Ambac Assurance Corporation, et al., Appellants,v.Countrywide Home Loans, Inc., et al., Respondents, Bank of America Corp., Defendant.BriefN.Y.June 6, 2018APL-2017-00156 New York County Clerk’s Index No. 651612/10 Court of Appeals OF THE STATE OF NEW YORK AMBAC ASSURANCE CORPORATION and THE SEGREGATED ACCOUNT OF AMBAC ASSURANCE CORPORATION, Plaintiffs-Appellants, —against— COUNTRYWIDE HOME LOANS, INC., COUNTRYWIDE SECURITIES CORP. and COUNTRYWIDE FINANCIAL CORP., Defendants-Respondents, —and— BANK OF AMERICA CORP., Defendant. BRIEF FOR AMICUS CURIAE PROFESSOR MARK J. BROWNE IN SUPPORT OF PLAINTIFFS-APPELLANTS d MICHAEL S. VOGEL JOHN S. CRAIG LAUREN J. PINCUS ALLEGAERT BERGER & VOGEL LLP 111 Broadway, 20th Floor New York, New York 10006 Telephone: (212) 571-0550 Facsimile: (212) 571-0555 Attorneys for Amicus Curiae Professor Mark J. Browne April 20, 2018 TABLE OF CONTENTS Page TABLE OF AUTHORITIES ii STATEMENT OF INTEREST OF AMICUS CURIAE 1 PRELIMINARY STATEMENT 2 ARGUMENT 6 I. THE FIRST DEPARTMENT’S RULE IMPOSES SUBSTANTIAL COSTS ON INSURERS, HONEST INSURANCE APPLICANTS, AND SOCIETY AS A WHOLE BY FRUSTRATING THE EFFICIENT OPERATION OF THE INSURANCE MARKET 6 Insurers’ Ability to Rely On the Truthfulness of Insurance Applicants Is Central to the Functioning of the Insurance Market A. 7 B. Requiring Proof of Reasonable Reliance Would Undermine the Function of the Insurance Market, to the Benefit of Fraud-Doers at the Expense of Both Insurers and Honest Insurance Applicants Imposing A Loss Causation Requirement Would Similarly Undermine the Function of the Insurance Market, to the Benefit of Fraud-Doers at the Expense of Both Insurers and Honest Insurance Applicants 15 C. 21 II. TREATING AMBAC AND OTHER RMBS INSURERS DIFFERENTLY FROM INSURERS PERMITTED TO RESCIND POLICIES WOULD INCREASE INSURANCE COSTS AND HARM THE MARKET 23 CONCLUSION 30 i TABLE OF AUTHORITIES Page(s) Cases A.M.I. Diamonds Co. v. Hanover Ins. Co., 397 F.3d 528 (7th Cir. 2005) Geer v. Union Mut. Life Ins. Co., 273 N.Y. 261 (1937) Ginsburg v. Pacific Mut. Life Ins. Co., 89 F.2d 158 (2d Cir. 1937) Glickman v. N.Y. Life Ins. Co., 291 N.Y. 45 (1943) Matter of Liquidation of Union Indem. Ins. Co. , 89 N.Y.2d 94(1996) MBIA Insurance Corp. v. Countrywide Home Loans, Inc., 105 A.D.3d 412 (1st Dep’t 2013) Nat’l Fed’n oflndep. Bus. v. Sebelius, 567 U.S. 519(2012) New England Mut. Life Ins. Co. v. Doe, 93 N.Y.2d 122 (1999) Stipcich v. Metro. Life Ins. Co., 277 U.S. 311 (1928) 13 22, 23 23 15,23 16 25 12 17, 24 16 Statutes New York Insurance Law § 3105, New York Insurance Law § 3106, 2, 15,25 2, 15 ii Page(s) Other Authorities Christopher M. James, Mortgage-Backed Securities: How Important Is “ Skin in the Game”?, FRJBSF Economic Letter (Dec. 13, 2010).... Emmett J. Vaughan & Therese M. Vaughan, Fundamentals of Risk and Insurance (11th ed. 2014) George A. Akerlof, The Market for “ Lemons Quality Uncertainty and the Market Mechanism, 84:3 Q.J. Econ. 488 (1970) George E. Rejda & Michael J. McNamara, Principles of Risk Management and Insurance (13th ed. 2017) George L. Priest, “The Current Insurance Crisis and Modern Tort Law,” 96 Yale L.J. 1521 (1987) Mark S. Dorfman & David A. Cather, Introduction to Risk Management and Insurance (10th ed. 2012) Mark V. Pauly, The Economics of Moral Hazard: Comment, 58:3 Am. Econ. Rev. 531 (1968) Michael Rothschild & Joseph Stiglitz, Equilibrium in Competitive Insurance Markets: An Essay on the Economics of Imperfect Information, 90:4 Q.J. Econ 629 (1976) Roberta Romano, “What Went Wrong with Directors’ and Officers’ Liability Insurance?” Steven L. Schwarcz, Structured Finance: A Gidde to the Principles of Asset Securitization § 2:3 (3d ed. 2010) 14 10, 15, 17, 22 10, 11, 12 8, 13, 15 11 21 10, 13 10, 11, 12 11 26 iii STATEMENT OF INTEREST OF AMICUS CURIAE Mark J. Browne, Ph.D., is a leading expert in the field of insurance and risk management. He holds the Robert Clements Distinguished Chair in Risk Management and Insurance and serves as Chair of the Faculty of the School of Risk Management in the Tobin College of Business at St. John’s University in New York City. He earned a B.S., M.A., and Ph.D. in economics from the University of Pennsylvania, in each case with a focus on the subjects of risk management and insurance. Professor Browne has published extensively in peer-reviewed academic journals, including on subjects directly relevant to the present appeal, such as adverse selection and the effect of legal rules on the operation of the insurance industry, and he has testified on insurance issues before the United States Senate Committee on Banking, Housing, and Urban Affairs. He has served as President of the American Risk and Insurance Association, the European Group of Risk and Insurance Economists, and the Risk Theory Society, and is a member of the Board of Directors of Mitsui Sumitomo Insurance USA, Mitsui Sumitomo Marine Management, and Mitsui Sumitomo Insurance Company of America. 1 A copy of Professor Browne’s curriculum vitae is being submitted along with this Brief. Affirmation of Michael S. Vogel, dated April 20, 2018, Exh. B. (footnote continued) PRELIMINARY STATEMENT In the decision below, the First Department rejected both the longstanding common-law rule (codified at Insurance Law §§ 3105 and 3106) that an insurer should not be liable on a policy induced by fraud or based upon a breach of warranty that materially increases the risk of loss, and its own prior jurisprudence applying that principle to cases such as this one, where a monoline insurer was required to make claims payments to innocent RMBS bondholders and then sought to recover those payments from the wrongdoer which had fraudulently induced the insurer to issue its policy.2 By treating Ambac’s claim as a garden-variety fraud claim and requiring proof of reasonable reliance and proximate causation linking specific misrepresentations to eventual losses, the First Department not only ignored longstanding common-law principles, but the unique characteristics of insurance that make those common-law principles critically important in the first place. Although Professor Browne is being compensated at his typical rate by Plaintiffs-Appellants Ambac Assurance Corporation and the Segregated Account of Ambac Assurance Corporation (together, “Ambac”) for his time spent preparing the proposed amicus curiae brief, the opinions and conclusions expressed in this Brief represent his own independent views. They do not represent the views of the institution at which he teaches or of any other entity with which he is affiliated. 2 Amicus Professor Browne does not take a position as to whether the policies at issue were induced by fraud. Instead, Professor Browne submits this Brief to assist the Court in deciding the legal consequences that should follow upon proof of fraudulent inducement. 2 The essence of insurance is the transfer of risk from an insured to an insurer, which is then able to pool those risks with other, similar risks and thereby spread the risk of valuable but uncertain activities in a way that benefits insureds, insurers, and society as a whole. Because insurance applicants will systematically know far more than the insurer about facts that are material to the issuance of a policy, they will have strong incentives to conceal or misrepresent facts that would prevent them from getting insurance or require them to pay higher premiums. The truthfulness of applicants thus goes to the heart of the insurance bargain, in which the insurer agrees to insure contractually-defined risks based on specific representations and warranties. Without adequate legal rules to protect insurers, the unscrupulous applicant could simply conceal material facts or disregard material warranties and obtain insurance on terms that the insurer never would have offered. This would unfairly shift the wrongdoer’s own risk onto the insurer and, by extension, honest policyholders who subsidize that risk through higher premiums- if such insurance remains available at all. For this reason, New York law has long recognized that an insurer is not liable on a policy that it has been induced to issue through fraud, without individualized proof of reasonable reliance or that the loss was caused by the falsity of the representation. This rule has benefited not just insurers, but honest policyholders, who benefit through lower 3 rates when they are not forced to subsidize policies issued to dishonest insurance applicants, and has supported New York’s role as a financial and insurance center. There are sound economic and public-policy reasons why insurance fraud is treated differently from fraud in other transactions. Insurance transactions are unique in that the only thing being transferred is risk of loss. In other transactions, risk of loss is allocated secondarily to the transfer of actual assets and liabilities, and rules of reasonable reliance and causation are necessary to avoid opportunistic behavior, where parties seek to use claims of fraud as a pretext to avoid a transaction that they are no longer satisfied with for other reasons. The same concern does not exist in the insurance context, because the essence of the bargain is that the insurance applicant identifies exactly what is being insured (here, structured securities whose cash flows are dependent on the actual characteristics of the loan pools) so that the insurer can evaluate the risks and pricing of insurance based on that information. If the information provided by the applicant is fraudulent, nothing is being transferred except a risk of loss that the insurer never would have accepted had it known the truth. The First Department’s ruling requiring Ambac to prove reasonable reliance and loss causation elevates form over substance and overturns the longstanding principle that insurers should not be liable on a policy induced by fraud. The lower court’s rationale was that the policies issued by Ambac based on the 4 misrepresentations of Defendants-Respondents Countrywide Home Loans, Inc., Countrywide Securities Corporation, and Countrywide Financial Corporation (collectively, “Countrywide”) were irrevocable and that Ambac is here seeking not rescission but instead compensation for performance of the wrongfully-induced insurance contract from the insurance applicant that committed the fraud and materially breached its warranties. R12.A. This is a distinction without a difference. Countrywide had the same incentives to lie as any insured, and Ambac had the same need for truthful information in deciding whether and on what terms to issue its policy. Indeed, Ambac’s agreement to make the policies irrevocable was itself a product of Countrywide’s fraud. The fact that Ambac seeks reimbursement for policy payments rather than to avoid making those payments in the first instance in no way changes the fundamental policy underlying the common-law rule, which places the costs of the wrongdoer’s false statement on the wrongdoer, not the insurer, and by extension, honest insureds. To the contrary, upholding the First Department’s distinction would irrationally create disincentives to issue socially- desirable policies that irrevocably benefit innocent insureds. The common-law allocation of risk properly aligns the insurers’ incentives by placing the risk of loss on wrongdoers. By reversing that allocation of risk, the First Department’s decision upends settled expectations and irrationally stifles the 5 useful economic activity of insurers, like Ambac, that operate in industries and situations where insurers are precluded from rescission. ARGUMENT I. THE FIRST DEPARTMENT’S RULE IMPOSES SUBSTANTIAL COSTS ON INSURERS, HONEST INSURANCE APPLICANTS, AND SOCIETY AS A WHOLE BY FRUSTRATING THE EFFICIENT OPERATION OF THE INSURANCE MARKET. The First Department’s rule, that an insurer must prove justifiable reliance and loss causation to recover claims payments made on a fraudulently-induced policy, should be reversed because it frustrates the operation of the insurance market and imposes costs that would ultimately be borne by honest insureds and society as a whole. In order to function efficiently, insurance depends on the premise that information provided by insurance applicants must be truthful. See Point I.A, below. In reliance on the established rule protecting insurers that are induced by fraud to issue a policy, insurance companies are able to set pricing and generally operate on the assumption that applicants’ information is provided truthfully. The First Department’s rule increases costs and frustrates the market’s ability to transfer risks and promote economically desirable activities. First, the requirement that insurers can recover policy payments induced by fraud only if they can prove reasonable reliance forces insurers to undertake the expense of verifying applicant’s representations on the front end, an expense that falls first on 6 insurers but ultimately falls on honest policyholders, who end up subsidizing the dishonest ones through higher premiums. See Point I.B, below. Second, the requirement that insurers prove that their policy payments were proximately caused by the information that was fraudulently concealed forces the insurer to stand behind a policy that it never would have issued. In effect, the fraudulent applicant would have obtained a policy that is exactly the same as the one the insurer might have issued to an honest and less risky applicant, subject only to a limited policy exclusion if the insurer can affirmatively prove a direct causal connection between the misrepresentation and the loss. Of course, given the potential complexities of proving loss causation and the inherent uncertainty of litigation, it is hardly guaranteed that the insurer would succeed in asserting this limited policy exclusion even when the loss was in fact caused by the misrepresentation. Again, the First Department’s rule unfairly shifts the cost of the fraud from the wrongdoer to the insurer and, in turn, honest insurance applicants. See Point I.C, below. A. Insurers’ Ability to Rely On the Truthfulness of Insurance Applicants Is Central to the Functioning of the Insurance Market. Since those seeking to transfer risk to an insurance company typically know more about the risk for which insurance is being sought than the insurer, insurers collect information on the risk prior to making coverage decisions. Gathering information on the risk and making a coverage decision based upon that 7 information is commonly referred to in the insurance industry as underwriting.3 Underwriting is one of the primary functions of insurance companies, without which they cannot operate effectively or efficiently, and in many cases cannot operate at all. As set forth below, there are strong economic reasons why the ability to rely on the truthfulness of insurance applicants during the underwriting process is essential to the underwriting that makes the provision of insurance possible. First, it is the nature of insurance that insurance companies manage pools of risks in the aggregate. By pooling homogenous risks, insurers are able to take advantage of the Law of Large Numbers and thereby reduce the overall level of risk faced by society.4 The Law of Large Numbers is a statistical concept that, in essence, holds that the larger the number of risks in a pool, the greater the likelihood is that the actual outcomes will closely approximate the expected outcomes. For example, if a fair coin is flipped ten times, we would not be very surprised to see seven (i.e., 70%) heads, as statistically there is a greater than 17% probability of seven or more heads coming up by random chance. In contrast, if we flip the same coin 1,000 times, there is essentially no chance that the fair coin 3 George E. Rejda & Michael J. McNamara, Principles of Risk Management and Insurance 46 (13th ed. 2017) (hereinafter, “Rejda & McNamara”). 4 Rejda & McNamara at 41. 8 will come up heads 70% of the time, and indeed there is a greater than 95% chance that the total number of heads will be between 468 and 532.3 As a result of the Law of Large Numbers, an insurer can prudently price insurance with a great deal of confidence as to the likely losses in a pool of risks with known characteristics. This is only possible, however, if the insurer can reliably create sufficiently homogenous risk pools. If it cannot do so, the insurer will face the risk of excessive claims, which can harm or even destroy the insurer’s financial condition. In order to create appropriate risk pools, information sharing is of fundamental importance. The applicant for insurance, who seeks to transfer a risk to an insurer, will systematically have knowledge about characteristics of that risk that the insurer does not possess, including information associated with its likelihood of resulting in a loss and the amount of the loss, should it occur. The insurer needs this information because it affects the insurer’s willingness to bear this risk, or if it is willing to bear the risk, the premium that the insurer will charge. However, applicants for insurance have an economic incentive to deceive insurers by providing incorrect or incomplete information in order to obtain insurance coverage that would otherwise not be provided or that would not be provided for 5 Calculations of this nature can readily be made with a binomial distribution calculator, many of which are available online. E.g., http://stattrek.com/online- calculator/binomial.aspx. 9 the same premium.6 Economic research provides evidence that individuals and entities will often act opportunistically to the detriment of a less-informed party when information is asymmetric in their favor.7 In the insurance context, insurance applicants opportunistically acting on an asymmetric information advantage can lead to two significant problems that are recognized in the academic literature and in case law: adverse selection and moral hazard. Insurance companies that are less informed about the characteristics of risks they cover than applicants for insurance are subject to adverse selection, which can cause poor insurance risks to predominate in the pool of insured risks. Adverse selection occurs when the insurer, lacking complete information, is unable to perceive differences in risk quality. Dissimilar risks are thus pooled at an average premium level with better risks subsidizing worse risks. Risk pools suffering from adverse selection can become unstable as higher-quality risks leave 6 See Emmett J. Vaughan & Therese M. Vaughan, Fundamentals of Risk and Insurance 174-75 (11th ed. 2014) (hereinafter, “Vaughan & Vaughan”) (“Many applicants for insurance are inclined to misrepresent important facts. The reason for this is to obtain insurance when otherwise no company would issue a policy or to obtain coverage at a lower cost”). 7 See, e.g., George A. Akerlof, The Market for “ Lemons Quality Uncertainty and the Market Mechanism, 84:3 Q.J. Econ. 488, 492-94 (1970) (hereinafter, “Akerlof’); Michael Rothschild & Joseph Stiglitz, Equilibrium in Competitive Insurance Markets: An Essay on the Economics of Imperfect Information, 90:4 Q.J. Econ 629, 634-37 (1976) (hereinafter, “Rothschild & Stiglitz”); Mark V. Pauly, The Economics of Moral Hazard: Comment, 58:3 Am. Econ. Rev. 531, 534 (1968) (hereinafter, “Pauly”). 10 the pool (whether seeking coverage elsewhere or foregoing insurance) to avoid the costs imposed on them by lower-quality risks entering the risk pool. Of course, as higher-quality risks leave the risk pool, the pool becomes even more skewed toward lower-quality risks, increasing the cost of insurance to remaining higher-quality risks, who will in turn flee the pool to avoid the further increasing premiums.8 This iterative process of better-than-average risks leaving the pool followed by premium increases is known as a “death spiral”.9 Death spirals can result in the financial failure of an insurer. Death spirals that occur across all insurers may be induced by regulations or other legal rules, and may even result in the collapse of an entire insurance market.10 For example, as the United States Supreme Court observed in upholding the Affordable Care Act, an individual mandate to buy health insurance was necessary because otherwise, 8 See, e.g., Akerlof at 492-94; Rothschild & Stiglitz at 638. 9 See generally Akerlof at 492-94; Rothschild & Stiglitz at 634-37. 10 See, e.g., Roberta Romano, “What Went Wrong with Directors’ and Officers’ Liability Insurance?”, 14 Delaware J. Corp. L. 1, 27, 29-30 (1989) (“When courts rewrite an insurance contract, the price insurers received will not have been commensurate with the risk they actually bore. Higher premiums are necessary on new policies, with terms identical to older, cheaper policies, to compensate the insurer for the court-added risk. [T]he new risks being placed on insurers may be so difficult to assess as to be uninsurable, which could lead insurers to withdraw from the market.”); see also George L. Priest, “The Current Insurance Crisis and Modem Tort Law,” 96 Yale L.J. 1521, 1524-26 (1987). 11 “Congress knew . . . that simply barring insurance companies from relying on an applicant’s medical history . . . would trigger an adverse-selection death spiral in the health-insurance market: Insurance premiums would skyrocket, the number of uninsured would increase, and insurance companies would exit the market.” Nat’l Fed’n oflndep. Bus. v. Sebelius, 567 U.S. 519, 619 (2012). The same principle applies here - if insurers like Ambac cannot recover claim payments on policies induced by fraud, insurance applicants like Countrywide will have an incentive to conceal risks, the risk pool will worsen, and higher-quality risks will be pushed out of the pool by higher premiums. In contrast, with complete information, insurers would not suffer from adverse selection, and higher-quality risks would not subsidize lower-quality risks.11 The ability of insurers to gather the necessary information to underwrite insurance is thus necessary for the successful operation of an insurance market. 11 See, e.g., Rothschild & Stiglitz at 633-38, 648 (analyzing difficulty of establishing equilibria in insurance markets with imperfect information and noting that “[i]f only the high-risk individuals would admit to their having high accident probabilities, all individuals would be made better off without anyone being worse off’ because this would cure the information asymmetry); see generally Akerlof. The insurer’s need for complete information justifies the common law’s imposition of special rules governing insurance fraud. Fraudulent misrepresentations pose the greatest risk of adverse selection because by definition they involve opportunistic behavior by a party with an informational advantage seeking to gain coverage at an unfair price. 12 Moral hazard occurs where a market participant does not bear the full cost of or responsibility for its actions.12 Similar to adverse selection, in the case of moral hazard, insureds with an information advantage (relative to an insurer) act opportunistically. However, moral hazard not only affects the risk allocation between the insured and the insurer, but it also can create incentives on the part of the insured to engage in risky, socially deleterious behavior. Thus, not only does moral hazard lead to inefficiency through higher insurance rates for honest insurance applicants, it can cause broader injury by causing an insured or prospective insured to behave inefficiently. See A277 (moral hazard is “the risk that counterparties will undertake excessively risky actions once they are not being perfectly monitored.... For instance, a car owner being more reckless once he has insurance than before, increasing the chances for a claim”); see also A.M.I. Diamonds Co. v. Hanover Ins. Co., 397 F.3d 528, 530 (7th Cir. 2005) (Posner, J.) (“Moral hazard refers to the effect of insurance in causing the insured to relax the care he takes to safeguard his property because the loss will be borne in whole or part by the insurance company.”). In the RMBS context, moral hazard can arise when an originator, like Countrywide, “makes insufficient efforts at selecting good borrowers” because it “does not bear the consequences in the originate-to distribute 12 See, e.g., Pauly at 34; Rejda & McNamara at 23. 13 model”. A277.13 Because lenders did not bear the costs of their risky lending, their incentive was to create volume at the expense of stability, without regard to the broader social cost. In the case of the housing market, “the costs of moral hazard and adverse selection can be so prohibitive that markets disappear.... This is what happened in 2007 and 2008. . ..” A287. For insurance to operate efficiently despite the challenges of adverse selection and moral hazard, insurers must protect themselves by requiring that the insured provide accurate information about the risks and their conduct. Verifying this information can be extremely costly. Placing the responsibility for fulfilling policy conditions on the insured imposes the cost of compliance on the party best able to verify and monitor the information provided at the lowest possible cost. If policies could not be voided or policy payments recovered where this information is not accurate, opportunistic behavior would result in riskier behavior by insureds, higher insurance costs, and higher societal costs (both from the unnecessarily high insurance costs and from the underlying risky behavior). 13 See also Christopher M. James, Mortgage-Backed Securities: How Important Is “ Skin in the Game”?, FRBSF Economic Letter (Dec. 13, 2010) (finding “significant performance differences based on the loss exposure of the mortgage originator”; “In short, skin in the game matters for performance.”) Available at https://www.frbsf.org/economic-research/publications/economic- letter/2010/december/mortgage-backed-securities/. 14 The common-law rule embodied in Insurance Law §§ 3105 and 3106 effectively implements the policy goal that information provided by insurance applicants during the underwriting process must be provided truthfully by protecting insurance companies from loss on policies obtained through fraud.14 This rule goes to the heart of the availability and use of insurance, and the First Department’s decision undermining this rule threatens to compromise both. B. Requiring Proof of Reasonable Reliance Would Undermine the Function of the Insurance Market, to the Benefit of Fraud-Doers at the Expense of Both Insurers and Honest Insurance Applicants. The common insurance industry practice of verifying information provided by insureds only after a claim has arisen is supported by strong public policy considerations. As set forth in Point I.A, above, insurance applicants have an incentive to conceal information from insurers in order to obtain coverage they could not otherwise obtain or to obtain coverage with a lower premium, while insurers need accurate information to decide whether and on what terms to offer insurance so that they can manage risk pools in a way that is economically sustainable for the insurer and fair to all insureds. Because of these competing incentives, insurers must verify the information that insurance applicants provide. In theory, this verification could occur either prior to policy inception or at a later 14 See, e.g., Glickman v. N.Y. Life Ins. Co., 291 N.Y. 45 (1943). See also Rejda & McNamara at 194-95; Vaughan & Vaughan at 174-75. 15 time. In practice, in reliance on the longstanding common-law rule protecting insurers who issue policies induced by fraud, insurers often defer verification until a loss occurs and a claim is made- a practice that efficiently reduces the cost of providing insurance and thereby reduces policy premiums. By requiring proof of reasonable reliance, the First Department essentially forces insurers to conduct that verification prior to policy inception, thereby imposing substantial and unjustified costs on both insurers and honest insureds. The common-law rule protecting insurers when they issue policies induced by fraud recognizes that insurers do reasonably rely on applicants’ statements, even without further investigation, because “[ijnsurance policies are traditionally contracts uberrimae fidei” requiring “the insured to disclose conditions affecting the risk”. Stipcich v. Metro. Life Ins. Co., 277 U.S. 311,316 (1928); see also Matter of Liquidation of Union Indem. Ins. Co., 89 N.Y.2d 94, 106 (1996) (“The phrase uberrimae fidei and its translation, ‘of the utmost good faith,’ has long been used to characterize the core duty accompanying reinsurance contracts.”). Indeed, the “deep historical roots” of the principle of uberrimae fidei may be traced to “the early days of marine insurance, [when] an underwriter was often called on to insure a ship that was halfway around the world and had to accept the word of the 16 applicant that the ship was still afloat.”15 While in the modem world verification may less frequently be actually impossible, the essential principle remains unchanged. Both the insurer and insurance applicant benefit from the law’s imposition of a duty of candor on which the insurer may reasonably rely because the insurer is protected from loss on a policy induced by fraud, while the applicant may obtain efficiently-priced insurance without the market-skewing costs of expensive and time-consuming pre-issuance verification. Under these long-established principles, insurers can accept the information provided by applicants as truthful and complete at the time the policy is underwritten, while applicants remain responsible for the truthfulness of their statements.16 If an applicant has not been truthful, the insurer retains the right to investigate and challenge the applicant’s statements when a claim arises. This rule 15 Vaughan & Vaughan at 173. 16 Certain limited exceptions exist, most notably with incontestability clauses, which provide a period of repose after which an insurer may no longer challenge the validity of the policy, which are mandated by statute in New York and many other states for life insurance and some other policies. See New England Mut. Life Ins. Co. v. Doe, 93 N.Y.2d 122, 128-29 (1999). Even under these statutes, however, the insurer is permitted to verify the applicant’s statements after the policy issues, albeit for a limited time, id. at 129, and some incontestability statutes “creat[e] an exception for ‘fraudulent misstatements’”, id. at 131. As this Court observed, those statutes that do not create such an express exception for fraud present a “difficult case”, because “[i]t would be difficult for us to conclude that the Legislature knowingly enacted a statute that would encourage fraud.” Id. This is not such a difficult case, as no statute compels the First Department’s rule, which would encourage fraud. See Point LA, above. 17 ensures that the applicant has an economic incentive to be truthful. As set forth in Point I.A, above, there are strong economic reasons for requiring truthfulness from applicants and for holding applicants responsible for their deception regardless of when it is discovered, because without truthfulness from applicants, the insurance market cannot function. Requiring that applicants be truthful or risk bearing the cost of future losses, either through the voiding of their insurance policies or liability for claims payments, minimizes the cost of insurance and promotes the availability and use of insurance. First, as described in Point I.A, above, incentives for applicants to be truthful help counterbalance the adverse selection and moral hazard problems that are an inherent challenge in the insurance industry. Second, if insurers were required to determine the truthfulness of applicants’ statements prior to the inception of the policy, they would need to incur expenses verifying the information provided by every applicant rather than just those who suffer a loss. This additional cost would harm insurers and honest applicants for insurance, and it would benefit only insurance applicants that fraudulently seek to obtain policies based on false representations. Consider the following example. If there are 100 applicants for insurance and verifying the information provided by each costs $5, the insurer would incur $500 in underwriting verification costs. If, in contrast and according to current 18 practice, the insurer does not bear the loss on a policy that would not have been issued had the applicant been truthful, expense will only be incurred verifying the truthfulness of statements made by applicants who experience a loss. If four have losses, for instance, the insurer will incur $20 in expense, rather than $500. If the information provided by all must be verified, this additional cost of $480 ($500- $20) will be priced into the cost of insurance that all policyholders must pay, increasing the cost of an insurance policy by $4.80, if all 100 buy the policy. All 100, however, likely will not buy the policy. Assume that 10 of the insurance applicants are dishonest and detected. They will not purchase coverage, as it will not be offered. Further, assume that 10 more of the applicants are honest and are offered insurance, but choose not to purchase it. The expense incurred by the insurer verifying the information of the dishonest applicants, as well as of the honest applicants who choose not to purchase insurance, must be priced into the insurance of those who were honest and purchase insurance. Thus, the additional $480 in underwriting costs must be divided among the 80 honest applicants who purchase coverage, meaning that each would pay an additional cost of $6. For their part, the dishonest applicant can move on to another insurer with another attempt at deception aimed at entering a homogenous risk pool to which it would be denied entry without deception. This process imposes an unnecessary cost on the truthful that they do not bear under the longstanding rule that statements are 19 assumed to be truthful at the point of policy inception and are subject to verification later, typically at the time of loss. Thus, the common-law rule is financially advantageous to the truthful and financially disadvantageous to the dishonest. The First Department’s rule is the opposite, and would increase the costs of financial guaranty insurance and other types of insurance, thereby undermining desirable economic activity. For example, insurers would also face increased front- and back-end costs in the automotive and employment insurance contexts. While Countrywide seeks to distinguish Ambac’s situation from such “high-volume insurers” (Countrywide Br. at 23), there is nothing about the economics of the situation that distinguishes high- volume from other insurers. The insurer’s need to establish homogenous risk pools, and the problems of adverse selection and moral hazard resulting from asymmetric information, are equally applicable to Ambac’s business, as the record below establishes. See, e.g., A276-80. Moreover, to the extent the First Department’s decision requires individualized proof of reliance and causation with respect to tens of thousands of individual mortgage loans, Ambac would be subject to the same practical challenges as the sort of “high-volume insurer” Countrywide invokes. The First Department’s rule, if it stands, would confound the settled expectations of insurers who have long understood that they were entitled to rely 20 upon material representations and warranties made by applicants.17 As such, it would frustrate the market’s ability to transfer risks, negatively affecting insureds (and potential insureds) and society as a whole, and thereby undermine New York’s role as a major hub of insurance and economic activity. C. Imposing A Loss Causation Requirement Would Similarly Undermine the Function of the Insurance Market, to the Benefit of Fraud-Doers at the Expense of Both Insurers and Honest Insurance Applicants. For similar reasons, the First Department’s imposition of a loss causation requirement would undermine the efficacy of the insurance markets and benefit fraud-doers at the expense of insurers and honest applicants. By requiring proof that the loss was caused by the fraudulently concealed facts, the First Department’s rule essentially allows the dishonest applicant to obtain a policy that it could never have obtained if it had been truthful - namely, the same policy an honest and less risky applicant could have obtained, albeit one with an exclusion for losses that the insurer can prove were caused by material concealed information. Moreover, because it will often be difficult as a practical matter to prove causation even when it in fact exists, the dishonest applicant will even have obtained some insurance for losses that are caused by the concealed fact. Such proof of causation imposes enforcement costs and, because of such expense and uncertainty, incentivizes 17 See, e.g., Mark S. Dorfman & David A. Cather, Introduction to Risk Management and Insurance 168-69 (10th ed. 2012). 21 insurers to verify applications upfront, with all the attendant costs discussed in Point I.B, above. Thus, the First Department’s rule creates exactly the wrong incentives and undermines the truthfulness of insurance applicants that is essential to the efficient function of the insurance market. See Vaughan & Vaughan at 174 (“The illogical result of such legislation [requiring that the misrepresented or concealed material fact contribute to the loss] may be demonstrated easily.... Such legislation seems to put a premium on fraud or at least make contracts based on fraudulent intent more common.”). Further, by creating a loss causation requirement, the First Department impinges on the insurer’s freedom to choose its own risks. See Geer v. Union Mut. Life Ins. Co., 273 N.Y. 261, 265 (1937). In Geer, the decedent failed to disclose certain medical conditions on his application for life insurance, and subsequently died of unrelated causes (inhaling carbon monoxide). Id. at 272-73 (Finch, J., dissenting). The majority held for the insurance company on the grounds that the nondisclosure was material as a matter of law. Id. at 270. As the Court explained, the relevant inquiry focuses not on causation or, indeed, any consideration other than whether the nondisclosure was material to the insurer’s decision to issue the policy: The salient features of the problem presented in this case are these: The insurance company did not agree that a jury might decide what risks the company should accept. It reserved that choice to itself, and in order to 22 determine whether in a given case it should exercise that choice, it required certain information of each applicant. Id. (emphasis added). Likewise, here, the salient feature of this case is that if Ambac proves that it was induced to issue the policies based on materially false representations, it should not be forced to let a jury’s view of loss causation decide whether the resulting loss is one that Ambac must bear. For this reason, New York law has long not required the insurer to demonstrate a causal connection between the loss and the fraudulently concealed facts. Glickman v. N.Y. Life Ins. Co., 291 N.Y. 45 (1943) (affirming judgment for insurer where “[t]he fact that the applicant died from another cause does not disprove the increase of risk” of insured’s misrepresentation); Geer, supra-, Ginsburg v. Pacific Mut. Life Ins. Co., 89 F.2d 158, 159 (2d Cir. 1937) (finding for insurer without proof of “causal connection” where misrepresentation concerned “an illness unrelated to the disease upon which the present claim for indemnity is based”) (citations omitted). II. TREATING AMBAC AND OTHER RMBS INSURERS DIFFERENTLY FROM INSURERS PERMITTED TO RESCIND POLICIES WOULD INCREASE INSURANCE COSTS AND HARM THE MARKET. The First Department’s decision is premised on a distinction between insurers that seek to void insurance contracts on the basis of fraud (or to defend against claims seeking coverage) versus those, like Ambac, that seek to recover 23 claims payments made on fraudulently-induced policies. R12.A. From an economic and risk-allocation perspective, this is a distinction without a difference. There is no sound economic justification why insurers which, like Ambac, are obligated to make payments to innocent beneficiaries and later seek to recover from fraudulent applicants should be treated materially differently from insurers which are permitted to rescind policies initially granted to dishonest applicants. In either case, the proper goal is the same: to place the costs of the wrongdoer’s dishonesty on the wrongdoer, not on the insurer, honest insurance applicants, or innocent beneficiaries. Manifestly, there are many good reasons why New York should want to promote the issuance of policies that irrevocably benefit innocent insureds-whether in financial guaranty insurance (where the irrevocability feature allows investors to invest with confidence), auto insurance (where it protects victims injured by drivers who may have improperly obtained a policy), or other contexts. Similarly, it is good policy to compel wrongdoers to compensate insurers for performance on wrongfully induced policies in order to discourage fraud. See New England Mut. Life Ins. Co., 93 N.Y.2d 122 at 131 (“[i]t would be difficult for us to conclude that the Legislature knowingly enacted a statute that would encourage fraud”). The First Department’s prior jurisprudence, which held that insurers like Ambac do not need to prove reasonable reliance or loss causation, 24 was consistent with both policies. See MBIA Insurance Corp. v. Countrywide Home Loans, Inc., 105 A.D.3d 412 (1st Dep’t 2013). Consider, for example, an applicant for individual medical insurance who lies about his age to obtain insurance. There is no question that the resulting policy could be avoided under Insurance Law § 3105. Now consider an employer that lies about its employees’ ages to obtain group medical insurance. There may be sound reasons for requiring the insurer to honor claims made by innocent employees, but as between the dishonest applicant and the insurer there is no reason that their relative legal positions should differ at all. To the contrary, just as the insurer is permitted to shift the loss back to the individual applicant upon proof that the applicant’s fraud caused it to issue a policy it would not otherwise have issued, the insurer should equally be able to shift the loss back to the dishonest applicant seeking group insurance for its employees, and it makes no sense to impose additional hurdles of reasonable reliance and causation on the insurer in such a case. The fact that rescission is unavailable in some contexts reflects considerations such as the need to protect innocent third-parties - whether they are bondholders, as here, or accident victims, or innocent employees whose employer lied on an insurance application- not to protect applicants who make fraudulent statements in order to obtain insurance. 25 As in this case, RMBS insurance policies are typically unconditional and irrevocable. R4811 (“Irrevocable policies are the cornerstone of the guaranteed- securities market. Unless policies are irrevocable, ratings agencies cannot base their ratings on an insurer’s AAA-credit rating.”). Without this feature, the policy could not serve its essential function of credit enhancement, which is to “provide[] the assurance needed to sell the . . . securities in the public markets at investment grade prices”.18 Thus, the irrevocability of Ambac’s policy was clearly designed to protect investors, and indirectly benefitted all participants in the securitization, including Countrywide, which was able to find a ready market for its loans because investors were more willing to purchase securitizations of its loans when they were backed by Ambac’s guarantee. Likewise, individual borrowers benefited from the lower interest rates that the rapid securitization of these loans made possible. Irrevocability was not intended to benefit fraudulent applicants who obtained that irrevocable insurance on the basis of false representations, and there is no economic rationale for the First Department’s rule interpreting it as doing so. If the First Department rule stands, monoline insurers (and other insurers that for sound reasons offer irrevocable policies to protect innocent parties or are required by law to do so) will be hindered in their efforts to recover losses from 18 Steven L. Schwarcz, Structured Finance: A Guide to the Principles of Asset Securitization § 2:3 (3d ed. 2010). 26 wrongdoers, and will inefficiently be forced to undertake additional, expensive underwriting before the policy is issued. To create residential mortgage-backed securitizations like the ones at issue here, Countrywide would typically cause a “sponsor” to aggregate thousands or tens of thousands of loans, which would ultimately be transferred to a securitization trust. R4708. The securitization trust would in turn issue securities, payments on which derived from the monthly payment obligations on the underlying mortgage loans. R4513. In this case, Countrywide represented that it had properly underwritten all of the mortgages in the securitizations. Loan underwriting is an involved process that includes obtaining and analyzing various types of applicant information to inform a decision about whether to offer credit and, if so, at what price. There are, of course, substantial costs associated with collecting and analyzing these data, which costs were ultimately borne by consumers in the form of underwriting or origination fees. Were Ambac, or insurers generally, required to repeat this process, it would inefficiently incur costs for conducting due diligence performed by Countrywide and would unnecessarily raise the cost of the securities ultimately issued. For example, underwriting fees associated with a single mortgage often fall between 27 $400 to $1,000.19 Duplicating any meaningful portion of Countrywide’s underwriting would have resulted in substantial additional expenses.20 Certainly, it would be prohibitively expensive and necessitate higher insurance prices. Instead, the representations and warranties made by Countrywide in the contracts it entered into with Ambac and third-party market participants (including that the RMBS complied in all material respects with securities law; that information disclosed regarding Countrywide’s operations and/or financial condition did not contain false or misleading material statements; that Countrywide did not know of any circumstances reasonably expected to cause a “Material Adverse Change” in its business or financial condition; and that the loans complied with Countrywide’s underwriting guidelines) were designed to ensure Ambac that it would not need to replicate the data collection and analysis performed in the initial mortgage underwriting. Moreover, even beyond cost, repeating the same underwriting tasks as Countrywide would have slowed Ambac’s underwriting substantially, undermining 19 See http://themortgageinsider.net/defmitions/underwriting-fee.html; see also http://homeguides.sfgate.com/underwriting-fees-associated-mortgage- 100286.html. 20 While these costs could potentially be reduced through targeted diligence and statistical sampling, the costs would nonetheless be substantial, and the insurer would bear the risk that a dishonest applicant might seek to manipulate the targeted diligence process to deceive the insurer. 28 its ability to compete in the marketplace.21 Of course, it would not just be Ambac who would suffer under this rule. All insurers subject to this regime would be forced to do the same inefficient and time-consuming re-verifications. The end result would be to dramatically slow the securitization process. Consumers, the individual borrowers, would be the first victims of this regime, because the additional costs would have to be accounted for in the terms of their loans. Moreover, sponsors and originators would be forced to incur increased carrying costs as they waited for the re-verification process, resulting in still more cost to consumers. In short, the sound economic reasons for allowing an insurer to avoid a policy that it is fraudulently induced to issue apply with full force to the present claim for damages. If the Court finds that Ambac was fraudulently induced by Countrywide to issue irrevocable policies to innocent policyholders , it should place responsibility for the loss where, as a policy matter, it must belong-on the party whose fraudulent misstatements and omissions and materially false warranties induced it to issue policies that it never would have issued (or issued on the same terms) had it known the truth. 21 Underwriting speed or speed of approval is important for the success of insurance agencies, as faster approval times are associated with increased revenues. See, e.g., https://guidingmetrics.com/content/insurance-industrys-18- most-critical-metrics/. 29 30 CONCLUSION For the foregoing reasons, the undersigned Amicus Curiae joins with Plaintiff-Appellant Ambac in respectfully requesting this Court to reverse the decision of the Appellate Division. Dated: April 20, 2018 New York, New York ALLEGAERT BERGER & VOGEL LLP By: ______________________________ Michael S. Vogel John S. Craig Lauren J. Pincus 111 Broadway, 20th Floor New York, New York 10006 (212) 571-0550 Attorneys for Professor Mark J. Browne as Amicus Curiae /s/ Michael S. Vogel CERTIFICATION I certify pursuant to 500.13(c)(1) that the total word count for all printed text in the body of the brief, exclusive of the statement of the status of related litigation; the corporate disclosure statement; the table of contents, the table of cases and authorities and the statement of questions presented required by subsection (a) of this section; and any addendum containing material required by subsection 500.1(h) of this Part is 6,915 words. Dated: April 20, 2018 New York, New York Respectfully submitted, ALLEGAERT BERGER & VOGEL LLP By: ______________________________ Michael S. Vogel John S. Craig Lauren J. Pincus 111 Broadway, 20th Floor New York, New York 10006 (212) 571-0550 Attorneys for Professor Mark J. Browne as Amicus Curiae /s/ Michael S. Vogel