirement in proposed § 226.43(e)(2)(iv). Several industry trade associations, banks, civil rights organizations, and consumer advocacy groups specifically supported the requirement. Several commenters requested clear guidance on the amounts to be included in the monthly payment amount, including mortgage-related obligations. In addition, a civil rights organization and several consumer advocacy groups argued that the creditor should also be required to consider recurring, non-debt expenses, such as medical supplies and child care.

As discussed above in the section-by-section analysis of § 1026.43(b)(8), the Bureau is adopting the proposed definition of mortgage-related obligations in renumbered § 1026.43(b)(8), with certain clarifying changes and additional examples.

For the reasons discussed above, the Bureau is adopting the mortgage-related obligations portion of § 226.43(e)(2)(vi) as proposed in renumbered § 1026.43(e)(2)(vi). The final rule does not contain a specific requirement that the creditor consider, when underwriting the consumer's monthly payment, recurring non-debt expenses, such as medical supplies and child care. However, such expenses, if known to the creditor at the time of consummation, may be relevant to a consumer's ability to rebut the presumption of compliance that applies to qualified mortgages that are higher-priced covered transactions. See section-by-section analysis of § 1026.43(e)(1)(ii)(B).

43(e)(2)(iv)(A)

Proposed § 226.43(e)(2)(iv)(A) would have implemented TILA section 129C(b)(2)(A)(iv) and (v), in part, and provided that, to be a qualified mortgage under proposed § 1026.43(e)(2), the creditor must underwrite the loan using the maximum interest rate that may apply during the first five years after consummation. However, the statute does not define the term “maximum rate,” nor does the statute clarify whether the phrase “the maximum rate permitted under the loan during the first 5 years” means the creditor should use the maximum interest rate that occurs during the first five years of the loan beginning with the first periodic payment due under the loan, or during the first five years after consummation of the loan. The former approach would capture the rate recast for a 5/1 hybrid adjustable-rate mortgage that occurs on the due date of the 60th monthly payment, and the latter would not.

The Board interpreted the phrase “maximum rate permitted” as requiring creditors to underwrite the loan based on the maximum interest rate that could occur under the terms of the loan during the first five years after consummation, assuming a rising index value. See proposed comment 43(e)(2)(iv)-1. The Board noted that this interpretation is consistent with current guidance contained in Regulation Z regarding disclosure of the maximum interest rate. See MDIA Interim Rule, 75 FR 58471 (Sept. 24, 2010). The Board further stated that this interpretation is consistent with congressional intent to encourage creditors to make loans to consumers that are less risky and that afford the consumer a reasonable period of time to repay (i.e., 5 years) on less risky terms. For the reasons described in the proposed rule, the Bureau is adopting the “maximum interest rate” provision in § 1026.43(e)(2)(iv) as proposed in renumbered § 1026.43(e)(2)(iv).

The Board proposed to interpret the phrase “during the first 5 years” as requiring creditors to underwrite the loan based on the maximum interest rate that may apply during the first five years after consummation. TILA section 129C(b)(2)(A)(v). The preamble to the proposed rule explains several reasons for this interpretation. First, the Board noted that a plain reading of the statutory language conveys that the “first five years” is the first five years of the loan once it comes into existence (i.e., once it is consummated). The Board believed that interpreting the phrase to mean the first five years beginning with the first periodic payment due under the loan would require an expansive reading of the statutory text.

Second, the Board noted that the intent of this underwriting condition is to ensure that the consumer can afford the loan's payments for a reasonable amount of time and that Congress intended for a reasonable amount of time to be the first five years after consummation.

Third, the Board proposed this approach because it is consistent with prior iterations of this statutory text and the Board's 2008 HOEPA Final Rule. As noted above, the Dodd-Frank Act codifies many aspects of the repayment ability requirements contained in existing § 1026.34(a)(4) of the Board's 2008 HOEPA Final Rule.

Fourth, the Board believed that interpreting the phrase “during the first five years” as including the rate adjustment at the end of the fifth year would be of limited benefit to consumers because creditors could easily structure their product offerings to avoid application of the rule. For example, a creditor could move a rate adjustment that typically occurs on the due date of the 60th monthly payment to due date of the first month that falls outside the specified time horizon, making any proposal to extend the time period in order to include the rate adjustment of diminished value.

Finally, the Board believed that the proposed timing of the five-year period could appropriately differ from the approach used under the 2010 MDIA Interim Final Rule, given the different purposes of the rules. The Board amended the 2010 MDIA Interim Final Rule to require that creditors base their interest rate and payment disclosures on the first five years after the due date of the first regular periodic payment rather than the first five years after consummation. See 75 FR 81836, 81839 (Dec. 29, 2010). The revision clarified that the disclosure requirements for 5/1 hybrid adjustable-rate mortgages must include the rate adjustment that occurs on the due date of the 60th monthly payment, which typically occurs more than five years after consummation. The disclosure requirements under the 2010 MDIA Interim Final Rule, as revised, are intended to help make consumers aware of changes to their loan terms that may occur if they choose to stay in the loan beyond five years and therefore, helps to ensure consumers avoid the uninformed use of credit. The Board believed a different approach is appropriate under proposed § 226.43(e)(2)(iv) because that requirement seeks to ensure that the loan's payments are affordable for a reasonable period of time. For the reasons stated above, the Board believed that Congress intended the first five years after consummation to be a reasonable period of time to ensure that the consumer has the ability to repay the loan according to its terms.

For all the above-listed reasons, the Board interpreted the statutory text as requiring that the creditor underwrite the loan using the maximum interest rate during the first five years after consummation. The Board solicited comment on its interpretation of the phrase “first five years” and the appropriateness of this approach. The Board also proposed clarifying commentary and examples, which are described below.

As described above, commenters generally supported the payment calculation requirements in the proposed rule, including the five-year payment calculation. A comment from a coalition of consumer advocates suggested that the period may not be long enough to assure a consumer's ability to repay given that the average homeowner holds their mortgage for approximately seven years, and suggested that the five-year payment calculation requirement be extended to reflect the average mortgage duration of the first ten years of the loan. Two industry commenters suggested that the time horizon in the required payment calculation for qualified mortgages be consistent with the proposed requirement in the 2011 QRM Proposed Rule that the payment calculation be based on the maximum rate in the first five years after the date on which the first regular periodic payment will be due. One such commenter noted that the payment calculation approach in the 2011 QRM Proposed Rule is more protective of consumers. Another industry commenter suggested that the final rule should measure the first five years from the first regularly scheduled payment, for consistency with the 2010 MDIA Interim Final Rule. An association of State bank regulators agreed with the Board's reasoning, noting that creditors could structure loans to recast outside any parameter set by the rule and that an effective way to prevent purposeful evasion of the payment calculation provision would require legislation.

Notwithstanding the Board's proposed approach, the Bureau interprets the phrase “during the first 5 years” as requiring creditors to underwrite the loan based on the maximum interest rate that may apply during the first five years after the first regular periodic payment will be due. Like the Board, the Bureau finds the statutory language to be ambiguous. However, the Bureau believes that the statutory phrase “during the first 5 years” could be given either meaning, and that this approach provides greater protections to consumers by requiring creditors to underwrite qualified mortgages using the rate that would apply after the recast of a five-year adjustable rate mortgage. Further, as noted, this approach is consistent with the payment calculation in the 2011 QRM Proposed Rule and in existing Regulation Z with respect to the disclosure requirements for interest rates on adjustable-rate amortizing loans.

Accordingly, § 1026.43(e)(2)(iv)(A) provides that a qualified mortgage under § 1026.43(e)(2) must be underwritten, taking into account any mortgage-related obligations, using the maximum interest rate that may apply during the first five years after the date on which the first regular periodic payment will be due. Although the Bureau is finalizing the commentary and examples to § 226.43(e)(2)(iv) as proposed in the commentary to renumbered § 1026.43(e)(2)(iv), the final rule makes conforming changes to the proposed commentary to reflect the adjusted time horizon. The proposed commentary and the changes to the proposed commentary as implemented in the final rule are described below.

The Bureau is finalizing comment 43(e)(2)(iv)-1 as proposed, but with conforming changes to reflect the new time horizon. In the final rule, the comment provides guidance to creditors on how to determine the maximum interest rate during the first five years after the date on which the first regular periodic payment will be due. This comment explains that creditors must use the maximum rate that could apply at any time during the first five years after the date on which the first regular periodic payment will be due, regardless of whether the maximum rate is reached at the first or subsequent adjustment during such five year period.

The Bureau is finalizing comment 43(e)(2)(iv)(A)-2 as proposed. That comment clarifies that for a fixed-rate mortgage, creditors should use the interest rate in effect at consummation, and provides a cross-reference to § 1026.18(s)(7)(iii) for the meaning of the term “fixed-rate mortgage.”

The Bureau is finalizing comment 43(e)(2)(iv)-3 as proposed, but with conforming changes to reflect the new time horizon. That comment provides guidance to creditors regarding treatment of periodic interest rate adjustment caps, and explains that, for an adjustable-rate mortgage, creditors should assume the interest rate increases after consummation as rapidly as possible, taking into account the terms of the legal obligation. The comment further explains that creditors should account for any periodic interest rate adjustment cap that may limit how quickly the interest rate can increase under the terms of the legal obligation. The comment states that where a range for the maximum interest rate during the first five years is provided, the highest rate in that range is the maximum interest rate for purposes of this section. Finally, the comment clarifies that where the terms of the legal obligation are not based on an index plus a margin, or formula, the creditor must use the maximum interest rate that occurs during the first five years after the date on which the first regular periodic payment will be due.

The Bureau is also adopting comment 43(e)(2)(iv)-3.i through .iii as proposed, but with conforming changes to the comment to reflect the new time horizon. Those comments provide examples of how to determine the maximum interest rate. For example, comment 43(e)(2)(iv)-3.1 illustrates how to determine the maximum interest rate in the first five years after the date on which the first regular periodic payment will be due for an adjustable-rate mortgage with a discounted rate for three years.

The Bureau is also finalizing comment 43(e)(2)(iv)-4 as proposed, but with conforming changes to reflect the new time horizon. Comment 43(e)(2)(iv)-4 clarifies the meaning of the phrase “first five years after the date on which the first regular periodic payment will be due.” This comment provides that under § 1026.43(e)(2)(iv)(A), the creditor must underwrite the loan using the maximum interest rate that may apply during the first five years after the date on which the first regular periodic payment will be due, and provides an illustrative example.

43(e)(2)(iv)(B)

Proposed § 226.43(e)(2)(iv)(B) would have implemented TILA section 129C(b)(2)(A)(iv) and (v), in part, by providing, as part of meeting the definition of a qualified mortgage under proposed § 1026.43(e)(2), that the creditor underwrite the loan using periodic payments of principal and interest that will repay either (1) the outstanding principal balance over the remaining term of the loan as of the date the interest rate adjusts to the maximum interest rate that occurs during the first five years after consummation; or (2) the loan amount over the loan term. See proposed § 226.43(e)(2)(iv)(B)(1) and (2).

TILA section 129C(b)(2)(A)(iv) and (v) states that underwriting should be based “on a payment schedule that fully amortizes the loan over the loan term.” The Board noted that unlike the payment calculation assumptions set forth for purposes of the general ability-to-repay rule, under TILA section 129C(a)(6), the underwriting conditions for purposes of meeting the definition of a qualified mortgage do not specify the loan amount that should be repaid, and do not define “loan term.” For consistency and to facilitate compliance, the Board proposed to use the terms “loan amount” and “loan term” in proposed § 226.43(b)(5) and (b), respectively, for purposes of this underwriting condition.

However, the Board also believed that a loan that meets the definition of a qualified mortgage and which has the benefit of other safeguards, such as limits on loan features and fees, merits flexibility in the underwriting process. Accordingly, the Board proposed to permit creditors to underwrite the loan using periodic payments of principal and interest that will repay either the outstanding principal balance as of the date the maximum interest rate during the first five years after consummation takes effect under the terms of the loan, or the loan amount as of the date of consummation. The Board believed the former approach more accurately reflects the largest payment amount that the consumer would need to make under the terms of the loan during the first five years after consummation, whereas the latter approach would actually overstate the payment amounts required. This approach would have set a minimum standard for qualified mortgages, while affording creditors latitude to choose either approach to facilitate compliance.

For the reasons described in the proposed rule, the Bureau is finalizing § 226.43(e)(2)(iv)(A) as proposed in renumbered § 1026.43(e)(2)(iv)(A). However, the final rule makes conforming changes to the proposed commentary to reflect the adjusted time-horizon to the first five years after the due date of the first regular periodic payment. The proposed commentary and the changes to the proposed commentary in the final rule are described below.

The Bureau is finalizing comment 43(e)(2)(iv)-5 as proposed, but with conforming changes to reflect the new time horizon. Comment 43(e)(2)(iv)-5 provides further clarification to creditors regarding the loan amount to be used for purposes of this second condition in § 1026.43(e)(2)(iv). The comment explains that for a creditor to meet the definition of a qualified mortgage under § 1026.43(e)(2), the creditor must determine the periodic payment of principal and interest using the maximum interest rate permitted during the first five years after the date on which the first regular periodic payment will be due that repays either (1) the outstanding principal balance as of the earliest date the maximum interest rate can take effect under the terms of the legal obligation, over the remaining term of the loan, or (2) the loan amount, as that term is defined in § 1026.43(b)(5), over the entire loan term, as that term is defined in § 1026.43(b)(6). This comment provides illustrative examples for both approaches.

The Bureau is finalizing comment 43(c)(2)(iv)-6 as proposed. That comment reiterates that § 1026.43(e)(2)(iv) requires creditors to take mortgage-related obligations into account when underwriting the loan and refers to § 1026.43(b)(8) and its associated commentary for the meaning of mortgage-related obligations.

The Bureau is also finalizing comment 43(e)(2)(iv)-7 as proposed, but with conforming changes to reflect the new time horizon. Comment 43(e)(2)(iv)-7 provides examples of how to determine the periodic payment of principal and interest based on the maximum interest rate during the first five years after the date on which the first regular periodic payment will be due under § 1026.43(e)(2)(iv). The final rule provides an additional example of how to determine the periodic payment of principal and interest based on the maximum interest rate during the first five years after the date on which the first regular periodic payment will be due under § 1026.43(e)(2)(iv) for an adjustable-rate mortgage with a discount of seven years, to illustrate how the payment calculation applies in a loan that adjusts after the five-year time horizon. Comment 43(e)(2)(iv)-7.iv provides an example of a loan in an amount of $200,000 with a 30-year loan term, that provides for a discounted interest rate of 6 percent that is fixed for an initial period of seven years, after which the interest rate will adjust annually based on a specified index plus a margin of 3 percent, subject to a 2 percent annual interest rate adjustment cap. The index value in effect at consummation is 4.5 percent. The loan is consummated on March 15, 2014, and the first regular periodic payment is due May 1, 2014. Under the terms of the loan agreement, the first rate adjustment is on April 1, 2021 (the due date of the 84th monthly payment), which occurs more than five years after the date on which the first regular periodic payment will be due. Thus, the maximum interest rate under the terms of the loan during the first five years after the date on which the first regular periodic payment will be due is 6 percent. Under this example, the transaction will meet the definition of a qualified mortgage if the creditor underwrites the loan using the monthly payment of principal and interest of $1,199 to repay the loan amount of $200,000 over the 30-year loan term using the maximum interest rate during the first five years after the date on which the first regular periodic payment will be due of 6 percent.

43(e)(2)(v)

43(e)(2)(v)(A)

TILA section 129C(b)(2)(A)(iii) provides that a condition for meeting the requirements of a qualified mortgage is that the income and financial resources relied upon to qualify the obligors on the residential mortgage loan are verified and documented. This requirement is consistent with requirement under the general ability-to-repay standard to consider and verify a consumer's income or assets using third-party records, pursuant to TILA section 129C(a)(1) and (3), as discussed above in the section-by-section analysis of § 1026.43(c)(2)(i) and (c)(4).

Proposed § 226.43(e)(2)(v) would have implemented TILA section 129C(b)(2)(A)(iii) by providing that for a covered transaction to be a qualified mortgage, the creditor must consider and verify the consumer's current or reasonably expected income or assets to determine the consumer's repayment ability, as required by proposed § 226.43(c)(2)(i) and (c)(4). The proposal used the term “assets” instead of “financial resources” for consistency with other provisions in Regulation Z and, as noted above, the Bureau believes that the terms have the same meaning. Proposed comment 43(e)(2)(v)-1 would have clarified that creditors may rely on commentary to proposed § 226.43(c)(2)(i), (c)(3) and (c)(4) for guidance regarding considering and verifying the consumer's income or assets to satisfy the conditions for a qualified mortgage under proposed § 226.43(e)(2)(v).

For the reasons discussed in the proposal, the Bureau is finalizing § 226.43(e)(2)(v)(A) as proposed in renumbered § 1026.43(e)(2)(v)(A), with additional clarification that the value of the dwelling includes any real property to which the dwelling is attached. Renumbered § 1026.43(e)(2)(v)(A) also provides that the creditor must consider and verify the consumer's current or reasonably expected income or assets other than the value of the dwelling (including any real property attached to the dwelling) that secures the loan, in accordance with appendix Q, in addition to § 1026.43(c)(2)(i) and (c)(4). Comment 43(e)(2)(v)-2 clarifies this provision, by explaining that, for purposes of this requirement, the creditor must consider and verify, at a minimum, any income specified in appendix Q. A creditor may also consider and verify any other income in accordance with § 1026.43(c)(2)(i) and (c)(4); however, such income would not be included in the total monthly debt-to-income ratio determination by § 1026.43(e)(2)(vi). As described below, appendix Q contains specific standards for defining “income,” to provide certainty to creditors as to whether a loan meets the requirements for a qualified mortgage. The final rule includes this reference to appendix Q and additional comment to clarify the relationship between the requirement to consider a consumer's current or reasonably expected income in § 1026.43(e)(2)(v)(A) and the definition of “income” in appendix Q. In other words, a creditor who considers “income” as defined in appendix Q meets the income requirement in § 1026.43(e)(2)(v)(A), so long as that income is verified pursuant to § 1026.43(c)(4). In addition, comment 43(e)(2)(v)-1 provides that for guidance on satisfying § 1026.43(e)(2)(v), a creditor may rely on commentary to § 1026.43(c)(2)(i) and (vi), (c)(3), and (c)(4).

43(e)(2)(v)(B)

The Board's proposed Alternative 2 would have required that creditors consider and verify the following additional underwriting requirements, which are also required under the general ability-to-repay standard: the consumer's employment status, the consumer's monthly payment on any simultaneous loans, the consumer's current debt obligations, the consumer's monthly debt-to-income ratio or residual income, and the consumer's credit history. The commentary would have provided that creditors could look to commentary on the general repayment ability provisions under proposed § 226.43(c)(2)(i), (ii), (iv), and (vi) through (viii), and (c)(3), (c)(4), (c)(6), and (c)(7) for guidance regarding considering and verifying the consumer's repayment ability to satisfy the conditions under § 226.43(e)(2)(v) for a qualified mortgage. See proposed comment 43(e)(2)(v)-1 under Alternative 2. The Board proposed these additions pursuant to its legal authority under TILA section 129C(b)(3)(B)(i). The Board believed that adding these requirements may be necessary to better ensure that the consumers are offered and receive loans on terms that reasonably reflect their ability to repay the loan.

In the final rule, § 1026.43(e)(2)(v)(B) provides that, to meet the requirements for a qualified mortgage under § 1026.43(e)(2), the creditor must consider and verify the consumer's current debt obligations, alimony, and child support, in accordance with appendix Q and § 1026.43(c)(2)(vi) and (c)(3). In addition, new comment 43(e)(2)(v)-3 clarifies that, for purposes of considering and verifying the consumer's current debt obligations, alimony, and child support pursuant to § 1026.43(e)(2)(v)(B), the creditor must consider and verify, at a minimum, any debt or liability specified in appendix Q. A creditor may also consider and verify other debt in accordance with § 1026.43(c)(2)(vi) and (c)(3); however, such debt would not be included in the total monthly debt-to-income ratio determination required by § 1026.43(e)(2)(vi). As described below, appendix Q contains specific standards for defining “debt,” to provide certainty to creditors as to whether a loan meets the requirements for a qualified mortgage. The final rule includes this reference to appendix Q and additional comment to clarify the relationship between the requirement to consider a consumer's current debt obligations, alimony, and child support in § 1026.43(e)(2)(v)(B) and the definition of “debt” in appendix Q. In other words, a creditor who considers “debt” as defined in appendix Q meets the requirement in § 1026.43(e)(2)(v)(B), so long as that income is verified pursuant to § 1026.43(c)(3).

The Bureau is incorporating the requirement that the creditor consider and verify the consumer's current debt obligations, alimony, and child support into the definition of a qualified mortgage in § 1026.43(e)(2) pursuant to its authority under TILA section 129C(b)(3)(B)(i). The Bureau finds that this addition to the qualified mortgage criteria is necessary and proper to ensure that responsible, affordable mortgage credit remains available to consumers in a manner that is consistent with the purposes of TILA section 129C and necessary and appropriate to effectuate the purposes of TILA section 129C, which includes assuring that consumers are offered and receive residential mortgage loans on terms that reasonably reflect their ability to repay the loan. The Bureau also incorporates this requirement pursuant to its authority under TILA section 105(a) to issue regulations that, among other things, contain such additional requirements, other provisions, or that provide for such adjustments for all or any class of transactions, that in the Bureau's judgment are necessary or proper to effectuate the purposes of TILA, which include the above purpose of section 129C, among other things. The Bureau believes that this addition to the qualified mortgage criteria is necessary and proper to achieve this purpose. In particular, as discussed above, the Bureau finds that incorporating the requirement that a creditor consider and verify a consumer's current debt obligations, alimony, and child support into the qualified mortgage criteria ensures that creditors consider, on an individual basis, and verify whether a consumer has the ability to repay a qualified mortgage. Furthermore, together with the requirement to consider and verify income, the Bureau believes this requirement to consider and verify debt obligations, alimony, and child support strengthens consumer protection and is fundamental to the underlying components of the requirement in § 1026.43(e)(2)(vi), which provides a specific debt-to-income ratio threshold.

Ultimately, the Bureau believes that the statute is fundamentally about establishing standards for determining a consumer's reasonable ability to repay and therefore believes it is appropriate to incorporate the ability-to-repay underwriting requirements into the qualified mortgage definition to ensure consistent consumer protections for repayment ability for a qualified mortgage. However, as described above, most of the ability-to-repay requirements must be considered and verified to satisfy the specific debt-to-income ratio requirement in § 1026.43(e)(2)(vi), which requires the creditor to follow the standards for “debt” and “income” in appendix Q, including the consumer's employment status, monthly payment on the covered transaction, monthly payment on simultaneous loans of which the creditor is aware, and monthly payment on mortgage-related obligations. For this reason, unlike the Board's proposed Alternative 2, the final rule does not separately require consideration and verification of these factors that are part of the general ability-to-repay analysis.

43(e)(2)(vi)

TILA section 129C(b)(2)(vi) states that the term qualified mortgage includes any mortgage loan “that complies with any guidelines or regulations established by the Bureau relating to ratios of total monthly debt to monthly income or alternative measure of ability to pay regular expenses after payment of total monthly debt, taking into account the income levels of the consumer and such other factors as the Bureau may determine relevant and consistent with the purposes described in paragraph (3)(B)(i).”

Board's Proposal

Under proposed § 226.43(e)(2)(v) under Alternative 1, creditors would not have been required to consider the consumer's debt-to-income ratio or residual income to make a qualified mortgage. The Board noted several reasons for proposing this approach. First, the Board noted that the debt-to-income ratio and residual income are based on widely accepted standards which, although flexible, do not provide certainty that a loan is a qualified mortgage. The Board believed this approach is contrary to Congress' apparent intent to provide incentives to creditors to make qualified mortgages, since they have less risky features and terms. Second, the Board noted that because the definition of a qualified mortgage under Alternative 1 would not require consideration of current debt obligations or simultaneous loans, it would be impossible for a creditor to calculate the debt-to-income ratio or residual income without adding those requirements as well. Third, the Board stated that data shows that the debt-to-income ratio generally does not have a significant predictive power of loan performance once the effects of credit history, loan type, and loan-to-value ratio are considered. Fourth, the Board noted that although consideration of the mortgage debt-to-income ratio (or “front-end” debt-to-income) might help consumers receive loans on terms that reasonably reflect their ability to repay the loans, the Board's outreach indicated that creditors often do not find that “front-end” debt-to-income ratio is a strong predictor of ability to repay. Finally, the Board stated its concern that the benefit of including the debt-to-income ratio or residual income in the definition of qualified mortgage may not outweigh the cost to certain consumers who may not meet widely accepted debt-to-income ratio standards, but may have other compensating factors, such as sufficient residual income or other resources to be able to reasonably afford the mortgage. A definition of qualified mortgage that required consideration of the consumer's debt-to-income or residual income could limit the availability of credit to those consumers.

However, under proposed § 226.43(e)(2)(v) under Alternative 2, a qualified mortgage would have been defined as a loan which, among other things, the creditor considers the consumer's monthly debt-to-income ratio or residual income, pursuant to proposed § 226.43(c)(2)(vii) and (c)(7). The Board noted that, without determining the consumer's debt-to-income ratio, a creditor could originate a qualified mortgage without any requirement to consider the effect of the new loan payment on the consumer's overall financial picture. The consumer could have a very high total debt-to-income ratio under reasonable underwriting standards, and be predicted to default soon after the first scheduled mortgage payment. Accordingly, the Board believed that including the debt-to-income ratio or residual income in the definition of qualified mortgage might ensure that the consumer has a reasonable ability to repay the loan.

The Board did not propose a quantitative standard for the debt-to-income ratio in the qualified mortgage definition, but solicited comment on the appropriateness of such an approach. The Board's proposal noted several reasons for declining to introduce a specific debt-to-income ratio for qualified mortgages. First, as explained in the 2008 HOEPA Final Rule, the Board was concerned that setting a specific debt-to-income ratio could limit credit availability without providing adequate off-setting benefits. 73 FR 4455 (July 30, 2008). The Board sought comment on what exceptions may be necessary for low-income consumers or consumers living in high-cost areas, or for other cases, if the Board were to adopt a quantitative debt-to-income standard.

Second, outreach conducted by the Board revealed a range of underwriting guidelines for debt-to-income ratios based on product type, whether creditors used manual or automated underwriting, and special considerations for high- and low-income consumers. The Board believed that setting a quantitative standard would require it to address the operational issues related to the calculation of the debt-to-income ratio. For example, the Board would need clearly to define income and current debt obligations, as well as compensating factors and the situations in which creditors may use compensating factors. In addition, the debt-to-income ratio is often a floating metric, since the percentage changes as new information about income or current debt obligations becomes available. A quantitative standard would require guidelines on the timing of the debt-to-income ratio calculation, and what circumstances would necessitate a re-calculation of the debt-to-income ratio. Furthermore, a quantitative standard may also need to provide tolerances for mistakes made in calculating the debt-to-income ratio. The rule would also need to address the use of automated underwriting systems in determining the debt-to-income.

For all these reasons, the Board did not propose a quantitative standard for the debt-to-income ratio. The Board recognized, however, that creditors, and ultimately consumers, may benefit from a higher degree of certainty surrounding the qualified mortgage definition that a quantitative standard could provide. Therefore, the Board solicited comment on whether and how it should prescribe a quantitative standard for the debt-to-income ratio or residual income for the qualified mortgage definition.

Comments

As noted above, the Bureau received comments in response to the Board's 2011 ATR Proposal and in response to the Bureau's May 2012 notice to reopen the comment period. The reopened comment period solicited comment specifically on new data and information obtained from the Federal Housing Finance Agency (FHFA) after the close of the original comment period. In the notice to reopen the comment period, the Bureau, among other things, solicited comment on data and information as well as sought comment specifically on certain underwriting factors, such as a debt-to-income ratio, and their relationship to measures of delinquency or their impact on the number or percentage of mortgage loans that would be a qualified mortgage. In addition, the Bureau sought comment and data on estimates of litigation costs and liability risks associated with claims alleging a violation of ability-to-repay requirements.

Comments on general debt-to-income ratio or residual income requirement. In response to the proposed rule, some industry commenters argued that the final rule should not require consideration and verification of a consumer's monthly debt-to-income ratio or residual income for a qualified mortgage. They argued that such an approach would create a vague, subjective definition of qualified mortgage. Certain industry commenters requested that if the Bureau added consideration and verification of the debt-to-income ratio or residual income to the definition of a qualified mortgage, the Bureau establish flexible standards. These commenters argued that imposing low debt-to-income ratio requirements would be devastating to many potential creditworthy homebuyers.

Other industry commenters suggested that if the Bureau added consideration and verification of the debt-to-income ratio or residual income to the definition of a qualified mortgage, the Bureau provide clear and objective standards. For example, one industry trade group commenter noted that, historically, the debt-to-income ratio has been a key metric used to assess a consumer's ability to repay a mortgage loan, and has been incorporated into both manual and automated underwriting systems used in the industry. Some industry commenters asked that the final rule adopt the VA calculation of residual income. See also the section-by-section analysis of section 1026.43(c)(7). Another industry commenter suggested that any mortgage with a residual income of at least $600 be sufficient for a qualified mortgage. Another industry commenter suggested that, at a minimum, residual income considerations would require a workable standard with clear, specific, and objective criteria and be explicitly limited to specific expense items. An industry trade group commenter recommended that if the Bureau requires the use of residual income, creditors be allowed flexibility in considering residual income along with other factors in loan underwriting. Comments that addressed a specific debt-to-income ratio are discussed below.

Several industry commenters recommended that if the Bureau required consideration and verification of the debt-to-income ratio or residual income for a qualified mortgage, creditors be permitted to take compensating factors into account. They suggested that the Bureau provide examples of compensating factors, such as: (1) The property being an energy-efficient home; (2) the consumer having probability for increased earnings based on education, job training, or length of time in a profession; (3) the consumer having demonstrated ability to carry a higher total debt-load while maintaining a good credit history for at least 12 months; (4) future expenses being lower, such as child-support payments to cease for child soon to reach age of majority; or (5) the consumer having substantial verified liquid assets.

Consumer advocates generally supported adding consideration and verification of the debt-to-income ratio or residual income to the definition of a qualified mortgage. They noted that such inclusion would help ensure that consumers receive mortgages they can afford and that such factors are basic, core features of common-sense underwriting that are clearly related to the risk of consumer default. To that point, these commenters contended that residual income is an essential component, especially for lower-income families. One consumer group commenter stressed that residual income standards should be incorporated, and pointed to the FHFA data in the Bureau's notice to reopen the comment period to demonstrate that relying solely on debt-to-income ratios is insufficient to ensure sound lending based on a consumer's ability to repay.

Many industry and consumer group commenters and interested parties supported use of a specific debt-to-income ratio threshold. For example, some suggested that if a consumer's total debt-to-income ratio is below a specified threshold, the mortgage loan should satisfy the qualified mortgage requirements, assuming other relevant conditions are met. At least one industry commenter supported allowing the use of FHA underwriting guidelines to define “debt” and “income.”

Although many commenters supported the use of a specific debt-to-income ratio threshold, both industry and consumer group commenters noted that relying on debt-to-income is only one element of underwriting, and that creditors have used other compensating factors and underwriting criteria. Some commenters acknowledged that a consumer's debt-to-income ratio is a useful measure of loan performance; however, they asserted that the year of origination (i.e., vintage) has more bearing on loan performance. In addition, some commenters argued that measures of consumer credit history and loan-to-value are more predictive, and that broader economic factors largely determine loan performance. Several industry commenters recommended a debt-to-income ratio cutoff that is at the upper end of today's relatively conservative lending standards, while permitting creditors to consider loans that exceed that debt-to-income ratio threshold if the consumer satisfies other objective criteria (such as reserves, housing payment history, and residual income), that help creditors assess the consumer's ability to repay the loan. These commenters argued that the FHFA data in the Bureau's notice to reopen the comment period demonstrate that when loans are properly underwritten, debt-to-income ratios can be relatively high without significantly affecting loan performance.

Numerous commenters argued that the Bureau should consider the costs and benefits of selecting a maximum debt-to-income ratio for qualified mortgages. Many industry and consumer group commenters argued that a debt-to-income threshold that is too low would unnecessarily exclude a large percentage of consumers from qualified mortgages. One joint industry and consumer group comment letter suggested a 43 percent total debt-to-income ratio. In addition to a debt-to-income requirement, some commenters and interested parties suggested that the Bureau should include within the definition of a “qualified mortgage” loans with a debt-to-income ratio above a certain threshold if the consumer has a certain amount of assets, such as money in a savings or similar account, or a certain amount of residual income. For example, an industry commenter suggested a 45 percent total debt-to-income ratio, with an allowance for higher total debt-to-income ratios of up to 50 percent for consumers with significant assets (e.g., at least one year's worth of reserves). This commenter asked that the Bureau carve out consumers who have shown ability to maintain a high debt-to-income ratio or who have a nontraditional credit history. This commenter explained that the higher the debt-to-income ratio, the more likely a brief interruption in income or unexpected large expense could compromise repayment ability. The commenter noted that only a numerical standard would provide sufficient certainty for creditors and investors, since they may otherwise end up litigating what is a reasonable debt-to-income ratio. Another industry commenter asked that a 50 percent back-end debt-to-income ratio be sufficient. This commenter noted that without clear and objective standards, creditors trying to make a qualified mortgage would fall back on the qualified residential mortgage standards.

Another industry trade association commenter argued that a total debt-to-income ratio threshold of 43 percent is problematic because according to the FHFA data in the Bureau's notice to reopen the comment period, there is no appreciable difference in performance for loans with a 43 percent debt-to-income ratio and loans with 46 percent debt-to-income ratio. In other words, commenters argued that the FHFA data supports a higher debt-to-income ratio threshold, such as 46 percent. Another commenter noted that the FHFA data does not include data on portfolio loans.

Some consumer group commenters suggested that the Bureau conduct further research into the role of debt-to-income ratios and the relationship between a consumer's debt-to-income ratio and residual income. One commenter noted that the Bureau should consider a tiered-approach for higher-income consumers who can support a higher debt-to-income ratio. Another consumer group commenter argued that residual income should be incorporated into the definition of qualified mortgage. Several commenters suggested that the Bureau use the general residual income standards of the VA as a model for a residual income test, and one of these commenters recommended that the Bureau coordinate with FHFA to evaluate the experiences of the GSEs in using residual income in determining a consumer's ability to repay.

Some commenters opposed including a specific debt-to-income ratio threshold into the qualified mortgage criteria. For example, one commenter argued that though the qualified mortgage criteria should be as objective as possible, a specific debt-to-income threshold should not be imposed because the criteria should be flexible to account for changing markets. Another commenter argued that creditors should be able to consider debt-to-income and residual income ratios, but creditors should not be restricted to using prescribed debt-to-income or residual income ratios. One industry commenter contended that if the Bureau were to impose a 45 percent total debt-to-income ratio, for example, most larger secondary market investors/servicers would impose a total debt-to-income ratio that is much lower (such as 43 percent or 41 percent) as a general rule of risk management.

Final Rule

The Bureau believes, based upon its review of the data it has obtained and the comments received, that the use of total debt-to-income as a qualified mortgage criterion provides a widespread and useful measure of a consumer's ability to repay, and that the Bureau should exercise its authority to adopt a specific debt-to-income ratio that must be met in order for a loan to meet the requirements of a qualified mortgage. The Bureau believes that the qualified mortgage criteria should include a standard for evaluating whether consumers have the ability to repay their mortgage loans, in addition to the product feature requirements specified in the statute. At the same time, the Bureau recognizes concerns that creditors should readily be able to determine whether individual mortgage transactions will be deemed qualified mortgages. The Bureau addresses these concerns by adopting a bright-line debt-to-income ratio threshold of 43 percent, as well as clear and specific standards, based on FHA guidelines, set forth in appendix Q for calculating the debt-to-income ratio in individual cases.

The Bureau believes that a consumer's debt-to-income ratio is generally predictive of the likelihood of default, and is a useful indicator of such. At a basic level, the lower the debt-to-income ratio, the greater the consumer's ability to pay back a mortgage loan would be under existing conditions as well as changed circumstances, such as an increase in an adjustable rate, a drop in future income, or unanticipated expenses or new debts. The Bureau's analysis of FHFA's Historical Loan Performance (HLP) dataset, data provided by the FHA, and data provided by commenters all bear this out. These data indicate that debt-to-income ratio correlates with loan performance, as measured by delinquency rate (where delinquency is defined as being over 60 days late), in any credit cycle. Within a typical range of debt-to-income ratios for prudent underwriting (e.g., under 32 percent debt-to-income to 46 percent debt-to-income), the Bureau notes that generally, there is a gradual increase in delinquency with higher debt-to-income ratio. The record also shows that debt-to-income ratios are widely used as an important part of the underwriting processes of both governmental programs and private lenders.

The Bureau recognizes the Board's initial assessment that debt-to-income ratios may not have significant predictive power once the effects of credit history, loan type, and loan-to-value are considered. In the same vein, the Bureau notes that some commenters suggested that the Bureau include compensating factors in addition to a specific debt-to-income ratio threshold. Even if a standard that takes into account multiple factors produces more accurate ability-to-pay determinations in specific cases, incorporating a multi-factor test or compensating factors into the definition of a qualified mortgage would undermine the goal of ensuring that creditors and the secondary market can readily determine whether a particular loan is a qualified mortgage. Further, the Bureau believes that compensating factors would be too complex to calibrate into a bright-line rule and that some compensating factors suggested by commenters as appropriate, such as loan-to-value ratios, do not speak to a consumer's repayment ability.

Therefore, as permitted by the statute, the Bureau is adopting a specific debt-to-income ratio threshold because this approach provides a clear, bright line criterion for a qualified mortgage that ensures that creditors in fact evaluate consumers' ability to repay qualified mortgages and provides certainty for creditors to know that a loan satisfies the definition of a qualified mortgage. A specific debt-to-income ratio threshold also provides additional certainty to assignees and investors in the secondary market, which should help reduce possible concerns regarding legal risk and potentially promote credit availability. As numerous commenters have urged, there is significant value to providing objective requirements that can be determined based on loan files. As described below, the final rule generally requires creditors to use the standards for defining “debt” and “income” in appendix Q, which are adapted from current FHA guidelines, to minimize burden and provide consistent standards. The standards set forth in appendix Q provide sufficient detail and clarity to address concerns that creditors may not have adequate certainty about whether a particular loan satisfies the requirements for being a qualified mortgage, and therefore will not deter creditors from providing qualified mortgages to consumers. The Bureau anticipates that the standards will facilitate compliance with the Dodd-Frank Act risk retention requirements, as the 2011 QRM Proposed Rule relied on FHA standards for defining “debt” and “income.” The Bureau has consulted with the Federal agencies responsible for the QRM rulemaking in developing this rule, and will continue to do so going forward.

Based on analysis of available data and comments received, the Bureau believes that 43 percent is an appropriate ratio for a specific debt-to-income threshold, and that this approach advances the goals of consumer protection and preserving access to credit. The Bureau acknowledges, based on its analysis of the data, that there is no “magic number” which separates affordable from unaffordable mortgages; rather, as noted above, there is a gradual increase in delinquency rates as debt-to-income ratios increase. That being said, the Bureau understands that 43 percent is within the range of debt-to-income ratios used by many creditors and generally comports with industry standards and practices for prudent underwriting. As noted above, 43 percent is the threshold used by the FHA as its general boundary. Although the Bureau notes that Fannie Mae's and Freddie Mac's guidelines generally require a 36 percent debt-to-income ratio, without compensating factors, the Bureau believes that a 43 percent debt-to-income threshold represents an appropriate method to define which loans merit treatment as qualified mortgages. In particular, the Bureau believes that 43 percent represents a prudent outer boundary for a categorical presumption of a consumer's ability to repay.

As discussed above, there was significant debate among the commenters about the precise debt-to-income ratio threshold to establish. Although a lower debt-to-income threshold would provide greater assurance of a consumer's ability to repay a loan, many commenters argued, and the Bureau agrees, that establishing a debt-to-income ratio threshold significantly below 43 percent would curtail many consumers' access to qualified mortgages. One commenter estimated that roughly half of conventional borrowers would not be eligible for qualified mortgage loans if the debt-to-income ratio was set at 32 percent, while 85 percent of borrowers would be eligible with a ratio set at 45 percent.

At the same time, the Bureau declines to establish a debt-to-income ratio threshold higher than 43 percent. The Bureau recognizes that some commenters suggested that debt-to-income ratios above 43 percent would not significantly increase the likelihood of default (depending to some extent on the presence of compensating factors), and that some consumers may face greater difficulty obtaining qualified mortgages absent a higher threshold. However, as the debt-to-income ratio increases, the presence of compensating factors becomes more important to the underwriting process and in ensuring that consumers have the ability to repay the loan. The general ability-to-repay procedures, rather than the qualified mortgage framework, is better suited for consideration of all relevant factors that go to a consumer's ability to repay a mortgage loan.

Thus, the Bureau emphasizes that it does not believe that a 43 percent debt-to-income ratio represents the outer boundary of responsible lending. The Bureau notes that even in today's credit-constrained market, approximately 22 percent of mortgage loans are made with a debt-to-income ratio that exceeds 43 percent and that prior to the mortgage boom approximately 20 percent of mortgage loans were made above that threshold. Various governmental agencies, GSEs, and creditors have developed a range of compensating factors that are applied on a case by case basis to assess a consumer's ability to repay when the consumer's debt-to-income ratio exceeds a specified ratio. Many community banks and credit unions have found that they can prudently lend to consumers with a higher debt-to-income ratio based upon their firsthand knowledge of the individual consumer. As discussed below, many of those loans will fall within the temporary exception that the Bureau is recognizing for qualified mortgages. Over the long term, as the market recovers from the mortgage crisis and adjusts to the ability-to-repay rules, the Bureau expects that there will be a robust and sizable market for prudent loans beyond the 43 percent threshold even without the benefit of the presumption of compliance that applies to qualified mortgages. In short, the Bureau does not believe that consumers who do not receive a qualified mortgage because of the 43 percent debt-to-income ratio threshold should be cut off from responsible credit, and has structured the rule to try to ensure that a robust and affordable ability-to-repay market develops over time.

The Bureau also believes that there would be significant negative consequences to the market from setting a higher threshold. For instance, if the qualified mortgage debt-to-income ratio threshold were set above 43 percent, it might sweep in many mortgages in which there is not a sound reason to presume that the creditor had a reasonable belief in the consumer's ability to repay. At a minimum, adopting a higher debt-to-income threshold to define qualified mortgages would require a corresponding weakening of the strength of the presumption of compliance—which would largely defeat the point of adopting a higher debt-to-income threshold. Additionally, the Bureau also fears that if the qualified mortgage boundary were to cover substantially all of the mortgage market, creditors might be unwilling to make non-qualified mortgage loans, with the result that the qualified mortgage rule would define the limit of credit availability. The Bureau believes that lending in the non-qualified mortgage market can and should be robust and competitive over time. The Bureau expects that, as credit conditions ease, creditors will continue making prudent, profitable loans in non-traditional segments, such as to consumers who have sufficient total assets or future earning potential to be able to afford a loan with a higher debt-to-income ratio or consumers who have a demonstrated ability to pay housing expenses at or above the level of a contemplated mortgage.

Finally, the Bureau acknowledges arguments that residual income may be a better measure of repayment ability in the long run. A consumer with a relatively low household income may not be able to afford a 43 percent debt-to-income ratio because the remaining income, in absolute dollar terms, is too small to enable the consumer to cover his or her living expenses. Conversely, a consumer with a relatively high household income may be able to afford a higher debt ratio and still live comfortably on what is left over. Unfortunately, however, the Bureau lacks sufficient data, among other considerations, to mandate a bright-line rule based on residual income at this time. The Bureau expects to study residual income further in preparation for the five-year review of this rule required by the Dodd-Frank Act. See also section-by-section analysis of § 1026.43(c)(7).

The Bureau believes that it is important that the final rule provide clear standards by which creditors calculate a consumer's monthly debt-to-income ratio for purposes of the specific debt-to-income threshold in § 1026.43(e)(2)(vi). For this reason, the final rule provides specific standards for defining “debt” and “income” in appendix Q. These standards are based on the definitions of debt and income used by creditors originating residential mortgages that are insured by the FHA. In particular, appendix Q incorporates the definitions and standards in the HUD Handbook 4155.1, Mortgage Credit Analysis for Mortgage Insurance on One-to-Four-Unit Mortgage Loans, to determine and verify a consumer's total monthly debt and monthly income, with limited modifications to remove portions unique to the FHA underwriting process, such as references to the TOTAL Scorecard Instructions. The use of FHA guidelines for this purpose provides clear, well-established standards for determining whether a loan is a qualified mortgage under § 1026.43(e)(2). This approach is also consistent with the proposed approach to defining debt and income in the 2011 QRM Proposed Rule, and therefore could facilitate compliance for creditors. The Bureau has consulted with the Federal agencies responsible for the QRM rulemaking and will continue to do so going forward as that rulemaking is completed, as well as to discuss changes to FHA guidelines that may occur over time.

Accordingly, § 1026.43(e)(2)(vi) provides that, as a condition to being a qualified mortgage under § 1026.43(e)(2), the consumer's total monthly debt-to-income ratio does not exceed 43 percent. For purposes of § 1026.43(e)(2)(vi), the consumer's monthly debt-to-income ratio is calculated in accordance with appendix Q, except as provided in § 1026.43(e)(2)(vi)(B). Section 1026.43(e)(2)(vi)(B) contains additional requirements regarding the calculation of “debt,” for consistency with other parts of the qualified mortgage definition and § 1026.43. Specifically, that section provides that the consumer's monthly debt-to-income ratio must be calculated using the consumer's monthly payment on the covered transaction, including mortgage-related obligations, in accordance with § 1026.43(e)(2)(iv), and any simultaneous loan that the creditor knows or has reason to know will be made, in accordance with § 1026.43(c)(2)(iv) and (c)(6). Comment 43(e)(2)(vi)-1 clarifies the relationship between the definition of “debt” in appendix Q and the requirements of § 1026.43(e)(2)(vi)(B). Specifically, the comment states that, as provided in appendix Q, for purposes of § 1026.43(e)(2)(vi), creditors must include in the definition of “debt” a consumer's monthly housing expense. This includes, for example, the consumer's monthly payment on the covered transaction (including mortgage-related obligations) and simultaneous loans. Accordingly, § 1026.43(e)(2)(vi)(B) provides the method by which a creditor calculates the consumer's monthly payment on the covered transaction and on any simultaneous loan that the creditor knows or has reason to know will be made.

The Bureau notes that the specific 43 percent debt-to-income requirement applies only to qualified mortgages under § 1026.43(e)(2). For the reasons discussed below, the specific debt-to-income ratio requirement does not apply to loans that meet the qualified mortgage definitions in § 1026.43(e)(4) or (f).

43(e)(3) Limits on Points and Fees for Qualified Mortgages

43(e)(3)(i)

TILA section 129C(b)(2)(A)(vii) defines a “qualified mortgage” as a loan for which, among other things, the total points and fees payable in connection with the loan do not exceed 3 percent of the total loan amount. TILA section 129C(b)(2)(D) requires the Bureau to prescribe rules adjusting this limit to “permit lenders that extend smaller loans to meet the requirements of the presumption of compliance.” The statute further requires the Bureau to “consider the potential impact of such rules on rural areas and other areas where home values are lower.” The statute does not define and the legislative history does not provide guidance on the term “smaller loan” or the phrase “rural areas and other areas where home values are lower.”

The Board proposed two alternative versions of § 226.43(e)(3)(i) to implement the 3 percent points and fees cap for qualified mortgages and the adjustment to the cap for smaller loans. For both alternatives, the Board proposed a threshold of $75,000, indexed to inflation, for smaller loans. For loans above the $75,000 threshold, the 3 percent points and fees cap for qualified mortgages would have applied. For loans below $75,000, different limits would have applied, depending on the amount of the loan.

The Board explained that it set the smaller loan threshold at $75,000 because it believed that Congress intended the exception to the 3 percent points and fees cap to apply to more than a minimal, but still limited, proportion of home-secured loans. The Board noted that HMDA data show that 8.4 percent of first-lien, home-purchase (site-built) mortgages in 2008 and 9.7 percent of such mortgages in 2009 had a loan amount of $74,000 or less. The Board also stated that outreach and research indicated that $2,250—3 percent of $75,000—is within range of average costs to originate a first-lien home mortgage. Thus, the Board concluded that $75,000 appears to be an appropriate benchmark for applying the 3 percent limit on points and fees, with higher limits below that threshold offering creditors a reasonable opportunity to recover their origination costs.

Both of the Board's proposed alternatives would have separated loans into tiers based on loan size, with each tier subject to different limits on points and fees. The Board's proposed Alternative 1 would have consisted of five tiers of loan sizes and corresponding limits on points and fees:

• For a loan amount of $75,000 or more, 3 percent of the total loan amount;
• For a loan amount greater than or equal to $60,000 but less than $75,000, 3.5 percent of the total loan amount;
• For a loan amount greater than or equal to $40,000 but less than $60,000, 4 percent of the total loan amount;
• For a loan amount greater than or equal to $20,000 but less than $40,000, 4.5 percent of the total loan amount; and
• For a loan amount less than $20,000, 5 percent of the total loan amount.

Alternative 2 would have consisted of three tiers of loan sizes and corresponding limits on points and fees. The first and third tiers were consistent with Alternative 1. The middle tier was a sliding scale that reduced the points and fees cap (as a percentage of the loan amount) with each dollar increase in loan size. The three tiers of Alternative 2 would have consisted of:

• For a loan amount of $75,000 or more, 3 percent of the total loan amount;
• For a loan amount greater than or equal to $20,000 but less than $75,000, a percentage of the total loan amount yielded by the following formula:

○ Total loan amount−$20,000 = $Z

○ $Z × 0.0036 basis points = Y basis points

○ 500 basis points−Y basis points = X basis points

○ X basis points × 0.01 = Allowable points and fees as a percentage of the total loan amount.

• For a loan amount less than $20,000, 5 percent of the total loan amount.

The approach in Alternative 2 would have smoothed the transition from one tier to another and fixed an anomaly of Alternative 1. Under Alternative 1, for loans just above and below the dividing line between tiers, a greater dollar amount of points and fees would have been allowed on the smaller loans than on the larger loans. For example, the allowable points and fees on a total loan amount of $76,000 would have been $2,280 (3 percent of $76,000), but the permissible points and fees on a total loan amount of $70,000 would have been $2,450 (3.5 percent of $70,000).

The Board noted that its proposal was designed to ensure that if a loan is a qualified mortgage it would not also be a high-cost mortgage based on the points and fees. The Board stated its belief that the statute is designed to reduce the compliance burden on creditors when they make qualified mortgages, in order to encourage creditors to make loans with stable, understandable loan features. The Board expressed concern that creating points and fees thresholds for small loans that might result in qualified mortgages also being high-cost mortgages would discourage creditors from making qualified mortgages because the requirements and limitations of high-cost loans are generally more stringent than for other loans.

The Board requested comment on the proposed alternative loan size ranges and corresponding points and fees limits for qualified mortgages. The Board also requested comment on whether the loan size ranges should be indexed for inflation.

The Board stated that, instead of using a smaller loan threshold with different tiers, it had considered adjusting the criteria for smaller loans by narrowing the types of charges that would be included in points and fees for smaller loans. The Board indicated that outreach participants disfavored this approach because it would have required different ways of calculating points and fees, depending on loan size, and thus likely would have increased the burden of complying with the rules and the risk of error. The Board also stated that it had considered proposing an alternative points and fees threshold for certain geographical areas. As the Board noted, however, property values shift over time, and there is substantial variation in property values and loan amounts within geographical areas. Thus, adjusting the limits on points and fees based solely on geographic areas would have been a less straightforward and less precise method of addressing the statute's concern with smaller loans. No commenters supported these approaches.

Several industry commenters argued that points and fees have little, if any, relationship to consumers' ability to repay their mortgage loans and that qualified mortgages should therefore not be subject to limits on points and fees. Although they acknowledged that the Dodd-Frank Act generally prescribed a 3 percent limit on points and fees for qualified mortgages, they urged the Bureau to use its authority to eliminate this requirement.

Several industry commenters contended that the 3 percent limit on points and fees for qualified mortgages is too low. They maintained that the 3 percent cap would require creditors to increase interest rates to recover their costs and would limit consumers' flexibility to arrange their optimal combination of interest rates and points and fees. Industry commenters also claimed that the 3 percent limit would have a negative impact on consumers' access to affordable credit. Some industry commenters noted that the GSEs' seller/servicer guides contain standards that limit points and fees for loans that the GSEs purchase or securitize, with the current standards limiting points and fees to the greater of 5 percent of the mortgage amount or $1,000. The commenters argued that Bureau should use its authority adopt the GSEs' standards instead of the requirements prescribed by the Dodd-Frank Act. One commenter argued that, because of the complexity of the points and fees test, the Bureau should adopt a tolerance of one-quarter of 1 percent or $250 for the 3 percent limit so that de minimis errors in calculating points and fees would not prevent a loan from retaining the legal protection of a qualified mortgage.

With respect to the two proposed alternative versions of section 43(e)(3)(i), industry commenters generally preferred Alternative 1. They explained that Alternative 2 was too complex, would be difficult to implement, and would increase compliance and litigation costs. Some consumer advocates preferred Alternative 2, stating that it would be more beneficial to consumers. Other consumer advocates preferred Alternative 1, asserting that its simplicity would minimize miscalculations that could harm consumers. They stated that the difference to the consumer between Alternative 1 and Alternative 2 was marginal. Some of these consumer advocates argued that the benefit afforded by simplicity would outweigh the small pricing distortions.

Commenters did not object to the Board's general approach of setting a threshold amount for smaller loans and adjusting the points and fees cap for loans below the threshold. Instead, the comments discussed what the threshold loan amount should be for smaller loans and what limits should be imposed on points and fees for loans below the threshold.

Industry commenters contended that the Board's proposed limits on points and fees for smaller loans would be too low and would not permit creditors to recover their costs. They stated that many origination costs are fixed regardless of loan size. They asserted that if a creditor could not cover those costs through points and fees, the creditor would either not make the mortgage or increase the interest rate to cover the costs. Industry commenters expressed concern that, for smaller loans, a rate increase might result in the loan becoming a high-cost mortgage or in some consumers no longer being eligible for the loan. They contended that creditors would be reluctant to make these loans and credit availability would be compromised, in particular for low-income, minority, and rural consumers, and first-time home buyers. One commenter reported that if a consumer were offered a high interest rate to cover costs and the rate were increased to offset the costs of a smaller loan, the consumer would pay thousands of dollars more over the life of the loan. Industry commenters asserted that the proposed alternatives did not capture the congressional intent of providing creditors sufficient incentives to make smaller loans. Industry commenters urged the Bureau to revise the proposal to allow creditors to recover more of their costs through points and fees, either by increasing the threshold for smaller loans or raising the limits for loans below the threshold or by doing both.

Many industry commenters recommended raising the threshold for smaller loans from the $75,000 threshold proposed by the Board. One industry commenter suggested setting the threshold at $100,000, indexed to inflation. Relying on loan balances for median home prices, another industry commenter asked that the Bureau raise the threshold to $125,000. Many other industry commenters recommended raising the threshold to $150,000. One commenter noted that the average loan size in the United States at the end of the second quarter of 2010 was $193,800 and suggested using 80 percent of the average loan size, rounding off to the nearest $10,000.

In addition to urging the Bureau to raise the smaller loan threshold, many industry commenters recommended that the Bureau revise the proposal to permit creditors to charge higher points and fees for loans below the smaller loan threshold for qualified mortgages. Several industry commenters asked that the Bureau set the cap between 3.5 and 5 percent, indexed to inflation, for all loans under the smaller loans threshold. One industry commenter noted that Fannie Mae and Freddie Mac permit points and fees up to 5 percent. An industry commenter suggested a cap equal to the greater of 3 percent or $2,000, indexed to inflation. A combination of industry commenters and consumer advocates recommended a cap equal to the greater of 3 percent or $3,000. One industry commenter advocated a 4 percent cap for all loans below $125,000. Several industry commenters recommended that the cap be set at a fixed amount plus a percentage to lessen the impact of moving from one tier to the next.

In support of their arguments to raise the smaller loan threshold and to raise the limits on points and fees for loans below the threshold, several industry commenters provided data showing that many smaller loans would have exceeded the proposed points and fees caps. For example, a trade association commenter drew on data submitted by a member bank that showed that the majority of loans under $100,000 would exceed the points and fees cap, assuming fees paid to an affiliate title company were included, and that many loans between $100,000 and $150,000 would also exceed the cap. A trade association industry commenter shared data from one of its members, a financial services provider. The member reviewed over 250,000 of its recent loans and found that none of the loans under $75,000 would meet the proposed cap and that 50 percent of the loans under $125,000 would meet the cap. Several industry commenters reported that if the Bureau raised the smaller loan threshold to $150,000, a significantly smaller percentage of loans would exceed the points and fees cap.

A trade association representing the manufactured housing industry noted the Board's concern about setting the points and fees cap so high that some qualified mortgages would be deemed high-cost mortgages under HOEPA. The commenter argued, however, that the Bureau has authority to change high-cost mortgage thresholds and urged the Bureau to exercise this authority. The commenter cited section 1431 of the Dodd-Frank Act for the proposition that the Board may increase the amount of origination costs above $1,000 for loans less than $20,000. The commenter also said that section 1022 of the Dodd-Frank Act may grant the Board authority to exempt certain smaller sized manufactured home loans from the 5 percent points and fees caps on high-cost mortgages for loans above $20,000, based on asset class, transaction volume, and existing consumer protections.

Consumer advocates generally endorsed the $75,000 threshold for smaller loans. They questioned industry concerns that the 3 percent threshold would limit the availability of credit for consumers with comparatively low loan amounts. Instead, the commenters emphasized the importance of ensuring that qualified mortgages are affordable. In their view, the 3 percent points and fees cap is a key factor in ensuring affordability, so the exception for smaller loans should apply to only a limited proportion of loans. Consumer advocates argued that the points and fees cap should not exceed the 5 percent HOEPA trigger. They asserted that points and fees should be reasonable, reflect actual origination costs, and not result in disparate pricing schemes disadvantaging consumers with smaller loans.

One consumer advocate recommended analyzing the impact of a 3 percent points and fees cap on access to credit for low- and moderate-income consumers, in particular for Community Reinvestment Act loans. The commenter asked that the Bureau describe in preamble the results of any analysis of points and fees by loan amount, and for Community Reinvestment Act and non-Community Reinvestment Act loans.

In light of these comments, the Bureau is adopting revised § 1026.43(e)(3)(i) to implement the limits on points and fees for qualified mortgages. As noted above, several industry commenters argued that points and fees have little if any bearing on consumers' ability to repay their mortgage loans and that the points and fees limits would result in higher interest rates and reduced access to credit. They urged the Bureau to use its authority to eliminate the limits on points and fees for qualified mortgages. As an alternative to eliminating the points and fees limits entirely, some industry commenters requested that the Bureau adopt the GSEs' standards limiting points and fees for loans that they purchase or securitize. Those standards currently limit points and fees to the greater of 5 percent of the loan amount or $1,000.

The Bureau does not believe it would be appropriate to eliminate the limits on points and fees for qualified mortgages. The Bureau also declines to adopt the GSEs' current standards and raise the general 3 percent limit on points and fees. The goal of TILA section 129C is to assure that consumers are able to repay their mortgages over the term of the loans. Originators that make large sums up front may be less careful in assuring the consumers' ability to repay over time. Moreover, Congress may have believed that the points and fees limits may deter originators from imposing unnecessary or excessive up-front charges. In the absence of persuasive evidence that the points and fees limits will undermine consumers' access to affordable credit, the Bureau does not believe it would be appropriate to eliminate the points and fees limits or to raise the general 3 percent limit. As discussed in more detail below, however, the Bureau is implementing revised points and fees limits for smaller loans. The Bureau also notes that the Dodd-Frank Act did not adopt a tolerance that would allow creditors to exceed the points and fees limits by small amounts and declines to adopt such a tolerance.

As noted above, a consumer advocate requested that the Bureau conduct an analysis of the 3 percent points and fees cap on access to credit for low- and moderate-income consumers, in particular for Community Reinvestment Act loans. Given the lack of available data, it has not been practicable for the Bureau to perform such an analysis while finalizing this and other title XIV rules. The Bureau will consider whether it is possible and valuable to conduct such an analysis in the future.

Revised § 1026.43(e)(3)(i) employs an approach similar to that proposed by the Board to implement the 3 percent cap on points and fees and the adjustment to the cap for smaller loans. Like the Board's proposal, § 1026.43(e)(3)(i) sets a threshold for smaller loans, establishes tiers based on loan size, and sets limits on points and fees within each tier. However, § 1026.43(e)(3)(i) uses a mix of percentage and flat dollar limits to avoid anomalous results at tier margins and also adjusts the definition of smaller loan to include more transactions.

Although most commenters favored this tiering methodology, as noted above, some commenters suggested that the Bureau reject the Board's tiered approach and instead adopt a simpler mechanism, with all loan amounts below the threshold subject to a single percentage cap or dollar amount cap on points and fees. Like the Board, the Bureau believes the tiered approach provides a more flexible and calibrated mechanism for implementing the limits on points and fees for smaller loans. A single percentage cap that would apply to all smaller loans may not allow creditors a reasonable opportunity to recover costs for very small loans. It also may create a distortion in which loans just below the smaller loan threshold would be permitted to have significantly higher points and fees than loans just above the smaller loan threshold. A single dollar amount cap (e.g., $3,000) could result in points and fees that are a very high percentage of the very smallest loans and, as a result, could result in qualified mortgages also triggering the obligations of high-cost mortgages.

Thus, as in the Board's proposal, the final rule sets a threshold for smaller loans and establishes tiers, based on loan size, with different limits on points and fees. Specifically, § 1026.43(e)(3)(i) provides that a transaction is not a qualified mortgage unless the total points and fees payable in connection with the loan do not exceed:

• For a loan amount greater than or equal to $100,000, 3 percent of the total loan amount;
• For a loan amount greater than or equal to $60,000 but less than $100,000, $3,000;
• For a loan amount greater than or equal to $20,000 but less than $60,000, 5 percent of the total loan amount;
• For a loan amount greater than or equal to $12,500 but less than $20,000, $1,000 of the total loan amount;
• For a loan amount of less than $12,500, 8 percent of the total loan amount.

The Bureau's final rule departs from the proposal in two ways. First, § 1026.43(e)(3)(i) raises the threshold for smaller loans to $100,000. Second, for loans below the $100,000 threshold, § 1026.43(e)(3)(i) revises the points and fees caps for smaller loans within the various tiers. The general effect of these revisions will be to increase the points and fees that creditors can charge for smaller loans while still permitting those loans to meet the standard for a qualified mortgage. These two changes are discussed at greater length below.

$100,000 Threshold for Smaller Loans

To fulfill the stated purpose of the adjustment for smaller loans, the threshold should be set at a level that is sufficient to permit creditors making smaller loans a reasonable opportunity to recoup their origination costs and still offer qualified mortgages but not so high as to cause loans to exceed the HOEPA threshold to become high-cost mortgages. As noted above, the Board proposed to set the smaller loan threshold so that three percent of that amount would have provided creditors with a reasonable opportunity to recover their costs, with loans below that threshold subject to higher caps on points and fees. Thus, the Board's proposed $75,000 threshold would have created a benchmark of $2,250. The Board stated that its outreach and research indicated that $2,250 would be within the range of average costs to originate a first-lien home mortgage. However, as noted above, several industry commenters reported, based on recent loan data, that creditors' points and fees often exceed $2,250 for smaller loans and that a significant number of loans above $75,000 would exceed the three percent cap.

This evidence suggests that the $2,250 benchmark (and the corresponding $75,000 smaller loan threshold) in the proposal could have been insufficient to permit creditors to recoup all or even most of their origination costs. The Bureau is aware that the commenters' loan data reflects creditors' points and fees, and not the underlying costs. Nevertheless, the evidence that substantial proportions of smaller loans would have exceeded the points and fees limits raises concerns that the creditors would not be able to recover their costs through points and fees and still originate qualified mortgages. Creditors that are unable to recover their origination costs through points and fees would have to attempt to recover those costs through higher rates. If the higher rates would trigger the additional regulatory requirements applicable to high-cost loans under HOEPA or would render some potential consumers ineligible, then access to credit for at least some consumers could be compromised. Moreover, for consumers who plan to remain in their homes (and their loans) for a long time, a higher interest rate would result in higher payments over the life of the loan.

Some commenters claimed that a substantial portion of loans up to $125,000 or $150,000 would exceed the 3 percent points and fees cap and that the Bureau should raise the threshold accordingly. The Bureau disagrees for two reasons. First, this would stretch the meaning of “smaller loans.” In 2011, slightly under 21 percent of first-lien home mortgages were below $100,000 and another 22 percent were between $100,000 and $150,000. Thus, increasing the threshold to $150,000 would more than double the number of loans entitled to an exception to the congressionally-established points and fees cap and would capture over 40 percent of the market. The Bureau believes that this would be an overly expansive construction of the term “smaller loans” for the purpose of the exception to the general rule capping points and fees for qualified mortgages at 3 percent. Such a broad definition of “smaller loans” could allow the exception to undermine the cap on points and fees and frustrate congressional intent that qualified mortgages include limited points and fees. The function of the smaller loan exception to the points and fees cap is to make it possible for creditors making smaller loans to originate qualified mortgages. The smaller loan exception should provide creditors a reasonable opportunity to recover most, if not all, of their origination costs for smaller loans and still originate qualified mortgages. It should not be transformed into a mechanism that ensures that creditors can continue to charge the same points and fees they have in the past and still have their loans meet the qualified mortgage standard.

The Bureau concludes that a $100,000 small loan threshold strikes an appropriate balance between congressional goals of allowing creditors offering smaller loans to meet the standard for qualified mortgages and ensuring that qualified mortgages include limited points and fees. The $100,000 threshold (and, as discussed below, the corresponding adjustments to the points and fees limits for loans under that threshold) should provide creditors with a reasonable opportunity to recover most, if not all, of their origination costs through points and fees, reducing the likelihood that any increase in rates would trigger obligations of high-cost loans or would cause loans to be higher-priced covered transactions under § 1026.43(b)(4). At the same time, the $100,000 threshold would not render the smaller loan exception so broad that it undermines the general 3 percent cap on points and fees. It would cover a significant but still limited proportion of mortgages. According to the 2011 Home Mortgage Disclosure Act (HMDA) data, 20.4 percent of first-lien home purchase mortgages and 20.9 percent of first-lien refinances were less than $100,000.

Limits on Points and Fees for Smaller Loans

In addition to raising the smaller loan threshold to $100,000, § 1026.43(e)(3)(i) also differs from the Board's proposal by setting higher limits on points and fees for smaller loans. As noted above, the Bureau is concerned that the Board's proposal would not have provided creditors with a reasonable opportunity to recover their origination costs. Thus, § 1026.43(e)(3)(i) allows creditors higher limits on points and fees for smaller loans. Specifically, for loans of $60,000 up to $100,000, § 1026.43(e)(3)(i) allows points and fees of no more than $3,000. For loans of $20,000 up to $60,000, § 1026.43(e)(3)(i) allows points and fees of no more than 5 percent of the total loan amount. For loans of $12,500 up to $20,000, § 1026.43(e)(3)(i) allows points and fees of no more than $1,000. For loan amounts less than $12,500, § 1026.43(e)(3)(i) allows points and fees of no more than 8 percent of the total loan amount.

In contrast with the Board's proposed Alternative 1, § 1026.43(e)(3)(i) creates smooth transitions between the tiers. As noted above, under Alternative 1, the one-half percent changes in the points and fees cap between tiers would have produced the anomalous result that some smaller loans would have been permitted to include a higher dollar amount of points and fees than larger loans. While proposed Alternative 2 would have avoided this problem, it would also have been somewhat more complex, thereby increasing the risk of errors. The tiers in § 1026.43(e)(3)(i) all feature easy-to-calculate limits, making compliance easier.

Finally, the three lower tiers are tied to the comparable thresholds for high-cost loans to ensure that the points and fees on loans that satisfy the qualified mortgage standard do not trigger the additional obligations of high-cost mortgages. Under TILA as amended, a high-cost mortgage has points and fees equal to 5 percent of the total transaction amount if the transaction is $20,000 or more, and points and fees equal to the lesser of 8 percent of the total transaction amount or $1,000, if the transaction is less than $20,000. See TILA section 103(bb)(1)(A)(ii)(I) and (II). Setting the maximum points and fees caps based on the HOEPA triggers will help ensure that a qualified mortgage is not a high-cost mortgage because of the points and fees.

Proposed comment 43(e)(3)(i)-1 would have cross-referenced comment 32(a)(ii)-1 for an explanation of how to calculate the “total loan amount.” The Bureau adopts comment 43(e)(3)(i)-1 substantially as proposed, but it adds an explanation for tiers in which the prescribed points and fees limit is a fixed dollar amount rather than a percentage and revises the cross-reference because the explanation of calculating “total loan amount” is moved to comment 32(b)(5)(i)-1.

Proposed comment 43(e)(3)(i)-2 would have explained that a creditor must determine which category the loan falls into based on the face amount of the note (the “loan amount”), but must apply the allowable points and fees percentage to the “total loan amount,” which may be an amount that is different than the face amount of the note. The Bureau adopts comment 43(e)(3)(i)-2 substantially as proposed, but it revises some of the limits to reflect the changes described above.

Proposed comment 43(e)(3)(i)-3 would have provided examples of calculations for different loan amounts. The Bureau adopts comment 43(e)(3)(i)-3 with revisions to reflect the changes to some of the limits described above.

Impact on Rural Areas and Other Areas Where Home Values Are Lower

TILA section 129C(b)(2)(D) requires the Bureau to consider the rules' potential impact on “rural areas and other areas where home values are lower.” The Bureau considered the concerns raised by industry commenters that if the limits on points and fees for smaller loans were set too low, access to credit could be impaired, in particular for low income, minority, and rural consumers, and first-time home buyers. Setting the threshold for smaller loans too low may also negatively affect access to credit for manufactured housing, which disproportionately serves lower-income consumers and rural areas. The higher threshold and higher limits on points and fees for smaller loans should help to ensure that creditors are able to offer qualified mortgages in rural areas and other areas where home values are lower.

The Bureau declines to adopt the recommendation of one commenter that it exempt smaller loans for manufactured homes from the points and fees triggers for high-cost mortgages. Section 1431 of the Dodd Frank Act provides that a loan of $20,000 or more is deemed a high-cost mortgage if total points and fees exceed 5 percent of the total transaction amount and that a loan of less than $20,000 is deemed a high-cost mortgage if total points and fees exceed the lesser of 8 percent of the total transaction amount or $1,000, or other such dollar amount as the Bureau may prescribe by regulations. Such a change is beyond the scope of this rulemaking and is more appropriately addressed in the parallel HOEPA rulemaking.

43(e)(3)(ii)

Bona Fide Third-party Charges and Bona Fide Discount Points

As discussed in the section-by-section analysis of § 1026.32(b)(1)(i), the Bureau is moving the provisions excluding certain bona fide third-party charges and bona fide discount points to § 1026.32(b)(1)(i)(D) through (F). The Board had proposed to implement these provisions in proposed § 226.43(e)(3)(ii) through (iv).

Indexing Points and Fees Limits for Inflation

The Board requested comment on whether the loan size ranges for the qualified mortgage points and fees limits should be indexed for inflation. A few industry commenters recommended that the loan size ranges or the permitted dollar amounts of points and fees be adjusted for inflation. The Bureau believes that it is appropriate to adjust the points and fees limits to reflect inflation. In addition, the Bureau notes that, as prescribed by TILA section 103(aa)(3), what was originally a $400 points and fees limit for high-cost loans has been adjusted annually for inflation, and that the dollar amounts of the new high-cost points and fees thresholds in TILA section 103(bb)(1)(A)(ii)(II) will also be adjusted annually for inflation. The Bureau believes the points and fees thresholds for high-cost loans and qualified mortgages should be treated consistently with respect to inflation adjustments. Accordingly, in new § 1026.43(e)(3)(ii), the Bureau provides that the dollar amounts, including the loan amounts, shall be adjusted annually to reflect changes in the Consumer Price Index for All Urban Consumers (CPI-U). The adjusted amounts will be published in new comment 43(e)(3)(ii)-1.

43(e)(4) Qualified Mortgage Defined—Special Rules

As discussed above, the Bureau is finalizing the general qualified mortgage definition in § 1026.43(e)(2). Under that definition, qualified mortgages would be limited to loans that satisfy the qualified mortgage product feature criteria in the statute (including prohibitions on certain risky loan features, limitations on points and fees, and the requirement to underwrite to the maximum rate in the first five years of the loan), for which the creditor considers and verifies the consumer's income and assets and current debt obligations, alimony, and child support, and for which the consumer's total (or “back-end”) debt-to-income ratio is less than or equal to 43 percent.

The Bureau believes this approach establishes an appropriate benchmark over the long term for distinguishing which loans should be presumed to meet the ability-to-repay requirements under the Dodd-Frank Act, while also leaving room for the provision of responsible mortgage credit over time to consumers with higher debt-to-income ratios under the general ability-to-repay requirements. However, the Bureau acknowledges it may take some time for the non-qualified mortgage market to establish itself in light of the market anxiety regarding litigation risk under the ability-to-repay rules, the general slow recovery of the mortgage market, and the need for creditors to adjust their operations to account for several other major regulatory and capital regimes. In light of these factors, the Bureau has concluded that it is appropriate to provide a temporary alternative definition of qualified mortgage. This will help ensure access to responsible, affordable credit is available for consumers with debt-to-income ratios above 43 percent and facilitate compliance by creditors by promoting the use of widely recognized, federally-related underwriting standards.

Under this temporary provision, as a substitute for the general qualified mortgage definition in § 1026.43(e)(2), which contains a 43 percent debt-to-income ratio threshold, the final rule provides a second definition of qualified mortgage in § 1026.43(e)(4) for loans that meet the prohibitions on certain risky loan features (e.g., negative amortization and interest only features) and the limitations on points and fees under § 1026.43(e)(2) and are eligible for purchase or guarantee by the GSEs, while under the conservatorship of the FHFA, or eligible to be insured or guaranteed by the U.S. Department of Housing and Urban Development under the National Housing Act (12 U.S.C. 1707 et seq.) (FHA), the VA, the USDA, or the Rural Housing Service (RHS). The FHA, VA, USDA, and RHS have authority under the statute to define qualified mortgage standards for their own loans, so coverage under § 1026.43(e)(4), will sunset once each agency promulgates its own qualified mortgage standards, and such rules take effect. See TILA section 129C(b)(3)(ii). Coverage of GSE-eligible loans will sunset when conservatorship ends.

Even if the Federal agencies do not issue additional rules or conservatorship does not end, the temporary qualified mortgage definition in § 1026.43(e)(4) will expire seven years after the effective date of the rule. The Bureau believes that this will provide an adequate period for economic, market, and regulatory conditions to stabilize. Because the Bureau is obligated by statute to analyze the impact and status of the ability-to-repay rule five years after its effective date, the Bureau will have an opportunity to confirm that it is appropriate to allow the temporary provision to expire prior to the sunset. Covered transactions that satisfy the requirements of § 1026.43(e)(4) that are consummated before the sunset of § 1026.43(e)(4) will retain their qualified mortgage status after the temporary definition expires. However, a loan consummated after the sunset of § 1026.43(e)(4) may only be a qualified mortgage if it satisfies the requirements of § 1026.43(e)(2) or (f).

The alternative definition of qualified mortgage recognizes that the current mortgage market is especially fragile as a result of the recent mortgage crisis. It also recognizes the government's extraordinary efforts to address the crisis; GSE-eligible loans, together with the other federally insured or guaranteed loans, cover roughly 80 percent of the current mortgage market. In light of this significant Federal role and the government's focus on affordability in the wake of the mortgage crisis, the Bureau believes it is appropriate, for the time being, to presume that loans that are eligible for purchase, guarantee, or insurance by the designated Federal agencies and the GSEs while under conservatorship have been originated with appropriate consideration of consumers' ability to repay, where those loans also satisfy the requirements of § 1026.43(e)(2) concerning restrictions on product features and total points and fees limitations. The temporary definition is carefully calibrated to provide a reasonable transition period to the general qualified mortgage definition, including the 43 percent debt-to-income ratio requirement. While this temporary definition is in effect, the Bureau will monitor the market to ensure it remains appropriate to presume that the loans falling within those programs have been originated with appropriate consideration of the consumer's repayment ability. The Bureau believes this temporary approach will ultimately benefit consumers by minimizing any increases in the cost of credit as a result of this rule while the markets adjust to the new regulations.

The Bureau believes this temporary alternative definition will provide an orderly transition period, while preserving access to credit and effectuating the broader purposes of the ability-to-repay statute during the interim period. The Bureau believes that responsible loans can be made above a 43 percent debt-to-income ratio threshold, and has consciously structured the qualified mortgage requirements in a way that leaves room for responsible lending on both sides of the qualified mortgage line. The temporary exception has been carefully structured to cover loans that are eligible to be purchased, guaranteed, or insured by the GSEs (while in conservatorship) or Federal agencies regardless of whether the loans are actually so purchased, guaranteed, or insured; this will leave room for private investors to return to the market and secure the same legal protection as the GSEs and Federal agencies. At the same time, as the market recovers and the GSEs and FHA are able to reduce their presence in the market, the percentage of loans that are granted qualified mortgage status under the temporary definition will shrink towards the long-term structure.

In addition to being a loan that is eligible to be made, guaranteed, or insured by the above-described Federal agencies or the GSEs while in conservatorship, to meet the definition of qualified mortgage under § 1026.43(e)(4), the loan must satisfy the statutory qualified mortgage criteria regarding prohibitions on certain risky loan features and limitations on points and fees. Specifically, § 1026.43(e)(4)(i) provides that, notwithstanding § 1026.43(e)(2), a qualified mortgage is a covered transaction that satisfies the requirements of § 1026.43(e)(2)(i) through (iii). As discussed above, those provisions require: that the loan provide for regular periodic payments that do not result in an increase of the principal balance, allow the consumer to defer repayment of principal, or result in a balloon payments; that the loan term does not exceed 30 years; and that the total points and fees payable in connection with the loan do not exceed the threshold set forth in § 1026.43(e)(3). As described further below, the temporary definition does not include requirements to (1) verify and document the consumer's income or assets relied upon in qualifying the consumer; (2) underwrite a fixed rate loan based on a payment schedule that fully amortizes the loan over the term and takes into account all applicable taxes, insurance, and assessments; or (3) underwrite an adjustable-rate loan using the maximum interest rate permitted in the first five years. The Bureau highlights that a loan need not be actually purchased or guaranteed by the GSEs or insured or guaranteed by the above-listed Federal agencies to qualify for the temporary definition in § 1026.43(e)(4). Rather, the loan need only be eligible for such purchase, guarantee, or insurance.

Notably, the temporary qualified mortgage definition does not include “jumbo loans.” The Bureau does not believe that creditors making jumbo loans need the benefit of the temporary exception, as the Bureau views the jumbo market as already robust and stable. Jumbo loans can still be qualified mortgages if they meet the general rule (i.e. are within the 43 percent debt-to-income ratio and underwritten in accordance with the general qualified mortgage requirements).

Section 1026.43(e)(4)(iii) contains the sunset provisions for the special qualified mortgage definition in § 1026.43(e)(4). Specifically, § 1026.43(e)(4)(iii)(A) provides that each respective special rule in § 1026.43(e)(4)(ii)(B) (FHA loans), (e)(4)(ii)(C) (VA loans), (e)(4)(ii)(D) (USDA loans); and (e)(4)(ii)(E) (RHS loans) shall expire on the effective date of a rule issued by each respective agency pursuant to its authority under TILA section 129C(b)(3)(ii) to define a qualified mortgage. Section 1026.43(e)(4)(iii)(B) provides that, unless otherwise expired under § 1026.43(e)(4)(iii)(A), the special rules in § 1026.43(e)(4) are available only for covered transactions consummated on or before a date that is seven years after the effective date of this rule.

Comment 43(e)(4)-1 provides additional clarification regarding the special qualified mortgage definition. Specifically, the comment provides that, subject to the sunset provided under § 1026.43(e)(4)(iii), § 1026.43(e)(4) provides an alternative definition of qualified mortgage to the definition provided in § 1026.43(e)(2). To be a qualified mortgage under § 1026.43(e)(4), the creditor must satisfy the requirements under §§ 1026.43(e)(2)(i) through (iii), in addition to being one of the types of loans specified in §§ 1026.43(e)(4)(ii)(A) through (E).

Comment 43(e)(4)-2 clarifies the effect that a termination of conservatorship would have on loans that satisfy the qualified mortgage definition under § 1026.43(e)(4) because of their eligibility for purchase or guarantee by Fannie Mae or Freddie Mac. The comment provides that § 1026.43(e)(4)(ii)(A) requires that a covered transaction be eligible for purchase or guarantee by Fannie Mae or Freddie Mac (or any limited-life regulatory entity succeeding the charter of either) operating under the conservatorship or receivership of the FHFA pursuant to section 1367 of the Federal Housing Enterprises Financial Safety and Soundness Act of 1992 (12 U.S.C. 4617), as amended by the Housing and Economic Recovery Act of 2008). The special rule under § 1026.43(e)(4)(ii)(A) does not apply if Fannie Mae or Freddie Mac (or any limited-life regulatory entity succeeding the charter of either) has ceased operating under the conservatorship or receivership of the FHFA. For example, if either Fannie Mae or Freddie Mac (or succeeding limited-life regulatory entity) ceases to operate under the conservatorship or receivership of the FHFA, § 1026.43(e)(4)(ii)(A) would no longer apply to loans eligible for purchase or guarantee by that entity; however, the special rule would be available for a loan that is eligible for purchase or guarantee by the other entity still operating under conservatorship or receivership.

Comment 43(e)(4)(iii)-3 clarifies that the definition of qualified mortgage under § 1026.43(e)(4) applies only to loans consummated on or before a date that is seven years after the effective date of the rule, regardless of whether Fannie Mae or Freddie Mac (or any limited-life regulatory entity succeeding the charter of either) continues to operate under the conservatorship or receivership of the FHFA. Accordingly, § 1026.43(e)(4) is available only for covered transactions consummated on or before the earlier of either: (i) The date Fannie Mae or Freddie Mac (or any limited-life regulatory entity succeeding the charter of either), respectively, cease to operate under the conservatorship or receivership of the FHFA pursuant to section 1367 of the Federal Housing Enterprises Financial Safety and Soundness Act of 1992 (12 U.S.C. 4617), as amended by the Housing and Economic Recovery Act of 2008; or (ii) a date that is seven years after the effective date of the rule, as provided by § 1026.43(e)(4)(iii).

Finally, comment 43(e)(4)(iii)-4 clarifies that, to satisfy § 1026.43(e)(4)(ii), a loan need not be actually purchased or guaranteed by the GSEs or insured or guaranteed by the FHA, VA, USFA, or RHS. Rather, § 1026.43(e)(4)(ii) requires only that the loan be eligible for such purchase, guarantee, or insurance. Rather, § 1026.43(e)(4)(ii) requires only that the loan be eligible for such purchase, guarantee, or insurance. For example, for purposes of § 1026.43(e)(4), a creditor is not required to sell a loan to Fannie Mae or Freddie Mac (or any limited-life regulatory entity succeeding the charter of either) to be a qualified mortgage. Rather, the loan must be eligible for purchase or guarantee by Fannie Mae or Freddie Mac (or any limited-life regulatory entity succeeding the charter of either), including satisfying any requirements regarding consideration and verification of a consumer's income or assets, current debt obligations, and debt-to-income ratio or residual income. To determine eligibility, a creditor may rely on an underwriting recommendation provided by Fannie Mae and Freddie Mac's Automated Underwriting Systems (AUSs) or written guide. Accordingly, a covered transaction is eligible for purchase or guarantee by Fannie Mae or Freddie Mac if: (i) The loan conforms to the standards set forth in the Fannie Mae Single-Family Selling Guide or the Freddie Mac Single-Family Seller/Servicer Guide; or (ii) the loan receives an “Approve/Eligible” recommendation from Desktop Underwriter (DU); or an “Accept and Eligible to Purchase” recommendation from Loan Prospector (LP).

The Bureau is finalizing § 1026.43(e)(4) pursuant to its authority under TILA section 129C(b)(3)(B)(i) to prescribe regulations that revise, add to, or subtract from the criteria that define a qualified mortgage upon the findings described above. The Bureau believes the temporary qualified mortgage definition is necessary and proper to ensure that responsible, affordable mortgage credit remains available to consumers in a manner consistent with the purposes of TILA section 129C and necessary and appropriate to effectuate the purposes of TILA section 129C, which includes assuring that consumers are offered and receive residential mortgage loans on terms that reasonably reflect their ability to repay the loan.

As described above, the Bureau believes that the provision of qualified mortgage status to loans that are eligible for purchase, guarantee, or to be insured by the Federal entities described above will provide a smooth transition to a more normal mortgage market. Similarly, the Bureau believes that including all loans that are eligible to be made, guaranteed, or insured by agencies of the Federal government and the GSEs while under conservatorship, will minimize the risk of disruption as the market adjusts to the ability-to-repay requirements of this rule. This adjustment to the qualified mortgage definition will also facilitate compliance with the ability-to-repay requirements. The Bureau is also finalizing § 1026.43(e)(4) pursuant to its authority under TILA section 105(a) to issue regulations with such requirements, classifications, differentiations, or other provisions, and that provide for such adjustments and exceptions for all or any class of transactions, as in the judgment of the Bureau are necessary and proper to effectuate the purposes of TILA, to prevent circumvention or evasion thereof, or to facilitate compliance therewith. For the reasons described above, the Bureau believes the adjustments to the definition of qualified mortgage are necessary to effectuate the purposes of TILA, which include the above-described purpose of TILA section 129C, among other things, and to facilitate compliance therewith.

The Bureau is exercising this authority to remove certain qualified mortgage statutory criteria, as discussed further below, and to add criteria related to eligibility for Federal agency programs and GSEs while conservatorship, as outlined above, in order to create this qualified mortgage definition.

As noted above, § 1026.43(e)(4) applies to loans that are eligible for guarantee or insurance by the Federal agencies listed above. The provisions of section 1412 apply to all residential mortgage loans, including loans that are eligible for and are guaranteed or insured by the Federal agencies listed above. However, TILA section 129C(b)(3)(B)(ii) provides the Federal agencies listed above with authority, in consultation with the Bureau, to prescribe rules defining the types of loans they insure, guarantee or administer, as the case may be, that are qualified mortgages and such rules may revise, add to, or subtract from the criteria used to define a qualified mortgage upon certain findings. Consistent with this authority, the Bureau leaves to these agencies, in consultation with the Bureau, further prescribing qualified mortgage rules defining the types of loans they respectively insure, guarantee or administer, and their rules may further revise the qualified mortgage criteria finalized in this rule with respect to these loans. In light of the Federal agencies' authority in TILA section 129C(b)(3)(B)(ii), § 1026.43(e)(4) will sunset once each agency has exercised its authority to promulgate their own qualified mortgage standards.

As noted above, the final rule does not specifically include in the temporary definition the statutory requirements to (1) verify and document the consumer's income or assets relied upon in qualifying the consumer; (2) underwrite a fixed rate loan based on a payment schedule that fully amortizes the loan over the term and takes into account all applicable taxes, insurance, and assessments; or (3) underwrite an adjustable-rate loan using the maximum interest rate permitted in the first five years. As discussed above, the Bureau believes it is appropriate, for the time being, to presume that loans that are eligible for purchase, guarantee, or insurance by the designated Federal agencies and the GSEs while under conservatorship have been originated with appropriate consideration of consumers' ability to repay where the loans satisfy the requirements of § 1026.43(e)(2) concerning restrictions on product features and total points and fees limitations. Layering additional and different underwriting requirements on top of the requirements that are unique to each loan program would undermine the purpose of the temporary definition, namely, to preserve access to credit during a transition period while the mortgage industry adjusts to this final rule and during a time when the market is especially fragile. Accordingly, as noted above, the Bureau is using its authority under TILA section 129C(b)(3)(B)(i) to remove these statutory requirements from the qualified mortgage definition in § 1026.43(e)(4). For similar reasons the Bureau is not requiring that loans that meet this qualified mortgage definition meet the 43 percent debt-to-income ratio requirement in § 1026.43(e)(2). The eligibility requirements of the GSEs and Federal agencies incorporate debt-to-income ratio thresholds. However, the GSEs and Federal agencies also permit consideration of certain compensating factors that are unique to each loan program. The Bureau declines to layer an additional debt-to-income ratio requirement to avoid undermining the purpose of the temporary qualified mortgage definition.

43(f) Balloon-Payment Qualified Mortgages Made by Certain Creditors

TILA section 129C(b)(2)(E) authorizes the Bureau to permit qualified mortgages with balloon payments, provided the loans meet four conditions. Specifically, those conditions are that: (1) The loan meets certain of the criteria for a qualified mortgage; (2) the creditor makes a determination that the consumer is able to make all scheduled payments, except the balloon payment, out of income or assets other than the collateral; (3) the loan is underwritten based on a payment schedule that fully amortizes the loan over a period of not more than 30 years and takes into account all applicable taxes, insurance, and assessments; and (4) the creditor meets four prescribed qualifications. Those four qualifications are that the creditor: (1) Operates predominantly in rural or underserved areas; (2) together with all affiliates, has total annual residential mortgage loan originations that do not exceed a limit set by the Bureau; (3) retains the balloon-payment loans in portfolio; and (4) meets any asset-size threshold and any other criteria the Bureau may establish, consistent with the purposes of this subtitle.

The four creditor qualifications are nearly identical to provisions in section 1461 of the Dodd-Frank Act, which authorizes the Bureau under TILA section 129D(c) to exempt small creditors that operate predominantly in rural or underserved areas from a requirement to establish escrow accounts for certain first-lien, higher-priced mortgage loans. Specifically, the statute authorizes creation of an exemption for any creditor that (1) operates predominantly in rural or underserved areas; (2) together with all affiliates has total annual residential mortgage transaction originations that do not exceed a limit set by the Bureau; (3) retains its mortgage debt obligations in portfolio; and (4) meets any asset-size thresholds and any other criteria that the Bureau may establish.

The Board interpreted the two provisions as serving similar but not identical purposes, and thus varied certain aspects of the proposals to implement the balloon-payment qualified mortgage and escrow provisions. Specifically, the Board interpreted the qualified mortgage provision as being designed to ensure access to credit in rural and underserved areas where consumers may be able to obtain credit only from community banks offering balloon-payment mortgages, and the escrow provision to exempt creditors that do not possess economies of scale to cost-effectively offset the burden of establishing escrow accounts by maintaining a certain minimum portfolio size from being required to establish escrow accounts on higher-priced mortgage loans. Accordingly, the two Board proposals would have used common definitions of “rural” and “underserved,” but did not provide uniformity in calculating and defining various other elements. For the balloon balloon-payment qualified mortgage provisions, for instance, the Board's proposed § 226.43(f) would have required that the creditor (1) in the preceding calendar year, have made more than 50 percent of its balloon-payment mortgages in rural or underserved areas; and (2) have assets that did not exceed $2 billion. The Board proposed two alternatives each for qualifications relating to (1) the total annual originations limit; and (2) the retention of balloon-payment mortgages in portfolio. The proposal also would have implemented the four conditions for balloon-payment qualified mortgages under TILA section 129C(b)(2)(E) and used its adjustment and exception authority to add a requirement that the loan term be five years or longer.

In contrast, the Board's proposal for the escrows exemption under proposed § 226.45(b)(2)(iii) would have required that the creditor have (1) in the preceding calendar year, have made more than 50 percent of its first-lien mortgages in rural or underserved areas; (2) together with all affiliates, originated and retained servicing rights to no more than 100 first-lien mortgage debt obligations in either the current or prior calendar year; and (3) together with all affiliates, not maintained an escrow account on any consumer credit secured by real property. The Board also sought comment on whether to add a requirement for the creditor to meet an asset-size limit and what that size should be.

In both cases, the Board proposed to use a narrow definition of rural based on the Economic Research Service (ERS) of the USDA's “urban influence codes” (UICs). The UICs are based on the definitions of “metropolitan statistical areas” of at least one million residents and “micropolitan statistical areas” with a town of at least 2,500 residents, as developed by the Office of Management and Budget, along with other factors reviewed by the ERS that place counties into twelve separately defined UICs depending on the size of the largest city and town in the county. The Board's proposal would have limited the definition of rural to certain “non-core” counties that are not located in or adjacent to any metropolitan or micropolitan area. This definition corresponded with UICs of 7, 10, 11, and 12, which would have covered areas in which only 2.3 percent of the nation's population lives.

In light of the overlap in criteria between the balloon-payment qualified mortgage and escrow exemption provisions, the Bureau considered comments responding to both proposals in determining how to finalize the particular elements of each rule as discussed further below. With regard to permitting qualified mortgages with balloon payments generally, consumer group commenters stated that the balloon-payment qualified mortgage exemption is a discretionary provision, as TILA section 129C(b)(2)(E) states that the Bureau “may” provide an exemption for balloon-payment mortgages to be qualified mortgages, and stated that such an exemption should not be provided in the final rule because such exemption would have a negative effect on consumers' access to responsible and affordable credit. Trade association and industry commenters generally supported the balloon-payment qualified mortgage exemption, with some comments related to the specific provisions that are discussed below. One trade association commented that the exemption should extend to all balloon-payment mortgages held in portfolio by financial institutions; as such a broader exemption would achieve Congress's intent as well as reduce the difficulty that creditors would have in complying with the requirements in the proposal. Three trade associations and several industry commenters commented that the balloon-payment qualified mortgage exemption was needed to ensure access to credit for consumers in rural areas because smaller institutions in those areas use balloon-payment mortgages to control interest rate risk.

The Bureau believes Congress enacted the exemption in TILA section 129C(b)(2)(E) because it was concerned that the restrictions on balloon-payment mortgages under the ability to repay and general qualified mortgage provisions might unduly constrain access to credit in rural and underserved areas, where consumers may be able to obtain credit only from a limited number of creditors, including some community banks that may offer only balloon-payment mortgages. Because Congress explicitly set out detailed criteria, indicating that it did not intend to exclude balloon-payment mortgages from treatment as qualified mortgages that meet those criteria, and the Bureau is implementing the statutory exemption for balloon-payment mortgages to be qualified mortgages provided they meet the conditions described below. The Bureau believes those criteria reflect a careful judgment by Congress concerning the circumstances in which the potential negative impact from restricting consumers' access to responsible and affordable credit would outweigh any benefit of prohibiting qualified mortgages from providing for balloon payments. The Bureau therefore believes that the scope of the exemption provided in this final rule implements Congress's judgment as to the proper balance between those two imperatives.

The Bureau believes that there are compelling reasons underlying Congress's decision not to allow balloon-payment mortgages to enjoy qualified-mortgage status except in carefully limited circumstances. It is the rare consumer who can afford to make a balloon payment when due. Thus, ordinarily a consumer facing a balloon payment must obtain new financing. Depending on market conditions at the time and also the consumer's own economic circumstances, consumers may find it difficult to obtain affordable credit. Some consumers may be forced to sell their homes to pay off the balloon-payment mortgage. Others may find it necessary to take on a new loan on terms that create hardships for the consumers. Unscrupulous lenders may seek to take advantage of consumers faced with the necessity of making a balloon payment by offering loans on predatory terms.

On the other hand, in rural and other underserved areas, it is not uncommon for consumers to seek a mortgage loan of a type that cannot be sold on the secondary market, because of special characteristics of either the property in question or the consumer. Many community banks make mortgages that are held in portfolio in these circumstances. To manage interest rate risk and avoid complexities in originating and servicing adjustable rate mortgages, these banks generally make balloon-payment mortgage loans which the banks roll over, at then current market interest rate, when the balloon-payment mortgage comes due. For example, data available through the National Credit Union Administration indicates that among credit unions which make mortgages in rural areas (using the definition of rural described below), 25 percent make only balloon-payment or hybrid mortgages.

There are also substantial data suggesting that the small portfolio creditors that are most likely to rely on balloon-payment mortgages to manage their interest rate risks (or to have difficulty maintaining escrow accounts) have a significantly better track record than larger creditors with regard to loan performance. As discussed in more depth in the 2013 ATR Concurrent Proposal, because small portfolio lenders retain a higher percentage of their loans on their own books, they have strong incentives to engage in thorough underwriting. To minimize performance risk, small community lenders have developed underwriting standards that are different than those employed by larger institutions. Small lenders generally engage in “relationship banking,” in which underwriting decisions rely at least in part on qualitative information gained from personal relationships between lenders and consumers. This qualitative information focuses on subjective factors such as consumer character and reliability which “may be difficult to quantify, verify, and communicate through the normal transmission channels of banking organization.” While it is not possible to disaggregate the impact of each of the elements of the community banking model, the combined effect is highly beneficial. Moreover, where consumers have trouble paying their mortgage debt obligations, small portfolio creditors have strong incentives to work with the consumers to get them back on track, both to protect the creditors' balance sheets and their reputations in their local communities. Market-wide data demonstrate that loan delinquency and charge-off rates are significantly lower at smaller banks than larger ones.

The Bureau believes that these kinds of considerations underlay Congress's decision to authorize the Bureau to establish an exemption under TILA section 129C(b)(2)(E) to ensure access to credit in rural and underserved areas where consumers may be able to obtain credit only from such community banks offering these balloon-payment mortgages. Thus, the Bureau concludes that exercising its authority is appropriate, but also that the exemption should implement the statutory criteria to ensure it effectuates Congress's intent. Accordingly, as discussed in more detail below, the Bureau adopts § 1026.43(f) largely as proposed but with certain changes described below to implement TILA section 129C(b)(2)(E).

In particular, the Bureau has concluded that it is appropriate to make the specific creditor qualifications much more consistent between the balloon-payment qualified mortgage and escrow exemptions than originally proposed by the Board. The Bureau believes that this approach is justified by several considerations, including the largely identical statutory language, the similar congressional intents underlying the two provisions, and the fact that requiring small creditors operating predominantly in rural or underserved areas to track overlapping but not identical sets of technical criteria for each separate provision could create unwarranted compliance burden that itself would frustrate the intent of the statutes. Although the Bureau has recast and loosened some of the criteria in order to promote consistency, the Bureau has carefully calibrated the changes to further the purposes of each rulemaking and in light of the evidence suggesting that small portfolio lenders' relationship banking model provides significant consumer protections in its own right.

For the foregoing reasons, the Bureau is adopting § 1026.43(f)(1)(vi) to implement TILA section 129C(b)(2)(E)(iv) by providing that a balloon loan that meets the other criteria specified in the regulation is a qualified mortgage if the creditor: (1) In the preceding calendar year made more than 50 percent of its covered transactions secured by a first lien in counties designated by the Bureau as “rural” or “underserved”; (2) together with all affiliates extended 500 or fewer first-lien covered transactions in the preceding calendar year; and (3) has total assets that are less than $2 billion, adjusted annually for inflation. The final rule also creates greater parallelism with the escrow provision with regard to the requirement that the affected loans be held in portfolio by requiring in both rules that the transactions not be subject to a “forward commitment” agreement to sell the loan at the time of consummation. These qualifications and the other requirements under the final rule are discussed in more detail below.

43(f)(1) Exemption

The Bureau believes that the provisions of TILA section 129C(b)(2)(E) are designed to require that balloon-payment qualified mortgages meet the same criteria for qualified mortgages as described in TILA section 129C(b)(2)(A), except where the nature of the balloon-payment mortgage itself requires adjustment to the general rules. In TILA section 129C(b)(2)(A), a qualified mortgage cannot allow the consumer to defer repayment of principal. Deferred principal repayment may occur if the payment is applied to both accrued interest and principal but the consumer makes periodic payments that are less than the amount that would be required under a payment schedule that has substantially equal payments that fully repay the loan amount over the loan term. The scheduled payments that fully repay a balloon-payment mortgage over the loan term include the balloon payment itself and, therefore, are not substantially equal. Thus, balloon-payment mortgages permit the consumer to defer repayment of principal. Additionally, a qualified mortgage must explicitly fully amortize the loan amount over the loan term and explicitly cannot result in a balloon payment under TILA section 129C(b)(2)(A). Since TILA section 129C(b)(2)(A) contains these provisions, TILA section 129C(b)(2)(E) exempts balloon-payment qualified mortgages from meeting those requirements. TILA section 129C(b)(2)(E) has additional requirements that a creditor consider the consumer's ability to repay the scheduled payments using a calculation methodology appropriate for a balloon-payment mortgage.

Accordingly, the Bureau is adjusting the ability-to-repay requirements generally applicable to qualified mortgages under § 1026.43(e)(2) for the balloon-payment qualified mortgage exemption. Requirements that are the same in both the generally applicable qualified mortgage requirements and the balloon-payment qualified mortgage exemption are specifically described in paragraph (f)(1)(i). The requirements in the generally applicable qualified mortgage requirements that are inapplicable, for the reasons described below, to the balloon-payment qualified mortgage exemption are replaced by requirements in paragraph (f)(1)(ii), (iii) and (iv) that specifically address the provisions inherent in balloon-payment mortgages.

43(f)(1)(i)

TILA section 129C(b)(2)(E)(i) requires that a balloon-payment qualified mortgage meet all of the criteria for a qualified mortgage, except for the provisions that require the loan to have: (1) Regular periodic payments that provide for the complete repayment of principal over the loan term, (2) terms that do not result in a balloon payment, and (3) a payment schedule that fully amortizes the mortgage over the loan term taking into account all applicable taxes, insurance and assessments. The Board's proposed § 226.43(f)(1)(i) would have implemented this provision by requiring that balloon-payment qualified mortgages meet the same requirements for other qualified mortgages, except for specific provisions of § 226.43(e)(2) that would not have to be considered. Commenters did not address these requirements specifically. The Bureau is adopting § 1026.43(f)(1)(i) to implement TILA section 129C(b)(2)(E)(i) by providing that a balloon-payment qualified mortgage must meet the criteria for a qualified mortgage as required by § 1026.43(e)(2)(i)(A), (e)(2)(ii), (e)(2)(iii), and (e)(2)(v). These requirements are similar to the requirements in the Board's proposal, except that they are stated as affirmative requirements instead of excluding qualified mortgage requirements that are not required to be considered for balloon-payment qualified mortgages.

Section 1026.43(f)(1)(i), by exclusion, exempts balloon-payment qualified mortgages from the requirements in § 1026.43(e)(2)(i)(B), (e)(2)(i)(C), (e)(2)(iv), and (e)(2)(vi), which use calculation methodologies that would make the origination of balloon-payment qualified mortgages difficult, if not impossible. The requirements in subsequent provisions of § 1026.43(f)(1) are adopted below to require the consideration of scheduled payments and the debt-to-income ratio made in conjunction with alternative calculation methodologies that are appropriate for balloon-payment qualified mortgages.

Comment 43(f)(1)(i)-1 clarifies that a balloon-payment qualified mortgage under this exemption must provide for regular periodic payments that do not result in an increase of the principal balance as required by § 1026.43(e)(2)(i)(A), must have a loan term that does not exceed 30 years as required by § 1026.43(e)(2)(ii), must have total points and fees that do not exceed specified thresholds pursuant to § 1026.43(e)(2)(iii), and must satisfy the consideration and verification requirements in § 1026.43(e)(2)(v).

43(f)(1)(ii)

TILA section 129C(b)(2)(E)(ii) requires a creditor making a balloon-payment qualified mortgage to determine that the consumer is able to make all scheduled payments, except the balloon payment, out of income and assets other than the collateral. TILA section 129C(b)(2)(E)(iii) requires a creditor making a balloon-payment qualified mortgage to determine, among other things, that the scheduled payments include mortgage-related obligations. Proposed § 226.43(f)(1)(ii) would have required that the creditor determine that the consumer can make all of the scheduled payments, except for the balloon payment, from the consumer's current or reasonably expected income or assets other than the dwelling that secures the loan. Commenters did not address this requirement specifically. The Bureau is adopting § 1026.43(f)(1)(ii) to implement TILA section 129C(b)(2)(E)(ii) and a portion of TILA section 129C(b)(2)(E)(iii) by requiring a creditor to determine that the consumer can make all of the payments under the terms of the legal obligation, as described in § 1026.43(f)(1)(iv)(A), together with all mortgage-related obligations and excluding the balloon payment, from the consumer's income or assets other than the dwelling that secures the loan. Comment 43(f)(1)(ii)-1 provides an example to illustrate the calculation of the monthly payment on which this determination must be based. Comment 43(f)(1)(ii)-2 provides additional clarification on how a creditor may make the required determination that the consumer is able to make all scheduled payments other than the balloon payment.

43(f)(1)(iii)

TILA section 129C(b)(3)(B)(i) permits the addition of additional requirements or revision of the criteria that define a qualified mortgage upon the finds discussed below. The Board's proposal did not include an explicit requirement to consider the consumer's debt-to-income ratio in relation to a balloon-payment qualified mortgage. The Board, however, sought comment on what criteria should be included in the definition of a qualified mortgage to ensure that the definition provides an incentive to creditors to make qualified mortgages, while also ensuring that consumers have the ability to repay qualified mortgages. One commenter advocated eliminating the balloon-payment qualified mortgage exemption completely as they recommended that balloon-payment mortgages should not be permitted at all, but rather suggested that the Board and Bureau take steps to make the balloon-payment qualified mortgage exemption rare.

As discussed above with regard to other categories of qualified mortgages, the Bureau believes consideration of debt-to-income ratio or residual income is fundamental to any determination of ability to repay. A consumer is able to repay a loan if he or she has sufficient funds to pay his or her other obligations and expenses and still make the payments required by the terms of the loan. Thus, debt-to-income comparisons provide a valuable predictive metric in assessing the consumer's repayment ability. The Bureau believes that it would be inconsistent with congressional intent to have balloon-payment qualified mortgages not meet those same requirements, as modified to the particular nature of a balloon-payment mortgage.

Accordingly, the Bureau is adopting § 1026.43(f)(1)(iii) to provide that, to make a balloon-payment qualified mortgage, a creditor must consider and verify the consumer's monthly debt-to-income ratio or residual income in accordance with § 1026.43(c)(7) by using the calculation methodology described in § 1026.43(f)(iv)(A), together with all mortgage-related obligations and excluding the balloon payment. Comment 43(f)(1)(iii)-1 clarifies that the calculation required under § 1026.43(c)(7)(i)(A) should be made using the payment calculation methodology under § 1026.43(f)(1)(iv)(A), together with all mortgage-related obligations and excluding the balloon payment, in order to comply with § 1026.43(f)(1)(iii).

At the same time, however, the Bureau declines to impose a specific debt-to-income or residual threshold for this category of qualified mortgages because, as discussed above, the Bureau believes that small creditors excel at making highly individualized determinations of ability to repay that take into consideration the unique characteristics and financial circumstances of the particular consumer. While the Bureau believes that many creditors can make mortgage loans with consumer debt-to-income ratios above 43 percent that consumers are able to repay, the Bureau believes that portfolio loans made by small creditors are particularly likely to be made responsibly and to be affordable for the consumer even if such loans exceed the 43 percent threshold. The Bureau therefore believes that it is appropriate to presume compliance even above the 43 percent threshold for small creditors who meet the other criteria in § 1026.43(f). The Bureau believes that the discipline imposed when small creditors make loans that they will hold in their portfolio is sufficient to protect consumers' interests in this regard. Because the Bureau is not proposing a specific limit on consumer debt-to-income ratio, the Bureau does not believe it is necessary to require creditors to calculate debt-to-income ratio in accordance with a particular standard such as that set forth in appendix Q.

In adopting this requirement, the Bureau is adding a condition for a balloon-payment qualified mortgage that is not established by TILA section 129C(b)(2)(E). The Bureau adds this condition pursuant to TILA section 129C(b)(3)(B)(i), which authorizes the Bureau “to revise, add to, or subtract from the criteria that define a qualified mortgage upon a finding that such regulations are necessary or proper to ensure that responsible, affordable mortgage credit remains available to consumers in a manner consistent with the purposes of this section, necessary and appropriate to effectuate the purposes of this section and Section 129B, to prevent circumvention or evasion thereof, or to facilitate compliance with such sections.” A purpose of TILA section 129C, among other things, is to ensure that consumers are offered and receive loans on terms that they are reasonably able to repay. See TILA section 129B(a)(2). The Bureau believes that a creditor considering and verifying the consumer's monthly debt-to-income ratio or residual income in order for the balloon-payment mortgage to qualify as a balloon-payment qualified mortgage is necessary, proper, and appropriate both to effectuate the purposes of TILA section 129C to prevent circumvention or evasion thereof and to ensure that responsible, affordable mortgage credit remains available to consumers in a manner consistent with the purposes of this section. For these reasons, the Bureau believes that § 1026.43(f)(1)(iii), in requiring a creditor considering and verifying the consumer's monthly debt-to-income ratio or residual income in order for the balloon-payment mortgage to qualify as a balloon-payment qualified mortgage, effectuates the purposes of TILA section 129C and prevents circumvention or evasion thereof.

In addition the Bureau invokes its authority under section 105(a) in order to add the above qualification for a balloon-payment qualified mortgage. Section 105(a) authorizes the Bureau to issue regulations that, among other things, contain such additional requirements, other provisions, or that provide for such adjustments for all or any class of transactions, that in the Bureau's judgment are necessary or proper to effectuate the purposes of TILA, which include the above purpose of section 129C, among other things. See 15 U.S.C. 1604(a). The Bureau believes that this addition to the qualified mortgage criteria is necessary and proper to achieve this purpose.

43(f)(1)(iv)

TILA section 126C(b)(2)(E)(iii) and the Board proposal require that the loan be underwritten with specific payment calculation methodologies to qualify as a balloon-payment qualified mortgage. The underwriting of a loan is based on the terms of the legal obligation. The general requirements of a qualified mortgage in § 1026.43(e)(2) govern loans secured by real property or a dwelling with multiple methods of payment calculations, terms, and conditions. However, unlike other the types of qualified mortgage, the balloon-payment qualified mortgage deals with a specific type of transaction, a balloon-payment mortgage, with specific characteristics that are described in the legal obligation. Therefore, the Bureau considers the requirement of TILA section 129C(b)(2)(E)(iii) to be requirements relating to the terms of the legal obligation of the loan. Accordingly, the Bureau is adopting § 1026.43(f)(1)(iv), requiring the legal obligation of a balloon-payment qualified mortgage to have the following terms: (1) Scheduled payments that are substantially equal and calculated on an amortization period that does not exceed 30 years; (2) the interest rate does not vary during the loan term, and (3) the loan term is for five years or longer.

Scheduled Payments

TILA section 129C(b)(2)(E)(iii) requires that a balloon-payment qualified mortgage must be underwritten based on a payment schedule that fully amortizes the loan over a period of not more than 30 years and takes into account all applicable taxes, insurance, and assessments. The Board's proposed § 226.43(f)(1)(iii) incorporated this statutory requirement. Commenters did not address this requirement specifically.

The Bureau is adopting the Board's proposal and implements § 1026.43(f)(1)(iv) to require that the scheduled payments, on which the determinations required by § 1026.43(f)(1)(ii) and (f)(1)(iii) are based, are calculated using an amortization period that does not exceed 30 years. The requirement that the payments include all mortgage-related obligations is required as part of § 1026.43(f)(1)(ii), above. The Bureau believes that the underwriting referenced in TILA section 129C(b)(2)(E)(iii) corresponds to the determination of the consumer's repayment ability referenced in TILA section 129C(b)(2)(E)(ii). Comment 43(f)(1)(iv)-1 clarifies that the amortization period used to determine the scheduled periodic payments that the consumer must pay under the terms of the legal obligation may not exceed 30 years.

In its proposal, the Board sought comment on whether a balloon-payment mortgage with interest-only payments should qualify for the balloon-payment exemption. One association of State bank regulators commented that loans with interest-only payments would be properly excluded from the exemption in order to permit the exemption to be available only to those institutions that appropriately utilize the balloon-payment mortgages to mitigate interest rate risk. The Bureau agrees with this assessment and believes that permitting interest-only payments would be contrary to the intent of Congress requiring amortizing payments as a requirement of a qualified mortgage, as interest-only payments do not provide any reduction in principal. Accordingly, the Bureau is adding comment 43(f)(1)(iv)-2 which clarifies that a loan that provides for interest-only payments cannot qualify for the balloon-payment qualified mortgage exemption, because it would not require the consumer to make any payments towards the principal balance of the loan contrary to the requirement that the scheduled payments result in amortization of the loan for a period that does not exceed 30 years.

Fixed Interest Rate

TILA section 129C(b)(3)(B)(i) permits the addition of additional requirements upon the finding that such regulations are necessary or proper to ensure that responsible, affordable mortgage credit remains available to consumers. The Board's proposal did not include any restrictions on the interest rate terms of the loan, but did observe that community banks appear to originate balloon-payment mortgages to hedge against interest-rate risk. The Board sought comment on what criteria should be included in the definition of a qualified mortgage to ensure that the definition provides an incentive to creditors to make qualified mortgages, while also ensuring that consumers have the ability to repay qualified mortgages.

The Bureau believes that the purpose of the exemption was to permit balloon-payment mortgages to be originated for those consumers that still need or want them, and to permit competition between creditors that address interest rate risk through the use of adjustable rate mortgages and those creditors that address interest rate risk through the use of balloon-payment mortgages. The Bureau believes that creditors that have the infrastructure and resources to originate adjustable rate mortgages do not need to resort to the use of balloon-payment mortgages to address interest rate risk. Accordingly, the Bureau is adopting § 1026.43(f)(1)(iv)(B), which requires that the legal obligation of a balloon-payment qualified mortgage must include an interest rate that will not increase during the term of the loan.

In adopting this requirement, the Bureau is adding a condition for a balloon-payment qualified mortgage that is not established by TILA section 129C(b)(2)(E). The Bureau adds this condition pursuant to TILA section 129C(b)(3)(B)(i), which authorizes the Bureau “to revise, add to, or subtract from the criteria that define a qualified mortgage upon a finding that such regulations are necessary or proper to ensure that responsible, affordable mortgage credit remains available to consumers in a manner consistent with the purposes of this section, necessary and appropriate to effectuate the purposes of this section and Section 129B, to prevent circumvention or evasion thereof, or to facilitate compliance with such sections.” A purpose of TILA section 129C is to ensure that consumers are offered and receive loans on terms that they are reasonably able to repay. See TILA section 129B(a)(2). The Bureau believes that requiring the legal obligation of a balloon-payment qualified mortgage to contain an interest rate that does not increase during the loan term is necessary, proper, and appropriate both to effectuate the purposes of TILA section 129C and to prevent circumvention or evasion thereof and to ensure that responsible, affordable mortgage credit remains available to consumers in a manner consistent with the purposes of this section. For these reasons, the Bureau believes that § 1026.43(f)(1)(iv)(B), in requiring the legal obligation of a balloon-payment qualified mortgage to contain an interest rate that does not increase during the loan term, effectuates the purposes of TILA section 129C and prevents circumvention or evasion thereof.

In addition the Bureau invokes its authority under section 105(a) in order to add the above qualification for a balloon-payment qualified mortgage. Section 105(a) authorizes the Bureau to issue regulations that, among other things, contain such additional requirements, other provisions, or that provide for such adjustments for all or any class of transactions, that in the Bureau's judgment are necessary or proper to effectuate the purposes of TILA, which include the above purpose of Section 129C, among other things. See 15 U.S.C. 1604(a). The Bureau believes that this addition to the qualified mortgage criteria is necessary and proper to achieve this purpose.

Loan Term of Five Years or Longer

TILA section 129C(b)(3)(B)(i) permits the adoption of additional requirements upon the finding that such regulations are necessary or proper to ensure that responsible, affordable mortgage credit remains available to consumers. The Board's proposed § 226.43(f)(1)(iv) would have included the addition of a requirement that a balloon-payment qualified mortgage must have a loan term of five years or longer. One association of State bank regulators and an industry trade group commented that the five-year term requirement was appropriate, as the time period is consistent with other provisions of the proposed rule. One industry trade group and one industry commenter commented that three years would be a more appropriate term because some of the creditors that would qualify under proposed § 226.43(f)(1)(v) utilize three-year terms. The Bureau is not persuaded that the exemption was meant by Congress to permit any current business practice of creditors that would satisfy the requirements of proposed § 226.43(f)(1)(v), rather the exemption was meant to provide a reasonable exemption for some balloon-payment mortgages that still meet other requirements of a qualified mortgage. The Bureau notes that the statute requires underwriting for an adjustable-rate qualified mortgage to be based on the maximum interest rate permitted during the first five years. See TILA Section 129C(b)(2)(A)(v). Therefore, the Bureau is adopting the Board's proposal by implementing § 1026.43(f)(1)(iv)(C) requiring a loan term of five years or longer because it reflects the statutory intent that five years is a reasonable period to repay a loan. Since other requirements of a qualified mortgage include a review of the mortgage over a five-year term, it would be more consistent with the intent of the exemption for the balloon-payment mortgage to have at least a five-year term.

The Bureau believes that it is appropriate to structure the exemption to prevent balloon-payment mortgages with very short loan terms from being qualified mortgages because such loans would present certain risks to consumers. A consumer with a loan term of less than five years, particularly where the amortization period is especially long, would face a balloon payment soon after consummation, in an amount virtually equal to the original loan amount. The consumer would establish little equity in the property under such terms, and if the pattern is repeated the consumer may never make any significant progress toward owning the home unencumbered. Thus, the greater the difference between a balloon-payment mortgage's amortization period and its loan term, the more likely the consumer would face this problem. The Bureau's requirement of a minimum term therefore complements the 30-year maximum amortization period prescribed by TILA section 129C(b)(2)(E)(iii).

In adopting this requirement, the Bureau is adding a condition for a balloon-payment qualified mortgage that is not established by TILA section 129C(b)(2)(E). The Bureau adds this condition pursuant to TILA section 129C(b)(3)(B)(i), which authorizes the Bureau “to revise, add to, or subtract from the criteria that define a qualified mortgage upon a finding that such regulations are necessary or proper to ensure that responsible, affordable mortgage credit remains available to consumers in a manner consistent with the purposes of this section, necessary and appropriate to effectuate the purposes of this section and Section 129B, to prevent circumvention or evasion thereof, or to facilitate compliance with such sections.” A purpose of TILA section 129C is to ensure that consumers are offered and receive loans on terms that they are reasonably able to repay. See TILA section 129B(a)(2). For the reasons discussed above, the Bureau believes that a minimum loan term for balloon-payment mortgages is necessary and appropriate both to effectuate the purposes of TILA section 129C and to prevent circumvention or evasion thereof. For these reasons, the Bureau believes that § 1026.43(f)(1)(iv)(C), in limiting the exemption for balloon-payment qualified mortgages to covered transactions with loan terms of at least five years and thus ensuring that such products truly support mortgage affordability, effectuates the purposes of TILA section 129C and prevents circumvention or evasion thereof. The Bureau also believes this minimum loan term for balloon-payment qualified mortgages is necessary, proper, and appropriate to ensure that responsible, affordable mortgage credit remains available to consumers in a manner consistent with the purposes of Section 129C.

In addition the Bureau invokes its authority under section 105(a) in order to add the above qualification for a balloon-payment qualified mortgage. Section 105(a) authorizes the Bureau to issue regulations that, among other things, contain such additional requirements, other provisions, or that provide for such adjustments for all or any class of transactions, that in the Bureau's judgment are necessary or proper to effectuate the purposes of TILA, which include the above purpose of Section 129C, among other things. See 15 U.S.C. 1604(a). The Bureau believes that this addition to the qualified mortgage criteria is necessary and proper to achieve this purpose.

43(f)(1)(v) and (vi)

TILA section 129C(b)(2)(E)(iv) includes among the conditions for a balloon-payment qualified mortgage that the creditor (1) operates predominantly in rural or underserved areas; (2) together with all affiliates, has total annual residential mortgage loan originations that do not exceed a limit set by the Bureau; (3) retains the balloon-payment loans in portfolio; and (4) meets any asset-size threshold and any other criteria as the Bureau may establish. The Board proposed § 226.43(f)(1)(v) to impose specific requirements to implement some of these elements and sought comment on alternatives to implement others. Specifically, the Board: (1) Proposed a requirement that the creditor in the preceding year made more than 50 percent of its balloon-payment mortgages in rural or underserved areas; (2) sought comment on whether to adopt an annual originations limit based on either the total volume of mortgages or the total number of mortgages made in the last year by the creditor, together with affiliates, without proposing a specific threshold; (3) sought comment on two alternatives to implement the portfolio requirement by revoking a creditor's ability to make balloon-payment qualified mortgages if the creditor sold any balloon-payment mortgages either in the last year or at any time after the final rule was adopted; and alternatives, and (4) did not have assets that exceeded $2 billion, adjusted annually for inflation.

In contrast, the Board's escrows proposal would have implemented nearly identical statutory requirements under TILA 129D(c) by requiring that the creditor (1) in the preceding calendar year, have made more than 50 percent of its first-lien mortgages in rural or underserved areas; (2) together with all affiliates, originated and retained servicing rights to no more than 100 first-lien mortgage debt obligations in either the current or prior calendar year; and (3) not be permitted to invoke the exception for any first-lien higher-priced mortgage loan that was subject to a “forward commitment” to sell the loan at the time of consummation. The Board also sought comment on whether to impose an asset limit without proposing a specific threshold, and proposed to impose a further requirement that the creditor and its affiliates not maintain escrow accounts for any other loans in order to be eligible for the exception.

As stated above, the Bureau has considered the comments received under both proposals regarding implementation of the largely identical statutory criteria, and has concluded that it is appropriate to create a much higher degree of consistency between the elements in the two individual rules. Implementation of each of the statutory elements is discussed further below.

Holding of Balloon-Payment Mortgages in Portfolio

TILA section 129C(b)(E)(iv) requires that the lender keep balloon-payment mortgages in portfolio. The Board proposed to implement this requirement by removing a creditor's eligibility for the exemption under proposed § 226.43(f)(1)(v)(C) if it sold a balloon-payment mortgage during two alternative periods, one that would cover any time after the adoption of the final rule and another that would look only to sales during the preceding or current calendar year. The Board concluded that this was the best approach to implement the statutory requirement in the qualified mortgage context because it would allow a creditor to determine at consummation whether a particular balloon-payment loan was eligible to be a qualified mortgage and allow the loan to maintain such status even if it were sold, while creating strong safeguards against gaming of the exception by revoking the creditor's ability to invoke the provisions if they began selling such loans to other holders.

In contrast, the Board's 2011 Escrows Proposal would have implemented a parallel statutory requirement under TILA section 129D(c)(3) by looking to whether the particular first-lien, higher-priced mortgage loan was subject to sale under a “forward commitment.” Forward commitments are agreements entered into at or before consummation of a transaction under which a purchaser is committed to acquire the specific loan or loans meeting specified criteria from the creditor after consummation. The Board believed that the proposal was a reasonable way to implement the statutory requirement because it would allow the creditor and consumer to determine at consummation whether an escrow requirement was required to be established; the Board reasoned that fashioning the rule in a way that would require that an escrow account be established sometime after consummation if the particular loan was transferred to a non-eligible holder would be potentially burdensome to consumers, since the consumer may not have the funds available to make a large lump-sum payment at that time. At the same time, the Board believed the rule would prevent gaming of the escrows exception because it thought that small creditors would be reluctant to make a loan that they did not intend to keep in their portfolios unless they had the assurance of a committed buyer before extending the credit.

Comments received on the escrows proposal had a divergence of opinion on how the forward commitment requirement would work in practice. One trade association commenter stated that the forward commitment requirement would prevent creditors from selling portfolio mortgage debt obligations in the future. This appears to be a misreading of the Board's 2011 Escrows Proposal, as it would not have restricted the sale of higher-priced mortgage loans. Instead, the proposed forward commitment requirement provided that, so long as the higher-priced mortgage loan was not subject to a forward commitment at the time of consummation, the higher-priced mortgage loan could be sold on the secondary market without requiring an escrow account to be established at that time. One consumer advocacy group, concerned about the possibility that creditors would use the provision to skirt the escrow requirements, suggested a blanket rule that higher-priced mortgage loans that are exempt must be maintained in the portfolio of the creditor or, alternatively, that upon sale secondary market purchasers must be required to establish escrow accounts for such mortgage debt obligations.

After consideration of these comments and further analysis of parallels between the two rulemakings, the Bureau believes that it is useful and appropriate to implement the no-forward-commitment requirement in both rules. Accordingly, the Bureau is adding § 1026.43(f)(1)(v) to provide that a loan is not eligible to be a balloon-payment qualified mortgage if it is subject, at consummation, to a commitment to be acquired by another person, other than a person that separately meets the requirements of § 1026.43(f)(1)(vi). Comment 43(f)(1)(v)-1 clarifies that a balloon-payment mortgage that will be acquired by a purchaser pursuant to a forward commitment does not satisfy the requirements of § 1026.43(f)(1)(v), whether the forward commitment refers to the specific transaction or the balloon-payment mortgage meets prescribed criteria of the forward commitment, along with an example. The Bureau believes the rationale for the balloon-payment qualified mortgage exemption is not present when a loan will be or is eligible to be acquired pursuant to a forward commitment, even if the creditor is exempt, as the creditor does not intend to retain the balloon-payment mortgage in its portfolio.

In adopting this requirement, the Bureau is adding a condition for a balloon-payment qualified mortgage that is not established by TILA section 129C(b)(2)(E). The Bureau is adopting § 1026.43(f)(1)(vi) pursuant to TILA section 129C(b)(3)(B)(i), which authorizes the Bureau “to revise, add to, or subtract from the criteria that define a qualified mortgage upon a finding that such regulations are necessary or proper to ensure that responsible, affordable mortgage credit remains available to consumers in a manner consistent with the purposes of this section, necessary and appropriate to effectuate the purposes of this section and Section 129B, to prevent circumvention or evasion thereof, or to facilitate compliance with such sections.” A purpose of TILA section 129C is to ensure that consumers are offered and receive loans on terms that they are reasonably able to repay. See TILA section 129B(a)(2). The Bureau believes that the prohibition on mortgages originated in conjunction with a forward commitment from qualifying as a balloon-payment qualified mortgage is necessary, proper, and appropriate both to effectuate the purposes of TILA section 129C and to prevent circumvention or evasion thereof. For these reasons, the Bureau believes that § 1026.43(f)(1)(v), in limiting the exemption for balloon-payment qualified mortgages to mortgages that are not originated in conjunction with a forward commitment, effectuates the purposes of TILA section 129C and prevents circumvention or evasion thereof and is necessary, proper, and appropriate to do so. Limiting balloon-payment qualified mortgages to those that are not originated in conjunction with a forward commitment effectively facilitates compliance with the statutory requirement that a balloon-payment qualified mortgage is extended by a creditor that retains the balloon-payment qualified mortgages in portfolio.

In addition the Bureau invokes its authority under section 105(a) in order to add the above qualification for a balloon-payment qualified mortgage. Section 105(a) authorizes the Bureau to issue regulations that, among other things, contain such additional requirements, other provisions, or that provide for such adjustments for all or any class of transactions, that in the Bureau's judgment are necessary or proper to effectuate the purposes of TILA, which include the above purpose of Section 129C, among other things. See 15 U.S.C. 1604(a). The Bureau believes that this addition to the qualified mortgage criteria is necessary and proper to achieve this purpose.

“Operates Predominantly in Rural or Underserved Areas”

Under TILA section 129C(b)(2)(E)(iv)(I), to qualify for the exemption, a creditor must “operate predominantly in rural or underserved areas.” The Board's proposed § 226.43(f)(1)(v)(A) would have required a creditor to have made during the preceding calendar year more than 50 percent of its total balloon-payment mortgages in “rural or underserved” areas. The Board sought comment generally on the appropriateness of the proposed approach to implement the phrase “operate predominantly.” Two trade group commenters commented that the balloon exemption should extend to all creditors that retain balloon-payment mortgages in their portfolio, and to eliminate this proposed requirement, which would have the same effect as the extension of the exemption proposed generally, discussed above.

Overall, the Bureau believes Congress enacted the exemption in TILA section 129C(b)(2)(E) to ensure access to credit in rural and underserved areas where consumers may be able to obtain credit only from such community banks or credit unions offering balloon-payment mortgages. The “operates predominantly in” requirement serves to limit the exemption to these institutions. To remove this portion of the qualifications of the creditor would be to circumvent Congress's stated requirement that the exemption was intended for creditors operating predominantly in rural and underserved areas and would potentially extend the exemption to, for example, a national bank that makes loans in rural areas and that is fully capable of putting on its balance sheet fixed rate 30-year mortgage loans or adjustable rate mortgage loans. The Bureau believes that “predominantly” indicates a portion greater than half, hence the regulatory requirement of more than 50 percent.

The Board also proposed § 226.43(f)(2) to implement this provision by defining the terms “rural” and “underserved,” which are not defined in the statute. The Board's proposed § 226.43(f)(2) established separate criteria for both rural and underserved areas. Commenters addressing the creditor qualifications under § 226.43(f)(2) discussed the definitions themselves, and did not comment on the necessity of creating definitions for the terms rural and underserved. The Bureau is adopting the Board's approach by implementing section 1026.43(f)(2) which establishes separate criteria for both “rural” and “underserved.” This means that a property could qualify for designation by the Bureau under either definition, and that covered transactions made by a creditor in either a rural or underserved area will be included in determining whether the creditor operates predominantly in such areas.

“Rural”

As described above, the Board's proposed definition of rural for purposes of both the balloon-payment qualified mortgage and escrows exception relied upon the USDA ERS “urban influence codes” (UICs). The UICs are based on the definitions of “metropolitan” and “micropolitan” as developed by the Office of Management and Budget, along with other factors reviewed by the ERS, which place counties into twelve separately defined UICs depending on the size of the largest city and town in the county. The Board's proposal would have limited the definition of rural to certain “non-core” counties that are not located in or adjacent to any metropolitan or micropolitan area. This definition corresponded with UICs of 7, 10, 11, or 12. The population that would have been covered under the Board's proposed definition was 2.3 percent of the United States population under the 2000 census. The Board believed this limited the definition of “rural” to those properties most likely to have only limited sources of mortgage credit because of their remoteness from urban centers and their resources. The Board sought comment on all aspects of this approach to defining rural, including whether the definition should be broader or narrower.

Many commenters in both rulemakings, including more than a dozen trade group commenters, several individual industry commenters, one association of State banking regulators, and a United States Senator, suggested that this definition of a rural area was too narrow and would exclude too many creditors from qualifying for the balloon-payment qualified mortgage exemption and constrain the availability of credit to rural properties. The comment from a United States Senator suggested using the eligibility of a property to secure a single-family loan under the USDA's Rural Housing Loan program as the definition of a rural property. A trade association argued that because community banks use balloon-payment mortgages to hedge against interest rate risk, the exemption should not be confined to rural areas.

The Bureau agrees that a broader definition of “rural” is appropriate to ensure access to credit with regard to both the escrows and balloon-payment qualified mortgage exemptions. In particular, the Bureau believes that all “non-core” counties should be encompassed in the definition of rural, including counties adjacent to a metropolitan area or a county with a town of at least 2,500 residents (i.e., counties with a UIC of 4, 6, and 9 in addition to the counties with the UICs included in the Board's definition). The Bureau also believes that micropolitan areas which are not adjacent to a metropolitan area should be included within the definition of rural, (i.e., counties with a UIC of 8). These counties have significantly fewer creditors originating higher-priced mortgage loans and balloon-payment mortgages than other counties. Including these counties within the definition of rural would result in 9.7 percent of the population being included within rural areas. Under this definition, only counties in metropolitan areas or in micropolitan areas adjacent to metropolitan areas would be excluded from the definition of rural.

The Bureau also considered adopting the definition of rural used to determine the eligibility of a property to secure a single family loan under the USDA's Rural Housing Loan program. For purposes of the Rural Housing Loan program, USDA subdivides counties into rural and non-rural areas. As a result, use of this definition would bring within the definition of rural certain portions of metropolitan and micropolitan counties. Given the size of some counties, particularly in western States, this approach may provide a more nuanced measure of access to credit in some areas than a county-by-county metric. However, use of the Rural Housing Loan metrics would incorporate such significant portions of metropolitan and micropolitan counties that 37 percent of the United States population would be within areas defined as rural. Based on a review of HMDA data and the location of mortgage transactions originated by HMDA reporting entities, the average number of creditors in the areas that would meet the USDA's Rural Housing Loan program definition of rural is ten. The Bureau believes that a wholesale adoption of the Rural Housing Loan definitions would therefore expand the definition of rural beyond the intent of the escrow and balloon-payment qualified mortgage exemptions under sections 1412 and 1461 of the Dodd-Frank Act by incorporating areas in which there is robust access to credit.

Accordingly, the final rule incorporates the provisions of the escrow final rule providing that a county is rural if it is neither in a metropolitan statistical area, nor in a micropolitan statistical area that is adjacent to a metropolitan statistical area. The Bureau intends to continue studying over time the possible selective use of the Rural Housing Loan program definitions and tools provided on the USDA Web site to determine whether a particular property is located within a “rural” area. For purposes of initial implementation, however, the Bureau believes that defining “rural” to include more UIC categories creates an appropriate balance to preserve access to credit and create a system that is easy for creditors to implement.

“Underserved”

The Board's proposed § 226.43(f)(2)(ii) would have defined a county as “underserved” during a calendar year if no more than two creditors extend consumer credit five or more times in that county. The definition was based on the Board's judgment that, where no more than two creditors are significantly active, the inability of one creditor to offer a balloon-payment mortgage would be detrimental to consumers who would have limited credit options because only one creditor would be left to provide the balloon-payment mortgage. Essentially, a consumer who could only qualify for a balloon-payment mortgage would be required to obtain credit from the remaining creditor in that area. Most of the same commenters that stated that the definition of rural was too narrow, as discussed above, also stated that the definition of underserved was too narrow, as well. The commenters proposed various different standards, including standards that considered the extent to which the property was in a rural area, as an alternate definition.

The Bureau believes the purpose of the exemption is to permit creditors that rely on certain balloon-payment mortgage products to continue to offer credit to consumers, rather than leave the mortgage loan market, if such creditors' withdrawal would significantly limit consumers' ability to obtain mortgage credit. In light of this rationale, the Bureau believes that “underserved” should be implemented in a way that protects consumers from losing meaningful access to mortgage credit. The Bureau is proposing to do so by designating as underserved only those areas where the withdrawal of a creditor from the market could leave no meaningful competition for consumers' mortgage business. The Bureau believes that the expanded definition of rural, as discussed above, and the purposes of the balloon-payment qualified mortgage exemption enable continued consumer ability to obtain mortgage credit.

Scope of Mortgage Operations

The Bureau has made one other change to the final rule to make the standards more consistent as between the balloon qualified mortgage and escrows exemption with regard to what type of mortgage loan operations are tracked for purposes of determining whether a creditor operates predominantly in rural or underserved areas. As noted above, the Board's proposed rule for balloon-payment qualified mortgages would have based a creditor's eligibility on the geographic distribution of its balloon-payment mortgages, while the escrows proposal focused on the distribution of first-lien mortgages. Given that the underlying statutory language regarding “operates predominantly” is the same in each instance and that tracking each type of mortgage separately would increase administrative burden, the Bureau believes it is appropriate to base the threshold for both rules on the distribution of all first-lien “covered transactions” as defined in § 1026.43(b)(1). The Bureau believes that counting all transactions will facilitate compliance, promote consistency in applying the two exemptions under both rulemakings, and be more useful in identifying which institutions truly specialize in serving rural and underserved areas. The Bureau also believes that it is appropriate to measure first-lien covered transactions because the balloon-payment mortgages that will meet the requirements of the balloon-payment qualified mortgage exemption will be first-lien covered transactions, as having subordinate financing along with the balloon-payment mortgage would be rare since it further constrains a consumer's ability to build equity in the property and able to refinance the balloon-payment mortgage when it becomes due. Accordingly, a creditor must have made during the preceding calendar year more than 50 percent of its total covered transactions secured by a first lien on property in a rural or underserved area, which is the same as the requirement of § 1026.35(b)(2)(iii)(A) in the 2013 Escrows Final Rule.

Total Annual Residential Mortgage Loan Origination

TILA section 129C(b)(2)(E)(iv)(II) requires the Bureau to establish a limitation on the “total annual residential mortgage loan originations” for a creditor seeking to fall within the balloon-payment qualified mortgage exemption. The Board's proposed § 226.43(f)(1)(v)(B) provided two alternatives to meet the statutory requirement that the creditor “together with all affiliates, has total annual residential mortgage originations that do not exceed a limit set by the Board.” TILA section 129C(b)(2)(E)(iv)(II). The first alternative was a volume based limit, and the second alternative was a total annual number of covered transactions limit. The Board's proposal did not propose any specific numeric thresholds for either alternative, but rather sought comment on the appropriate volume or number of loans originated based on the alternatives described in the proposal.

In contrast, the Board's escrow proposal would have restricted eligibility to creditors that, along with their affiliates, originate and service no more than 100 new first-lien loans per calendar year. Although the Dodd-Frank Act requirement to establish escrow accounts applies only to higher-priced mortgage loans that are secured by first liens, the Board reasoned that it was appropriate to base the threshold on all first-lien originations because creditors are free to establish escrow accounts for all of their first-lien mortgages voluntarily in order to achieve the scale necessary to escrow cost effectively. The Board estimated that a minimum servicing portfolio size of 500 is necessary to escrow cost effectively, and assumed that the average life expectancy of a mortgage loan is about five years. Based on this reasoning, the Board reasoned that creditors would no longer need the benefit of the exemption if they originated and serviced more than 100 new first-lien loans per year.

In response to the balloon-payment qualified mortgage loan proposal, two trade groups and one association of State bank regulators argued that other criteria, such as the asset-size limit or portfolio requirement, were sufficient and neither a volume nor a total annual number of covered transactions limit would be necessary. One trade group commenter suggested combining the proposed alternatives and permit creditors to pick which limit they would operate under. Other trade group and industry commenters indicated that it would be preferable to base the annual originations limit on the number of transactions rather than volume because of the varying dollar amount of loans originated, which would constrain the number of consumers with limited credit options which could obtain balloon-payment mortgages in rural or underserved areas. Four trade group and industry commenters suggested increasing the threshold for the total annual number of covered transactions by various amounts ranging from 250 to 1,000 transactions. The commenters did not articulate any particular reason or data to support the suggested limits, other than one commenter who indicated its suggestion was intended to be higher than its own amount of total annual covered transactions.

Similarly in the escrows rulemaking, commenters asserted that the 100-loan threshold was not in fact sufficient to make escrowing cost-effective. Suggestions for higher thresholds ranged from 200 to 1,000 mortgage debt obligations per year originated and serviced, though no commenters provided data to support their suggestions for alternative thresholds or to refute the Board's cost analysis. One consumer advocacy commenter suggested the proposed threshold was too high because it counted only first-lien mortgage transactions, instead of all mortgage debt obligations, but offered no specific alternative amount. Two industry commenters also suggested that the origination limit should measure only the number of higher-priced mortgage loans originated and serviced by the creditor and its affiliates.

The Bureau believes that the requirement of TILA section 129C(b)(2)(E)(iv)(II) reflects Congress's recognition that larger creditors who operate in rural or underserved areas should be able to make credit available without resorting to balloon-payment mortgages. Similarly, the requirement of TILA section 129C(d) reflects a recognition that larger creditors have the systems capability and operational scale to establish cost-efficient escrow accounts. In light of the strong concerns expressed in both rulemakings about the potential negative impacts on small creditors in rural and underserved areas, the Bureau conducted further analysis to try to determine the most appropriate thresholds, although it was significantly constrained by the fact that data is limited with regard to mortgage originations in rural areas generally and in particular with regard to originations of balloon-payment mortgages.

The Bureau started with the premise that it would be preferable to use the same annual originations threshold in both rules in order to reflect the consistent language in both statutory provisions focusing on “total annual mortgage loan originations,” to facilitate compliance avoiding requiring institutions to track multiple metrics, and to promote consistent application of the two exemptions. This requires significant reconciliation between the two proposals, however, because the escrows proposal focused specifically on loans originated and serviced in order to best gauge creditors' ability to maintain escrow accounts over time, while servicing arrangements are not directly relevant to the balloon-payment qualified mortgage. However, to the extent that creditors chose to offer balloon-payment mortgages to manage their interest rate risk without having to undertake the compliance burdens involved in administering adjustable rate mortgages over time, the Bureau believes that both provisions are focused in a broad sense on accommodating creditors whose systems constraints might otherwise cause them to exit the market.

With this in mind, the Bureau ultimately has decided to adopt a threshold of 500 or fewer annual originations of first-lien loans for both rules. The Bureau believes that this threshold will provide greater flexibility and reduce concerns that the specific threshold that had been proposed in the escrows rulemaking (100 loans originated and serviced annually) would reduce access to credit by excluding creditors who need special accommodations in light of their capacity constraints. At the same time, the increase is not as dramatic as it may first appear because the Bureau's analysis of HMDA data suggests that even small creditors are likely to sell off a significant number of loans to the secondary market. Assuming that most loans that are retained in portfolio are also serviced in house, the Bureau estimates that a creditor originating no more than 500 first-lien loans per year would maintain and service a portfolio of about 670 mortgage debt obligations over time, assuming a life expectancy of five years per mortgage debt obligation. Thus, the higher threshold will help to assure that creditors who are subject to the escrow requirements do in fact maintain portfolios of sufficient size to maintain the accounts on a cost efficient basis over time, in the event that the Board's estimate of a minimum portfolio of 500 loans was too low. However, the Bureau believes that the 500 annual originations threshold in combination with the other requirements will still assure that the balloon-payment qualified mortgage and escrow exceptions are available only to small creditors that focus primarily on a relationship-lending model and face significant systems constraints.

Asset-Size Threshold

Under TILA section 129C(b)(2)(E)(iv)(IV), to qualify for the exemption, a creditor must meet any asset-size threshold established by the Bureau. The Board's proposed § 226.43(f)(1)(v)(D) would have established the threshold for calendar year 2013 at $2 billion, with annual adjustments for inflation thereafter. Thus, a creditor would satisfy this element of the test for 2013 if it had total assets of $2 billion or less on December 31, 2012. This number was based on the limited data available to the Board at the time of the proposal. Based on that limited information, the Board reasoned that none of the entities it identified as operating predominantly in rural or underserved areas had total assets as of the end of 2009 greater than $2 billion, and therefore, the limitation should be set at $2 billion. The Board expressly proposed setting the asset-size threshold at the highest level currently held by any of the institutions that appear to be smaller institutions that served areas with otherwise limited credit options. The Board sought comment on what threshold would be appropriate and whether the asset-size test is necessary at all. Conversely, in the escrows proposal the Board did not propose an asset threshold, but rather simply requested comment on whether a threshold should be established and, if so, what it should be.

In response to the Board's 2011 ATR Proposal, one association of State bank regulators suggested that the asset-size threshold be included and be the only requirement for a creditor to qualify for the balloon-mortgage qualified mortgage exemption. Two trade group commenters suggested that a $2 billion asset-size threshold was appropriate, with one also suggesting that the asset-size threshold be the only requirement for a creditor to qualify for the balloon-mortgage qualified mortgage exemption. One industry commenter suggested that the asset-size threshold be $10 billion.

In response to the Board's 2011 Escrows Proposal, the association of State bank regulators again suggested that an asset-size threshold be the only requirement to qualify for the escrow exception, but did not propose a specific dollar threshold. A trade association suggested a threshold of $1 billion, but did not provide a rational for that amount.

For reasons discussed above, the Bureau is adopting a mortgage origination limit as contemplated by the statute. Given that limitation, restricting the asset size of institutions that can claim the exemption is of limited importance. Nonetheless, the Bureau believes that an asset limitation is still helpful because very large institutions should have sufficient resources to adapt their systems to provide mortgages without balloon payments and with escrow accounts even if the scale of their mortgage operations is relatively modest. A very large institution with a relatively modest mortgage operation also does not have the same type of reputational and balance-sheet incentives to maintain the same kind of relationship-lending model as a smaller community-based lender. Accordingly, the Bureau believes that the $2 billion asset limitation by the Board remains an appropriate limitation and should be applied in both rulemakings. Accordingly, the creditor must have total assets of less than $2 billion as of December 31, 2012, which is the same as the requirement of § 1026.35(b)(2)(iii)(C) in the 2013 Escrows Final Rule.

Criteria Creditor Also Must Satisfy in the Final Rule Adopted From the Board's 2011 Escrows Proposal

The Bureau notes that the three criteria discussed above are the same in both TILA 129C(b)(2)(E)(iv) and 129D(c). Commenters in both the Board's 2011 ATR Proposal and the Board's 2011 Escrows Proposal also made a note of the need to have consistent application of requirements and definitions across the Title XIV Rulemakings. The comments received in both of the Board's proposals identified the same concerns and made similar suggestions for each of the criteria in both the Board's 2011 ATR Proposal and 2011 Escrows Proposal. The Bureau believes the balloon-payment qualified mortgage exemption is designed to ensure access to credit in rural and underserved areas where consumers may be able to obtain credit only from a limited number of creditors. One way to ensure continued access to credit for these consumers is to reduce and streamline regulatory requirements for creditors so that creditors maintain participation in or enter these markets. One method by which this can be accomplished is by having one set of requirements that are consistent between differing regulatory purposes. These criteria, since they are identical in TILA, can be adopted once in one section of Regulation Z and referenced by the other section.

Accordingly, the Bureau is adopting § 1026.43(f)(1)(vi) to require the creditor to meet the satisfy the requirements stated in § 1026.35(b)(iii)(A), (B), and (C), adopted in the 2013 Escrows Final Rule, in order to originate a balloon-payment qualified mortgage under § 1026.43(f)(1). Comment 43(f)(1)(vi)-1.i clarifies that the Bureau publishes annually a list of counties that qualify as rural or underserved in accordance with § 1026.35(b)(2)(iii)(A). The comment further clarifies that the Bureau's annual determination of rural or underserved counties are based on the definitions set forth in § 1026.35(b)(2)(iv). Comment 43(f)(1)(vi)-1.ii clarifies that the creditor along with all affiliates must not originate more than 500 first lien transactions during the preceding calendar year in accordance with § 1026.35(b)(2)(iii)(B). Comment 43(f)(1)(vi)-1.iii clarifies that the initial asset-size threshold for a creditor is $2 billion for calendar year 2013 and will be updated each December to publish the applicable threshold for the following calendar year in accordance with§ 1026.35(b)(2)(iii)(C). The comment further clarifies that a creditor that had total assets below the threshold on December 31 of the preceding year satisfies this criterion for purposes of the exemption during the current calendar year.

43(f)(2) Post-Consummation Transfer of Balloon-Payment Qualified Mortgage

As noted in the discussion related to paragraph (f)(1)(v) above, TILA section 129C(b)(E)(iv) requires that the lender keep balloon-payment mortgages in portfolio, which addressed in both the Board's 2011 ATR Proposal and 2011 Escrows Proposal in different ways. In light of the differences between the two rulemakings and in particular the important ramifications of qualified mortgage status over the life of the loan, however, the Bureau believes that it is also appropriate for this final rule to contain additional safeguards concerning post-consummation sales that are not pursuant to a forward commitment in order to prevent gaming of the balloon-payment qualified mortgage exception. As noted above, the Board had proposed an approach under which the creditor would lose its eligibility to originate balloon-payment qualified mortgages once it sold any balloon-payment mortgages. Under one alternative, a single sale after the effective date of the rule would have permanently disqualified the creditor from invoking the exception, while the other alternative would have disqualified the creditor from invoking the exception for two calendar years.

In addition to the comments received on the Board's 2011 Escrows Proposal related to the forward commitment requirement discussed in paragraph (f)(1)(v), above, two trade group commenters and one industry commenter indicated that the second alternative was preferable, but urged the Bureau only to look at the last calendar year, instead of the current or prior years. Of these commenters, one trade group and the industry commenter suggested adding a de minimis number of permitted transfers of balloon-payment qualified mortgages. One trade group commenter noted that the statute requires that only balloon-payment qualified mortgages be kept in portfolio. Another trade group commenter questioned the impact that either of the Board's alternatives would have on a rural creditors' ability to sell a balloon-payment mortgage if the creditor was directed to do pursuant to action requirements of prudential regulators, such as a prompt corrective action notice.

The Bureau agrees with commenters that the first alternative would work against the stated purpose of the balloon-payment qualified mortgage exemption, as creditors that would not qualify would forever be excluded from this exemption in the future. Over time, this would further reduce the creditors originating balloon-payment qualified mortgages and thereby reduce the availability of credit to those markets. In addition, the Bureau believes the Board's second alternative mitigates but does not eliminate these difficulties. Under the second alternative the disqualification from originating balloon-payment qualified mortgages would be temporary rather than permanent, but even so creditors who found it necessary to sell off a balloon-payment mortgage would pay a steep price in terms of their ability to originate loans in the future, and credit availability would be negatively impacted. Commenters that supported the second alternative did so with the stated preference for the second alternative to the first, instead of the requirements of the second alternative itself.

The Bureau believes these concerns can be eliminated or reduced by providing, as a general rule, that if a balloon-payment qualified mortgage is sold, that mortgage loses its status as a qualified mortgage, but the creditor does not lose its ability to originate balloon-payment qualified mortgages in the future. The rule would be subject to four exceptions to permit a balloon-payment qualified mortgage to be sold in narrowly defined circumstances without losing its qualified mortgage status. The first exception would allow for a sale to any person three years after consummation; this would require the creditor to keep the balloon-payment qualified mortgage for the same period of time that a consumer could bring a claim for violation of § 1026.43 under TILA section 130(e). This facilitates managing interest rate risk by selling seasoned balloon-payment qualified mortgages, but encourages responsible underwriting because the originating creditor would keep all risk of affirmative claims while those claims could be asserted. The second exemption would permit creditors to sell to other qualifying creditors, which would provide flexibility and consistency with the portfolio requirement. The third exception would address the need of creditors to sell loans to comply with requirements of prudential regulators, conservators, receivers and others who have the responsibility to ensure creditors are operating within the bounds of the law. The fourth exemption addresses changes in the ownership of the creditor itself, so that the balloon-payment qualified mortgages held by the creditor do not lose their qualified mortgage status solely because of the change in ownership of the creditor.

Accordingly, the Bureau is adopting § 1026.43(f)(2) to require a creditor to retain a balloon-payment qualified mortgage in its portfolio, otherwise the balloon-payment qualified mortgage will no longer be a qualified mortgage, with four exceptions as set forth above. Comment 43(f)(2)-1 clarifies that creditors must generally hold a balloon-payment qualified mortgage in portfolio, subject to four exceptions. Comment 43(f)(2)-2 clarifies that the four exceptions apply to all subsequent transfers, and not just the initial transfer of the balloon-payment qualified mortgage, and provides an example. Comment 43(f)(2)(i)-1 clarifies the application of the exception relating to transfers of the balloon-payment qualified mortgage three years or more after consummation. Comment 43(f)(2)(ii)-1 clarifies the application of the exemption relating to the transfer of a balloon-payment qualified mortgage to a creditor that meets the requirements of § 1026.43(f)(1)(vi). Comment 43(f)(2)(iii)-1 clarifies the application of the exemption related to the transfer of a balloon-payment qualified mortgage pursuant to the requirements of a supervisory regulator and provides an example. Comment 43(f)(2)(iv)-1 clarifies the application of the exemption related to the transfer of a balloon-payment qualified mortgage as a result or the merger or sale of the creditor and provides an example.

43(g) Prepayment Penalties

As discussed above regarding treatment of prepayment penalties under the points and fees test for qualified mortgages and for high-cost loans under HOEPA in § 1026.32(b)(1) and the definition of prepayment penalty under § 1026.32(b)(6), the Dodd-Frank Act restricts prepayment penalties in a number of ways. Section 1026.43(g) implements TILA section 129C(c), which establishes general limits on prepayment penalties for all residential mortgage loans. Specifically, TILA section 129C(c) provides that:

• Only a qualified mortgage may contain a prepayment penalty;
• A qualified mortgage with a prepayment penalty may not have an adjustable rate and may not have an annual percentage rate that exceeds the threshold for a higher-priced mortgage loan;
• The prepayment penalty may not exceed three percent of the outstanding balance during the first year after consummation, two percent during the second year after consummation, and one percent during the third year after consummation;
• There can be no prepayment penalty after the end of the third year after consummation; and
• A creditor may not offer a consumer a loan with a prepayment penalty without offering the consumer a loan that does not include a prepayment penalty.

Taken together, the Dodd-Frank Act's amendments to TILA relating to prepayment penalties mean that most closed-end, dwelling-secured transactions: (1) May provide for a prepayment penalty only if the transaction is a fixed-rate, qualified mortgage that is neither high-cost nor higher-priced under §§ 1026.32 and 1026.35; (2) may not, even if permitted to provide for a prepayment penalty, charge the penalty more than three years following consummation or in an amount that exceeds two percent of the amount prepaid; and (3) may be required to limit any penalty even further to comply with the points and fees limitations for qualified mortgages, or to stay below the points and fees trigger for high-cost mortgages. Section 1026.43(g) now reflects these principles.

The Board proposal implemented TILA section 129C(c) in § 226.43(g) without significant alteration, except that under proposed § 226.43(g)(2)(ii), the Board proposed to apply the percentage tests outlined in the statute to the amount of the outstanding loan balance prepaid, rather than to the entire outstanding loan balance, to provide tighter restrictions on the penalties allowed on partial prepayments.

Commenters generally supported the Board's proposal, though some industry commenters expressed concern that limitations on prepayment penalties would reduce prices on the sale of mortgages in the secondary market due to increased prepayment risk. Consumer advocates generally supported limiting prepayment penalties, as described by amended TILA section 129C(c), as an important element in ensuring affordability. Other industry commenters expressed concern that such a limitation on the imposition of prepayment penalties would lead to fewer creditors conditionally waiving closing costs, noting that this implication might limit access to credit. At least one industry commenter argued that the Board's proposal to limit prepayment penalties was too broad in scope, stating the legislative history demonstrated that the true target of the prepayment penalty prohibition of TILA section 129C(c) was limited to mortgages with teaser rates and/or balloon payments and to protect subprime consumers, not those consumers who chose a product with a lower interest rate in exchange for a prepayment penalty provision. The Bureau does not find this argument persuasive, given the plain language of amended TILA section 129C(c).

After review, the Bureau is adopting most of the Board's proposal, although as discussed below the Bureau is altering the prepayment limitation in the first year after consummation to reflect the separate limitations enacted in sections 1431 and 1432 of the Dodd-Frank Act, regarding high-cost mortgages.

Scope; Reverse Mortgages and Temporary Loans

Section 1026.43(g) implements TILA section 129C(c), which applies to a “residential mortgage loan,” that is, to a consumer credit transaction secured by a dwelling, including any real property attached to the dwelling, other than an open-end credit plan or a transaction secured by a consumer's interest in a timeshare plan. See TILA section 103(cc)(5). Consequently, the regulation refers to “covered transaction,” which as defined in § 1026.43(b)(1) and discussed further in the section-by-section analysis of § 1026.43(a) excludes open-end credit plans and transactions secured by timeshares from coverage consistent with statutory exclusions. However, neither the definition of “residential mortgage loan” nor the TILA section 129C(c)(1) prepayment penalty prohibition excludes reverse mortgages or temporary or “bridge” loans with a term of 12 months or less, such as a loan to finance the purchase of a new dwelling where the consumer plans to sell a current dwelling. See TILA sections 103(cc)(5), 129C(a)(8), 129C(c). Moreover, because under TILA section 129C(c)(1)(A), only a qualified mortgage may have a prepayment penalty and reverse mortgages and temporary loans are excluded from the ability-to-repay and qualified mortgage requirements of the Dodd-Frank Act (and thus may not be qualified mortgages), prepayment penalties would not be permitted on either product absent further accommodation.

The Board proposal sought comment on whether further provisions addressing the treatment of reverse mortgages were warranted. Because reverse mortgages are not subject to the ability-to-repay requirements, the Board did not propose to define a category of closed-end reverse mortgages as qualified mortgages, though it sought comment on the possibility of using its authority to do so, given that qualified mortgage status affects both application of the Dodd-Frank Act prepayment penalty provisions and certain provisions concerning securitization and “qualified residential mortgages.” See TILA section 129C(b)(2)(A)(ix) and (b)(3)(B). The Board specifically requested comment on whether special rules should be created to permit certain reverse mortgages to have prepayment penalties. In particular, the Board sought comment on how it might create criteria for a “qualified mortgage” reverse mortgage that would be consistent with the purposes of qualified mortgages under TILA section 129C(b), and requested any supporting data on the prepayment rates for reverse mortgages.

Consumer advocates generally supported the Board's proposal to apply the prepayment penalty requirements to reverse mortgages, and industry commenters did not object. Moreover, commenters did not provide data or other advocacy to refute the Board's reasoning for including reverse mortgages within the scope of § 1026.43(g): (1) That the overwhelming majority of reverse mortgages being originated in the current market are insured by the FHA, which does not allow reverse mortgages to contain prepayment penalties; and (2) excluding “qualified” reverse mortgages from coverage of the prepayment penalty prohibition would not be necessary or appropriate to effectuate the purposes of TILA section 129C, absent an articulated reason why such exclusion would “assure that consumers are offered and receive residential mortgage loans on terms that reasonably affect their ability to repay the loans and that are understandable and not unfair, deceptive, or abusive.” See TILA section 129B(a)(2).

While the Board did not specifically seek comment with respect to whether further provisions addressing the treatment of bridge loans under § 1026.43(g) were warranted, commenters nevertheless discussed the intersection of bridge loans and prepayment penalties. As discussed in the section-by-section analysis of § 1026.32(b)(6), some industry commenters expressed concern that the availability of, or cost of, construction-to-permanent loans might suffer, should the rule restrict the permissible prepayment penalty charges levied by a creditor if a consumer does not convert the construction loan into a permanent loan with the same creditor within a specified time period. As discussed in the section-by-section analysis of § 1026.32(b)(6), some commenters may have been mistaken with respect to whether certain fees were, in fact, a prepayment penalty. To the extent fees charged by a bridge loan are a prepayment penalty, however, they are prohibited as of the effective date. According to § 1026.43(a)(3)(iii), the construction phase of a construction-to-permanent loan cannot be a qualified mortgage, and thus under § 1026.43(g)(1)(ii)(B) such a loan cannot include a prepayment penalty. Construction-to-permanent loans are discussed in more detail in the section-by-section analysis of § 1026.43(a).

Accordingly, the Bureau is finalizing the rule at this time without special provisions to otherwise alter the general scope of this rule, as discussed in the section-by-section analysis of § 1026.43(a), such as by allowing the application of prepayment penalties for either reverse mortgages or temporary loans. The Bureau may revisit the issue in subsequent years, either as part of a future rulemaking to evaluate application of all title XIV requirements to reverse mortgages or as part of the five-year review of significant rules required under section 1022(d) of the Dodd-Frank Act.

43(g)(1) When Permitted

TILA section 129C(c)(1)(A) provides that a covered transaction must not include a penalty for paying all or part of the principal balance before it is due unless the transaction is a qualified mortgage as defined in TILA section 129C(b)(2). TILA section 129C(c)(1)(B) further restricts the range of qualified mortgages on which prepayment penalties are permitted by excluding qualified mortgages that have an adjustable rate or that meet the thresholds for “higher-priced mortgage loans” because their APRs exceed the average prime offer rate for a comparable transaction by a specified number of percentage points.

To implement TILA section 129C(c)(1), the Board proposed § 226.43(g)(1), which provided that a covered transaction may not include a prepayment penalty unless the prepayment penalty is otherwise permitted by law, and the transaction: (1) Has an APR that cannot increase after consummation; (2) is a qualified mortgage, as defined in § 226.43(e) or (f); and (3) is not a higher-priced mortgage loan, as defined in § 226.45(a). The Board proposed under § 226.43(g)(1)(i) that a prepayment penalty must be otherwise permitted by applicable law. The Board reasoned that TILA section 129C(c) limits, but does not specifically authorize, including a prepayment penalty with a covered transaction. Thus, TILA section 129C(c) does not override other applicable laws, such as State laws, that may be more restrictive with respect to prepayment penalties, so a prepayment penalty would not be permitted if otherwise prohibited by applicable law. This approach is consistent with prepayment penalty requirements for high-cost mortgages under § 1026.32(d)(7)(i) and higher-priced mortgage loans under § 1026.35(b)(2)(i).

The Board proposed § 226.43(g)(1)(ii)(A) to interpret the statutory language to apply to covered transactions for which the APR may increase after consummation. This regulatory language is consistent with other uses of “variable-rate” within Regulation Z, such as comment 17(c)(1)-11, which provides examples of variable-rate transactions.

Some consumer advocates did not support the Board's proposal, arguing that for certain mortgages (specifically step-rate mortgages) the interest rate can increase after consummation without affecting the APR. These commenters argued that the purpose of TILA section 129C(c)(1)(B)(i) is to avoid allowing a creditor to lock a consumer into a rising-cost mortgage via a prepayment penalty and a rising interest rate. Consumer groups expressed concern that a consumer might become “trapped” by a prepayment penalty on the one hand, and a rising interest rate on the other. The Bureau does not find this argument persuasive. TILA section 129C(1)(B)(i) prohibits a transaction with “an adjustable rate” from including a prepayment penalty. Longstanding rules under Regulation Z for closed-end transactions generally categorize transactions based on the possibility of APR changes, rather than interest rate changes. This distinction is relevant because covered transactions may have an APR that cannot increase after consummation even though a specific interest rate, or payments, may increase after consummation. For example, the APR for a “step-rate mortgage” without a variable-rate feature does not change after consummation, because the rates that will apply and the periods for which they will apply are known at consummation. See § 1026.18(s)(7)(ii) (defining “step-rate mortgage” for purposes of transaction-specific interest rate and payment disclosures). Thus, the danger of an interest rate/prepayment penalty “trap” is mitigated in a step-rate loan because the consumer knowledge of the exact payments to expect each month for the 36 months following consummation during which a prepayment penalty might apply. The Bureau therefore is adopting § 1026.43(g)(1)(ii)(A) as proposed. A fixed-rate mortgage or a step-rate mortgage therefore may have a prepayment penalty, but an adjustable-rate mortgage may not have a prepayment penalty. See § 1026.18(s)(7)(i) through (iii) (defining “fixed-rate mortgage,” “step-rate mortgage,” and “adjustable-rate mortgage”).

Balloon-Payment Mortgages

Under TILA section 129C(c)(1)(A), a covered transaction may not include a prepayment penalty unless the transaction is a qualified mortgage under TILA section 129C(b)(2). The Board proposed to implement TILA section 129C(c)(1)(A) in § 226.43(g)(1)(ii)(B) and noted that, under section 129C(b)(2)(e), a covered transaction with a balloon payment may be a qualified mortgage if the creditor originates covered transactions primarily in “rural” or “underserved” areas, as discussed in detail above in the section-by-section analysis of § 1026.43(f); thus, a consumer could face a prepayment penalty if the consumer attempts to refinance out of a balloon-payment qualified mortgage before the balloon payment is due. The Board solicited comment on whether it would be appropriate to use its legal authority under TILA sections 105(a) and 129B(e) to provide that a balloon-payment qualified mortgage may not have a prepayment penalty in any case. Most commenters generally supported the Board's decision not to extend the prepayment penalty ban to all balloon-payment loans, noting the need for such financial products in rural and underserved areas. In light of the access concerns, the Bureau declines to exercise its exception authority under TILA sections 105(a) and 129B(e) to add a blanket prohibition of prepayment penalties for all balloon-payment loans. Accordingly, the Bureau is adopting § 1026.43(g)(1)(ii)(B) as proposed. The Bureau will continue to monitor the use of balloon-payment qualified mortgages and their use of prepayment penalties.

Threshold for a Higher-Priced Mortgage Loan

Under TILA section 129C(c)(1)(B), a covered transaction may not include a prepayment penalty unless the transaction's APR is below the specified threshold for “higher-priced mortgage loans.” As discussed above, those thresholds are determined by reference to the applicable average prime offer rate. The Board proposed under § 226.43(g)(1)(ii)(C) that a creditor would determine whether a transaction is a higher-priced mortgage loan based on the transaction coverage rate rather than the APR, for purposes of the prepayment penalty restriction, because APRs are based on a broader set of charges, including some third-party charges such as mortgage insurance premiums, than average prime offer rates. The Board expressed a concern that using the APR metric posed a risk of over-inclusive coverage beyond the subprime market and instead proposed using the transaction coverage rate.

In August 2012, the Bureau extended the notice-and-comment period for comments relating to the proposed adoption of the more inclusive finance charge, including the transaction coverage rate. At that time, the Bureau noted that it would not be finalizing the more inclusive finance charge in January 2013. See 77 FR 54843 (Sept. 6, 2012). The Bureau therefore does not address in this rulemaking the numerous public comments that it received concerning the proposed alternatives for the APR coverage test. The Bureau instead will address such comments in connection with its finalization of the 2012 TILA-RESPA Integration Proposal, thus resolving that issue together with the Bureau's determination whether to adopt the more inclusive finance charge. The Bureau is thus adopting the definition of a higher-priced loan as defined in § 1026.35(a), which corresponds to the thresholds specified in TILA section 129C(1)(B)(ii).

43(g)(2) Limits on Prepayment Penalties

TILA section 129C(c)(3) provides that a prepayment penalty may not be imposed more than three years after the covered transaction is consummated and limits the maximum amount of the prepayment penalty. Specifically, TILA section 129C(c)(3) limits the prepayment penalty to (1) three percent of the outstanding principal balance during the first year following consummation; (2) two percent during the second year following consummation; and (3) one percent during the third year following consummation.

The Board's proposed § 226.43(g)(2) was substantially similar to TILA section 129C(c)(3) except that the Board proposed to determine the maximum penalty amount by applying the percentages established in the statute to the amount of the outstanding loan balance prepaid, rather than to the entire outstanding loan balance. The Board reasoned that calculating the maximum prepayment penalty based on the amount of the outstanding loan balance that is prepaid, rather than the entire outstanding loan balance, would effectuate the purposes of TILA section 129C(c) to facilitate partial (and full) prepayment by more strictly limiting the amounts of prepayment penalties imposed.

The Board noted in its proposal that under HOEPA as amended by the Dodd-Frank Act, TILA section 103(bb)(1)(A)(iii) now defines a “high-cost mortgage” as any loan secured by the consumer's principal dwelling in which the creditor may charge prepayment fees or penalties more than 36 months after the closing of the transaction, or in which the fees or penalties exceed, in the aggregate, more than two percent of the amount prepaid. Moreover, under amended TILA section 129(c)(1), high-cost mortgages are prohibited from having prepayment penalties. Accordingly, any prepayment penalty in excess of two percent of the amount prepaid on any closed-end mortgage would both trigger and violate HOEPA's high-cost mortgage protections. The Board did not propose to implement these limitations on prepayment penalties in § 226.43(g)(2), but did solicit comment on whether the proposed text should be modified to incorporate the limitation of prepayment penalty amounts to two percent of the amount prepaid, as provided under TILA sections 103(bb)(1)(A)(iii) and 129(c)(1). The Board also solicited comment on whether to adopt some other threshold to account for the limitations on points and fees, including prepayment penalties, to satisfy the requirements for “qualified mortgages,” under TILA section 129C(b)(2)(A)(vii) and proposed § 226.43(e)(2)(iii).

The Bureau did not receive significant comment on the proposed adjustment of determining the maximum penalty amount by applying the percentages established in the statute to the amount of the outstanding loan balance prepaid, rather than to the entire outstanding loan balance, and therefore is adopting § 1026.43(g)(2) to measure prepayment penalties using the outstanding loan balance prepaid, as proposed. The Bureau is making this adjustment pursuant to its authority under TILA section 105(a) to issue regulations with such requirements, classifications, differentiations, or other provisions, and that provide for such adjustments and exceptions for all or any class of transactions, as in the judgment of the Bureau are necessary and proper to effectuate the purposes of TILA, to prevent circumvention or evasion thereof, or to facilitate compliance therewith. For instance, the Bureau believes that it would be inconsistent with congressional intent to strong disfavor and limit prepayment penalties for the Bureau to allow creditors to charge one or two percent of the entire outstanding loan balance every time that a consumer pays even a slightly greater amount than the required monthly payment due.

The Bureau did not receive significant comment on how to resolve the differing prepayment thresholds for high-cost mortgages and qualified mortgages, as described by the Board. But the Bureau believes that it is imperative to provide clear guidance to creditors with respect to all new limitations on prepayment penalties in dwelling-secured credit transactions, as imposed by the Dodd-Frank Act. As noted by the Board, new TILA section 129C(c)(3) limits prepayment penalties for fixed-rate, non-higher-priced qualified mortgages to three percent, two percent, and one percent of the outstanding loan balance prepaid during the first, second, and third years following consummation, respectively. However, amended TILA sections 103(bb)(1)(A)(iii) and 129(c)(1) for high-cost mortgages effectively prohibit prepayment penalties in excess of two percent of the amount prepaid at any time following consummation for most credit transactions secured by a consumer's principal dwelling by providing that HOEPA protections (including a ban on prepayment penalties) apply to mortgage loans with prepayment penalties that exceed two percent of the outstanding loan balance prepaid. The Bureau concludes that, to comply with both the high-cost mortgage provisions and the qualified mortgage provisions, creditors originating most closed-end mortgage loans secured by a consumer's principal dwelling would need to limit the prepayment penalty on the transaction to: (1) No more than two percent of the amount prepaid during the first and second years following consummation, (2) no more than one percent of the amount prepaid during the third year following consummation, and (3) zero thereafter.

Accordingly, the Bureau is modifying the final rule to reflect the two percent cap imposed by the Dodd-Frank Act amendments to HOEPA. As adopted in final form, § 1026.43(g)(2) amends the maximum prepayment penalty threshold for qualified mortgages during the first year following consummation, specified as three percent in TILA section 129C(c), to two percent, to reflect the interaction of the qualified mortgage and HOEPA revisions. In addition to finalizing this provision as a matter of reasonable interpretation of how the statutory provisions work together, the Bureau is making this adjustment pursuant to its authority under TILA section 105(a) to issue regulations with such requirements, classifications, differentiations, or other provisions, and that provide for such adjustments and exceptions for all or any class of transactions, as in the judgment of the Bureau are necessary and proper to effectuate the purposes of TILA, to prevent circumvention or evasion thereof, or to facilitate compliance therewith. The Bureau is exercising this adjustment to prevent creditor uncertainty regarding the interaction of qualified mortgages and high-cost mortgage rules, thus facilitating compliance. For example, assume a creditor issues a loan that meets the specifications of a § 1026.43(e) qualified mortgage. The loan terms specify that this creditor may charge up to three percent of any prepaid amount in the year following consummation. If the Bureau implements TILA section 129C(c) and sections 103(bb)(1)(A)(iii) and 129(c)(1) for high-cost mortgages, which effectively prohibit prepayment penalties in excess of two percent of the amount prepaid at any time following consummation, then the creditor will have complied with certain provisions of TILA while violating others. Thus, to avoid this complex interaction, the Bureau is eliminating the possibility of simultaneous compliance with and violation of TILA by reducing the maximum prepayment penalty allowed in the year following consummation to two percent under § 1026.43(g)(2)(ii)(A).

Comment 43(g)(2)-1 clarifies that a covered transaction may include a prepayment penalty that may be imposed only during a shorter period or in a lower amount than provided in § 1026.43(g)(2). Comment 43(g)(2)-1 provides the example of a prepayment penalty that a creditor may impose for two years after consummation that is limited to one percent of the amount prepaid. The Bureau is changing the prepayment example in comment 43(g)(2)-1 to reflect the Bureau's adjustment in § 1026.43(g)(2)(ii)(A) of the maximum prepayment penalty in the first year after consummation from three percent to two percent.

The Bureau recognizes that TILA section 129C(b)(2)(A)(vii) indirectly limits the amount of a prepayment penalty for a qualified mortgage, by limiting the maximum “points and fees” for a qualified mortgage to three percent of the total loan amount. See § 1026.43(e)(2)(iii), discussed above. The definition of “points and fees” includes the maximum prepayment penalty that may be charged, as well as any prepayment penalty incurred by the consumer if the loan refinances a previous loan made or currently held by the same creditor or an affiliate of the creditor. See TILA section 103(bb)(4)(E), § 1026.32(b)(1), and accompanying section-by-section analysis. Thus, if a creditor wants to include the maximum two percent prepayment penalty as a term of a qualified mortgage, it generally would have to forego any other charges that are included in the definition of points and fees. See the section-by-section analysis of § 1026.32(b)(1).

43(g)(3) Alternative Offer Required

Under TILA section 129C(c)(4), if a creditor offers a consumer a covered transaction with a prepayment penalty, the creditor also must offer the consumer a covered transaction without a prepayment penalty. The Board proposed § 226.43(g)(3), which contained language to implement TILA section 129C(c)(4) and added provisions to ensure comparability between the two alternative offers. Specifically, the proposed rule would mandate that the alternative covered transaction without a prepayment penalty must: (1) Have an APR that cannot increase after consummation and the same type of interest rate as the covered transaction with a prepayment penalty (that is, both must be fixed-rate mortgages or both must be step-rate mortgages); (2) have the same loan term as the covered transaction with a prepayment penalty; (3) satisfy the periodic payment conditions for qualified mortgages; and (4) satisfy the points and fees conditions for qualified mortgages. Proposed § 226.43(g)(3) also provided that the alternative covered transaction must be a transaction for which the consumer likely qualifies.

The Bureau did not receive significant comment on the proposal and is adopting § 1026.43(g)(3) as proposed. The Bureau is adding the additional conditions proposed by the Board to those specified in TILA section 129C(c)(4) to ensure that the alternative covered transactions is a realistic alternative for the consumer: A loan under substantially similar terms as the loan with a prepayment penalty for which the consumer likely qualifies. The Bureau is including these additional requirements pursuant to the Bureau's authority under TILA section 105(a) to prescribe regulations that contain such additional requirements, classifications, differentiations, or other provisions, or provide for such adjustments or exceptions for all or any class of transactions, as in the judgment of the Bureau are necessary or proper to effectuate the purposes of TILA, to prevent circumvention or evasion thereof, or to facilitate compliance therewith.

The Bureau believes that requirements designed to ensure that the alternative covered transactions effectuate the purposes of TILA section 129C(c)(4) by enabling the consumer to focus on a prepayment penalty's risks and benefits without having to consider or evaluate other differences between the alternative covered transactions. For example, under final § 1026.43(g)(3), a consumer is able to compare a fixed-rate mortgage with a prepayment penalty with a fixed-rate mortgage without a prepayment penalty, rather than with a step-rate mortgage without a prepayment penalty. Also, the Bureau believes requiring that the alternative covered transaction without a prepayment penalty be one for which the consumer likely qualifies effectuates the purposes of and prevents circumvention of TILA section 129C(c)(4), by providing for consumers to be able to choose between options that likely are available.

Under § 1026.43(g)(1)(i), a covered transaction with an APR that may increase after consummation may not have a prepayment penalty. The Board proposed in § 226.43(g)(3)(i) that, if a creditor offers a covered transaction with a prepayment penalty, the creditor must offer an alternative covered transaction without a prepayment penalty and with an APR that may not increase after consummation. The Board also proposed that the covered transaction with a prepayment penalty and the alternative covered transaction without a prepayment penalty must have the same type of interest rate. The Board offered these proposals to ensure that a consumer is able to choose between substantially similar alternative transactions. The Bureau did not receive significant comment on the proposal and is adopting the Board's proposal regarding the APR and the type of interest rate for the alternative transaction.

Higher-priced mortgage loans. The Board proposed that, under § 226.43(g)(3), if a creditor offers a covered transaction with a prepayment penalty, which may not be a higher-priced mortgage loan, the creditor may offer the consumer an alternative covered transaction without a prepayment penalty that is a higher-priced mortgage loan. The Board reasoned that TILA section 129C(c)(4) is intended to ensure that a consumer has a choice whether to obtain a covered transaction with a prepayment penalty, not to limit the pricing of the alternative covered transaction without a prepayment penalty that the creditor must offer. In fact, all things being equal, one would expect a creditor to cover the increased risk of prepayment by increasing the rate, thereby increasing the likelihood that the transaction might be a higher-priced mortgage loan. Furthermore, the Board noted that restricting the pricing of the required alternative covered transaction without a prepayment penalty might result in some creditors choosing to offer fewer loans. The Board thus did not propose to limit rate increases for the alternative covered transaction. The Bureau did not receive significant comment on this aspect of the proposal and is adopting the rule as proposed.

Timing of offer. The Board proposal concerning the alternative offer without a prepayment penalty that a creditor is required to offer under TILA section 129C(c)(4) did not specify that the creditor makes this alternative offer at or by a particular time. The Board proposal was consistent with § 1026.36(e)(2) and (3), which provide a safe harbor for the anti-steering requirement if a loan originator presents certain loan options to the consumer. These rules also do not contain a timing requirement. The Board solicited comment on whether it would be appropriate to require that creditors offer the alternative covered transaction without a prepayment penalty during a specified time period, such as before the consumer pays a non-refundable fee or at least fifteen calendar days before consummation. The Board also solicited comment on whether, if a timing requirement were included for the required alternative offer, whether a timing requirement should also be included under the safe harbor for the anti-steering requirement, for consistency. The Bureau did not receive significant comment on the proposal and is not including a specific timing requirement. The Bureau will continue to study required alternative offers to ensure that creditors offer consumers a meaningful alternative transaction that does not contain a prepayment penalty, in accordance with the purposes of TILA section 129C(c)(4). In the course of its review, if the Bureau determines that more specific timing requirements would provide more consumer choice, the Bureau may propose to revise § 1026.43(g)(3) accordingly.

The Board proposed comment 43(g)(3)(i)-1 to clarify that the covered transaction with a prepayment penalty and the alternative covered transaction without a prepayment penalty both must be either fixed-rate mortgages or step-rate mortgages. The Bureau did not receive significant comment on the proposal and is adopting the comment with some revisions for clarification only. For purposes of § 1026.43(g)(3)(i), the term “type of interest rate” means whether the covered transaction is a fixed-rate mortgage, as defined in § 1026.18(s)(7)(iii), or a step-rate mortgage, as defined in § 1026.18(s)(7)(ii).

Substance of offer. As discussed above, § 1026.43(g)(1)(ii)(B) provides that a covered transaction with a prepayment penalty must be a qualified mortgage, as defined in § 1026.43(e)(2), (e)(4), or (f). The Board proposal concerning the alternative offer without a prepayment penalty that a creditor is required to offer under TILA section 129C(c)(4) did not mandate that the alternative covered transaction offered without a prepayment penalty must also be a qualified mortgage. But under proposed § 226.43(g)(3)(ii) through (iv), the Board proposed to incorporate three conditions of qualified mortgages on the alternative offer, so that consumers may choose between alternative covered transactions that are substantially similar. Accordingly, the Board proposed that the alternative covered transaction without a prepayment penalty must: (1) Have the same loan term as the covered transaction with a prepayment penalty; (2) satisfy the periodic payment conditions in § 1026.43(e)(2)(i); and (3) satisfy the points and fees condition under § 1026.43(e)(2)(iii), based on the information known to the creditor at the time the transaction is offered. The Bureau did not receive significant comment on the proposal and is adopting the Board's proposal. The Bureau is including this provision both as part of its interpretation of TILA section 129C(c)(4) and using its authority under TILA sections 105(a), which provides that the Bureau's regulations may contain such additional requirements, classifications, differentiations, or other provisions, and may provide for such adjustments and exceptions for all or any class of transactions as in the Bureau's judgment are necessary or proper to effectuate the purposes of TILA, prevent circumvention or evasion thereof, or facilitate compliance therewith. 15 U.S.C. 1604(a), 1639b(e). This approach is further supported by the authority under TILA section 129B(e) to condition terms, acts or practices relating to residential mortgage loans that the Bureau finds necessary and proper to ensure that responsible, affordable mortgage credit remains available to consumers in a manner consistent with the purposes and to effectuate the purposes of section 129B and 129C, and that are in the interest of the consumer, among other things. 15 U.S.C. 1639b(e). The purposes of TILA include the purposes that apply to 129B and 129C, to assure that consumers are offered and receive residential mortgage loans on terms that reasonably reflect their ability to repay the loan. See 15 U.S.C. 1639b(a)(2). The Bureau believes that requiring the creditor that offers the consumer a loan with a prepayment penalty to also offer the consumer the ability to choose an alternative covered transaction that is otherwise substantially similar, besides not including a prepayment penalty, is necessary and proper to fulfill such purposes by ensuring that the consumer is offered a reasonable alternative product that the consumer can repay and which does not include a prepayment penalty. For this reason, this provision is also in the interest of the consumer.

The Board proposed comment 43(g)(3)(iv)-1 to provide guidance for cases where a creditor offers a consumer an alternative covered transaction without a prepayment penalty under § 1026.43(g)(3) and knows only some of the points and fees that will be charged for the loan. For example, a creditor may not know that a consumer intends to buy single-premium credit unemployment insurance, which would be included in the points and fees for the covered transaction. Proposed comment 43(g)(3)(iv)-1 clarified that the points and fees condition is satisfied if the creditor reasonably believes, based on the information known to the creditor at the time the offer is made, that the amount of points and fees to be charged for an alternative covered transaction without a prepayment penalty will be less than or equal to the amount of points and fees allowed for a qualified mortgage under § 1026.43(e)(2)(iii). The Bureau did not receive significant comment on the proposal and is adopting the comment largely as proposed.

The Board proposed comment 43(g)(3)(v)-1 to clarify what is meant by an alternative transaction for which the consumer likely qualifies. In this example, the creditor has a good faith belief the consumer can afford monthly payments of up to $800. If the creditor offers the consumer a fixed-rate mortgage with a prepayment penalty for which monthly payments are $700 and an alternative covered transaction without a prepayment penalty for which monthly payments are $900, the requirements of § 1026.43(g)(3)(v) are not met. Proposed comment 43(g)(3)(v)-1 also clarified that, in making the determination the consumer likely qualifies for the alternative covered transaction, the creditor may rely on information provided by the consumer, even if the information subsequently is determined to be inaccurate. The Bureau did not receive significant comment on the proposal and is adopting the Board's comment as proposed. Comment 43(g)(3)(v)-1 is substantially similar to comment 36(e)(3)-4, which provides clarification under the rules providing a safe harbor for the anti-steering requirements if, among other things, a loan originator presents the consumer with loan options for which the consumer likely qualifies. In addition to agreeing with the Board's reasoning, the Bureau is adopting this rule and comment to promote consistency and further the Bureau's initiative to provide streamlined regulatory guidance.

43(g)(4) Offer Through a Mortgage Broker

The requirement to offer an alternative covered transaction without a prepayment penalty applies to a “creditor.” See TILA section 129C(c)(4). TILA section 103(f), in relevant part, defines “creditor” to mean a person who both: (1) Regularly extends consumer credit which is payable by agreement in more than four installments or for which the payment of a finance charge is or may be required, and (2) is the person to whom the debt arising from the consumer credit transaction is initially payable on the face of the evidence of indebtedness (or, if there is no such evidence of indebtedness, by agreement). 15 U.S.C. 1602(f).

The Board proposed § 226.43(g)(4), which would apply when a creditor offers a covered transaction with a prepayment penalty through a mortgage broker, as defined in § 1026.36(a)(2), to account for operational differences in offering a covered transaction through the wholesale channel versus through the retail channel. The Board proposed under § 226.43(g)(4) that, if a creditor offers a covered transaction to a consumer through a mortgage broker, as defined in § 1026.36(a)(2), the creditor must present to the mortgage broker an alternative covered transaction without a prepayment penalty that meets the conditions in § 1026.43(g)(3). The Board reasoned that the requirement to offer an alternative covered transaction without a prepayment penalty properly is applied to creditors and not to mortgage brokers, because creditors “offer” covered transactions, even if mortgage brokers present those offers to consumers. Further, the Board noted that, if Congress had intended to apply TILA section 129C(c)(4) to mortgage brokers, Congress would have explicitly applied that provision to “mortgage originators” in addition to creditors. The Board's proposal also provided under proposed § 226.43(g)(4)(ii) that the creditor must establish, by agreement, that the mortgage broker must present the consumer an alternative covered transaction without a prepayment penalty that meets the conditions in § 1026.43(g)(3) offered by (1) the creditor, or (2) another creditor, if the transaction has a lower interest rate or a lower total dollar amount of origination points or fees and discount points.

The Bureau did not receive significant comment on proposed § 226.43(g)(4) and is adopting § 1026.43(g)(4) largely as proposed. By providing for the presentation of a loan option with a lower interest rate or a lower total dollar amount of origination points or fees and discount points than the loan option offered by the creditor, § 1026.43(g)(4) facilitates compliance with § 1026.43(g)(3) and with the safe harbor for the anti-steering requirement in connection with a single covered transaction, as governed by § 1026.36(e)(3)(i). Section 1026.43(g)(4) does not affect the conditions that a loan originator must meet to take advantage of the safe harbor for the anti-steering requirement, however. Thus, if a loan originator chooses to use the safe harbor, the originator must present the consumer with: (1) The loan option with the lowest interest rate overall, (2) the loan option with the lowest interest rate without certain risky features, including a prepayment penalty, and (3) the loan option with the lowest total origination points or fees and discount points. See § 1026.36(e)(3)(i). The Bureau believes that requiring a mortgage broker to present to a consumer the creditor's alternative covered transaction without a prepayment penalty could confuse the consumer if he or she is presented with numerous other loan options under § 1026.36(e). Presenting a consumer with four or more loan options for each type of transaction in which the consumer expresses an interest may not help the consumer to make a meaningful choice. When compared with other loan options a mortgage broker presents to a consumer, a creditor's covered transaction without a prepayment penalty might not have the lowest interest rate (among transactions either with or without risky features, such as a prepayment penalty) or the lowest total dollar amount of origination points or fees and discount points, and thus might not be among the loan options most important for consumers to evaluate. Also, the creditor may have operational difficulties in confirming whether or not a mortgage broker has presented to the consumer the alternative covered transaction without a prepayment penalty.

The Board proposed comment 43(g)(4)-1 to clarify that the creditor may satisfy the requirement to present the mortgage broker such alternative covered transaction without a prepayment penalty by providing the mortgage broker a rate sheet that states the terms of such an alternative covered transaction without a prepayment penalty. The Board proposed comment 43(g)(4)-2 to clarify that the creditor's agreement with the mortgage broker may provide for the mortgage broker to present both the creditor's covered transaction and a covered transaction offered by another creditor with a lower interest rate or a lower total dollar amount of origination points or fees and discount points. Comment 43(g)(4)-2 also cross-references comment 36(e)(3)-3 for guidance in determining which step-rate mortgage has a lower interest rate. The Board proposed comment 43(g)(4)-3 to clarify that a creditor's agreement with a mortgage broker for purposes of § 1026.43(g)(4) may be part of another agreement with the mortgage broker, for example, a compensation agreement. The comment clarifies that the creditor thus need not enter into a separate agreement with the mortgage broker with respect to each covered transaction with a prepayment penalty. The Bureau did not receive significant comment on proposed comments 43(g)(4)-1 through -3 and is adopting these comments largely as proposed.

Provisions Not Adopted

As explained in the preamble to the Board's proposal, the Board did not propose specific rules under proposed § 226.43(g)(4) to apply in the case where the loan originator is the creditor's employee. The Bureau did not receive significant comment on that omission and likewise is not adopting special provisions under § 1026.43(g)(4) to apply where the loan originator is the creditor's employee. The Bureau believes that, in such cases, the employee likely can present alternative covered transactions with and without a prepayment penalty to the consumer without significant operational difficulties.

The Board solicited comment on whether additional guidance was needed regarding offers of covered transactions through mortgage brokers that use the safe harbor for the anti-steering requirement, under § 226.36(e)(2) and (3). The Bureau did not receive significant comment on the proposal and concludes that additional guidance is not currently required. The Bureau will continue to study the interaction between prepayment penalty restrictions, as applied to mortgage brokers under § 1026.43(g)(4) and the safe harbor for the anti-steering requirement, under § 1026.36(e)(2) and (3) to ensure that brokers are operating with sufficient guidance. In the course of its review, if the Bureau determines that more guidance would provide clarity or otherwise reduce compliance burden, then the Bureau may propose to add additional guidance.

43(g)(5) Creditor That Is a Loan Originator

The Board proposed § 226.43(g)(5) to address table funding situations, where a creditor does not provide the funds for a covered transaction out of its own resources but rather obtains funds from another person and, immediately after consummation, assigns the note, loan contract, or other evidence of the debt obligation to the other person. Such a creditor generally presents to a consumer loan options offered by other creditors, and this creditor is a loan originator subject to the anti-steering requirements in § 1026.36(e). See § 1026.36(a)(1); comment 36(a)(1)-1. Like other loan originators, such a creditor may use the safe harbor for the anti-steering requirements under § 1026.36(e)(2) and (3). The Board proposed that, if the creditor is a loan originator, as defined in § 1026.36(a)(1), and the creditor presents a consumer a covered transaction with a prepayment penalty offered by a person to which the creditor would assign the covered transaction after consummation, the creditor must present the consumer an alternative covered transaction without a prepayment penalty offered by (1) the prospective assignee, or (2) another person, if the transaction offered by the other person has a lower interest rate or a lower total dollar amount of origination points or fees and discount points. The Board reasoned that its proposal provided flexibility with respect to the presentation of loan options, which facilitates compliance with § 1026.43(g)(3) and with the safe harbor for the anti-steering requirement in connection with the same covered transaction. See § 1026.36(e)(3)(i).

The Bureau did not receive significant comment on the proposal and is adopting the Board's proposal. Like § 1026.43(g)(4), § 1026.43(g)(5) does not affect the conditions that a creditor that is a loan originator must meet to take advantage of the safe harbor for the anti-steering requirement. Accordingly, if a creditor that is a loan originator chooses to use the safe harbor, the creditor must present the consumer (1) the loan option with the lowest interest rate overall, (2) the loan option with the lowest interest rate without certain risky features, including a prepayment penalty, and (3) the loan option with the lowest total origination points or fees and discount points. See § 1026.36(e)(3)(i).

The Board proposed comment 43(g)(5)-1 to clarify that a loan originator includes any creditor that satisfies the definition of the term but makes use of “table-funding” by a third party. The Bureau did not receive significant comment on the proposed comment and is adopting it as proposed. The Board proposed comment 43(g)(5)-2 to cross-reference guidance in comment 36(e)(3)-3 on determining which step-rate mortgage has a lower interest rate. The Bureau did not receive significant comment on the proposal and is adopting the Board's proposed comment.

43(g)(6) Applicability

TILA section 129C(c)(1)(A) provides that only a qualified mortgage may contain a prepayment penalty and TILA section 129C(c)(4) further requires the creditor to offer the consumer an alternative offer that does not contain a prepayment penalty. The Board proposed § 226.43(g)(6) to provide that § 226.43(g) would apply only if a transaction is consummated with a prepayment penalty and would not be violated if (1) a covered transaction is consummated without a prepayment penalty or (2) the creditor and consumer do not consummate a covered transaction. The Bureau did not receive significant comment on the proposal and is adopting the Board's proposal under § 1026.43(g)(6).

Section 1026.43(g)(2) limits the period during which a prepayment penalty may be imposed and the amount of any prepayment penalty. As provided in § 1026(g)(6), those prepayment penalty limitations apply only if a covered transaction with a prepayment penalty is consummated. Similarly, § 1026.43(g)(3) requires a creditor that offers a consumer a covered transaction with a prepayment penalty to offer the consumer an alternative covered transaction without a prepayment penalty. Where a consumer consummates a covered transaction without a prepayment penalty, § 1026(g)(6) states that it is unnecessary to require that the creditor offer the consumer an alternative covered transaction without a prepayment penalty. Thus § 1026.43(g) applies only if the consumer consummates a covered transaction with a prepayment penalty.

43(h) Evasion; Open-End Credit

TILA section 129C, which addresses the ability-to-repay requirements and qualified mortgages, applies to residential mortgage loans. TILA section 103(cc)(5) defines “residential mortgage loans” as excluding open-end credit plans, such as HELOCs. In its proposal, the Board recognized that the exclusion of open-end credit plans could lead some creditors to attempt to evade the requirements of TILA section 129C by structuring credit that otherwise would have been structured as closed-end as open-end instead.

The Board proposed § 226.43(h) to prohibit a creditor from evading the requirements of § 226.43 by structuring a transaction that does not meet the definition of open-end credit in § 226.2(a)(20) as open-end credit, such as a HELOC. The Board proposed comment 43(h)-1 to explain that where a loan is documented as open-end credit but the features and terms, or other circumstances, demonstrate that the loan does not meet the definition of open-end credit, then the loan is subject to the rules for closed-end credit, including § 226.43. The Board proposed these provisions using its authority under TILA sections 105(a) and 129B(e) to prevent circumvention or evasion. The Board noted that an overly broad anti-evasion rule could limit consumer choice by casting doubt on the validity of legitimate open-end plans, and the Board thus solicited comment on whether to limit the anti-evasion rule's application, for example, to HELOCs secured by first liens where the consumer draws down all or most of the entire line of credit immediately after the account is opened.

Consumer groups generally supported the proposed anti-evasion provision; some consumer groups suggested that the provision should be expanded to require all HELOCs to comply with all Dodd-Frank Act requirements, expressing concern over the potential for consumer abuse. Industry commenters generally sought clarification on the anti-evasion rule, noting that ambiguity with respect to the provision might limit creditors' ability, or willingness, to offer HELOCs or other open-end credit products.

The Bureau is adopting the Board's proposal largely as proposed. Section 1026.43(h) is also consistent with the Board's 2008 HOEPA Final Rule, § 1026.35(b)(4), which provides a similar anti-evasion provision with respect to higher-priced mortgage loans. The Bureau is including this provision both as part of its interpretation of TILA section 129C and using its authority under TILA section 105(a), which provides that the Bureau's regulations may contain such additional requirements, classifications, differentiations, or other provisions, and may provide for such adjustments and exceptions for all or any class of transactions as in the Bureau's judgment are necessary or proper to effectuate the purposes of TILA, prevent circumvention or evasion thereof, or facilitate compliance therewith, and TILA section 129B(e) to prevent circumvention or evasion. 15 U.S.C. 1604(a), 1639b(e). The purposes of TILA include the purposes that apply to section 129C, to assure that consumers are offered and receive residential mortgage loans on terms that reasonably reflect their ability to repay the loan. See 15 U.S.C. 1639b(a)(2). While some industry commenters requested further clarification on this provision, so as to avoid limiting consumer choice, the Bureau believes that no further commentary is required. A creditor that offers a consumer an open-end line of credit in the ordinary course of business need not be concerned with running afoul of the anti-evasion requirement, and a creditor need not undertake any additional compliance or reporting steps to do so. A creditor only violates § 1026.43(h) when the creditor structures credit secured by a consumer's dwelling that does not meet the definition of open-end credit in § 1026.2(a)(20) as an open-end plan in order to evade the ability-to-repay requirements. The Bureau's approach should allow creditors acting in good faith to continue to provide credit to consumers in the manner best fit for business needs and consumer demand, without concern of accidentally running afoul of the anti-evasion requirement.

VI. Effective Date

This final rule is effective on January 10, 2014. The rule applies to transactions for which the creditor received an application on or after that date. As discussed above in part III.C, the Bureau believes that this approach is consistent with the timeframes established in section 1400(c) of the Dodd-Frank Act and, on balance, will facilitate the implementation of the rules' overlapping provisions, while also affording creditors sufficient time to implement the more complex or resource-intensive new requirements.

As noted above, in response to the proposal, some industry commenters requested that the Bureau provide additional time for compliance because the Bureau is finalizing several mortgage rules at the same time. These commenters expressed concern over both the breadth and complexity of new rules expected from the Bureau and from other regulators. Some commenters stated that small institutions, in particular, might face a higher cost of compliance under the timeframes established in section 1400(c) of the Dodd-Frank Act. One industry commenter explained that the new rules would require creditors to alter financial products, modify compliance systems, and train staff. Another industry commenter noted that some credit unions and other institutions that rely on third-party providers, such as software vendors, to assist with compliance might face particular challenges with implementing necessary changes over a short time period since such third parties will need time to incorporate necessary updates and conduct testing, and include the changes in their scheduled releases. Some commenters urged the Bureau to coordinate publishing and effective dates among the title XIV rules and the QRM rulemaking, in order to assist creditors in minimizing compliance burden.

For the reasons already discussed above, the Bureau believes that an effective date of January 10, 2014 for this final rule and most provisions of the other title XIV final rules will ensure that consumers receive the protections in these rules as soon as reasonably practicable, taking into account the timeframes established by the Dodd-Frank Act, the need for a coordinated approach to facilitate implementation of the rules' overlapping provisions, and the need to afford creditors and other affected entities sufficient time to implement the more complex or resource-intensive new requirements.

VII. Dodd-Frank Act Section 1022(b)(2) Analysis

A. Overview

In developing the final rule, the Bureau has considered potential benefits, costs, and impacts. In addition, the Bureau has consulted, or offered to consult with, the prudential regulators, SEC, HUD, FHFA, the Federal Trade Commission, and the Department of the Treasury, including regarding consistency with any prudential, market, or systemic objectives administered by such agencies. The Bureau also held discussions with or solicited feedback from the United States. Department of Agriculture, Rural Housing Service, the Federal Housing Administration, and the Department of Veterans Affairs regarding the potential impacts of the final rule on those entities' loan programs.

The Board issued the 2011 ATR Proposal prior to the transfer of rulemaking authority to the Bureau. As the Board was not subject to Dodd-Frank Act section 1022(b)(2), the 2011 ATR Proposal did not contain a proposed Dodd-Frank Act section 1022 analysis.

The Dodd-Frank Act and the final rule establish minimum standards for consideration of a consumer's repayment ability for creditors originating certain closed-end, residential mortgage loans. These underwriting requirements are similar, but not identical, to the ability-to-repay requirements that apply to high-cost and higher-priced mortgage loans under current regulations. In general, the Act and the final rule prohibit a creditor from making a covered transaction unless the creditor makes a reasonable and good faith determination, based on verified and documented information, that the consumer has a reasonable ability to repay the loan according to its terms.

These documentation and verification requirements effectively prohibit no documentation and limited documentation loans that were common in the later years of the housing bubble. The final rule generally requires the creditor to verify the information relied on in considering a consumer's debts relative to income or residual income after paying debts, using reasonably reliable third-party records, with special rules for verifying a consumer's income or assets. The creditor must calculate the monthly mortgage payment based on the greater of the fully-indexed rate or any introductory rate, assuming monthly, fully amortizing payments that are substantially equal. The final rule provides special payment calculation rules for loans with balloon payments, interest-only loans, and negative amortization loans.

The final rule provides special rules for complying with the ability-to-repay requirements for a creditor refinancing a “non-standard mortgage” into a “standard mortgage.” Under the final rule, a non-standard mortgage is defined as an adjustable-rate mortgage with an introductory fixed interest rate for a period of one year or longer, an interest-only loan, or a negative amortization loan. Under this provision, a creditor refinancing a non-standard mortgage into a standard mortgage does not have to consider the specific underwriting criteria a lender must otherwise consider under the general ability-to-repay option, if certain conditions are met.

To provide creditors more certainty about their potential liability under the ability-to-pay standards while protecting consumers from unaffordable loans, the Dodd-Frank Act creates a presumption of compliance with the ability-to-pay requirement when creditors make “qualified mortgages.” According to the statute, covered transactions, in general, are qualified mortgages where: the loan does not contain negative amortization, interest-only payments, or balloon payments (except in certain limited circumstances); the term does not exceed 30 years; points and fees (excluding up to two bona fide discount points) do not exceed three percent of the total loan amount; the income or assets and debt obligations are considered and verified; the underwriting is based on the maximum rate during the first five years, uses a payment schedule that fully amortizes the loan over the loan term, and takes into account all mortgage-related obligations.

Under the final rule creditors have three options for originating a qualified mortgage. Under the first option, the loan must satisfy basic documentation and verification requirements for income or assets and debt, and the consumer must have a total (or “back-end”) debt-to-income ratio that is less than or equal to 43 percent. With respect to a loan that satisfies these criteria and is not a higher-priced covered transaction, there is a conclusive presumption that the creditor satisfied the ability-to-pay requirements so that the loan qualifies for a legal safe harbor under the ability-to-repay requirements. A loan that satisfies these criteria and is a higher-priced covered transaction receives a rebuttable presumption of compliance with the ability-to-repay requirements.

The second option for originating a qualified mortgage provides a temporary expansion of the general definition. Through this option, a loan is a qualified mortgage if it meets the prohibitions on certain loan features, the limitations on points and fees and loan terms that apply under the general definition and also meets one of the following requirements: is eligible for purchase or guarantee by the Federal National Mortgage Association (Fannie Mae) or the Federal Home Loan Mortgage Corporation (Freddie Mac) (collectively, the GSEs), while operating under the conservatorship or receivership of the FHFA; is eligible to be purchased or guaranteed by any limited-life regulatory entity succeeding the charter of either the GSEs; or is eligible to be insured by the FHA, VA or USDA or USDA RHS. This temporary provision expires with respect to GSE-eligible loans when conservatorship of the GSEs ends and expires with respect to each other category of loans on the effective date of a rule issued by each respective Federal agency pursuant to its authority under TILA section 129C(b)(3)(ii) to define a qualified mortgage. Alternatively, if GSE conservatorship continues or the Federal agencies do not issue rules defining qualified mortgage pursuant to TILA section 129C(b)(3)(ii), the temporary qualified mortgage definition expires seven years after the effective date of the rule.

Unlike loans that are qualified mortgages under the general definition, there is no specific monthly debt-to-income ratio threshold to be a qualified mortgage under this temporary provision, except as may be required to be eligible for purchase or guarantee or to be insured by the GSEs or Federal agencies. The temporary qualified mortgage definition does not specifically include documentation and verification requirements or a specific payment calculation requirement. The Bureau understands that, to be eligible for purchase or guarantee by the GSE's or to be eligible to be guaranteed or insured by the Federal agencies, a loan must first satisfy certain payment calculation requirements and repayment ability analyses (which include consideration of a consumer's total monthly debt-to-income ratio) and the information on which the calculation is based must be documented and verified. As is true with respect to the first category of qualified mortgages described above, a loan that satisfies these criteria and is not a higher-priced covered transaction receives a legal safe harbor under the ability-to-repay requirements. A loan that satisfies these criteria and is a higher-priced covered transaction receives a rebuttable presumption of compliance with the ability-to-repay requirements.

The third option for qualified mortgages exists only for small creditors operating predominantly in rural or underserved areas, who are allowed under the rule to originate a balloon-payment qualified mortgage. Specifically, this option exists for lenders originating 500 or fewer covered transactions, secured by a first lien, in the preceding calendar year, with assets equal to or under $2 billion (to be adjusted annually), and who made more than 50 percent of their total covered transactions secured by first liens on properties in counties that are “rural” or “underserved.” These creditors are allowed to offer loans with balloon payments assuming the loan also meets certain loan-specific criteria: the creditor must satisfy the requirements under the general qualified mortgage definition regarding consideration and verification of income or assets and debt obligations; the loan cannot permit negative amortization; the creditor must determine that the consumer can make all of the scheduled payments (other than the final balloon payment) under the terms of the legal obligation from the consumer's current or reasonably expected income or assets other than the dwelling that secures the transaction; the loan must have a term of least five years and no more than 30 years; the interest rate is fixed during the term of the loan; the creditor must base the payment calculation on the scheduled periodic payments, excluding the balloon payment; and the loan must not be subject to a forward commitment at the time of consummation.

Unlike loans that are qualified mortgages under the general definition, there is no specific debt-to-income ratio requirement for balloon-payment qualified mortgages. However, creditors must generally consider and verify a consumer's monthly debt-to-income ratio. Like the other qualified mortgage definitions, a loan that satisfies the criteria for a balloon-payment qualified mortgage and is not a higher-priced covered transaction receives a legal safe harbor under the ability-to-repay requirements for as long as the loan is held in portfolio by the creditor who originated the loan. The safe harbor also applies to balloon-payment qualified mortgages which are sold three years or more after consummation. A loan that satisfies the balloon payment qualified mortgage criteria and is a higher-priced covered transaction receives a rebuttable presumption of compliance with the ability-to-repay requirements.

As discussed above, the final rule provides a conclusive presumption of compliance with the ability-to-repay requirements for loans that satisfy the definition of a qualified mortgage and are not higher-priced covered transactions (i.e., APR does not exceed Average Prime Offer Rate (APOR) + 1.5 percentage points for first liens or 3.5 percentage points for subordinate liens). The final rule provides a rebuttable presumption of compliance with ability-to-repay requirements for all other qualified mortgage loans, meaning qualified mortgage loans that are higher-priced covered transactions. A consumer who seeks to rebut the presumption must prove that, at the time of consummation, in light of the consumer's income and debt obligations, the consumer's monthly payment (including mortgage-related obligations) on the covered transaction and any simultaneous loans of which the creditor was aware, would leave the consumer with insufficient residual income to pay living expenses, including recurring and material obligations or expenses of which the creditor was aware.

Finally, the final rule implements the Dodd-Frank Act limits on prepayment penalties, lengthens the time creditors must retain records that evidence compliance with the ability-to-repay and prepayment penalty provisions, and prohibits evasion of this rule, in connection with credit that does not meet the definition of open-end credit, by structuring a closed-end extension of credit as an open-end plan.

A consumer who brings an action against a creditor for a violation of the ability-to-repay requirements within three years from when the violation occurs may be able to recover special statutory damages equal to the sum of all finance charges and fees paid by the consumer, unless the creditor demonstrates that the failure to comply is not material; actual damages; statutory damages in an individual action or class action, up to a prescribed threshold; and court costs and attorney fees that would be available for violations of other TILA provisions. After the expiration of the three-year time period, the consumer is precluded from bringing an affirmative claim against the creditor. At any time, when a creditor or an assignee initiates a foreclosure action, a consumer may assert a violation of these provisions “as a matter of defense by recoupment or setoff.” There is no time limit on the use of this defense, although the recoupment or setoff of finance charge and fees is limited to the first three years of finance charges and fees paid by the consumer under the mortgage.

B. Data and Quantification of Benefits, Costs and Impacts

Section 1022 of the Dodd-Frank Act requires that the Bureau, in adopting the rule, consider potential benefits and costs to consumers and covered persons resulting from the rule, including the potential reduction of access by consumers to consumer financial products or services resulting from the rule, as noted above; it also requires the Bureau to consider the impact of proposed rules on covered persons and the impact on consumers in rural areas. These potential benefits and costs, and these impacts, however, are not generally susceptible to particularized or definitive calculation in connection with this rule. The incidence and scope of such potential benefits and costs, and such impacts, will be influenced very substantially by economic cycles, market developments, and business and consumer choices, that are substantially independent from adoption of the rule. No commenter has advanced data or methodology that it claims would enable precise calculation of these benefits, costs, or impacts. Moreover, the potential benefits of the rule on consumers and covered persons in creating market changes anticipated to address market failures are especially hard to quantify.

In considering the relevant potential benefits, costs, and impacts, the Bureau has utilized the available data discussed in this preamble, where the Bureau has found it informative, and applied its knowledge and expertise concerning consumer financial markets, potential business and consumer choices, and economic analyses that it regards as most reliable and helpful, to consider the relevant potential benefits and costs, and relevant impacts. The data relied upon by the Bureau includes the public comment record established by the proposed rule, as well as the data described in the Bureau's Federal Register notice reopening the comment for this rule, and the public comments thereon.

However, the Bureau notes that for some aspects of this analysis, there are limited data available with which to quantify the potential costs, benefits, and impacts of the final rule. For example, data on the number and volume of various loan products originated for the portfolios of bank and non-bank lenders exists only in certain circumstances. Data regarding many of the benefits of the rule such as the benefits from prevented defaults or from prevented injuries to the financial system are also limited.

In light of these data limitations, the analysis below generally provides a qualitative discussion of the benefits, costs, and impacts of the final rule. General economic principles, together with the limited data that are available, provide insight into these benefits, costs, and impacts. Where possible, the Bureau has made quantitative estimates based on these principles and the data that are available. For the reasons stated in this preamble, the Bureau considers that the rule as adopted faithfully implements the purposes and objectives of Congress in the statute. Based on each and all of these considerations, the Bureau has concluded that the rule is appropriate as an implementation of the Act.

C. Baseline for Analysis

The provisions of Dodd Frank concerning minimum loan standards and the ability-to-repay requirement are self-effectuating, and the Dodd-Frank Act does not require the Bureau to adopt a regulation to implement these amendments. The Act does require the Bureau to issue regulations to “carry out the purposes of” the subsection governing qualified mortgages, which includes the “presumption of compliance” accorded those mortgages. In the absence of such regulations, the statutory provisions would take effect on January 21, 2013, and there would be no clarification beyond the statute as to the meaning of the ability-to-repay requirement, which mortgages meet the statutory criteria for a qualified mortgage, and the nature of the presumption of compliance with respect to such mortgages. Thus, many costs and benefits of the final rule considered below would arise largely or entirely from the statute, not from the final rule. The final rule would provide substantial benefits compared to allowing these provisions to take effect alone by clarifying parts of the statute that are ambiguous. Greater clarity on these issues should reduce the compliance burdens on covered persons by reducing costs for attorneys and compliance officers as well as potential costs of over-compliance and unnecessary litigation.

Section 1022 of the Dodd-Frank Act permits the Bureau to consider the benefits and costs of the rule solely compared to the state of the world in which the statute takes effect without an implementing regulation. To provide the public better information about the benefits and costs of the statute, however, the Bureau has nonetheless chosen to evaluate the benefits, costs, and impacts of the major provisions of the final rule against a pre-statutory baseline. That is, the Bureau's analysis below considers the benefits, costs, and impacts of the relevant provisions of the Dodd-Frank Act combined with the final rule implementing those provisions relative to the regulatory regime that pre-dates the Act and remains in effect until the final rule takes effect. As noted, current regulations have parallel but not identical ability-to-repay rules applied to higher-price and high-cost mortgage loans.

In the analysis, in addition to referring to present market conditions, the Bureau refers at times to data from other historical periods—the market as it existed from 1997 to 2003 and the years of the bubble and the collapse—to provide the public a fuller sense of the potential impacts of the rule in other market conditions. Considering the current state of the market makes clear the near term benefits and costs of the provisions. However, at this point in the credit cycle, the market is highly restrictive and operating under very tight credit conditions. Against this background, the benefits and the costs of the rule may appear smaller than otherwise.

The Bureau considers the mortgage market as it existed from 1997 through 2003 useful to assess some of the rule's possible effects when credit conditions, and the economy more generally, return to normal. During this period, home prices were generally rising and the housing market was in a positive phase. Notably, interest rates were falling in 2002 and 2003, which created a very large surge in refinancing activity. This period may not be perfectly representative of an “average” market, but these years span almost a full business cycle, capturing the end of 1990's expansion, the early 2000's recession and the beginning of the next expansion.

The analysis also uses data from the period 2004 through 2009. Beginning in 2004, the mortgage market in the United States was in the height of the housing bubble. In 2007 home prices, mortgage lending, and the economy more generally collapsed. The period that covers the “bubble” years and the crash that followed is also useful to gauge the impacts of the final rule. It is exactly the lending conditions during those years, and the damage they caused, that the statute and the final rule are primarily designed to prevent. Examining the performance and effects of the mortgages offered during this period, loans that were largely originated based on the perceived value of collateral, offers insights into the potential benefits and costs of the rule.

D. Coverage of the Final Rule

The provisions of the final rule require creditors to determine a consumer's ability to repay a “residential mortgage loan, ” excluding reverse mortgages and temporary bridge loans of 12 months or less, (referred to as “covered transactions”) ”and establish new rules and prohibitions on prepayment penalties. For these purposes, this rule covers with some exceptions, any dwelling-secured consumer credit transaction, regardless of whether the consumer credit transaction involves a home purchase, refinancing, home equity loan, first lien or subordinate lien, and regardless of whether the dwelling is a principal residence, second home, vacation home (other than a timeshare residence), a one- to four-unit residence, condominium, cooperative, mobile home, or manufactured home. However, the Dodd-Frank Act specifically excludes from these provisions open-end credit plans or extensions of credit secured by an interest in a timeshare plan. The final rule generally also excludes reverse mortgages, residential construction loans, and bridge loans with a term of 12 months or less.

E. Potential Benefits and Costs to Consumers and Covered Persons

In the analysis of benefits, costs and impacts, the Bureau has chosen to consider the ability-to-repay provisions together with the various qualified mortgage provisions. The discussion below first addresses the economics of an ability-to-repay standard, and considers the specific market failures that the statute and the rule aim to address. In general, market failures may include incomplete markets, externalities, imperfect competition, imperfect information, or imperfect information processing by consumers and several of those are discussed here. The benefits and costs of the requirement to assess ability to repay based upon documented and verified information are then discussed along with the impacts of the new liabilities, and the presumption of compliance that mitigates those liabilities established under the Dodd-Frank Act.

Additional provisions of the rule are considered including the impacts of the provisions related to points and fees, prepayment penalties and the definition of “rural or underserved”. The relationship between these provisions and other mortgage related rulemakings is discussed. The benefits, costs and impacts of the final rule in relation to several major alternatives are then discussed.

1. Economics of Ability To Repay

The basic requirement of Section 1411 of the Dodd-Frank Act is that a covered transaction may only be made when the creditor has made a “reasonable and good faith” determination that the consumer will be able to repay the loan. In the absence of any market imperfections, when negotiating a loan, both the lender and borrower would understand and consider the probability of default and the related costs should such a default occur. Creditors would extend credit if, and only if, the “price” of the loan, i.e., the risk-adjusted return (the return taking into account the expected loss from default) is high enough to justify the investment. Informed consumers would accept the loan if, and only if, the benefits of financing the property are worth the costs, including any expected costs in the likelihood that they default and cannot maintain access to the specific property.

The primary benefits or costs from an ability-to-repay requirement therefore derive from situations, where, absent such a requirement, these conditions are not met or where certain externalities may exist. These may include situations where the originator or creditor is not fully informed or has incorrect information about the transaction. More likely, a fully informed originator or creditor may not fully internalize all of the relevant costs, and is willing to extend credit even though the consumer may lack the ability to repay. Since the consumer willingly enters into the transaction, he or she must also be uninformed of either the true likelihood or true costs of default, or must not fully internalize all of the relevant costs. As discussed below, some of these situations arise when the lender or the borrower, fully understanding the risks of the loan and the inherent costs to themselves, do not factor costs borne by parties outside the transactions into their decisions.

Collateral based or “hard money” lending is one possible case where such lending could occur. If the lender is assured (or believes he is assured) of recovering the value of the loan by gaining possession of the asset, the lender may not pay sufficient attention to the ability of the borrower to repay the loan or to the impact of default on third parties. For very low loan-to-value (LTV) mortgages, i.e., those where the value of the property more than covers the value of the loan, the lender may not care at all if the borrower can afford the payments. Even for higher LTV mortgages, if prices are rising sharply, borrowers with even limited equity in the home may be able to gain financing since lenders can expect a profitable sale or refinancing of the property as long as prices continue to rise.

Other cases may involve loan originators who do not bear the credit risk of the loan, and therefore do not bear the ultimate costs of default. The common case is lenders who sell their loans: these lenders earn upfront origination fees from consumers and gains on sale but (absent complete contracts that provide otherwise) may not generally bear the costs of a later borrower default. As the relative size of the upfront fees increase, the potential agency problems do as well. The market recognizes the informational issues in these transactions and has developed mechanisms to mitigate adverse selection and moral hazard. For example, purchasers of loans engage in due diligence, either directly or by hiring third parties, validating the information provided about the loans and ensuring that the seller has provided only loans that meet agreed upon criteria. In addition, contracts provide that ex-post, should a loan perform poorly, the originator may have to repurchase the loan. This contracting feature is also designed to ensure that the initial creditor of the loan has the proper incentives to verify the borrower's ability to repay or the collateral value. Still, not all information about the loan may be captured and passed among sequential owners of the loan; some tacit information, not passed on, may give the creditor an informational advantage over others and diminishes the creditors' incentives to verify the consumer's ability to repay.

However, even lenders who maintain loans in portfolio may pay insufficient attention to the borrower's ability to repay. Cases where the loan creditor can earn sufficiently high up-front compensation, or where incentives of the individual loan originators and the creditor differ, may lead to lending that does not include a realistic assessment of the borrower's ability to repay. For example, a retail loan originator who earns commission may not have the same incentives as the owners of the bank that employs the loan originator and who will bear the ultimate cost of the loan once on portfolio. Even if such loan originators do not have final decision-making authority as to whether the creditor will make the loan, the loan originator controls the information that the underwriter receives and may have an information advantage that could systematically bias underwriting decisions. This information problem, and therefore the risk of poorly underwritten portfolio loans, may be even greater where the originator is not an employee of the creditor as is true in the brokerage and correspondent lending contexts.

In all these cases, the common problem is the failure of the originator or creditor to internalize particular costs, often magnified by information failures and systematic biases that lead to underestimation of the risks involved. The first such costs are simply the pecuniary costs from a defaulted loan—if the loan originator or the creditor does not bear the ultimate credit risk, he or she will not invest sufficiently in verifying the consumer's ability to repay. Even in cases where the lender does bear those costs, he or she will usually not fully internalize the private costs that a defaulting borrower will incur should default occur. Further, there are social costs from default that creditors may not internalize, as discussed below.

As noted earlier, the borrower also must decide whether to enter into the mortgage, and fully informed, perfectly rational consumers should consider their own risk of default and private costs in the event of default. However, as with lenders, borrowers may not fully anticipate the future probability or costs of default, either because they are uninformed or for other reasons. Consumers may underestimate the true costs of homeownership or be overly optimistic about their own future (or even current) financial condition. This can be exacerbated in the case of less sophisticated consumers negotiating with more informed mortgage professionals who have an interest in closing the loan and who may falsely reassure consumers about the consumers' ability to repay.

Consumers (and as noted above, creditors) may also misjudge the current or future value of the property securing the loan. This latter phenomenon was very much in evidence during the later years of the housing bubble as many consumers simply assumed that in times of financial stress, they could always sell or refinance. Further, consumers may not understand or may underestimate the costs they will incur in the event of default, such as the loss of the borrower's own home, costs of relocation, and the borrower's loss of future credit, employment and other opportunities for which credit reports or credit scores weigh in the decision.

As noted above, neither party to the transaction is likely to internalize costs to third parties. Even among very informed consumers and creditors, most will not internalize the social costs that delinquency or foreclosure can have. Research has consistently shown that a foreclosure will have a negative effect on the other homeowners in the vicinity either through the displacement of demand that otherwise would have increased the neighborhood prices, reduced valuations of future sales if the buyers and/or the appraisers are using the sold foreclosed property as a comparable, vandalism, and disinvestment. While the estimated magnitudes and the breadth of the impact differ, researchers seem to agree that there is a negative impact on houses in the vicinity of the foreclosure, and this impact is the highest for the houses that are the closest to the foreclosed house and for the houses that get sold within a short period of time of the foreclosed sale.

Research is also beginning to examine other spillover effects from foreclosures including increases in neighborhood crime and social effects on family members such as hampered school performance. Social policy has long favored homeownership for the societal benefits that may ensue; the negative spillovers from foreclosures can be seen as the inverse of this dynamic.

The Dodd-Frank Act and the final rule address these potential market failures through minimum underwriting requirements at origination and new liability for originators and assignees in cases where the standards are found not to be met. For qualified mortgages that have earned the conclusive presumption, meeting the qualified mortgage product criteria and underwriting requirements and pricing of the loan at a prime rate are judged in the rule to be enough to ensure that the lender made a reasonable and good faith determination that the borrower will be able to repay the loan. For loans where the final rule creates a presumption of compliance but leaves room for the borrower to rebut the presumption of compliance, or loans for which there is no presumption (i.e., loans that are not qualified mortgages) the lender may exert greater care in underwriting the loan than would be true in the absence of any liability for extending a loan which the consumer cannot afford to repay. Lenders therefore face an initial market tradeoff when choosing the optimal level of costs to bear in documenting and underwriting the loan and assessing the ability to repay (subject to the minimum standards all loans must meet): some increased effort (and therefore increased cost) at the time of origination may lower costs resulting from possible liability should the borrower become delinquent or default. Since assignees now share this liability, they have an additional incentive to monitor the behavior of the original creditor. The ex-post liability to the consumer mitigates the incentives for the creditor to shirk on the ex-ante investments in the underwriting.

Even creditors making the optimal choice of effort when documenting, verifying and underwriting the loan may still face some legal challenges from consumers ex-post. This will occur when a consumer proves unable to repay a loan and wrongly believes (or chooses to assert) that the creditor failed to properly assess the consumer's ability to repay before making the loan. This will likely result in some litigation expense, although the Bureau believes that over time, that expense will likely diminish as experience with litigation resolves more precise guidelines regarding what level of compliance is considered complete. After some experience, litigation expense will most likely result where compliance is insufficient or from limited novel sets of facts and circumstances where some ambiguity remains. Regardless of which party incurs the costs, the economic costs of these actions are the resources used to litigate these cases, thereby helping to ensure compliance and limiting the incidence of loosely documented originations. The reimbursement of interest and fees, along with the statutory damages, paid to the borrower, constitute, in economic terms, a transfer—a cost to the originator or assignee and a benefit to the compensated borrower.

2. Potential Benefits of the Ability-To-Repay Provisions for Consumers and Covered Persons

The final rule will help to ensure that loans are not made without regard for the borrower's ability to repay and thereby protect consumers and as noted above, others affected by defaults and foreclosures. (These others are themselves consumers and the adverse spillover effect from defaults and foreclosures very much impacts their economic well being.) Historically, the conditions under which credit is extended have been cyclical in nature. Periods of tight credit, such as the conditions that exist in the current mortgage market, are marked by reduced loan activity, very stringent lending standards, and extreme care in underwriting. In such periods, the benefits of a regime designed to require prudent underwriting, may be less apparent, and, in the near term, adopting such a regime, as the final rule does, will likely have little direct and immediate effect either on consumers or covered persons. As explained further in the discussion of costs to consumers and covered persons, lenders generally are already doing what the rule requires and a large majority of their loans will qualify for the conclusive presumption of compliance.

However, as credit expands, as it almost inevitably will, the final rule will help to ensure that loans are made properly and with regard for the borrower's ability to repay. To assess the benefits of the final rule, therefore, it is useful to examine the provisions of the final rule in the context of the recent housing bubble and its collapse in 2007.

There is growing evidence that many of the market failures in the previous discussion were in play in the years leading up to the housing collapse. In some cases, lenders and borrowers entered into loan contracts on the misplaced belief that the home's value would provide sufficient protection. These cases included subprime borrowers who were offered loans because the lender believed that the house value either at the time of origination or in the near future could cover any default. Some of these borrowers were also counting on increased housing values and a future opportunity to refinance; others likely understood less about the transaction and were at an informational disadvantage relative to the lender. These cases also included Alt-A loans taken by borrowers hoping to speculate on housing values.

In both of these situations, these loans frequently involved less traditional products, loans structured with minimal monthly payments in order to allow the borrower to qualify and to carry the loan for a period of time with minimal expense. Many of these loans were sold into the secondary market, limiting the lenders' credit risk, but many lenders also retained these loans on their own portfolios either with the intent of earning the full anticipated profits from such loans over time or with the intent to hold the loans for a period of time before selling them. And throughout the housing boom, most lenders and borrowers entering into such agreements failed to consider the costs that default would inflict on other properties (and the consumers who inhabited them) and on the financial system and economy writ large.

The benefits from the ability-to-repay requirements therefore come from further limiting and deterring unaffordable lending, above and beyond the current ability-to-pay requirements for higher-priced mortgage loans, and thereby reducing the ensuing private and social costs of excess delinquency and default. For example, the basic requirement that all loans be underwritten based on documented income and debt would have eliminated many of the loans made later in the bubble that led to crisis. Described as “stated-income” loans or “liar-loans,” these mortgages became very prevalent in the later years of the expansion and had very poor, and worse than expected, performance when the markets collapsed. There is also growing evidence that incomes on many mortgage applications were overstated in the years before the crash. Importantly, while limited and reduced documentation loans were a large segment of the subprime market, many of these loans were also made to prime, higher credit score borrowers and on properties with lower loan-to-value ratios. This suggests a substantial benefit to the documentation and verification requirements across all segments of the market, particularly the substantial majority of covered transactions that current ability-to-pay requirements do not cover now and are not expected to cover in the future.

As prices rose, aspiring homeowners borrowed money by misstating their income; many loan originators were at least indifferent to or even complicit or proactive in these endeavors. The systemic effects were evident: the extension of credit against inflated incomes expanded the supply of credit, which in turn continued the rapid rise of house prices in the later years of the housing boom and exacerbated the eventual crash.

The statute and the final rule also require that creditors must underwrite based on an amortizing payment using the fully indexed rate (or the maximum rate in five years for qualified mortgages) and including, with limited exceptions, any balloon payments in the first five years. This effectively bans the practice of underwriting loans based upon low upfront payments, either the lower interest-only payments on interest-only loans or negatively amortizing option ARMs or the teaser rates on hybrid ARMs.

In their later incarnations, interest-only and negatively amortizing loans (along with loans with terms greater than 30 years) were often sold on the basis of the consumer's ability to afford the initial payments and without regard to the consumer's ability to afford subsequent payments once the rate was recast. At the peak of the market, between 2004 and 2006, the percentage of loans that were interest-only, option ARMs or 40-year mortgages rose from just 7 percent of originations to 29 percent. The lower payment possibility for these loans allows borrowers to qualify for loans that they otherwise may not have been able to afford; but this comes with the same risks just described. The performance of many of these loans was also very poor, and worse than expected, with the onset of the downturn. The final rule does not ban such products outright, but rather requires that lenders that make such loans have a “reasonable and good faith” belief in the borrower's ability to repay and that in formulating such a belief the lender must calculate the monthly payment based on the fully indexed rate and fully amortizing payments, and does not allow these loans to enjoy the presumption of compliance associated with qualified mortgage status. The new underwriting requirements, coupled with the liability for violating these rules, should deter improper loans and ensure proper underwriting and diligence when making such loans; again limiting cases of personal or social harm.

Underwriting hybrid ARMs to the teaser rate was also a very common practice, in particular among subprime loans of the early 2000's. So called “2/28” and “3/27” loans were often underwritten based on the low initial payment, and exposed the borrower to potential payment shocks, and a need to refinance, two or three years into the mortgage. For example, in 2005, the teaser rate on subprime ARMs with an initial fixed-rate period of two or three years was 3.5 percentage points below the fully indexed rate. As a result, mortgages originated in that year faced a potentially large change in the interest rate and payment, or “payment shock,” at the first adjustment even absent any change in the index.

The evidence is mixed on whether payment shock at the initial interest rate adjustment causes default. And indeed, for some borrowers, these loans can be efficient contracts that allow for the extension of credit (see discussion below). However, the widespread use of the product put many borrowers in precarious financial positions and may also have fueled the systemic rise in home prices. The elimination of these products should limit both the individual and the systemic harms which ultimately translate, in the largest part, into harms to individual consumers.

The final rule reduces the likelihood that these products will reemerge on a broad scale and thus should limit the potential for individual and the systemic harms. The final rule bans no-doc and the old low-doc loans since the level of documentation is lower than that required by the rule). * * * The rule reduces the incentive to offer these other alternative mortgage products by requiring that underwriting be done assuming a fully amortizing payment at the fully indexed rate. The final rule also does not provide any legal protection for the lender that makes these loans (or the investor that acquires or guarantees them) as the loans are categorically disqualified from being qualified mortgage. These non-amortizing products will likely persist only in narrow niches for more sophisticated borrowers who want to match their mortgage payment to changes in their expected income stream and who have the resources to qualify for the products under the stringent underwriting assumptions the statute and regulation require. But these products will not likely be marketed as broadly as they were during the bubble.

In addition to the products just described, loans with points and fees (except for bona fide discount points) that exceed three percent of the total amount cannot be qualified mortgages, except as applicable for smaller loans as defined. Creditors may take more care in originating a loan when more of the return derives from performance over time (interest payments) rather from upfront payments (points and fees). As such, this provision may offer lenders more incentive to underwrite these loans carefully. As loans with higher points and fees are usually assumed to be offered to borrowers in weaker financial circumstances, this provision offers protection to that class of borrowers.

As discussed above, the various liability provisions provide the incentives for lenders to take proper care judging the borrower's ability to repay. This incentive is strongest for loans that are not qualified mortgages. Within the qualified mortgage space, higher priced mortgage loans (HPMLs) are still subject to ability-to-repay liability but afforded a rebuttable presumption of compliance. This liability already exists under rules that took effect in October 2009 for HPMLs, so that relative to existing rules, there are few benefits (or costs) associated with the liability provisions for such loans. However, there are some material differences in the underwriting requirements and smaller differences in the scope of the presumption where the liability now applies where it did not in the past. The new assignee liability may also strengthen the incentives relative to the existing rules.

Comparing the rebuttable presumption for higher priced qualified mortgages to the conclusive presumption (safe harbor) provision for qualified mortgages below the higher-priced threshold highlights the benefit of leaving the possibility of rebuttal in place. Borrowers paying higher rates on mortgage loans that meet the qualified mortgage product features are most likely to have lower credit scores, lower incomes and/or other risk factors; as such, it is among these subprime borrowers that a greater possibility exists for lenders to place the borrower into a loan that he or she may not have the ability to repay. The ability of the borrower to rebut the presumption of compliance leaves lenders with the additional incentive to “double check” the loan to examine further the borrower's financial condition and residual income, and to ensure that these higher risk borrowers have the means to live in the home they just purchased or refinanced.

Where a consumer is unable to afford his or her mortgage—and proves that the lender lacked a reasonable and good faith belief in the consumer's repayment ability at the time the loan was made—the damages the borrower recovers are a benefit to that party. The same damages should also be considered a cost to the lender and as such, estimates regarding the frequency of such actions and the dollar amounts involved are in the next section discussing costs.

Another impact of the differentiated structure of the final rule, where certain loans enjoy a conclusive presumption, others are given a rebuttable presumption and still others are subject to ability to repay scrutiny without the benefit of a presumption, is that some borrowers may gain “better” loans as lenders choose to make loans that qualify for the highest level of legal protection. Lenders in less competitive environments who have some flexibility over product offerings and/or pricing power may find it more profitable to offer a borrower a qualified mortgage rather than a non-qualified mortgage if, for such lenders, the expected value of the heightened legal protection is enough of an expected cost savings to offset any revenue reduction from making the qualified mortgage. For example, a creditor may restructure the price of a transaction with points and fees otherwise just above the points and fees limit for a qualified mortgage to have fewer upfront costs, and a higher interest rate, so that the loan is then under the limit and a qualified mortgage. Similarly, situations could exist where lowering the price on a loan would make the loan eligible for the safe harbor rather than the rebuttable presumption. The prevalence of these situations, or others similar situations, is hard to predict and depends on the future prices for mortgages in each of these segments, the competitive nature of the segments, and the individual lender's and borrower's situation.

The benefits of the rule, as discussed above, will be widely shared among individual borrowers, creditors, investors, and the public (consumers) generally. As discussed above, the loss that occurs when a consumer is unable to repay a loan is felt by the consumer, the holder(s) of that loan, and other parties outside the transaction including other consumers and would-be-consumers. Ensuring that lenders make a reasonable and good faith determination of the borrower's ability to repay should prevent a widespread deterioration of underwriting standards, the extension of excess credit and the broader negative effects that can have on these parties. To the extent lenders are deterred from making unaffordable loans, or encouraged to make more affordable loans, all of these parties will benefit.

3. Potential Costs of the Ability-To-Repay Provisions to Consumers and Covered Persons

In this part the Bureau considers costs to consumers and covered persons of the ability to repay provisions of the statute and final rule, including any potential cost in the form of reduced access to credit for consumers. The primary ongoing costs of the requirements of the final rule rest in the underwriting costs, including costs at origination to verify information on which the lender relies in the underwriting decision and the increased liability on lenders and assignees. As previously noted, in the current environment, lenders are already largely complying with these requirements and thus the rule should impose minimal, if any, ex ante costs. But in other credit environments, when creditors may wish to lower their underwriting criteria and require less documentation and perform less verification, the rule would require them to make a good faith and reasonable determination of ability to repay and to require them to incur ex-ante costs to document, verify and consider income and debt (and credit history). This should increase the quality of underwriting of mortgages at origination and thereby limit the prevalence of future delinquency and default, and the level of ex-post costs. (Of course, exogenous or unanticipated events and borrower behavior will still result in some delinquent and defaulting loans and some possible legal actions.) In this scheme, the possibility of legal recourse by the borrower serves as an incentive for better lender assessment of repayment ability as well as offering borrowers redress for wrongdoing. Lenders will determine the optimal combination of upfront underwriting cost and ex-post liability costs; to the extent these costs increase and competitive conditions allow lenders to pass this cost onto borrowers, some borrowers will pay more for their loans. At the margin, certain loans that were made in the past, namely those where the borrower has limited ability to repay, will not be made.

a. Costs of the Documentation and Underwriting Requirements

Two distinct requirements of the final rule—the requirement to verify income or assets, debt, and credit history, and the requirement to underwrite a mortgage based on an assessment of debt load using the fully indexed rate and fully amortizing payment—create costs for certain creditors and consumers. The final rule follows the statute in requiring that all creditors verify borrowers' income, debt and credit history. Reduced documentation loans were originally offered to high credit quality borrowers with substantial incomes. However, in the 2000's, the prevalence of these loans increased substantially and the borrowers to whom they were offered changed. Anecdotally, some of these loans could have been made with full documentation; however, for that subset of loans, it was precisely the reduced processing times and paperwork costs of originating these loans that made them popular among mortgage brokers and originators during the boom.

From this perspective, for certain consumers and creditors, requiring full documentation and verification may result in the loan being made with a less efficient contractual form, or possibly in the loan not being made. In these latter cases, consumers would lose the benefits they get from the mortgage (the benefits of owning a home, for example, or the benefits of obtaining better terms on a loan through a refinancing) and creditors would lose any profits on the loan. However, for most other originators, and consumers, reduced documentation loans were a way to grant credit to unqualified borrowers who did not have the means to afford the mortgage. As discussed in the benefits section, the elimination of these loans in these circumstances is a principal benefit of the rule.

For borrowers for whom the most efficient outcome (from a societal perspective) is, in fact, a reduced documentation mortgage, the requirements in the final rule have two possible costs. The time and material to verify the required underwriting elements with documents are true resource costs; depending on competitive conditions, the lender or the borrower may bear the actual costs. Precise estimates of these costs from time and motion studies or cost function analyses are not available, but the required pay stubs or tax records should not be a large burden. The final rule allows income to be verified utilizing copies of tax returns which the consumer can provide the creditor and permits debts to be verified utilizing a credit report. For those with more idiosyncratic income sources that would somehow not be reflected on a tax return, the costs may be slightly higher. However, it is also possible that certain loans that would be made absent the documentation requirements would not be made under the rule. This could happen, for example, in cases where the cost of documenting the required factors is sufficiently high or where the borrower pays an exorbitant “privacy” cost in disclosing the documents. The final rule only requires that income or assets be verified to the extent they are relied upon by the creditor in assessing the consumer's ability to repay; thus the consumer is not required to disclose or document income or assets except if the consumer prefers to have her ability to repay assessed without regard to the undisclosed information. In the event that there are cases in which, despite these rules, a consumer who could qualify for a mortgage is unwilling to incur the privacy cost in documenting income or assets, the transaction will not occur: and the benefit to consumers and lenders from these `lost' transactions is the relevant cost.

Relative to industry practice today, these requirements are likely to impose only a very limited burden for creditors. With the exception of the two situations discussed below, most loans today are made under very stringent, and perhaps inefficiently high, documentation requirements. The Bureau understands that full documentation is required for all purchase loans and many refinance loans being supported by government programs such as FHA. In addition, both Fannie Mae and Freddie Mac currently require full documentation. The Bureau believe that only a small subset of loans that creditors intend to hold on portfolio are underwritten today without the documentation that meets or is very close to the documentation required by the final rule. For this limited set of loans, the rule imposes the costs already described: The direct compliance costs to collect the required documentation in order to verify the information provided by the consumer and any costs from forgone transactions.

One exception to the stringent documentation requirements now prevailing in the market (and exceeding the requirements of the rule) are certain streamlined refinance programs aimed at aiding the housing market recovery and certain targeted housing support programs offered to low and moderate income borrowers. The Bureau recognizes that the requirements of the final rule could greatly increase costs for these programs and hinder their success. It also recognizes that the possibility of consumer harm is likely limited in these contexts. As a result, elsewhere in today's Federal Register the Bureau is proposing certain exemptions from these requirements and seeking comment on the scope of such exemptions.

There may also be some situations where lenders may have systems to document and verify the required information, but who do so in a manner that varies slightly from the provisions of the rule. These lenders may have to bear some costs to modify their systems or practices, but as noted above the Bureau understands there to be few such cases. Lenders who do collect information as required by the final rule, but who may use it differently may also incur some costs. For example, certain lenders may have systems or procedures in which the calculation of the DTI ratio does not conform to the requirements in appendix Q. Such a creditor could continue its current practices, which should they satisfy the ability-to-repay requirements, albeit without the benefit of a presumption of compliance. Lenders that prefer to make qualified mortgages with a presumption of compliance would have to bear the costs to modify systems or make other changes in order to calculate the required figures according to the rule. Modifications to information technology systems may also be necessary to enable lenders to label and track qualified mortgages.

More broadly, the Bureau also recognizes that the establishment of the ability-to-pay requirements and the related distinction for qualified mortgages under the Act, will require modifications to existing compliance systems and to creditors' other management policies and procedures. For example, review and monitoring procedures may have to be altered to ensure compliance with the new requirements. Again, given the current state of the mortgage market, it is likely that many of these procedures are largely already in place.

If measured relative to the benchmark of the earlier periods, either the period from 1997 to 2003 or the later years of the bubble, the requirements of the final rule could be seen to impose more substantial costs. Over the former period, there were more limited documentation loans than today, however it appears that many of these arose in the situations described where such lending is efficient. By the latter period, there were even more such loans and the balance appears to have shifted to one where many if not most of the limited documentation loans had misstated income and other deficiencies.

During those periods there were likely some lenders, as evidenced by the existence of no-income, no-asset (NINA) loans, that used underwriting systems that did not look at or verify income, debts, or assets, but rather relied primarily on credit score and LTV. Under the final rule, these lenders would be impacted in two ways: They would have to collect and verify income, assets and debts; and more importantly, they would have to change much of their underlying business model to consider the required factors. As noted, the Bureau does not believe such lending is currently being practiced, and the benefits of preventing such lending may be substantial (as discussed above).

The requirements that all loans be underwritten assuming a fully amortizing payment and the fully indexed rate (or to obtain qualified mortgage status the maximum rate within 5 years of origination) have costs similar in nature to the documentation requirements. There are some individuals or households with projected increases in income that will match the projected increased housing costs; the final rule allows the creditor to factor expected future income into the denominator of the debt-to-income calculation but does require that the numerator be calculated on the fully-indexed payment. There also may be individuals with constant income but a housing need that is shorter than the introductory period. In at least these latter cases, there may be some loans where it is efficient to qualify the borrower only on the current payment or some other amount. It is difficult to quantify the set of borrowers affected in this way, however to the extent that those loans are not made, both the lender and borrower will incur the costs of lost profits and lost consumer benefits, respectively.

The provisions of the rule requiring extended retention times for documentation sufficient to show compliance with the rule (from two years to three years) will also impose some very limited costs on creditors. Electronic storage, communication and backup are very inexpensive and are likely to decrease in costs further.

b. Liability Costs

Creditor may trade off the ex-ante underwriting cost just discussed with ex-post liability costs that stem from TILA's liability provisions and their interaction with the rule's qualified mortgage and presumption of compliance provisions. Qualified mortgages with interest rates below the threshold for higher-priced covered transactions enjoy a conclusive presumption of compliance (although disputes may arise as to whether a particular loan meets the qualified mortgage test); qualified mortgages above the specified interest rate threshold enjoy a rebuttable presumption of compliance with the ability-to-repay requirements; and, loans that are not qualified mortgages are subject to general ability-to-repay provisions, under which the borrower will bear the burden of proof for establishing a violation. Within each segment, lenders and borrowers (or their attorneys in contingency arrangements) must pay for the costs of litigation, whether such litigation arises in the context of a private right of action brought by the borrower, or a defense raised by the borrower to a foreclosure. Originators and assignees also face various contingencies that may arise if such a claim is raised or succeeds.

Within each segment, the additional costs increase proportionally with borrowers' probability of delinquency or default. For example, the additional cost for qualified mortgages with a rebuttable presumption of compliance is smallest for lower debt-to-income (DTI) ratio loans (since these borrowers are less likely to be in a position to need or want to bring claims) and increases as the DTI ratio (keeping other factors constant) rises. The same is true as the interest rate of a loan increases, assuming that interest rate is accurately calibrated to risk.

In estimating empirically the long-run additional liability costs from alleged or actual violations of the final rule, the Bureau examines the mortgage market as it existed from 1997 to 2003. The Bureau applies that market data and the pre-statute baseline to compare the liability for creditors under the final rule to the liability they would have incurred under the legal regime that existed under federal law just before passage of the Act.

i. Size of the Market Segments

The data used in estimating liability costs comes from several sources. Data regarding the loans guaranteed or purchased by Fannie Mae and Freddie Mac are from the Historical Loan Performance (HLP) dataset maintained by FHFA. The FHFA shared a one percent random sample of these loans with the Bureau, along with information about their characteristics and performance. In the notice to reopen the comment period for this rulemaking, the Bureau detailed these data and requested comment. Commenters were generally supportive of using these data, but suggested looking at other sources as well including proprietary industry datasets available for sale. These data cover a large but select portion of GSE loans. In contrast, the HLP data cover the entire universe of GSE loans and even the one percent sample is more representative. As such, the Bureau believes the HLP data are the better data for the GSE segment of the market and has consulted with the suggested sources in other parts of the analysis. Over the 1997-2003 period loans guaranteed or purchased by the GSEs comprised roughly 47 percent of the mortgage market.

Similarly, information on loans insured by the FHA was provided by the FHA in response to the June 5, 2012 notice. The data cover the years from 1997 to 2011 and exclude Home Equity Conversion Mortgages (HECM) as well as mortgages with seller-funded downpayment. Combined with loan insured by the Veterans Administration or the Rural Housing Service, these loans comprised an estimated 9 percent of the market during this period. The Bureau did not get loan-level data from the VA or RHS.

Data on mortgages in non-agency securitizations were taken from proprietary industry sources that the Bureau has licensed. While less complete than the HLP files, these data also include data on the characteristics and performance of individual loans. Over the 1997 to 2003 period, this segment comprised roughly 13 percent of originations. The remaining loans are those held on the balance sheets of banks, thrifts and credit unions. While aggregate data regarding the performance of these portfolios is available, comprehensive loan level data similar to the enterprise, FHA and private-label loans is not. As a result, the actual characteristics of individual loans are not available.

Without the temporary provisions granting qualified mortgage status to certain loans that are eligible to be purchased by the GSEs or insured by FHA, VA and RHS, of the mortgages originated during the 1997 to 2003 period, the Bureau estimates that roughly 70 percent of would have been qualified mortgages. Most of these loans would qualify for the safe harbor, and perhaps one to four percent points of these loans would have been qualified mortgages subject to the rebuttable presumption. Another 22 percent of loans would have been non-qualified mortgages subject to the ability-to-repay requirements. The remaining 8 percent of loans made over that period were appear to have been made without sufficient documentation to be permitted under TILA section 129C documentation or were subprime hybrid adjustable rate mortgages underwritten to teaser rates in a way that is no longer allowed under the final rule. An important caveat is that these estimates are not adjusted to account for: (1) Loans with total points and fees above the thresholds and therefore not eligible to be qualified mortgages; (2) the exception of rural balloon loans to qualified mortgages; or the exception for streamlined refinancings of non-traditional loans.

Based on data from 2011, the Bureau estimates that without the temporary provisions granting qualified mortgage status to certain loans purchasable by the GSEs or insurable by FHA, VA and RHS, 76 percent of mortgages would have been qualified mortgages inside the safe harbor, 2 percent of mortgages would have been qualified mortgages with a rebuttable presumption, and 22 percent of mortgages would have been subject to the ability-to-repay requirements. These estimates are subject to the same limitation stated above.

ii. Liability Costs for Qualified Mortgages

For qualified mortgages claimed to be within the safe harbor, borrowers will have no claim against the lender for ability-to-repay violations unless the loan does not in fact meet the requirements for safe harbor treatment. Based on the experience of loans originated during the 1997-2003 period, the Bureau estimates that roughly four percent of qualified mortgages loans will ever be 60 days delinquent and less than one percent are expected to result in foreclosure. The performance of the qualified mortgages that have a conclusive presumption of compliance is expected to be slightly better than these averages.

The Bureau believes that only a very small fraction of these delinquent or foreclosed-upon borrowers would seek to raise an ability-to-repay claim. The conclusive presumption precludes liability for loans which meet the eligbility criteria for a safe haror, i.e. loans whose product features make them eligible; for which the lender verified income, assets, and debts and properly calculated the DTI ratio to be 43 percent or less; and which are not higher priced. And even if a loan is erroneously categorized as a qualified mortgage with a safe harbor, a borrower still cannot recover unless the lender has violated the general ability-to-repay requirements, including the requirement that the lender make a “reasonable and good faith” determination that the consumer had the ability to repay. Generally, only a small percentage of borrowers contest foreclosure and even smaller percentage do so with the benefit of legal representation. This fact, and the limited chance of success for borrowers to raise successful claims, makes it very unlikely that many claims will arise from borrowers with these qualified mortgages.

For qualified mortgage loans above the higher-priced threshold, costs (as well as benefits) of the final rule derive from the differences, including differences with respect to the originator and assignee liability, between the existing liability rules and the final rule. Under existing rules, creditors that make a higher-priced mortgage loan (HPML) are not allowed to extend credit without regard to “the consumer's repayment ability as of consummation, including the consumer's current and reasonably expected income, employment, assets other than the collateral, current obligations, and mortgage-related obligations.” Further, a creditor is presumed to have complied if the creditor properly verifies and documents income and assets, made the determination using the largest payment of principal and interest scheduled in the first seven years following consummation, and took into account the ratio of total debt obligations to income, or the income the consumer had after paying debt obligations.

As noted, 1 to 4 percent of loans, based on data from the 1997- 2003 period, are estimated to be qualified mortgages with a rebuttable presumption. As just described, the delinquency rates and default rates are expected to be just around 4 percent and 1 percent respectively.

Nearly all of the mortgages that will be qualified mortgages above the higher-priced threshold are currently covered by the existing HPML presumption of compliance, because the requirements in the final rule that qualified mortgage loans be fully documented, have verified income and be underwritten to the maximum payment in the first five years of the loan (with the exception for rural balloon loans) will in most cases also satisfy the requirements for obtaining the presumption under the 2008 HOEPA Final Rule. The final rule's requirements for obtaining the status of a qualified mortgage (and thus the rebuttable presumption) are slightly more prescriptive than the existing rules for gaining that presumption and this difference in the criteria for qualification may leave borrowers with slightly less opportunity to rebut the presumption of compliance.

For the subset of these borrowers that are in default more than three years into the mortgage, that seek to and are able to successfully rebut the lender's presumption of compliance (when seeking an offset during foreclosure), and that are therefore entitled to compensation, the returns from this action are in fact reduced relative to the existing rules which do not limit the recovery period in a claim for offset in a foreclosure proceeding brought by the creditor. As such, the probability that lenders will have to defend such an action is reduced relative to current rules although the subset described above is likely to be so small that the impact will be immaterial. As discussed below, relative to the existing rules lenders may face increased putback risk from investors although that, too, is small.

For the set of borrowers that are in default within the first three years, potential damages are not reduced; however, the increased requirements at origination to qualify for qualified mortgage status, and the correspondingly more limited grounds on which to rebut the presumption reduce the probability of a successful challenge. So here too, the probability that lenders will have to defend such an action may be reduced or at least held constant relative to current rules. Overall, therefore the ex-post liabilities for lenders are likely reduced for these loans.

Relative to current rules for HPMLs, the current rule extends liability to assignees. The establishment of assignee liability does not increase the amount that a borrower can obtain from a successful legal action; however, it does increase the number of parties from whom the borrower can seek redress. Borrowers in a foreclosure action in a judicial state can now assert their claim against the assignee bringing the foreclosure action, rather than having to initiate an affirmative lawsuit against the originator that no longer holds the loan. The effect is to reduce the costs of bringing these defensive actions and therefore increasing their likely number. For loans that are not sold, or for borrowers wishing to bring affirmative actions, the establishment of assignee liability has little or no effect.

The extension of liability to assignees may also increase the cost of contracting between the two parties. Under the final rule, the borrower now has a contingent claim against two parties. As a result, the two parties will want to contract ex-ante about the extent of each party's liability under the various contingencies. This increase in contracting costs should be small for two reasons. First, even in the absence of assignee liability, the market has already included these contingencies in standard contracts. For example, following the Board's 2008 rule, the Fannie Mae seller servicer guide was amended to include provisions that HPMLs are “eligible for delivery to Fannie Mae provided [that] * * * lenders represent and warrant when they sell an HPML to Fannie Mae that the mortgage complies in all respects with Regulation Z requirements for HPMLs, including the underwriting and consumer protection requirements. ” The Freddie Mac seller servicer guide has similar provisions. With contracts like these already in place, it appears that amending contracts for the particulars of the final rule should be small. Second, underwriting guidelines, pooling and servicing agreements and other contracts in the mortgage market are currently being reworked and refined. Among the myriad of changes, addenda to manage the ability-to-repay liabilities of the current rule should be only a small cost.

iii. Non-Qualified Mortgages and Estimation of Costs

The remaining loans are not qualified mortgages. These include for example, mortgage loans with a back-end DTI ratio over 43 percent, loans with points and fees above three percent of the loan balance, mortgages with a term over 30 years, or balloon loans that do not qualify for qualified mortgage balloon definition. For loans in this segment priced below the higher-priced threshold, the obligation to assess the consumer's ability to repay and the liability where the lender fails to do so is a new liability for both the originator and any assignees. For loans in this segment above the higher-priced threshold, lenders cannot invoke a rebuttable presumption of compliance and for those loans that are not high-cost loans, assignees are subject to expanded liability as compared to current rules.

The Bureau has estimated litigation costs under the new ability to pay standards for non-qualified mortgages. Estimating costs for non-qualified mortgages should reasonably serve an upper bound for the costs for qualified mortgages. Costs for putbacks, or loans the buyers of which force the sellers to take back on their books because they do not satisfy the final rule are also estimated.

Estimating the increased liability costs involves a series of assumptions about the performance of these loans, the probability that borrowers will bring particular actions, and the subsequent behavior of lenders and courts. Some assumptions about costs are also necessary.

Under the ability-to-repay provisions, consumers can bring an action against the lender at any point during the first three years of the loan or as an offset to foreclosure at any time. In the latter cases, the recovery of interest and finance charges is capped at the amount paid during the first three years.

The Bureau has estimated these costs as follows. To begin, assume an average loan balance of $210,000 (just below the mean balance for first lien loans reported in HMDA in 2011), an average interest rate of 7 percent (the average mortgage rate for 30 yr. mortgages from 1997 to 2003) , and an average of $3,150 (1.5 points) paid up front in fees. Further, assume that, on average, affirmative cases and contested early foreclosures happen at the midpoint of the period, 18 months after consummation. This implies that for the affirmative cases, and the early foreclosures borrowers contest, successful borrowers are reimbursed for fees and interest an average of roughly $29,200. (The Bureau assumes in this calculation that all prevailing borrowers receive $4,000 in statutory TILA damages.) For the later foreclosures, defined here as foreclosure that occur three or more years after loan consummation, borrowers who contest foreclosure are reimbursed for 36 months of interest or roughly $51,250.

Based on data from the FHFA for 1997-2003 for loans with DTI ratios above 43 percent, it is reasonable to assume, 3.5 percent of loans reach 60 day delinquency during the first three years of the loan but do not start a foreclosure process, an additional 1.5 percent of loans start the foreclosure process within the first three years, and an additional 1.5 percent of loans start the foreclosure process after three years. The Bureau believes that consumers who have fallen behind on their mortgage payments are unlikely to initiate an ability to repay claim in court prior to foreclosure. Rather, they will likely seek to work with their servicer and the owner of the loan to cure the delinquency through, e.g., forbearance or some form of loan modification, or where that is not possible, to reach an agreement to enable the consumer to walk away from the property and the loan (i.e., deed in lieu or short sale). Once a foreclosure proceeding is commenced, however, it will then be in the interest of consumers to assert ability-to-repay claims where there is a plausible basis to do so; this is especially true in judicial foreclosure states because an ability-to-repay claim can be asserted as a defense by way of offset against whoever holds the loan at the time of the foreclosure (i.e., the originator or assignee).

The ability of consumers to assert such claims either defensively or, in non-judicial foreclosure states, in affirmative actions will depend to some extent upon their ability to obtain legal representation. In its notice reopening the comment period for the rule, the Bureau specifically requested information and data regarding the frequency of such actions. In general, industry commenters asserted, that even under the rebuttable presumption standard, future legal actions under the rule would be very common. In contrast, consumer and community groups pointed to the available evidence and experience to suggest that only a very small minority of consumers in foreclosure are represented and that very few claims are brought. Consumer group commenters pointed out the practical limitations of consumers to bring an ability-to-repay claim, noting that few distressed homeowners would be able to afford and obtain legal representation often necessary to mount a successful rebuttal in litigation. Consumer groups also provided percentages of borrowers in foreclosure who are represented by lawyers, noting the difficulty of bringing a TILA violation claim, and addressed estimates of litigation costs, such as attorney's fees. The data provided however are quite limited: two commenters (both representing industry) suggest that during the recent years there were roughly 900 mortgage-related TILA cases filed each year in Federal court while data regarding the number of TILA claims brought in state courts were not provided.

More specifically the Bureau has considered the available evidence with respect to the extent of litigation under laws potentially analogous to this one, such as the 2008 HOEPA Final Rule (which does not provide assignee liability, except as applicable to high cost mortgages) and under HOEPA and state anti-predatory lending laws (which generally do provide for assignee liability). So far as the Bureau is aware, claims under these rules have been very infrequent. Industry participants likely have access to the most complete information about litigation activity, much of which activity is not reported in legal databases such as Lexis and Westlaw. Industry commenters, however, did not bring forth any evidence to suggest that claims have been anything but rare. Thus, relative to the one to two million annual foreclosure starts from 2009 through 2011, the record supports a conclusion that litigation under TILA generally and under the most directly analogous federal and state laws has been very limited.

Industry commenters maintained that past experience is not a guide because new liability under the Dodd-Frank Act will increase incentives for litigation. The Bureau recognizes that the availability of new ability-to-repay remedies may make it easier for consumers to obtain representation (by providing those consumers whose loans are not currently covered by the Board rule with new rights; and those consumers whose loans are covered, with more easily asserted, and to that extent more valuable claims). Thus, the analysis below of litigation costs relies on very conservative (likely unrealistic) assumptions about the extent to which the Dodd-Frank liability provisions will increase litigation levels above levels under current laws.

Among the three percent of borrowers that are in foreclosure, the Bureau assumes that 20 percent will bring an action against the lender for failing to meet the ability-to-repay requirements; that implies that 0.6 percent of borrowers will bring claims. As noted, this value is many times higher than recent experience with the 2008 HOEPA Final Rule or analogous state laws would suggest and is a very conservative upper bound. One half of these borrowers, should they prevail, are assumed to be entitled to 18 months of interest and the other half to 36 months of interest. Based on our assumed loan size ($210,000), interest rate (7%), and origination fees ($3,150) as discussed above, on average a successful borrower will have a claim of $40,225 (including the statutory TILA damages, before legal costs).

To estimate legal costs, assume that in each case, the lender will move for summary judgment based upon what they are likely to claim to be undisputed evidence documenting their consideration of borrowers' ability to pay. The consumer would likely claim that he or she was unable to pay the mortgage from its inception, and would have to present evidence from which it could be inferred that the creditor did not make a “reasonable and good faith determination” of the consumer's ability to repay. To estimate legal costs, assume that in each case, following any discovery permitted, the lender will move for summary judgment, which is a written request for a judgment in the moving party's favor (along with a written legal brief in support of the motion with supporting documents and affidavits) before a lawsuit goes to trial, claiming that all factual and legal issues can be decided in the moving party's favor, as a means to avoid trial altogether. The opposing party (i.e., the consumer) would need to show that there are triable issues of fact. The analysis assumes that, in these motions, the lender will succeed four-fifths of the time. In the remaining one fifth of cases, the lender settles prior to summary judgment and pays the full value of the claim. This assumption is also conservative. In evidence provided by industry commenters which the commenters suggested were analogous, lenders prevailed in nearly all of the cases cited.

To litigate these cases, the borrower is assumed to spend 60 hours of attorney time up to and including responding to the motion for summary judgment while the lender, given its resources, is assumed to spend 170 hours up to and including filing the relevant motions. In 2011, the average wage for lawyers in the legal services industry was $68.75/hr; adjusting that figure to reflect benefits and other forms of compensation, and a 50 percent mark-up for firm yields an hourly rate for legal services of $150/hr. With these assumptions, borrowers are willing to bring cases, and lenders will defend them, since on average both sides are ahead relative to simply dropping the claim or paying it in full. To reflect the expected value of these costs, the costs of non-qualified mortgages would increase by 10 basis points (0.1 percent of the loan amount, or roughly $212 for the $210,000 loan). Assuming loans with a weighted average life of four years, this could add roughly 2.5 basis points (0.025 percentage points) to the rate of each loan. Were the whole cost passed on to the consumer, increasing the rate from 7.0 percent to 7.025 percent, the monthly payment would rise by roughly $3.50. The resource cost to litigate this case is also roughly 10 basis points since it includes the lenders' and the borrowers' legal expenses of $25,500 and $9,000, respectively, and excludes the transfer of $40,225 that occurs in successful cases.

iv. Sensitivity Analysis

As part of a sensitivity analysis, the Bureau has estimated these costs under different assumptions. Notably, industry commenters provided estimates of the costs for various types of cases related to mortgage actions. These comments suggest a much higher cost for legal expenses of $300 per hour and closer to 300 hours to litigate cases that involve motions for summary judgment. Using these figures (and the assumption that borrowers' legal expenses include a proportionally higher 150 hours at $300/hr), the increased cost of each loan is approximately 31 basis points or an increase in the interest rate of just under 8 basis points (0.08 percentage points). Importantly, in this scenario, using the assumptions set forth previously about loan size and other factors, lenders would spend $107,000 to defend claims worth substantially less than the legal costs ($40,225). It is possible, however, that lenders would be willing to litigate such cases in order to discourage future litigation but, if so, one would expect a corresponding diminution of litigation over time.

As a second sensitivity test, going back to the original legal cost estimates, one can assume that of the 3.5 percent of borrowers who find themselves behind on their payments during the first three years, 84 percent (or 3 percent of total borrowers) chose to bring affirmative claims. This would quintuple the original estimates on a per loan basis to fifty basis points spread over a four-year average life. Similarly, one could assume that a larger percentage of borrowers in default bring claims. Raising that assumption from 20 percent to 40 percent results in estimated costs of 20 basis points per loan.

Originators and assignees share the liability for ability-to-repay violations. Depending on the contract in place, lenders will bear some repurchase risk for those loans that are sold into the secondary market. For example, sellers of loans to the GSEs already bear this risk for HPMLs since the enterprises have the right to put the loan back in case of ability-to-repay violations. In cases where the lender is defunct or there are other issues affecting the lender's capacity to reassume the risk, the purchaser of the loan may be unable to exercise that right and will bear the additional liability costs. The need of both the seller and the buyer to budget for expected capital and liquidity charges in these situations, and to negotiate the specific transactions, will also add some costs. However, in recent work, some economists have estimated that even for loans from the 2005 to 2008 vintage repurchase risk added conservatively about 19 basis points (or 0.19 percent of the loan amount) to the cost of a loan. Given the much lower default rates in the coming years (based on the default rates during the 1997-2003 period), and the increased underwriting requirements mandated by the final rule even for non-qualified mortgages, these costs are likely to be closer to 1-3 basis points at most.

v. Summary of Litigation Costs

Combining liability costs and repurchase costs, estimated costs for non-qualified mortgage loans (loans made under the ability-to-repay standard without any presumption of compliance) are estimated to increase by approximately twelve basis points (or 3 basis points (0.03 percentage points) on the rate); under very conservative estimates, this figure could be as high as forty basis points (or ten basis points (0.01 percentage points) on the rate). Depending on the competitive conditions in the relevant product and geographic markets, some of this increase will be passed on to borrowers and the rest will be absorbed by lenders. Certain borrowers may be priced out of the market as a result of the price increase. However, the number of such borrowers is likely to be very small given the values above since an increase of even ten basis points on the rate on an average mortgage would increase the monthly payment by less than $10.

vi. Temporary Provisions for Qualified Mortgages

As described in the preamble, the final rule recognizes the fragility of the current mortgage market and therefore includes temporary measures extending qualified mortgage status to loans that in the long run may not be qualified mortgages. These include loans with a DTI above 43 percent and that nonetheless can be purchased or guaranteed by the GSEs, insured by the FHA, VA or RHS. Based on the data as of year-end 2011, such loans are approximately 18 percent of the market. Without fuller data on the points and fees and product features associated with most loans, it is hard to estimate precisely the size of this segment or predict how large it would be several years from now with, or without, the statute taking effect. Ignoring those features, based on information about the rates and fees on these loans we believe roughly 97 percent of these loans should qualify for the legal safe harbor with the conclusive presumption of compliance (i.e., they are not higher-priced covered transactions) and 3 percent are estimated to qualify for the rebuttable presumption (i.e., they are higher-priced covered transactions). The temporary expansion of the definition of a qualified mortgage results in over 95 percent of the market being granted qualified mortgage status.

Extending qualified mortgage status to these loans reduces costs to lenders as described above and limits some of the consumer protections that an increased possibility of liability would create if a creditor were able to satisfy the GSE or federal agency underwriting standards without having a reasonable and good faith believe in the consumer's ability to repay. However, the added certainty from this reduced liability should benefit both consumers and covered persons. The mortgage market is still fragile, even four plus years past the most turbulent portions of the financial crisis. With lenders and the markets in general adjusting to new regulations designed to counter the forces behind the crisis, extending qualified mortgage status to these segments of loans should limit any disruption to the supply of mortgage credit with only limited effects on consumers. The extension of qualified mortgage status to these loans should allow the market time to digest the rules and for any increase in premia associated with uncertainty about litigation and putback costs to diminish.

c. Access to Credit

Overall, the Bureau believes that the final rule will not lead to a significant reduction in consumers' access to consumer financial products and services, namely mortgage credit. The Bureau notes the potential for the ability to repay requirements, including increased documentation and amortization requirements, to prevent some consumers from qualifying for a loan. First, the final rule generally bans no-doc and low-doc loans to the extent the level of documentation is lower than that required by the rule. The final rule would by definition prevent borrowers who would only qualify for these types of loans from receiving a mortgage; as discussed, that is one of the benefits of the rule. Second, the final rule generally increases documentation requirements for mortgage loans and requires underwriting to be done based on an assumed fully amortizing loan at the fully indexed rate.

As noted above, when measured against the current marketplace, the Bureau anticipates the effect of these requirements on access to credit to be very small. The Bureau anticipates that, as the economy recovers, the currently restrictive credit environment will loosen. Indeed, if anything, the Bureau anticipates that the immediate effect of the rule may be to contribute to the recovery of the mortgage market by reducing legal uncertainty which may be affecting lending. This is especially true if the impact of the rule were compared to a post-statutory baseline (i.e. to the implementation of the Dodd-Frank ability to pay and qualified mortage provisions without implementing regulations.)

Measured against the years leading up to the financial crisis, when lending standards were quite loose, the effects of the final rule on access to credit would of course have been significantly larger. The final rule will set a floor to the loosening of credit in order to prevent the deterioration of lending standards to dangerous levels. A primary goal of the statute was to prevent a repeat of the deterioration of lending standards that contributed to the financial crisis, which harmed consumers in various ways and significantly curtailed their access to credit. Such a goal will, by definition, entail some potential diminution of access to credit as market standards change over time. The Bureau believes that, to the extent the final rule reduces credit access, it will primarily reduce inefficient lending that ignores or inappropriately discounts a consumer's ability to repay the loan, thereby preventing consumer harm, rather than impeding access to credit for borrowers that do have an ability to repay. The Bureau notes that the rule may have a disproportionate impact on access to credit for consumers with atypical financial characteristics, such as income streams that are inconsistent over time or particularly difficult to document.

There also exists the potential for both increased documentation requirements and increased liability to increase the price of mortgage loans for some consumers. As discussed above, price increases from both increased documentation requirements and increased liability should be small. The documentation requirements, such as providing a pay stub or tax return, will impose relatively little additional cost to most consumers. Similarly, the increased documentation costs for creditors should not be significant, or result in more than relatively small increases in the cost of mortgage loans.

With respect to liability costs, the Bureau notes that over 95 percent of the current market is estimated to satisfy one of the definitions of a qualified mortgage, greatly reducing the expected cost of litigation. The Bureau also notes that the clear standards established for determining whether a loan is a qualified mortgage should reduce uncertainty regarding litigation costs, which will mitigate any resulting impact on access to credit. In light of the foregoing considerations, the Bureau believes that the ability to repay requirements and the accompanying potential litigation costs will create, at most, relatively small price increases for mortgage loans. These small price increases, in turn, are not likely to result in the denial of credit to more than a relatively small number of borrowers, some of whom commenters pointed out could be low income, at the margin.

The Bureau notes that concerns have been raised concerning the application of increased documentation and amortization requirements to such entities as certain nonprofits and state housing finance agencies, as well as certain refinancing programs. As applied to such entities and programs, the final rule may restrict access to mortgage credit, including for consumers who may otherwise have limited credit options, while doing little to further the consumer protection purposes of the statute. To address these concerns, the Bureau has proposed separately to exempt some such entities and programs from these documentation and amortization requirements.

The Bureau also notes that concerns have been raised regarding the application of the qualified mortgage criteria and the general ability to repay requirements to certain small creditors. These concerns arise from the observation that for many community banks and credit unions, for example, compliance resources are scarce and compliance costs as a percentage of revenue can be high. At the same time, these institutions employ a traditional model of relationship lending that did not succumb to the general deterioration in lending standards that contributed to the financial crisis. Moreover, because this business model may be based on particularized knowledge of customers and the development of durable customer relationships, the resulting loans may be beneficial to customers even when they do not conform to the general standards set forth in the final rule. Further, these institutions have particularly strong incentives not only to maintain positive reputations in their communities, but also, because they often keep the loans they make in their own portfolios, to pay appropriate attention to the borrower's ability to repay the loan. Accordingly, the Bureau has proposed separately to provide additional criteria by which certain small portfolio lenders may make qualified mortgages.

Greater access to credit can be associated with higher home prices and higher homeownership rates, and as discussed in the section on costs, there is some evidence of positive social effects from home ownership. As such, were the rule to overly restrict credit, it is important to note that these positive spillovers would also be limited. However, the Bureau does not believe that the rule will result in an inappropriate reduction in access to credit; rather, over time, the final rule should ensure that lending standards do not deteriorate to dangerous levels, while at the same time ensuring that lending not be too restrictive.

4. Potential impacts of other provisions

Below, the Bureau discusses the impacts of several other provisions of the final rule and notes their interaction with other rulemakings. These include the points and fees provisions (which interact with the HOEPA rulemaking), the provisions of the statute regarding prepayment penalties, and the definition of rural or underserved areas (which interacts with the current rulemaking regarding escrow account requirements for certain higher-priced mortgage loans and with the 2013 HOEPA final rule). The interagency rule on appraisal requirements for high-risk mortgage loans also interacts with the QM definition.

a. Points and Fees Provisions

To be a “qualified mortgage,” the statute requires (among the other requirements already discussed) that the total points and fees payable in connection with the loan do not exceed 3 percent of the total loan amount and requires the Bureau to prescribe rules adjusting this limit to “permit lenders that extend smaller loans to meet the requirements of the presumption of compliance.” As noted earlier, such a restriction may have the effect of limiting cases where creditors, having received more funds up front, are less concerned about the long-term performance of the loan.

In the final rule, that limit is amended to a tiered approach with the following limits: for a loan amount greater than or equal to $100,000, three percent of the total loan amount; for a loan amount greater than or equal to $60,000 but less than $100,000, $3,000; for a loan amount greater than or equal to $20,000 but less than $60,000, five percent of the total loan amount; for a loan amount greater than or equal to $12,500 but less than $20,000, $1,000 of the total loan amount; and, for a loan amount of less than $12,500, eight percent of the total loan amount.

The higher limits for smaller dollar loans should allow more loans to be made as qualified mortgages. Data on the points and fees associated with a representative set of loans is not currently available. As a result, the Bureau cannot estimate precisely how many loans are impacted by this change. Under TILA as amended, a high-cost mortgage has points and fees equal to five percent of the total transaction amount if the transaction is $20,000 or more, and points and fees equal to the lesser of eight percent of the total transaction amount or $1,000, if the transaction is less than $20,000. Setting the maximum points and fees caps based on the HOEPA triggers will help ensure that a qualified mortgage is not a high-cost mortgage because of the points and fees.

The Dodd-Frank Act substantially expanded the scope of compensation included in points and fees for both the qualified mortgage and high-cost mortgage points and fees limits. In addition to compensation paid to mortgage brokerage firms and individual brokers, points and fees also includes compensation paid to other mortgage originators, including employees of a creditor (i.e., loan officers). Under the existing rule, only consumer payments to mortgage brokers are included in points and fees for the high-cost mortgage threshold. Also under the Act, any fees paid to and retained by affiliates of the creditor must be included in points and fees (except for any bona fide third-party charge not retained by the creditor, loan originator, or an affiliate of either, unless otherwise required under the rule). The final rule restates these provisions.

In a concurrent proposal published elsewhere in today's Federal Register, the Bureau proposed one alternative which would permit loan originator compensation to be netted against other upfront charges paid by the consumer and one that would not. Still, the inclusion of loan originator compensation in points and fees under the Final Rule (together with the statutory provisions implementing in the Final Rule regarding the treatment of charges due to third parties affiliated with the creditor) could have the effect of limiting the number of loans eligible to be qualified mortgages. For most prime loans, the Bureau believes that this change will not have a major impact: current industry pricing practices and the exemption for bona fide discount points suggest that few of these loans will be constrained by the points and fees limits.

For loans near the border of higher-priced loans (i.e. loans one percentage point above APOR), the exemption for bona-fide discount points is reduced and for loans priced at two percentage points or more above APOR the exemption is eliminated. For these loans, the inclusion of loan originator compensation and affiliate fees could limit qualified mortgage status for certain loans. Loans that will qualify for the safe harbor, but where the borrower pays for these charges through a higher interest rate, may lose the conclusive presumption of compliance and instead have only the rebuttable presumption. This impact is most likely greater for lenders with affiliated companies whose charges must be included in the points and fees calculations.

b. Prepayment Penalties

The Final Rule implements the provisions of Dodd-Frank with respect to prepayment penalties. Specifically, in accordance with the statute, the rule prohibits prepayment penalties for any mortgage other than a fixed-rate mortgage that is a qualified mortgage and not a higher-priced mortgage. Where the Final Rule permits prepayment penalties, it limits these penalties to 2 percent of the outstanding balance on the loan during the first year after consummation and 1 percent of the outstanding balance during the second year after consummation.

Available information from the sources described above suggests that loans originated today do not contain prepayment penalties, and this is likely to be true for the foreseeable future. Neither loans originated for sale to Fannie Mae and Freddie Mac, nor loans insured by FHA generally contain prepayment penalties. Moreover, the Bureau understands that prime loans, which make up the vast majority of originations today, have in recent years rarely had prepayment penalties. Some originators may make subprime loans they hold on portfolio for which they charge prepayment penalties, but data on terms of loans on portfolio are not available and at least in the current market, this is likely to be a very small number of loans. With the low interest rates that prevail today, lenders see little reason to limit prepayment risk by charging prepayment penalties.

Prepayment penalties by design impose costs on consumers to switch from their current loans to loans with lower interest rates. This cost can be particularly high for consumers with potentially increasing payments and who seek to refinance to avoid the increases. Moreover, these penalties are complex and often not transparent to consumers. Consumers may not focus on prepayment penalty terms because they are more focused on the terms they find more salient, such as interest rate and payment amount. Leading up to the mortgage crisis, some loan originators sometimes took advantage of consumers' lack of awareness or understanding of prepayment penalties. Originators could sell unsuspecting consumers loans with substantial expected payment increases as well as substantial prepayment penalties that would prevent the consumer from refinancing.

By limiting prepayment penalties to prime, fixed-rate qualified mortgages, the Final Rule benefits consumers by limiting these cases and lowering the cost of exiting a mortgage. Consumers will be able to refinance at lower cost, either when market rates drop or when the consumer's risk profile improves. In other cases, consumers who are sold mortgages with rates higher than their risk profile warrants will be able to refinance their mortgages to a market rate at lower cost. In still other cases, consumers will be able to sell their homes and move at lower cost. This cost reduction from restriction of prepayment penalties is particularly important to consumers who incur drops in income or increases in expenses that cause them to struggle to make their mortgage payments.

However, to the extent prepayment penalties compensate investors for legitimate prepayment risk, restricting penalties will reduce the value of certain mortgages and limit the returns to creditors and investors (which includes entities that are covered persons as well as entities that are not covered persons). In these cases, the cost of credit for some consumers will rise as creditors raise prices to compensate for increased prepayment risk. Currently, the number of loans that would have prepayment penalties but for the Final Rule appears to be very small, however, so costs to consumers and covered persons are expected to be de minimis.

c. Definition of Small Lenders, Rural and Underserved

The final rule allows certain small creditors operating predominantly in rural or underserved areas to originate balloon-payment qualified mortgages. Specifically, this option exists for lenders originating 500 or fewer covered transactions (including their affiliates), secured by a first lien, in the preceding calendar year, with assets under $2 billion (to be adjusted annually), and who made more than 50 percent of their total covered transactions secured by first liens on properties in counties that are “rural” or “underserved.” For the purposes of the final rule, and the 2013 Escrow rule published elsewhere in today's Federal Register, the Bureau has defined rural to include noncore counties and those micropolitan counties that are not adjacent to metropolitan statistical areas using the Department of Agriculture's urban influence codes. Relative to the proposed rule that only included a subset of rural counties, the final rule expands the exemption. The Bureau has not altered the definition of underserved from that contained in the proposed rule.

Although there is no comprehensive evidence with respect to the prevalence of balloon loans, the Bureau understands anecdotally from outreach that in these rural areas, creditors sometimes have difficulty selling certain loans on the secondary market either because of unique features of the rural property or of the rural borrower. In these instances, the creditors will make a portfolio loan. Because of their small size, some of these creditors eschew ARMs and manage interest rate risk by making balloon payment loans which the creditors then roll-over based on then-current interest rate when the balloon payment comes due.

Relative to a pre-statutory baseline, the rural balloon provisions of the rule have minimal effect. Relative to a post-statutory baseline in which the statute was implemented without the exception for rural lenders, the provisions of the rule have the following impacts on consumers and covered persons. Creditors covered by the rule's definition are permitted to make balloon loans which are qualified mortgages, potentially mitigating consumer access to credit issues that might arise if balloon payment mortgages were restricted. The rule creates certain minimum, consumer-protective requirements with respect to such balloon loans, such as a minimum term of five years and a requirement that the interest rate be fixed for that period of time. The rule also requires that creditors verify and consider income and debts before making such loans (albeit without a fixed debt-to-income requirement). However, to the extent these creditors rely on this permission to make balloon loans rather than other types of qualified mortgages, the rule also denies these consumers the consumer protections associated with not giving balloon loans qualified mortgage status.

According to the definition used in the final rule, approximately 10 percent of the U.S. population lives in areas that the Bureau defines as rural or underserved: the Bureau estimates that 2,707 small creditors, currently issuing first-lien mortgages and operating predominantly in rural or underserved areas, will be able to originate balloon qualified mortgages as a result of the provision. Given the low population density of the areas currently defined as rural, the corresponding limits on the number of creditors, and the challenges of making loans that could be sold in the secondary market, keeping this source of credit in the community with the safeguards added by the rule is likely more important to consumers than the consumer protections associated with not allowing balloon loans to be qualified mortgages. In somewhat less rural areas, for example the micropolitan counties not covered by the definition in the final rule, there are more creditors that can provide alternative forms of credit, such as ARM loans, and more creditors in general.

d. Qualified Mortgages and Appraisals

One impact of the current definition of qualified mortgage is related to higher-risk mortgages as defined in the Act. The Act contains special appraisal requirements with respect to higher-risk mortgages; those requirements are the subject of an interagency rulemaking process which resulted in a proposed rule in August which the agencies expect to finalize shortly. The Act generally defines a higher-risk mortgage as a closed-end consumer credit transaction secured by a principal dwelling with an APR exceeding rate thresholds substantially similar to rate triggers currently in Regulation Z for higher-priced mortgage loans, but excluding qualified mortgages. In general, as the number of loans defined as qualified mortgages increases, the number of loans that would be covered by the proposed appraisal requirements decreases. Based on the general definition of qualified mortgage in the final rule, those higher priced mortgage loans with a debt-to-income ratio of 43 or less would be exempt from the new requirements for interior appraisals. The temporary provision allowing additional loans (e.g. loans with a higher debt to income ratio and that are purchasable by the GSEs or insurable by FHA), to be qualified mortgages could further remove mortgages from that requirement. The impact of this reduction in the scope of appraisal requirements is relatively muted for first lien mortgages because of the small number of high-risk mortgages to begin with and the fact that most lenders already do a full interior appraisal and share the results with the consumer.

E. Potential Specific Impacts of the Final Rule

1. Depository Institutions and Credit Unions With $10 Billion or Less in Total Assets, as Described in Section 1026

Some depository institutions and credit unions with $10 billion or less in total assets as described in Section 1026 may see different impacts from the final rule than larger institutions. These differences are driven by the lending practices and portfolios at smaller depository institutions and credit unions, notably those below roughly $2 billion in assets, and by the nature of these institutions' relationship to the secondary market.

The Bureau understands that lending practices at many smaller institutions (according to comment letters and outreach) are based on a more personal relationship-based model, and less on automated systems, at least when the lender plans to keep the loan on portfolio rather than sell it. To the extent that the documentation and verification requirements in the final rule differ from current practice at these institutions, the final rule may impose some new compliance costs. However, unless these institutions keep all of the loans they originate on portfolio, which seems unlikely, they are already subject to documentation requirements from the secondary market so that any incremental costs are likely to be small. In addition, data from HMDA indicate that, on average, a larger proportion of loan originations at smaller institutions are higher-priced mortgage loans and will therefore have the rebuttable presumption of compliance rather than the safe harbor. These loans already are subject to an obligation to assess repayment ability and a rebuttable presumption under the Board's 2009 rule, so any new effects on these loans from the final rule, at least the loans these institutions keep on portfolio, are expected to be limited. Historically, delinquency rates on mortgages at smaller institutions are lower than the average in the industry and as such, the expected litigation costs for these loans are also probably quite low. Nevertheless, the proposal posted elsewhere in today's Federal Register asks for comment on whether the safe harbor should be extended to additional loans at particular smaller institutions.

The establishment of assignee liability for violation of the ability-to-repay provisions may also differentially impact smaller institutions by increasing counterparty risk for entities purchasing mortgages from these institutions. As described above, creditors and secondary market purchasers are expected to contract around the new ability-to-repay liability. For example, both Fannie Mae and Freddie Mac require lenders to represent and warrant that loans sold to the enterprises meet the current ability-to-repay requirements and to repurchase loans in cases where violations are found. Under such an arrangement, should a consumer bring a claim, the purchaser will look to the originator to repurchase the loan; if the originator is no longer in business or does not have the financial means to do so, the purchaser will have to bear the risk. This places greater incentive on purchasers to vet potential counterparties and may impact some smaller institutions' ability to sell loans. The impact is likely greatest for loans made under the general ability-to-repay standard rather than for qualified mortgages. In the near term, the temporary provisions expanding the number of qualified mortgages, will greatly mitigate costs for these institutions.

2. Impact of the Provisions on Consumers in Rural Areas

The final rule should have minimal differential impacts on consumers in rural areas. In these areas, a greater fraction of loans are made by smaller institutions and carried on portfolio. The availability or pricing for fixed rate or adjustable-rate loans that are qualified mortgages is likely to be unaffected. Notably, the liability for these loans is nearly unchanged; those below the threshold will be subject to the safe harbor while those above the threshold have a rebuttable presumption similar to the one in place under existing regulation. Only the very small number of loans made by these institutions and then sold may be impacted by the changes in counterparty risk. Consumers constrained to borrow from these lenders may see a small increase in the price of credit, either from the lenders now having to fund the loan on the balance sheet or facing reduced prices in the secondary market. The possible increases in compliance costs just described may also lead to very small increases in rates.

An important difference between the rural and the non-rural consumers is the availability of balloon loans following the rule. While the balloon loans in the non-rural areas that are not underserved cannot be qualified mortgages, small lenders operating predominantly in the rural or underserved areas can, under certain conditions, originate balloons loans that are qualified mortgages. Thus, rural consumers will preserve access to credit, while potentially experiencing the lack of protection associated with prohibiting balloon transactions from being qualified mortgages. Despite the fact that excluding a small creditor from the balloon loan market generally does not significantly disrupt the price-setting process, this might not be true for rural markets. In particular, there are 567 counties that have three creditors or fewer (that originate five or more covered transactions per year), according to HMDA 2011. Going from three creditors to two could significantly increase prices for consumers.

Data regarding the specific mortgages originated and held on bank and credit union portfolios is very limited; the exception is the data on the credit union call report showing the total number and amount of balloon loans together with hybrid adjustable-rate mortgages. According to these data, there appear to be few institutions, and therefore very few consumers affected in this way. In counties where the problem should be worst, namely micropolitan counties not covered by the rural or underserved definition, there are just under 50 credit unions that extend balloon loans and not ARMs; in total they originate 1,200 balloon loans. Consumers seeking credit at these institutions, or similarly situated banks or thrifts, may face some costs in taking a different product or in switching institutions depending on the product offerings and prices in the market. The Bureau believes any price increase is likely not significant as these areas are served by multiple lenders. On average, according to the 2011 HMDA data, 16 lenders on average made higher-priced mortgage loans in these counties, a proxy for what could be balloon loans.

F. Alternatives Considered

Two factors are most relevant when comparing the benefits, costs and impacts of the final rule to alternative regulatory implementations: the requirements for underwriting each loan and the eventual legal liability attached to that loan. The current rule differs from the Board's proposal along both dimensions, particularly in regard to qualified mortgages, as it uses a slightly different structure overall, such as incorporating a specific debt-to-income ratio requirement. It also varies in structure from some other proposals offered by commenters. However, even within the structure developed in the final rule, the parameters within the rule (e.g. the DTI ratio threshold) could have been different. In order to more fully illuminate the impacts of the final rule, this section first considers the final rule in comparison to the proposals and then to other reasonable alternatives.

In the 2011 ATR Proposal, the Board proposed two alternative definitions for a qualified mortgage. The Board's Alternative 1 proposed to define a qualified mortgage using only the statutory provisions (except for the discretionary requirement to consider the consumer's debt-to-income ratio or residual income). That is, the definition of a qualified mortgage would be based on product features, cost limitations (points and fees limit) and income verification but would not require the creditor to follow any other specific underwriting procedures. Alternative 1 would have operated as a legal safe harbor with the conclusive presumption of compliance.

The final rule maintains a minimum standard for documenting and verifying loans and varies the legal liability with the perceived consumer risk. Alternative 1, on the other hand, placed more emphasis on the restrictions on product features to protect consumers. Loans without interest-only, negative amortization or balloon features, or where total points and fees do not exceed three points were assumed safe and therefore had limited requirements for documenting income and debt (relative to other loans) and were afforded the conclusive presumption of compliance.

Compared to this alternative, the final rule with the temporary provisions likely offers qualified mortgage status to a similar number of loans: without the effects of the temporary provisions, fewer loans would qualify as qualified mortgages. The final rule also mandates stricter documentation and verification of qualified mortgages and limits the presumption of compliance in the case of higher-priced covered transactions. Compared to Alternative 1, only those loans that meet the product, features and point-and-fee limitations and that have a DTI ratio less than or equal to 43 percent are qualified mortgages. This approach limits the reliance on compensating factors when underwriting high DTI ratio loans and recognizes that while such loans may be in the creditor's interest, there is a greater possibility that the consumer may not have the ability to repay the loan. This change likely increases costs slightly in order to provide this consumer protection. Requiring the additional verification of debts for qualified mortgages also provides additional consumer protection. Since this is current practice in the market today, this likely adds very little cost for the time being; however, it does impose costs as credit expands to the point that the market would otherwise relax verification requirements—as well as benefits to consumers and society at large from preventing loans based on unverified (or no) data. Compared to Alternative 1, the only difference in the strength of the liability protection for qualified mortgages is for those loans above the higher-priced threshold. In the final rule, these loans have a rebuttable presumption of compliance rather than a conclusive presumption. However, given that the legal standard today is a rebuttable presumption, the final rule nearly maintains the status quo for borrowers with HPMLs; adopting Alternative 1 would have been a slight diminution of these borrower's legal rights.

The Board's Alternative 2 would have provided the lender with a rebuttable presumption of compliance and would have defined a “qualified mortgage” as including the statutory criteria as well the additional underwriting requirements from the general ability-to-repay standard. The Board proposed to permit, but not require, creditors to comply with the underwriting requirements by looking to “widely accepted governmental and non-governmental underwriting standards” (such as the FHA's standards). The important difference between this aspect of Alternative 2 and Alternative 1 is that, under Alternative 2, the relative weights for such tradeoffs had to be derived from widely accepted standards.

Compared to Alternative 2, the final rule with the temporary provisions likely offers qualified mortgage status to a similar number of loans; without the effects of the temporary provisions, fewer loans would be eligible to be qualified mortgages. Under the final rule, there is little difference in the documentation and verification requirements; however, the presumption of compliance is strengthened for the majority of qualified mortgages. Compared to Alternative 2 (and to Alternative 1), only those loans that meet the product, features and cost limitations and that have a DTI ratio less than or equal to 43 percent are qualified mortgages. This limits the use of compensating factors for high DTI loans and recognizes that while such loans may be in the creditor's interest, there is a greater possibility that the consumer may not have the ability to repay the loan. This change likely increases costs slightly in order to provide this consumer protection. Both Alternative 2 and the final rule have very similar documentation and verification standards so there is little difference in the benefits and costs along that dimension. Relative to Alternative 2, the difference in the liability standard is for those qualified mortgages below the higher-priced threshold. In the final rule, these loans have a conclusive presumption of compliance rather than just a rebuttable presumption.

As noted in the preamble, a coalition of industry and consumer advocates presented another alternative proposal to the Bureau that would have provided a tiered approach to defining a qualified mortgage. Under the first tier, if the consumer's total debt-to-income ratio is 43 percent or less, the loan would be a qualified mortgage, and no other tests would be required. Under the second tier, if the consumer's total debt-to-income ratio is more than 43 percent, the creditor would apply a series of tests related to the consumer's front-end debt-to-income ratio (housing debt to income), stability of income and past payment history, availability of reserves, and residual income to determine if a loan is a qualified mortgage. This would have allowed some loans with up to 50 percent DTI ratios to meet the qualified mortgage definition. To the extent that it relies on additional factors beyond the DTI ratio, this alternative is similar to the Board's approach. However, the coalition's proposal generally restricted the factors considered to be factors related to ability to repay, rather than other factors related to credit or collateral in its determination. These commenters also supported a rebuttable presumption standard for qualified mortgages.

Relative to this alternative, the final rule will likely include fewer loans as qualified mortgages. The loans that will not be qualified mortgages are those that would qualify only under one or more of the additional factors besides DTI ratio that the alternative included: housing expenses, stability of income, reserves etc. As a result, these loans will have to meet the ability-to-repay standard of the final rule, providing additional consumer protections with the minor added costs described above. Relative to a rule including these factors, the final rule is simpler and easier to implement for industry, lowering costs overall. In addition, creditors are free to include such factors in their own credit decisions and to develop the best models for their inclusion. The Bureau views this more dynamic outcome as a benefit relative to a more prescriptive rule detailing how such factors should be traded off against each other. This alternative did include a rebuttable presumption of compliance for all qualified mortgages; as such, the final rule's safe harbor limits liability costs and consumer benefits, as already discussed, for those qualified mortgages that are not higher priced covered transactions.

As noted, the Bureau also considered certain alternatives to its own version of the final rule. One such alternative would have used a threshold of a 36 percent DTI ratio to define qualified mortgages. This would have left roughly an additional 15 percent of loans, both during the 1997-2003 period and during 2011, without a presumption of compliance. As noted however, the Bureau believes that 43 percent is a more efficient threshold: it is an accepted market standard, rates of delinquency and default for borrowers between 36 and 43 percent are still modest, and many borrowers—particularly in higher cost housing markets—borrow at these levels.

The Bureau also considered whether all qualified mortgages should have the same degree of presumption with the qualified mortgage standard—either all being afforded a conclusive presumption of compliance or all being afforded a rebuttable presumption. As discussed in the section-by-section analysis, the Bureau determined that the bifurcated approach in which only higher-priced covered transactions provide the consumer with the opportunity to rebut the presumption of compliance best balances the concerns of costs, certainty, and consumer protection.

VIII. Final Regulatory Flexibility Act

The Regulatory Flexibility Act (RFA) generally requires an agency to conduct an initial regulatory flexibility analysis (IRFA) and a final regulatory flexibility analysis (FRFA) of any rule subject to notice-and-comment rulemaking requirements, unless the agency certifies that the rule will not have a significant economic impact on a substantial number of small entities. The Bureau also is subject to certain additional procedures under the RFA involving the convening of a panel to consult with small business representatives prior to proposing a rule for which an IRFA is required.

In the 2011 ATR Proposal, the Board did not certify that the rule would not have a significant economic impact on a substantial number of small entities and therefore prepared an IRFA. In this IRFA the Board solicited comment on any costs, compliance requirements, or changes in operating procedures arising from the application of the proposed rule to small businesses, comment regarding any state or local statutes or regulations that would duplicate, overlap, or conflict with the proposed rule, and comment on alternative means of compliance for small entities with the ability-to-repay requirements and restrictions on prepayment penalties. Comments addressing the ability-to-repay requirements and restrictions on prepayment penalties are addressed in the section-by-section analysis above. Comments addressing the impact on the cost of credit are discussed below.

1. A Statement of the Need for, and Objectives of, the Rule

The Bureau is publishing a final rule to establish new ability-to-repay requirements related to mortgage origination. As discussed in the preamble, the final rule's amendments to Regulation Z implement certain amendments to TILA that were added by sections 1411, 1412, 1413, and 1414 of the Dodd-Frank Act in response to the recent foreclosure crisis to address certain lending practices (such as low- or no-documentation loans or underwriting mortgages without including any principal repayments in the underwriting determination) that led to consumers having mortgages they could not afford, thereby contributing to high default and foreclosure rates.

A full discussion of the market failures motivating these provisions of the Dodd-Frank Act and the final rule is included in the preamble and in the Bureau's section 1022 analysis above. Those discussions also describe the specific ways the final rule addresses these issues. However, in general, the purpose of the Dodd-Frank Act ability-to-repay requirements is to assure that consumers are offered and receive residential mortgage loans on terms that reasonably reflect their ability to repay the loans and that are understandable and not unfair, deceptive or abusive. Prior to the Dodd-Frank Act, existing Regulation Z provided ability-to-repay requirements for high-cost and higher-priced mortgages. Accordingly, new TILA section 129C generally prohibits a creditor from making a residential mortgage loan unless the creditor makes a reasonable and good faith determination, based on verified and documented information, that the consumer has a reasonable ability to repay the loan according to its terms, including any mortgage-related obligations (such as property taxes and mortgage insurance). Consistent with the statute, the final rule applies the ability-to-repay requirements of TILA section 129C to any consumer credit transaction secured by a dwelling, except an open-end credit plan, timeshare plan, reverse mortgage, or temporary loan.

Congress also recognized the importance of maintaining access to responsible, affordable mortgage credit. To provide creditors more certainty about their potential liability under the ability-to-repay standards while protecting consumers from unaffordable loans, the Dodd-Frank Act creates a presumption of compliance with the ability-to-repay requirement when creditors make “qualified mortgages.” Qualified mortgages do not contain certain features that Congress deemed to create a risk to consumers' ability to repay, and must be underwritten using standards set forth in the statute that are designed to assure that consumers will have the ability to repay these loans. The final rule establishes standards for complying with the ability-to-repay requirements, including defining “qualified mortgage.” The final rule provides three options for originating a qualified mortgage: under the general definition in § 1026.43(e)(2), for loans where the consumer's monthly debt-to-income ratio would not exceed 43 percent; under the definition § 1026.43(e)(4), for a maximum of seven years, for loans that are eligible for purchase by the GSEs while in conservatorship or certain other Federal agencies, and under § 1026.43(f), for loans that have balloon-payment features if the creditor operates predominantly in rural or underserved areas and meets certain asset-size and transaction volume limits.

Congress did not explicitly define the nature of the presumption of compliance that attaches to a qualified mortgage. Congress also left some contours of a qualified mortgage undefined, such as whether there should be a minimum debt-to-income ratio. Congress left these decisions to the Bureau and granted broad authority to revise, add to, or subtract from the qualified mortgage criteria upon a finding that doing so is “necessary or proper” or “necessary and appropriate” to achieve certain specified standards, such as ensuring that responsible, affordable mortgage credit remains available to consumers.

As discussed above, the final rule recognizes both the need to assure that consumers are offered and receive loans based on a reasonable and good faith determination of their repayment ability and the need to ensure that responsible, affordable mortgage credit remains available to consumers. The Bureau believes, based upon its analysis of the data available to it, that, under the final rule, the vast majority of loans originated today can meet the standards for a qualified mortgage so long as creditors follow the required procedures, such as verifying income or assets, and current debt obligations, alimony and child support. The Bureau also believes, based upon its analysis of the historical performance of loans meeting the rule's definition of “qualified mortgages,” that consumers will be able to repay these loans. The Bureau believes that the final rule will not restrict creditors' ability to make responsible loans, both within and outside the qualified mortgage space.

The final rule provides special rules for complying with the ability-to-repay requirements for a creditor refinancing a “non-standard mortgage” into a “standard mortgage.” The purpose of this provision is to provide flexibility for creditors to refinance a consumer out of a risky mortgage into a more stable one without undertaking a full underwriting process.

In addition to the ability-to-repay and qualified mortgage provisions, the final rule implements the Dodd-Frank Act limits on prepayment penalties and lengthens the time creditors must retain records that evidence compliance with the ability-to-repay and prepayment penalty provisions.

2. Summary of Significant Issues Raised by Comments in Response to the Initial Regulatory Flexibility Analysis, Statement of the Assessment of the Bureau of Such Issues, and a Statement of Any Changes Made as a Result of Such Comments

The Board's IRFA estimated the possible compliance costs for small entities from each major component of the rule against a pre-statute baseline. The Board requested comments on the IRFA.

The Board did not receive any comments in its IRFA. Industry commenters generally expressed concern with respect to the costs they anticipated from the 2011 ATR Proposal. The Bureau received numerous comments describing in general terms the impact of the proposed rule on small creditors and the need for the qualified mortgage definition to be structured as a safe harbor with clear, well-defined standards to ensure that the largest number of consumers possible can access credit. Small creditors are particularly concerned about the litigation risk associated with the requirement to make a reasonable and good faith determination of consumers' ability to repay based on verified and documented information. Because of their size, small creditors note that they are particularly unsuited to bear the burden and cost of litigation and would find it particularly difficult to absorb the cost of an adverse judgment. Indeed, small creditors insist that they will not continue to make mortgage loans unless they are protected from liability for violations of the ability-to-repay rules by a conclusive presumption of compliance or “safe harbor.” These small creditors' concerns about compliance with the ability-to-repay rule and associated litigation risk have been repeatedly expressed to the Bureau by their trade associations and prudential regulators.

Several commenters on the proposal urged the Bureau to adopt less stringent regulatory requirements for small creditors or for loans held in portfolio by small creditors. For example, at least two commenters on the proposal, a credit union and a state trade group for small banks, urged the Bureau to exempt small portfolio creditors from the ability-to-repay and qualified mortgage rule. Two other trade group commenters urged the Bureau to adopt less stringent regulatory requirements for small creditors than for larger creditors at least in part because mortgage loans made by small creditors often are held in portfolio and therefore historically have been conservatively underwritten.

Some industry commenters supported not including quantitative standards for such variables as debt-to-income ratios and residual income because they argued that underwriting a loan involves weighing a variety of factors, and creditors and investors should be allowed to exercise discretion and weigh risks for each individual loan. To that point, one industry trade group commenter argued that community banks, for example, generally have conservative requirements for a consumer's debt-to-income ratio, especially for loans that are held in portfolio by the bank, and consider many factors when underwriting for mortgage loans, such as payment history, liquid reserves, and other assets. Because several factors are considered and evaluated in the underwriting process, this commenter asserted that community banks can be flexible when underwriting for mortgage loans and provide arrangements for certain consumers that fall outside of the normal debt-to-income ratio for a certain loan. This commenter contended that strict quantitative standards would inhibit community banks' relationship lending and ability to use their sound judgment in the lending process. Some commenters contended that requiring specific quantitative standards could restrict credit access and availability for consumers.

A number of other commenters expressed concerns that the availability of portfolio mortgage loans from small creditors would be severely limited because the proposed exception for rural balloon loans was too restrictive. Some industry commenters urged the Bureau to allow balloon mortgage loans held in portfolio by the originating banks for the life of the loan to be included under this safe harbor so that small creditors could continue to meet the specific needs of their customers.

These comments, and the responses, are discussed in the section-by-section analysis and element 6-1 of this FRFA.

3. Response to the Small Business Administration Chief Counsel for Advocacy

The SBA Office of Advocacy (Advocacy) provided a formal comment letter to the Bureau in response to the Bureau's reopening of the comment period for certain issues relating to the ability-to-repay/qualified mortgage rulemaking. Among other things, this letter expressed concern about the following issues: the qualified mortgage definition and the use of data as a means for measuring a consumer's ability to repay.

First, Advocacy expressed concern that the qualified mortgage definition will have major implications on the viability of community banks. Advocacy pointed to the assertion made by small banks that they will no longer originate mortgage loans if they are only provided with a rebuttable presumption of compliance. In addition, according to Advocacy, small banks contend that establishing the qualified mortgage as a rebuttable presumption of compliance will reduce the availability and affordability of mortgages to consumers due to increased litigation and compliance costs, and the exit by certain small lenders unable to manage the risk. According to Advocacy, small banks assert that one way to enable them to compete effectively (and to ensure consumers can obtain affordable loans) is to establish the qualified mortgage as a safe harbor and allow for non-traditional loans such as mortgages with balloon payments to continue to be made.

The Bureau carefully considered the arguments for establishing the qualified mortgage as a safe harbor or rebuttable presumption of compliance in light of the proposed rule, and a complete discussion of the consideration of the Bureau's final rule can be found in the respective section of the section-by-section analysis, the Bureau's section 1022(b)(2) discussion, and in element 6-1 of this FRFA.

As discussed in more detail elsewhere, the final rule provides a safe harbor under the ability-to-repay requirements for mortgage loans that satisfy the definition of a qualified mortgage and are not higher-priced covered transactions (i.e., APR does not exceed Average Prime Offer Rate (APOR) + 1.5 percentage points for first liens or 3.5 percentage points for subordinate liens). The final rule provides a rebuttable presumption for all other qualified mortgage loans, meaning qualified mortgage loans that are higher-priced covered transactions (i.e., APR exceeds APOR + 1.5 percentage points for first lien or 3.5 percentage points for subordinate lien). The Bureau believes that a bifurcated approach to the presumption of compliance provides the best way of balancing consumer protection and access to credit considerations and is consistent with the purposes of the statute, while calibrating consumer protections and risk levels to match the historical record of loan performance. To reduce uncertainty in potential litigation, the final rule defines the standard by which a consumer may rebut the presumption of compliance afforded to higher-priced qualified mortgages.

The Bureau notes that the Board's proposed § 1026.43 did not include special provisions for portfolio loans made by small creditors and the Board's proposal did not address such an accommodation. However, this final rule is related to a proposed rule published elsewhere in today's Federal Register. As discussed in more detail below, in that proposal, the Bureau is proposing certain amendments to this final rule, including a proposal to define as a qualified mortgage a larger category of loans made and held in portfolio by small creditors than this final rule defines as a qualified mortgage.

Second, Advocacy expressed concern about using loan performance, as measured by the delinquency rate, as an appropriate metric to evaluate whether consumers had the ability to repay at the time their loans were consummated. Advocacy noted that a consumer's circumstances might change after the loan was made due to unemployment or illness. The Bureau agrees that consumers' circumstances can change and lead to delinquency or default. However, the Bureau also believes that DTI is an indicator of the consumer's ability to repay. All things being equal, consumers carrying loans with higher DTI ratios will be less able to absorb any such shocks and are more likely to default.

4. A Description of and an Estimate of the Number of Small Entities to Which the Rule Will Apply

The final rule will apply to creditors that engage in originating or extending certain dwelling-secured credit. The credit provisions of TILA and Regulation Z have broad applicability to individuals and businesses that originate and extend even small numbers of home-secured credit. See 1026.1(c)(1). Small entities that originate or extend closed-end loans secured by a dwelling are potentially subject to at least some aspects of the final rule.

For purposes of assessing the impacts of the final rule on small entities, “small entities” is defined in the RFA to include small businesses, small nonprofit organizations, and small government jurisdictions. 5 U.S.C. 601(6). A “small business” is determined by application of SBA regulations and reference to the North American Industry Classification System (NAICS) classifications and size standards. 5 U.S.C. 601(3). Under such standards, banks and other depository institutions are considered “small” if they have $175 million or less in assets, and for other financial businesses, the threshold is average annual receipts (i.e., annual revenues) that do not exceed $7 million.

The Bureau can identify through data under the Home Mortgage Disclosure Act, Reports of Condition and Income (Call Reports), and data from the National Mortgage Licensing System (NMLS) the approximate numbers of small depository institutions that will be subject to the final rule. Origination data is available for entities that report in HMDA, NMLS or the credit union call reports; for other entities, the Bureau has estimated their origination activities using statistical projection methods.

The following table provides the Bureau's estimate of the number and types of entities to which the rule will apply:

5. Projected Reporting, Recordkeeping, and Other Compliance Requirements

The final rule does not impose new reporting requirements. The final rule does, however, impose new recordkeeping and other compliance requirements on certain small entities. The requirements on small entities from each major component of the rule are presented below.

The Bureau discusses impacts against a pre-statute baseline. This baseline assumes compliance with the Federal rules that overlap with the final rule. The impact of the rule relative to the pre-statute baseline will be smaller than the impact would be if not for compliance with the existing Federal rules. In particular, creditors have already incurred some of the one-time costs necessary to comply with the final rule when they came into compliance with the 2008 HOEPA Final Rule on higher-priced mortgage loans. And creditors already have budgeted for some of the ongoing costs of the final rule to the extent those are costs necessary to remaining in compliance with the 2008 HOEPA Final Rule. These expenses attributable to the 2008 HOEPA Final Rule will facilitate and thereby reduce the cost of compliance with this final rule.

Recordkeeping Requirements

The final rule imposes new record retention requirements on covered persons. As discussed above, the final rule requires creditors to retain evidence of compliance with § 1026.43 (containing the ability-to-repay/qualified mortgage provisions and prepayment penalty restrictions) for three years after consummation. The final rule clarifies that creditors need not maintain actual paper copies of the documentation used to underwrite a transaction. For most covered persons, the required records will be kept in electronic form and creditors need retain only enough information to reconstruct the required records. This should limit any burden associated with the record retention requirement for creditors.

Other Compliance Requirements

As discussed in detail in the section-by-section analysis and the Bureau's section 1022(b)(2) discussion above, the final rule imposes new compliance requirements on creditors. In general, creditors will have to update their policies and procedures; additionally, creditors may have to update their systems, for example, to store flags identifying qualified mortgages, and to ensure compliance. The Bureau believes that small creditors' major one-time costs will be to learn about the final rule, consider whether they need to modify their underwriting practices and procedures to comply with the rule and, if necessary, modify their practices and procedures. The precise costs to small entities of modifying their underwriting practices, should they need to do so, are difficult to predict. These costs will depend on a number of factors, including, among other things, the current practices and systems used by such entities to collect and analyze consumer income, asset, and liability information, the complexity of the terms of credit products that they offer, and the range of such product offerings. To the extent that most small creditors' processes already align with the rule, any additional compliance costs should be minimal.

When originating mortgages, the creditor must calculate the monthly mortgage payment based on the greater of the fully indexed rate or any introductory rate, assuming monthly, fully amortizing payments that are substantially equal. The final rule provides special payment calculation rules for loans with balloon payments, interest-only loans, and negative amortization loans. The final rule may therefore increase compliance costs for small entities, particularly for creditors that offer products that contain balloon payments, interest-only loans, and negative amortization loans. The precise costs to small entities of updating their processes and systems to account for these additional calculations are difficult to predict, but these costs are mitigated, in some circumstances, by the presumption of compliance or safe harbor for qualified mortgages.

The Final Rule also includes requirements for documentation and verification of certain information that the creditor must consider in assessing a consumer's repayment ability. The final rule provides special rules for verification of a consumer's income or assets, and provides examples of records that can be used. Different verification requirements apply to qualified mortgages. Creditors that originate qualified mortgages under the general definition must verify a consumer's income or assets, current debt obligations, alimony, and child support, and must also verify a consumer's monthly debt-to-income ratio. The final rule does not contain specific verification requirements for creditors originating qualified mortgages under the temporary provisions; however, such loans must comply with eligibility requirements (including underwriting requirements) of the GSEs or the Federal agency program applicable to the loan.

The final rule also provides special rules for complying with the ability-to-repay requirements for a creditor refinancing a “non-standard mortgage” into a “standard mortgage.” This provision is based on TILA section 129C(a)(6)(E), which contains special rules for the refinance of a “hybrid loan” into a “standard loan.” The purpose of this provision is to provide flexibility for creditors to refinance a consumer out of a risky mortgage into a more stable one without undertaking a full underwriting process. Under the final rule, a non-standard mortgage is defined as an adjustable-rate mortgage with an introductory fixed interest rate for a period of one year or longer, an interest-only loan, or a negative amortization loan. Under this option, a creditor refinancing a non-standard mortgage into a standard mortgage does not have to consider the eight specific underwriting criteria under the general ability-to-repay option, if certain conditions are met, thus reducing compliance costs for small entities.

Prepayment limitations, as discussed in detail in the section-by-section analysis and the Bureau's section 1022 analysis, are also included in the final rule.

Estimate of the Classes of Small Entities Which Will Be Subject to the Requirement

Section 603(b)(4) of the RFA requires an estimate of the classes of small entities which will be subject to the requirement. The classes of small entities which will be subject to the reporting, recordkeeping, and compliance requirements of the final rule are the same classes of small entities that are identified above in part VIII.B.4.

Section 604(a)(5) of the RFA also requires an estimate of the type of professional skills necessary for the preparation of the reports or records. The Bureau anticipates that the professional skills required for compliance with the final rule are the same or similar to those required in the ordinary course of business of the small entities affected by the final rule. Compliance by the small entities that will be affected by the final rule will require continued performance of the basic functions that they perform today: Managing information about consumers and conducting sound underwriting practices for mortgage originations.

6-1. Description of the Steps the Agency has Taken To Minimize the Significant Economic Impact on Small Entities

The Bureau understands the new provisions will impose a cost on small entities, and has attempted to mitigate the burden consistent with statutory objectives. The Bureau has also taken numerous additional steps that are likely to reduce the overall cost of the rule. Nevertheless, the rule will certainly create new one-time and ongoing costs for creditors. The section-by-section analysis of each provision and the Bureau's section 1022 analysis contain a complete discussion of the following steps taken to mitigate the burden.

The final rule provides small creditors with the option of offering only qualified mortgages, which will enjoy either a presumption of compliance with respect to the repayment ability requirement (for higher-priced covered transactions) or a safe harbor from the repayment ability requirement, thus reducing litigation risks and costs for small creditors.

The Bureau believes that a variety of underwriting standards can yield reasonable, good faith ability-to-repay determinations. The Bureau is permitting creditors to develop and apply their own underwriting standards (and to make changes to those standards over time in response to empirical information and changing economic and other conditions) as long as those standards lead to ability-to-repay determinations that are reasonable and in good faith. In addition, the Bureau will permit creditors to use their own definitions and other technical underwriting criteria and notes that underwriting guidelines issued by governmental entities such as the FHA are a source to which creditors may refer for guidance on definitions and technical underwriting criteria. The Bureau believes this flexibility is necessary given the wide range of creditors, consumers, and mortgage products to which this rule applies. The Bureau believes this increased flexibility will reduce the burden on small creditors by allowing them to determine the practices that fit best with their business model.

Qualified Mortgage Provisions

The general definition of the qualified mortgage includes a very clear standard of 43 percent for the debt-to-income threshold and clear methods to compute that figure. The clarity of this provision, and others, should make implementation of and compliance with these provisions of the rule. The Bureau carefully considered the arguments for establishing the qualified mortgage as a safe harbor or rebuttable presumption of compliance in light of the proposed rule, and a complete discussion of the consideration of the Bureau's final rule can be found in the respective section of the section-by-section analysis. The final rule establishes standards for complying with the ability-to-repay requirements, including defining “qualified mortgage.” The final rule provides three options for originating a qualified mortgage: under the general definition in § 1026.43(e)(2), for loans where the consumer's monthly debt-to-income ratio would not exceed 43 percent; under the definition § 1026.43(e)(4), for a maximum of seven years, for loans that are eligible for purchase by the GSEs while in conservatorship or certain other Federal agencies, and under § 1026.43(f), for loans that have balloon-payment features if the creditor operates predominantly in rural or underserved areas and meets certain asset-size and transaction volume limits. The final rule provides a safe harbor under the ability-to-repay requirements for mortgage loans that satisfy the definition of a qualified mortgage and are not higher-priced covered transactions (i.e., APR does not exceed Average Prime Offer Rate (APOR) + 1.5 percentage points for first liens or 3.5 percentage points for subordinate liens). The final rule provides a rebuttable presumption for all other qualified mortgage loans, meaning qualified mortgage loans that are higher-priced covered transactions (i.e., APR exceeds APOR + 1.5 percentage points for first lien or 3.5 percentage points for subordinate lien).

The Bureau believes that a bifurcated approach to the presumption of compliance provides the best way of balancing consumer protection and access to credit considerations and is consistent with the purposes of the statute, while calibrating consumer protections and risk levels to match the historical record of loan performance. To reduce uncertainty in potential litigation, the final rule defines the standard by which a consumer may rebut the presumption of compliance afforded to higher-priced qualified mortgages. The Bureau's approach to the standards with which a consumer can rebut the presumption that applies to higher-priced transactions is further designed to ensure careful calibration.

The Bureau considered several alternatives, including only the safe harbor standard and only the rebuttable presumption standard. In its rulemaking, the Bureau tried to balance consumers' access to credit concerns with the consumer protection associated with reducing consumers' cost of litigation. Compared to the final rule, only the safe harbor standard marginally increased consumers' access to credit, but significantly reduced consumer protection. Conversely, only the rebuttable presumption standard marginally increased consumer protection, but significantly decreased consumers' access to credit.

Balloon-Payment Qualified Mortgage Provisions

The Bureau has also provided an exception to the general provision that a qualified mortgage may not provide for a balloon payment for loans that are originated by certain small creditors and that meet specified criteria. The Bureau understands that community banks originate balloon-payment loans to hedge against interest rate risk, rather than making adjustable-rate mortgages, and that community banks hold these balloon-payment loans in portfolio virtually without exception because they are not eligible for sale in the secondary market. Under the final rule, the Bureau is permitting small creditors operating predominantly in rural or underserved areas to originate a balloon-payment qualified mortgage.

Unlike loans that are qualified mortgages under the general definition, there is no specific debt-to-income ratio requirement for balloon-payment qualified mortgages. However, creditors must consider and verify a consumer's monthly debt-to-income ratio. Like the other qualified mortgage definitions, a loan that satisfies the criteria for a balloon-payment qualified mortgage and is not a higher-priced covered transaction receives a legal safe harbor under the ability-to-repay requirements. A loan that satisfies those criteria and is a higher-priced covered transaction receives a rebuttable presumption of compliance with the ability-to-repay requirements. The Bureau believes that this exception will decrease the economic impact of the final rule on small entities. In response to concerns regarding the proposed provisions for holding balloon-payment loans in portfolio, the final rule provides more flexible portfolio requirements which permit certain transfers.

Concurrent Proposal for Portfolio Loans Made by Small Creditors

The Bureau notes that the Board's proposal did not include special provisions for portfolio loans made by small creditors and the Board's proposal did not address such an accommodation.

The Bureau understands that creditors generally have in place underwriting policies, procedures, and internal controls that require verification of the consumer's reasonably expected income or assets, employment status, debt obligations and simultaneous loans, and debt-to-income or residual income. Notably, in response to the proposal, commenters stated that most creditors today are already complying with the full ability-to-repay underwriting standards. For these institutions, there would be no additional burden as a result of the verification requirements in the final rule, since those institutions collect the required information in the normal course of business. To the extent small creditors do not verify and document some or all of the information required by the proposed rule in the normal course of business, they will need to engage in certain one-time implementation efforts and system adjustments. These one-time costs might include expenses related to creditors needing to reanalyze their product lines, retrain staff, and reorganize the processing and administrative elements of their mortgage operations.

In a related proposed rule published elsewhere in today's Federal Register, the Bureau is proposing certain amendments to this final rule, including an additional definition of a qualified mortgage for certain loans made and held in portfolio by small creditors. The proposed new category would include certain loans originated by small creditors that: (1) Have total assets less than $2 billion at the end of the previous calendar year; and (2) together with all affiliates, originated 500 or fewer covered transactions, secured by first-liens during the previous calendar year. These loans generally conform the requirements under the general definition of a qualified mortgage except the 43 percent limit on monthly debt-to-income ratio. Under the proposed additional definition, a creditor would not have to use the instructions in the appendix to the final rule to calculate debt-to-income ratio, and a loan with a consumer debt-to-income ratio higher than 43 percent could be a qualified mortgage if all other criteria are met.

The Bureau also is proposing to allow small creditors to charge a higher annual percentage rate for first-lien qualified mortgages in the proposed new category and still benefit from a conclusive presumption of compliance or “safe harbor.” In addition, the Bureau also is proposing to allow small creditors operating predominantly in rural or underserved areas to offer first-lien balloon loans with a higher annual percentage rate and still benefit from a conclusive presumption of compliance with the ability to repay rules or “safe harbor.” The Bureau is proposing these changes because it believes they may be necessary to preserve access to credit for some consumers. The regulatory requirement to make a reasonable and good faith determination based on verified and documented evidence that a consumer has a reasonable ability to repay may entail significant litigation risk for small creditors. The Bureau believes that small creditors have historically engaged in responsible mortgage underwriting that includes thorough and thoughtful determinations of consumers' ability to repay, at least in part because they bear the risk of default associated with loans held in their portfolios. The Bureau also believes that because small creditors' lending model is based on maintaining ongoing, mutually beneficial relationships with their customers, they therefore have a more comprehensive understanding of their customers' financial circumstances and are better able to assess ability to repay than larger creditors.

Further, the Bureau understands that the only sources of mortgage credit available to consumers in rural and underserved areas may be small creditors because larger creditors may be unable or unwilling to lend in these areas. For these reasons, the Bureau is proposing a new category of qualified mortgages that would include small creditor portfolio loans and is also proposing to raise the annual percentage rate threshold for the safe harbor to accommodate small creditors' higher costs. The Bureau believes these steps may be necessary to preserve some rural and underserved consumers' access to non-conforming credit.

6-2. Description of the Steps the Agency Has Taken To Minimize Any Additional Cost of Credit for Small Entities

Section 603(d) of the RFA requires the Bureau to consult with small entities regarding the potential impact of the proposed rule on the cost of credit for small entities and related matters. 5 U.S.C. 603(d). The Bureau notes that the Board was not subject to this requirement when it issued its IRFA.

The Bureau does not believe that the final rule will result in an increase in the cost of business credit for small entities. Instead, the final rule will apply only to mortgage loans obtained by consumers primarily for personal, family, or household purposes and the final rule will not apply to loans obtained primarily for business purposes. Given that the final rule does not increase the cost of credit for small entities, the Bureau has not taken additional steps to minimize the cost of credit for small entities.

IX. Paperwork Reduction Act Analysis

Certain provisions of this final rule contain “collection of information” requirements within the meaning of the Paperwork Reduction Act of 1995 (44 U.S.C. 3501 et seq.) (Paperwork Reduction Act or PRA).

This final rule amends 12 CFR part 1026 (Regulation Z). Regulation Z currently contains collections of information approved by the Office of Management and Budget (OMB). The Bureau's OMB control number for Regulation Z is 3170-0015. The PRA (44 U.S.C 3507(a), (a)(2) and (a)(3)) requires that a Federal agency may not conduct or sponsor a collection of information unless OMB approved the collection under the PRA and the OMB control number obtained is displayed. Further, notwithstanding any other provision of law, no person is required to comply with, or is subject to any penalty for failure to comply with, a collection of information that does not display a currently valid OMB control number (44 U.S.C. 3512).

This final rule contains information collection requirements that have not been approved by the OMB and, therefore, are not effective until OMB approval is obtained. The unapproved information collection requirements are contained in sections 1026.25(c)(3) and 1026.43(c)-(f) of these regulations. The Bureau will publish a separate notice in the Federal Register announcing the submission of these information collection requirements to OMB as well as OMB's action on these submissions; including, the OMB control number and expiration date.

On May 11, 2011, the Board of Governors of the Federal Reserve System (Board) published notice of the proposed rule in the Federal Register (76 FR 27390). The information collection requirements in §§ 1026.25(c)(3) and 1026.43(c)-(f) were contained in the Board's proposal; however, these requirements were not separately discussed in the proposal's PRA section. For full public transparency, the Bureau now claims these requirements as information collections. The Bureau received no PRA-related comments to the Board's proposal on the information collections in §§ 1026.25(c)(3) and 1026.43(c).

A. Overview

As described below, the final rule amends the collections of information currently in Regulation Z to implement amendments to TILA made by the Dodd-Frank Act. The Dodd-Frank Act prohibits a creditor from making a mortgage loan unless the creditor makes a reasonable and good faith determination, based on verified and documented information, that the consumer will have a reasonable ability to repay the loan, including any mortgage-related obligations (such as property taxes). TILA section 129C(a); 15 U.S.C. 1639c(a). The Dodd-Frank Act provides special protection from liability for creditors who make “qualified mortgages.” TILA section 129C(b); 15 U.S.C. 1639c(b). The purpose of the Dodd-Frank Act ability-to-repay requirement is to assure that consumers are offered and receive residential mortgage loans on terms that reasonably reflect their ability to repay the loans and that are understandable and not unfair, deceptive or abusive. TILA section 129B(a)(2); 15 U.S.C. 1639b(a)(2). Prior to the Dodd-Frank Act, existing Regulation Z provided ability-to-repay requirements for high-cost and higher-priced mortgage loans. The Dodd-Frank Act expanded the scope of the ability-to-repay requirement to cover all residential mortgage loans.

The final rule establishes standards for complying with the ability-to-repay requirement, including defining “qualified mortgage.” The final rule provides three options for originating a qualified mortgage: under the general definition in § 1026.43(e)(2), for loans where the consumer's monthly debt-to-income ratio do not exceed 43 percent; under the definition § 1026.43(e)(4), for a maximum of seven years, for loans that are eligible for purchase by the GSEs while in conservatorship or certain other Federal agencies, and under § 1026.43(f), for loans that have a balloon-payment if the creditor operates predominantly in rural or underserved areas and meets certain underwriting requirements, and asset-size and transaction volume limits.

In addition to the ability-to-repay and qualified mortgage provisions, the final rule implements the Dodd-Frank Act limits on prepayment penalties and lengthens the time creditors must retain records that evidence compliance with the ability-to-repay and prepayment penalty provisions. Currently, Regulation Z requires creditors to retain evidence of compliance for two years after disclosures must be made or action must be taken. The final rule amends Regulation Z to require creditors to retain evidence of compliance with the ability-to-repay/qualified mortgage provisions and prepayment penalty restrictions in § 1026.43 for three years after consummation for consistency with statute of limitations on claims under TILA section 129C. See generally the section-by-section analysis of §§ 1026.25 and 1026.43, above.

The information collection in the final rule is required to provide benefits for consumers and would be mandatory. See 15 U.S.C. 1601 et seq.; 12 U.S.C. 2601 et seq. Because the Bureau does not collect any information under the final rule, no issue of confidentiality arises. The likely respondents would be depository institutions (i.e., commercial banks/savings institutions and credit unions) and non-depository institutions (i.e., mortgage companies or other non-bank lenders) subject to Regulation Z.

Under the final rule, the Bureau generally accounts for the paperwork burden associated with Regulation Z for the following respondents pursuant to its administrative enforcement authority: insured depository institutions with more than $10 billion in total assets, their depository institution affiliates, and certain nondepository lenders. The Bureau and the FTC generally both have enforcement authority over non-depository institutions for Regulation Z. Accordingly, the Bureau has allocated to itself half of the estimated burden to non-depository institutions. Other Federal agencies are responsible for estimating and reporting to OMB the total paperwork burden for the institutions for which they have administrative enforcement authority. They may, but are not required to, use the Bureau's burden estimation methodology.

Using the Bureau's burden estimation methodology, the total estimated burden under the changes to Regulation Z for all of the nearly 14,300 institutions subject to the final rule, including Bureau respondents, would be approximately 14,300 hours for one-time changes. The aggregate estimates of total burdens presented in this part VIII are based on estimated costs that are weighted averages across respondents. The Bureau expects that the amount of time required to implement each of the changes for a given institution may vary based on the size, complexity, and practices of the respondent.

B. Information Collection Requirements

The Bureau believes the following aspects of the final rule would be information collection requirements under the PRA.

1. Ability-To-Repay Verification and Documentation Requirements

Section 1026.43(c)(2) of the final rule contains eight specific criteria that a creditor must consider in assessing a consumer's repayment ability. Section 1026.43(c)(3) of the final rule requires creditors originating residential mortgage loans to verify the information that the creditor relies on in determining a consumer's repayment ability under § 1026.43(c)(2) using reasonably reliable third-party records. Section 1026.43(c)(4) of the final rule provides special rules for verification of a consumer's income or assets, and provides examples of records that can be used to verify the consumer's income or assets (for example, tax-return and payroll transcripts).

If a creditor chooses to make a qualified mortgage, different verification requirements apply to qualified mortgages. Creditors that originate qualified mortgages under § 1026.43(e)(2) or (f) must verify a consumer's income or assets, and current debt obligations, alimony and child support and must also verify a consumer's monthly debt-to-income ratio (or, in the case of qualified mortgages under § 1026.43(f), residual income). The final rule does not contain specific verification requirements for creditors originating qualified mortgages under § 1026.43(e)(4); however, such loans must comply with eligibility requirements (including underwriting requirements) of the GSEs or the Federal agency program applicable to the loan.

The Bureau estimates one-time and ongoing costs to respondents of complying with the requirements in § 1026.43 as follows.

One-time costs. The Bureau estimates that covered persons will incur one-time costs associated with reviewing the final rule. Specifically, the Bureau estimates that, for each covered person, one attorney and one compliance officer will each take 21 minutes (42 minutes in total) to read and review the sections of the Federal Register that describe the verification and documentation requirements, based on the length of the sections.

The Bureau estimates the one-time costs to the 135 depository institutions (including their depository affiliates) that are mortgage originator respondents of the Bureau under Regulation Z would be $7,700, or 94 hours. For the estimated 2,787 nondepository institutions and 77 privately insured credit unions that are subject to the Bureau's administrative enforcement authority, the Bureau is taking the half the burden for purposes of this PRA analysis. Accordingly, the Bureau estimates the total one-time costs across all relevant providers of reviewing the relevant sections of the Federal Register to be about 1000 hours or roughly $81,000.

Ongoing costs. The Bureau does not believe that the verification and documentation requirements of the final rule will result in additional ongoing costs for most covered persons. The Bureau understands that creditors generally have in place underwriting policies, procedures, and internal controls that require verification of the consumer's reasonably expected income or assets, employment status, debt obligations and simultaneous loans, credit history, and debt-to-income or residual income. Notably, in response to the 2011 ATR Proposal, commenters stated that most creditors today are already complying with the full ability-to-repay underwriting standards. For these institutions, there would be no additional burden as a result of the verification requirements in the final rule, since those institutions collect the required information in the normal course of business.

2. Record Retention Requirement

The final rule imposes new record retention requirements on covered persons. As discussed above in part V, the final rule requires creditors to retain evidence of compliance with § 1026.43 (containing the ability-to-repay/qualified mortgage provisions and prepayment penalty restrictions) for three years after consummation. See part V above, section-by-section analysis of § 1026.25.

The Bureau estimates one-time and ongoing costs to respondents of complying with the record retention requirement in § 1026.25 as follows.

One-time costs. The Bureau estimates that covered persons will incur one-time costs associated with reviewing the final rule. Specifically, the Bureau estimates that, for each covered person, one attorney and one compliance officer will each take 9 minutes (18 minutes in total) to read and review the sections of the final rule that describe the record retention requirements, based on the length of the sections.

The Bureau estimates the one-time costs to the 135 depository institutions (including their depository affiliates) that are mortgage originator respondents of the Bureau under Regulation Z would be $3,300, or 40 hours. For the estimated 2,787 nondepository institutions and 77 privately insured credit unions that are subject to the Bureau's administrative enforcement authority, the Bureau is taking the half the burden for purposes of this PRA analysis. Accordingly, the Bureau estimates the total one-time costs across all relevant providers of reviewing the relevant sections of the Federal Register to be about 430 hours or roughly $35,000.

Ongoing costs. The Bureau believes that any burden associated with the final rule's record keeping requirement will be minimal or de minimis. Under current rules, creditors must retain evidence of compliance with Regulation Z for two years after consummation; the final rule extends that period to three years after consummation for evidence of compliance with the ability-to-repay/qualified mortgage provisions and the prepayment penalty limitations in this final rule. The final rule clarifies that creditors need retain only enough information to reconstruct the required records.

The final rule clarifies that creditors need not maintain actual paper copies of the documentation used to underwrite a transaction. See comments 25(a)(2) and 25(c)(3)-1. For most covered persons, the required records will be kept in electronic form. This further reduces any burden associated with the final rule's record retention requirement for creditors that keep the required records in electronic form, as the only additional requirement will be to store data for an additional year, to the extent such creditors are currently storing such data for the minimum period required by Regulation Z.

Furthermore, the Bureau believes that many creditors will retain such records for at least three years in the ordinary course of business, even in the absence of a change to record retention requirements, due to the Dodd-Frank Act's extension of the statute of limitations for civil liability for violations of the prepayment penalty provisions or ability-to-repay provisions (including the qualified mortgage provisions) to three years after the date of a violation. Even absent the rule, the Bureau believes that most creditors will retain records of compliance with § 1026.43 for the life of the loan, given that the statute allows borrowers to bring a defensive claim for recoupment or setoff in the event that a creditor or assignee initiates foreclosure proceedings.

C. Summary of Burden Hours

The below table summarizes the one time and annual burdens under Regulation Z associated with information collections affected by the final rule for Bureau respondents under the PRA. For the two collections, the one-time burden for Bureau respondents is approximately 1,570 hours.

The Consumer Financial Protection Bureau has a continuing interest in the public's opinions of our collections of information. At any time, comments regarding the burden estimate, or any other aspect of this collection of information, including suggestions for reducing the burden, may be sent to:

The Consumer Financial Protection Bureau (Attention: PRA Office), 1700 G Street NW., Washington, DC, 20552, or by the internet to CFPB_Public_PRA@cfpb.gov.

List of Subjects in 12 CFR Part 1026

• Advertising
• Consumer protection
• Mortgages
• Reporting and recordkeeping requirements
• Truth in Lending

Authority and Issuance

For the reasons set forth in the preamble, the Bureau amends Regulation Z, 12 CFR part 1026, as set forth below:

PART 1026—TRUTH IN LENDING (REGULATION Z)

1. The authority citation for part 1026 continues to read as follows:

Authority: 12 U.S.C. 2601; 2603-2605, 2607, 2609, 2617, 5511, 5512, 5532, 5581; 15 U.S.C. 1601 et seq.

Subpart D—Miscellaneous

2. Section 1026.25 is amended by:

A. Revising paragraph (a); and

B. Adding and reserving paragraphs (c)(1) and (c)(2).

C. Adding paragraph (c)(3).

The additions and revisions read as follows:

Subpart E—Special Rules for Certain Home Mortgage Transactions

3. Section 1026.32 is amended by:

A. Revising the section heading;

B. Revising paragraph (b)(1);

C. Removing and reserving paragraph (b)(2);

D. Adding paragraph (b)(3) through (6)

The additions and revisions read as follows:

4. Add § 1026.43 to read as follows: