Large-Bank Deposit Insurance Determination Modernization Proposal

AGENCY:

Federal Deposit Insurance Corporation (“FDIC”).

ACTION:

Advance notice of proposed rulemaking (“ANPR”).

SUMMARY:

The FDIC is seeking comment on whether and how the largest insured depository institutions should be required to modify their deposit account systems to speed depositor access to funds in the event of a failure. Today, insured institutions do not track the insured status of their depositors yet the FDIC must make deposit insurance coverage determinations in the event of failure. The current process might result in unacceptable delays if used for an FDIC-insured institution with a large volume of deposit accounts. Such delays would have an impact on depositors' ability to access their funds and are likely to result in a resolution (of the failed institution) significantly more costly to the Deposit Insurance Fund. As currently contemplated, the options discussed in the ANPR would apply only to the 152 insured depository institutions with more than 250,000 deposit accounts and more than $2 billion in domestic deposits, as well as seven additional institutions with total assets over $20 billion, less than 250,000 deposit accounts and at least $2 billion in domestic deposits. In December 2005 the FDIC issued a prior advance notice of proposed rulemaking on this subject (“2005 ANPR”). This ANPR is a follow-up to that issuance. The FDIC is seeking comment on all aspects of the ANPR.

DATES:

Comments must be submitted on or before March 13, 2007.

ADDRESSES:

You may submit comments by any of the following methods:

• Agency Web site: http://www.FDIC.gov/regulations/laws/federal/propose.html. Follow the instructions for submitting comments.
• E-mail: comments@FDIC.gov.
• Mail: Robert E. Feldman, Executive Secretary, Attention: Comments/Legal ESS, Federal Deposit Insurance Corporation, 550 17th Street, NW., Washington, DC 20429.
• Hand Delivered/Courier: The guard station at the rear of the 550 17th Street Building (located on F Street), on business days between 7 a.m. and 5 p.m.
• Public Inspection: Comments may be inspected and photocopied in the FDIC Public Information Center, Room E-1002, 3501 North Fairfax Drive, Arlington, Virginia, between 9 a.m. and 5 p.m. on business days.
• Internet Posting: Comments received will be posted without change to http://www.FDIC.gov/regulations/laws/federal/propose.html , including any personal information provided.

FOR FURTHER INFORMATION CONTACT:

James Marino, Project Manager, Division of Resolutions and Receiverships, (202) 898-7151 or jmarino@fdic.gov, Joseph A. DiNuzzo, Counsel, Legal Division, (202) 898-7349 or jdinuzzo@fdic.gov or Catherine Ribnick, Counsel, Legal Division, (202) 898-3728 or cribnick@fdic.gov.

SUPPLEMENTARY INFORMATION:

I. Background

When handling a depository institution failure the FDIC is required to structure the least costly of all possible resolution transactions, except in the event of systemic risk. In addition, the FDIC is required to pay insured deposits “as soon as possible” after an institution fails and places a high priority on providing access to insured deposits promptly. In view of the significant industry consolidation in recent years, the FDIC is exploring new methods to modernize the process to determine the insurance status of each depositor in the event of a depository institution failure. The FDIC's current procedures to determine deposit insurance coverage may result in unacceptable delays if used for an FDIC-insured institution with a large volume of deposit accounts. In developing a new system to determine insurance coverage in a large-bank failure, the FDIC's goals are to minimize disruption to depositors and communities and to minimize costs to the Deposit Insurance Fund.

The ANPR's focus is on FDIC-insured institutions with complex deposit systems. These include those institutions with the largest volume of deposit accounts, currently expected to include 152 insured institutions with over 250,000 deposit accounts and total domestic deposits of at least $2 billion, as well as seven additional institutions with total assets over $20 billion, with less than 250,000 deposit accounts and total domestic deposits of at least $2 billion (“Covered Institutions”). One of the assumptions underlying this ANPR is that no institution would be required to submit detailed customer deposit data to the FDIC unless the institution was in danger of failing.

Insurance Coverage and Insurance Coverage Determination Procedures

The basic FDIC insurance limit is $100,000 per depositor, per insured institution. Deposits maintained by a person or entity in different ownership rights and capacities at one institution are separately insured up to the insurance limit. All types of deposits (for example, checking accounts, savings accounts, certificates of deposit, interest checks and cashier's checks) held by a depositor in the same ownership category at an institution are added together before the FDIC applies the insurance limit for that category. The FDIC generally relies upon the deposit account records of a failed institution in making deposit insurance determination.

To achieve accurate deposit insurance determinations, the FDIC uses a specialized system to analyze depositor data and apply insurance rules. As part of its normal practice, the FDIC obtains depositor data only at the time an insured institution is in danger of failing. These data are received in the weeks or months prior to failure, and the FDIC uses them to determine the insurance status of their depositors and to estimate the total amount of insured funds in the institution. The current FDIC deposit insurance determination process has several steps. Each step varies in time and complexity, depending on the institution's characteristics (primarily the number of deposit accounts and type of deposit account system). The following is a summary of the usual steps involved in the insurance coverage determination process where deposits are passed to an acquiring institution:

• Closing out the day's business. In the event of failure, it is the FDIC's practice to close out the insured institution's daily business prior to obtaining the account balances upon which the insurance determination is based. Generally, this process is completed according to the bank's existing procedures. All of the day's check processing and deposit transactions are completed, and end-of-day account balances are determined. This process can require varying lengths of time, across Covered Institutions. For larger institutions this process can run into the early morning hours.
• Obtain deposit data. A data file is obtained from the institution or its servicer. Obtaining usable data from the institution or its servicer frequently is a time-consuming process. The FDIC will provide the institution or its servicer with a standard data request. The standard data request requires the institution to provide approximately 45 data fields for each deposit account along with electronic copies of trial balances and deposit application reconciliations. FDIC technical staff works with the insured institution until the standard data set requirements are met and the files provided the FDIC can be processed properly. Generally, the FDIC has at least 30 days advance warning to plan and prepare for a failure. Data are requested in advance to test delivery capabilities, prove the balancing and reconciliation processes and make certain that all required data fields have been included.
• Process deposit data. Data are received and validated (including reconciliation to supporting subsidiary systems). Using its Receivership Liability System (“RLS”), the FDIC determines which accounts are fully insured, which are definitely uninsured and which are possibly uninsured (pending the collection of further information). The RLS automatically groups accounts based on the ownership category and the name(s), address, and tax identification number for each account. This process is part of the insurance determination performed on the depositor data received from a failed institution.
• FDIC holds/debits based on insurance determination results. Funds deemed insured are passed in full to the acquiring institution. Accounts definitely uninsured are debited for the uninsured amount and a receivership certificate (“RC”) is issued for the debited amount. Holds are placed on accounts deemed potentially uninsured for amounts over the insurance limit, and the account owner is contacted. If additional information is required from the depositor, a meeting is scheduled. These meetings afford the opportunity to collect information necessary to finalize the insurance determination on the possibly uninsured depositors. The typical institution resolved by the FDIC does not have the capability to post a large volume of holds electronically by batch. However, this is an essential requirement for an effective depositor claims process for larger institutions.

Least-Cost Resolution Requirements

As noted above, when handling a depository institution failure the FDIC is required by statute to structure the least costly of all possible resolution transactions, except in the event of systemic risk. Even with systemic-risk failures, the FDIC must conserve costs. Since the introduction of the systemic risk exception in 1991, no exceptions to the least-cost requirement have been made. The FDIC's least-cost requirement was intended to reduce resolution cost and instill a greater degree of market discipline by requiring losses to be borne by uninsured depositors and non-deposit creditors.

When an insured institution fails the FDIC may pay insured depositors up to the insurance limit (a “pay-off”) or the FDIC may sell the failed institution to another FDIC-insured institution (a “purchase and assumption transaction”). Another option is to establish a bridge bank or a conservatorship and transfer deposits to that institution. Preservation of the deposit franchise of a failed institution is an important facet of minimizing resolution costs.

Complexities Caused by Industry Consolidation

Historically, most insured institution closures occur on a Friday. In almost all cases, the FDIC has made funds available to the majority of insured depositors by the next business day, usually the Monday following a Friday closing. All of the insured institution failures of the past ten years have been of modest size, the largest being Superior Bank, FSB with total deposits at the time of closure of about $2 billion and roughly 90,000 deposit accounts.

Industry consolidation raises practical concerns about the FDIC's current business model for handling institution failures. In most instances, larger institutions are considerably more complex, have more deposit accounts, are more geographically dispersed and have more diverse systems and data-integration issues than small institutions. This is especially true of large institutions that have recently engaged in merger activity. Implications of industry consolidation over the past ten years can be seen in Table 1. If such trends continue, deposits will become even more concentrated in the foreseeable future.

The single most important facet determining the complexity of the claims process for depositors of a failed institution is the number of deposit accounts. Other factors are important as well, including the volume of daily transactions, the amount of uninsured funds, the number of separate computer systems or “platforms” on which deposit accounts are maintained, the speed at which the institution's deposit operations must be resumed following failure and the potential spillover implications of the failure. The FDIC's analysis of these factors as applied to larger banks indicates that the industry can be divided into two segments as shown in Table 2.

Large institutions typically have more accounts, more complex deposit systems and require a rapid resumption of deposit operations in the event of failure to protect the institution's franchise value. With Covered Institutions the speed of the claims process could be greatly enhanced by the FDIC obtaining a timely data download and by improving the institution's capability to automatically post holds or debit uninsured funds.

Covered Institutions are more likely to fail due to liquidity reasons prior to becoming critically undercapitalized under prompt corrective action. Most likely, this will be a more rapid and less orderly event. Institutions more susceptible to a liquidity insolvency pose greater problems for the FDIC. Such institutions have a less predictable failure date. The failure could occur on any day of the week, and pre-failure access to the institution may be limited because liquidity insolvency oftentimes is difficult to anticipate, and because liquidity insolvency can occur in a very compressed period of time.

Covered Institutions present unique challenges in the event of failure. For the smaller, less-complex Covered Institutions these challenges may be only modest; for the larger, more complex members of the group they are more severe. As noted, the FDIC is concerned about both the size and complexity of the deposit operations of Covered Institutions and the necessary speed of the claims process to make funds available quickly to depositors and maximize the institution's franchise (or re-sale) value.

II. The 2005 ANPR

The 2005 ANPR requested comment on three options for enhancing the speed at which depositors of the larger, more complex insured institutions would receive access to their funds in the event of failure. All of the options entailed modifications to the deposit account systems of Covered Institutions to facilitate the insurance determination process. Option 1 was to require the institution to install on its deposit system a capability that, in the event of failure, would place a temporary hold on a portion of the balances of large deposit accounts. The percentage hold amount would be determined by the FDIC at the time of failure, depending mainly on estimated losses to uninsured depositors. Such provisional holds would be placed immediately prior to the day the institution reopens for business (generally expected to be the next business day) as a bridge bank (discussed above). The institution also would need to be able to automatically remove these holds and replace them with the results of the deposit insurance determination when they become available. The insurance determination would be facilitated by certain depositor data (such as the depositor's name, address, and tax identification number) maintained by the institution in a standard format. The data would include a unique identifier for each depositor and the insurance ownership category of each account.

Option 2 was similar to Option 1 except that the standard data set would have included only information that institutions currently possessed. The option would not have required institutions to create a unique identifier for each depositor or to classify each account by ownership category.

Option 3 was to require the largest ten or twenty insured institutions (in terms of the number of deposit accounts) to know the insurance status of their depositors and to be able to deduct expected losses to uninsured depositors in the event of failure.

Comments on the 2005 ANPR

The FDIC received 28 comments on the 2005 ANPR. Six were from trade organizations, fourteen from large institutions, four from community banks and four from others. Most commenters expressed an appreciation of the objectives set forth in the 2005 ANPR. The letter submitted jointly by American Bankers Association, America's Community Bankers and The Financial Services Roundtable “recognize[d] that the Federal deposit insurance system's viability depends on the principle that no financial institution is either too big or too small to fail. The development of prudent systems to prepare for and respond to the failure of any size institution is an important component of the Corporation's receivership functions.” Nevertheless, the majority of commenters generally opposed implementation of any of the options offered in the 2005 ANPR. Eighteen of the twenty-eight comment letters (sixty-four percent) indicated opposition to the 2005 ANPR, citing high costs and regulatory burden. The aforementioned joint comment letter from three trade associations “urge[d] the Corporation to reconsider its program to implement the 2005 ANPR.” A complete summary of the comments received on the 2005 ANPR is provided in Appendix B.

III. The Revised ANPR

Process Overview

Under the process discussed in the ANPR, in the event of failure a Covered Institution would complete its nightly processing cycle according to the institution's normal practices. After completion of this nightly processing cycle provisional holds would be placed on large deposit accounts through the institution's deposit systems as specified by the FDIC. The placement of provisional holds will allow the opening of a bridge bank the day following failure, yet guard against the loss of uninsured deposit funds subject to loss. A standard set of data files reconciled to the institution's supporting subsidiary systems will then be provided to the FDIC, to be used as the basis for making deposit insurance determinations. The results of the insurance determination will be returned to the bridge bank, likely within several days. At this point the provisional holds will be removed en masse to be replaced with the results of the deposit insurance determination. The FDIC requests comment on all aspects of this contemplated approach, including cost/benefit issues and alternative approaches that would allow the FDIC to accomplish its objectives of affording a timely deposit insurance determination and a prompt release of funds to depositors.

Continuation of Business Operations

For the purposes of implementing the possible requirements explained in the ANPR, Covered Institutions should assume that their deposit operations would continue post failure in a bridge bank or a federally chartered mutual association. In the event of failure the bank would complete the nightly deposit processing cycle according to the institution's normal practices. For insurance determination purposes, the FDIC would use the deposit account balance generated at the end of the nightly processing cycle. This is the account balance against which provisional holds would be calculated.

Tiered Approach

Based on the comments received on the 2005 ANPR and additional analysis, the FDIC has refined its thinking in terms of how to approach the issues discussed in the 2005 ANPR. The FDIC is putting forward for comment an approach under which each insured depository institution would fall into one of three categories: Tier 1 Covered Institutions, Tier 2 Covered Institutions and Non-Covered Institutions. Tier 1 Institutions would include the largest, most complex institutions among those having at least 250,000 deposit accounts and more than $2 billion in domestic deposits. Tier 2 Institutions would include institutions of lesser complexity among those having at least 250,000 deposit accounts and more than $2 billion in domestic deposits, and those with at least $20 billion in domestic assets and $2 billion in domestic deposits not falling under the definition of a Tier 1 Institution. Non-Covered Institutions would be any insured depository institution not meeting the definition of a Tier 1 or 2 Covered Institution. Non-Covered Institutions would be exempt from the requirements discussed in the ANPR.

Compared to the 2005 ANPR, the definition of a Covered Institution has been expanded to include insured institutions with at least $20 billion in domestic assets and $2 billion in domestic deposits, regardless of the number of accounts. While some such institutions may have far fewer than 250,000 deposit accounts, the FDIC is concerned that—for such institutions—a Friday closure date cannot be expected, a bridge institution will need to be established quickly and that a high percentage of deposit accounts may involve uninsured funds. The FDIC is interested in comments on the challenges presented by such institutions in the event of failure compared to other institutions with a comparable number of deposit accounts. Should the definition of Covered Institutions be expanded to include institutions with fewer than 250,000 deposit accounts?

Requirements for Different Tiers/Explanation of Requirements

As explained more fully below, under the approach being put forward for comment, a Tier 1 Covered Institution would be required to have in place systems that could: (1) Provide a unique depositor identification (“ID”) for each depositor; (2) implement automated provisional holds against deposit accounts; (3) supply a standard data framework (where the form and content of this data structure will be developed in cooperation with insured institutions); (4) remove provisional holds; (5) supply an agreed upon standardized data structure to compute a trial balance; and (6) post holds and debits in batch mode resulting from the deposit insurance determination results. A Tier 2 Covered Institution would be subject to the same requirements as a Tier 1 Covered Institution except it would not have to provide a unique depositor ID for each depositor. Each of these requirements is described below, along with specific questions on which the FDIC requests comment.

(a) Unique Depositor ID

Tier 1 Covered Institutions would be required to uniquely identify each depositor. The FDIC requests comments on all aspects of this possible requirement. In particular:

• To what extent can Covered Institutions uniquely identify depositors using current systems and procedures?
• What would be the best method(s) to use for depositor identification? Should the FDIC specify the format to be used for depositor identification, or should this be left to the Covered Institution to determine?
• How expensive would it be for Covered Institutions to supply a unique identifier for each depositor? Is this something that Covered Institutions are considering for internal business purposes? If not, how do Covered Institutions determine common ownership for relationship management, cross-selling, risk management or other purposes? How long would it take to implement a unique depositor identification process? To what extent is the answer to that question a function of running deposit accounts on more than one platform?
• How reliable would the data be in identifying each depositor? To what extent are Covered Institutions able to identify account owners (as opposed to trustees, managers, beneficiaries, etc.) from source files being supplied to the FDIC for insurance determination purposes? Does this differ by types of accounts; for example, checking accounts versus (brokered) CDs?
• Could Covered Institutions uniquely identify depositors within a single legacy data system? Is there an accompanying Customer Information File (“CIF”) available for each legacy data system? Could the Covered Institutions provide instructions or rules to assist the FDIC to integrate depositor records across these legacy data sources?

(b) Provisional Holds Against Deposit Accounts

Under the suggested approach, Tier 1 and 2 Covered Institutions would be required to have in place an automated process for implementing a one-time FDIC provisional hold immediately following the completion of the nightly deposit processing cycle following a failure. The contemplated provisional hold algorithm contains variables that would be supplied by the FDIC only on the day of failure. Provisional holds would be applied to individual accounts (commonly owned deposits are not aggregated). Provisional holds would vary by individual account balance and type. Under one approach: (1) Deposit accounts with balances below $X dollars would not be subject to a provisional hold; (2) deposit accounts with balances between $X and $100,000 would be subject to a provisional hold of Y percent; and (3) deposit accounts with balances above $100,000 would be subject to a provisional hold of Z percent.

The FDIC would supply the values X, Y and Z to the institution on the day of failure. Those values could differ depending on whether the account is a demand deposit/NOW account, money market deposit/savings account or time deposit. X could be set at a higher level for DDA systems than for time deposit systems, for example. The values X, Y and Z also could differ depending on whether the institution categorizes the account as consumer or business. For these purposes, the account category would be the one normally used by the institution, rather than a definition more consistent with FDIC insurance rules. FDIC research indicates the likely value of X would fall between $30,000 and $80,000. Based on account-size distributions provided by a sample of insured institutions, this potential threshold range is expected to exclude over 90 percent of deposit accounts from the provisional hold process at most institutions. Given the historical loss experience for large institutions and their general liability structure, the FDIC expects that the values of Y and Z would be less than 15 percent.

The FDIC requests comments on all aspects of these possible requirements concerning provisional holds on deposits. In particular:

• What more would Covered Institutions need to know to design and implement such a system?
• What would be the overall cost to a Covered Institution for developing the capability to automatically post provisional holds?
• The deposit systems of many Covered Institutions use software purchased from a small group of vendors. To what extent would vendor-based software changes help mitigate the overall implementation costs of this program?
• Some Covered Institutions use a servicer to process deposit accounts, and some Covered Institutions share the same deposit servicer. To what extent would implementation changes made by the servicer mitigate the costs of this program?
• A provisional hold could potentially trigger complications in the back office of the bridge bank due to an increase in returned items. This might be mitigated if a large percentage of a depositor's checking account balance is made available immediately. If, for example, fifteen percent holds were placed on transaction accounts with balances over $50,000, how significant would the impact be for the back office of the bridge bank? Would overdraft facilities already in place with depositors mitigate this potential impact? If the impact is expected to be significant, how could it be mitigated? Would there be any potential complications in the back office of the bridge bank due to holds placed on MMDA, savings accounts or time deposits? If so, what types of complications, and how could they be mitigated?
• The FDIC may set Y and Z to the same percentage. If the FDIC required institutions to be prepared for only one ratio rather than two, would that reduce the system development costs, the reliability of the algorithm or the speed of running the algorithm? If so, by how much? If only one ratio were used, the FDIC might choose to apply the ratio to the entire balance of accounts with over $X dollars, or it might apply the ratio to only the portion of the balance that exceeds $X. The FDIC does not anticipate requiring institutions to be prepared for both options. Would this choice influence the system development costs, the reliability of the algorithm or the speed of running the algorithm? If so, which choice would be better, and to what degree would it be better?
• The FDIC may choose to set the same X, Y and Z for all deposit systems (as opposed to different thresholds or ratios for transaction account systems, MMDA/Savings systems and time deposit systems). If the FDIC required institutions to be prepared for only one set of thresholds and ratios, would that reduce the system development costs, the reliability of the algorithm or the speed of running the algorithm? If so, by how much?
• The FDIC may choose to set the same X, Y and Z for all account categories. If the FDIC required institutions to be prepared for only one set of thresholds and ratios, would that reduce the system development costs, the reliability of the algorithm or the speed of running the algorithm? If so, by how much?
• Where do individual retirement accounts (“IRAs”) reside? Are they clearly coded or otherwise identified on bank records in a way that would allow their ready identification? Are all IRAs generally found in time deposit systems, in other systems, or are they distributed across multiple systems?
• Since the FDIC would want to continue operating the institution on the business day after failure, the provisional hold process must be completed quickly. The time thresholds may be challenging especially if the institution does not fail on a Friday. Are there ways to structure the provisional hold requirements that would make it easier for institutions to meet the associated timing requirements? For example, would it be helpful if the FDIC agreed that $X would never fall below a predetermined amount (say $30,000 or $40,000)?
• How long would you expect such a program to run?
• What problems would occur if holds were placed during the first day (that is, before the evening check-clearing process) rather than before opening for business on the first day?

(c) The Generation of a Standard Data Structure Reconciled to the Supporting Subsidiary Systems

A fundamental aspect of this ANPR is the development of a standard data framework which does not place an onerous burden on Covered Institutions, while ensuring that the FDIC is provided with an optimum set of data structures within that framework that enable a timely and accurate insurance determination process. The FDIC seeks industry input into the development of this standard data framework. Industry participation will be important in assuring that the FDIC specifies standards that are adequate for making deposit insurance determinations without being unduly burdensome to Covered Institutions. Consequently, the FDIC seeks comment on all aspects pertinent to the development of this standard. Appendix C provides representative standard data elements.

• What would be the overall cost to a Covered Institution to develop a capability to produce a standard data structure complete with associated linked data sources for information such as account ownership or other maintained information relationships required to define a deposit account, as well as provide a data structure to facilitate the generation of a trial balance and reconciliations of accounts? Could a Covered Institution develop and deploy this standard in 18 months? Does the Covered Institution have a standard deposit account data framework that they would recommend the FDIC adopt as a standard to support this deposit account definition process?
• The deposit systems supporting many Covered Institutions use software purchased from a small group of vendors and servicers. Could a vendor or servicer develop the standard data structure and the necessary processing logic to pull the data into the specified standard format for multiple institutions or does your institution have unique details that would prevent this from occurring?
• To meet the proposed standard data structure requirement, institutions may have to link records from the CIF with the deposit systems or provide a key for linking elements so data from the CIF could be linked to individual account owner records. This would be more complex than a standard data structure that only included items from the deposit systems, but it would enable the FDIC to make timely insurance determinations. Once the systems had been developed and tested, how much longer would it take for an institution to prepare a standard data structure that included CIF and deposit system items, compared to one that included only deposit system items?
• The FDIC would require transmitted deposit balances to reconcile to the actual trial balance, both principal and interest dollar amounts and the deposit record counts. How does reconciliation affect timeliness? Can the process be developed in advance and automated?
• The standard data set should not contain records for foreign deposits or international banking facility (“IBF”) accounts, since they are not defined as deposits for insurance purposes. Do foreign deposits reside on separate deposit systems? Would your institution have any problems creating a data set that excludes foreign deposits not payable in the U.S.? If so, how might these problems be mitigated? Would your institution have problems placing a blanket freeze on all foreign deposits and IBF accounts so that the funds could not be drawn on the bridge bank?
• Deposits held by the institution's subsidiaries and affiliates should be included in the standard data set. For deposit insurance purposes all deposits owned by the same FDIC charter, whether an affiliate or subsidiary, should be included in the data call if the account is held at the institution. Would your institution have any problems complying with the standard data structure described above that includes the full balance of deposits held by subsidiaries and affiliates? If so, how might these problems be mitigated?
• Would Covered Institutions have difficulty supplying complete and reliable data for any of the items listed in Appendix C? If so, which ones? Do problems arise because the data are incomplete (available for some accounts but not others) or for other reasons?
• One of the items envisioned in the standard data structure is a flag for bank-owned accounts (the institution's payroll accounts, for example), but not accounts owned by others and managed by the institution (trust accounts, for example). These accounts are not deposits and thus should be excluded from the deposit insurance determination process. How costly would it be for institutions to provide a reliable flag for these accounts or remove them from the standard data set prior to transferring it to the FDIC? If no flag were available, the FDIC might place provisional holds on these accounts. Would such an action cause problems in the back office? If so, how serious a problem might it cause?
• In the event of failure, depositor data may be transmitted to the FDIC or its designee. One method for data transfer of the deposit file(s) is via secure FTP, requiring financial institutions or their servicers to use VPN to communicate with the FDIC over the Internet. What are the relative costs and benefits of using a secure FTP? Are there more effective, less costly ways of transmitting data to the FDIC?
• The transmission method may depend on the number of accounts in the transmission data sets. For some Covered Institutions the FDIC may have to deploy hardware to the failed institution. How would Covered Institution suggest this process be handled and which location would be optimum to support FDIC requirements?

(d) Posting the Insurance Determination Results and Removal of Provisional Holds

The FDIC would forward insurance results to be incorporated into the institution's deposit systems as soon as possible, perhaps as quickly as the day following the receipt of the standard data set. The results would dictate debits and holds to be placed by batch in an automated fashion on deposit accounts. The processing stream would be as follows: FDIC would notify Operations/IT that results are available. This notification would trigger a process whereby all provisional holds are removed en masse. After provisional holds have been removed, the bridge bank would run replacement transactions. Depending on the depositor's insurance status, the replacements could include: (1) No replacement (that is, just release the provisional hold); (2) a debit of the account by the amount specified by the FDIC; (3) a debit and credit of the account (that is, debit the uninsured balance and credit an advance dividend); and (4) placement of a FDIC hold that might not be the same amount as the provisional hold. In a few cases, new FDIC debits or holds may be placed on accounts that did not have a provisional hold. Both the removal of provisional holds and the placement of new FDIC transactions would have to be accomplished in the same nightly processing schedule and the institution would have to be open for business as usual on the next business day.

As to this proposed procedure, the FDIC requests responses to these specific questions:

• What would be the overall cost to a Covered Institution for developing the capability to remove provisional holds and automatically process account debits and holds based on the insurance determination results?
• Would the en masse removal of provisional holds, coupled with the placement of FDIC debits, credits and holds during the same processing schedule, raise operational issues? If so, what types of issues, and how might they be mitigated? Would the system development costs or operational risk be reduced if this process were only scheduled on a weekend?
• The FDIC is contemplating providing institutions with an ASCII/EBCDIC text file with debit, credit and hold transactions based on the insurance determination. Could the data contained in such a file be readily reformatted so that the transactions can be processed on the institution's deposit systems? Is there a format other than ASCII/EBCDIC that is easier and less costly for institutions? If so, what is it? Would it be helpful for the FDIC to provide institutions a sample data set (for testing) during the implementation period?
• In some cases, all accounts with debits would also have credits. The FDIC anticipates that this would simplify the reconciliation process and the settlement process between the insurance fund and the bridge bank, since the debits relate to uninsured balances and the credits relate to advance dividends. This policy would, however, increase the number of required transactions. Is the larger number of transactions problematic? If so, what are the problems and how might they be mitigated?
• One possible way to reduce the number of transactions in a given processing schedule would be to segment the process; for example, release provisional holds and replace them for only one system (or for selected accounts) per night until they are all completed. The FDIC anticipates that the costs associated with segmenting this process in some way would exceed the associated benefits. Do you believe this would be the case? If not, what benefits and costs would accrue for a segmented process and how should it be segmented?

Debiting time deposits may be operationally more difficult than transaction or savings accounts. It might not be possible to debit a certificate of deposit (“CD”) to reflect a loss resulting from the insurance determination results. Debiting a CD may require that the existing CD be closed and new one opened with the lesser dollar amount.

• What are the operational difficulties of requiring a cancellation of a large number of CDs? What is the best way to automate this process? Are there ways to build upon processes that are already in place for rolling over or paying out CDs? If so, how? The FDIC expects that, in the event of a large institution failure, its new claims system will create a file that contains the data needed by institutions to cancel an uninsured CD and replace it with a smaller CD. What information should be included in that file? What format should it take? Would it be helpful for the FDIC to provide institutions a sample data set (for testing) during the implementation period?

IV—Implementation and Testing Requirements

The FDIC is considering an approach under which an insured institution meeting the definitional requirements of a given tier for the two quarters prior to the effective date of the requirements discussed in the ANPR would have eighteen months to fully implement the respective requirements. The FDIC asks specific comment on whether more time would be needed to implement Tier 1 requirements. For example, should the implementation period be fifteen months for Tier 2 Covered Institutions and eighteen months for Tier 1 Covered Institutions?

Also, under the contemplated approach, regarding a merger of two or more Non-Covered Institutions resulting in Covered Institution status, the requirements of the new tier would have to be fully implemented within, for example, eighteen months following the completion of the merger. Would this be a reasonable way to handle the situation?

Under the contemplated approach, the FDIC would conduct an initial test at each Covered Institution sometime after the initial implementation period ends. Once the initial test is completed successfully, the FDIC anticipates that it would conduct additional tests infrequently at healthy institutions that do not make major changes to their deposit systems—perhaps only once every three-to-six years. More frequent testing may be necessary for institutions that move to Tier 1 from Tier 2, make major acquisitions, experience financial distress (even if the distress is unlikely to result in failure) or undertake major system conversions.

To reduce the frequency of FDIC testing and ensure ongoing compliance, the FDIC might consider requiring that Covered Institutions conduct tests in-house on a regular basis (perhaps every year) and provide the FDIC with evidence that the test was conducted and a summary of the test results. If the FDIC chose to do this, what type of protocols should be set? Should the FDIC prepare a standard report format for the summarized test results? Would it be less costly for institutions to submit test results to the FDIC regularly to reduce the FDIC testing frequency (say from every three years to every five-to-six years)? Which testing option would result in a more reliable process? Why?

In addition, the FDIC would have to test certain other requirements inside the institution, including but not limited to the ability to remove provisional holds en masse and place new holds and debits using a data set that meets the FDIC standards. The testing of processes involving transmittal of data to or from the FDIC would use dummy or scrambled data.

To protect financial privacy, the FDIC's testing process would not require that Covered Institutions transmit any sensitive customer data outside of the institution's premises. Therefore, all testing involving sensitive customer data would be conducted on the institution's premises. The FDIC does not intend to remove sensitive data from the institution's premises under the proposed testing process. These items include, but might not be limited to the completeness and reliability of the standard data structure, the format requirements of the standard data structure and the accuracy and effectiveness of the provisional holds.

V—New Deposit Accounts

Covered Institutions currently are not required to know the insurance status of depositors or inform them of this status when a new account is opened. The FDIC is interested in comments on whether Covered Institutions should be encouraged or required to know the insurance status of each new deposit account and/or notify customers of this status when a new account is opened.

Knowing the identity of each depositor is an important aspect of a deposit insurance determination. If Tier 1 Covered Institutions are not required to have a unique ID for each depositor, should the FDIC require a unique depositor ID to be assigned by Covered Institutions when a new account is opened? The insurance category of each account is necessary for the insurance determination process, but is not a requirement proposed in this ANPR. Should the FDIC require that Covered Institutions determine the insurance category of each new deposit account?

VI. Request for Comments

The FDIC realizes that the requirements discussed in the ANPR could not be implemented without some regulatory and financial burden on the industry. The FDIC is seeking to minimize these costs while at the same time ensuring it can effectively carry out its mandates to make insured funds available quickly to depositors and provide a least-cost resolution for Covered Institutions. The FDIC would like comment on the potential industry costs and feasibility of implementing the options in the ANPR. The FDIC also is interested in comments on whether there are other ways to accomplish its goals that might be more effective or less costly or burdensome. In other words, what approach or combination of approaches (which may include new alternatives) most effectively meets this cost/benefit tradeoff? The FDIC seeks comments on all aspects of the ANPR.

Between 2004 and 2006 the FDIC met with six would-be Covered Institutions and four software vendors/servicers for Covered Institutions. These meetings took place at various stages in the development process. The FDIC found these meetings to be extremely helpful and is requesting additional meetings with interested parties. FDIC staff is willing to travel to facilitate the meeting or structure a teleconference. Any such meetings will be documented in the FDIC's public files to note the institution's general views on the ANPR or answers to questions that have been posed. In past meetings, the institutions and software vendors/servicers discussed proprietary information. Such confidential information would not be made public. The record of the meeting could be prepared by the institution or the FDIC. Any institution or organization wishing to discuss this proposal in more detail or influence the way in which it is implemented should contact James Marino, Project Manager, Division of Resolutions and Receiverships, (202) 898-7151 or jmarino@fdic.gov.

During 2006 the FDIC met with several major software vendors/servicers to discuss an earlier version of the proposal outlined in this ANPR. These meetings provided useful insights into the operations of different deposit software and resulted in changes to the proposal. A previous version of the FDIC's proposal included a “freeze” on time deposits rather than the use of provisional holds against these accounts. The discussions with the software vendors resulted in an elimination of the “freeze” in favor of using provisional holds against all accounts. Further, an earlier version of the FDIC's proposal included three tiers for Covered Institutions rather than two. The third tier—to be comprised of the least complex of the Covered Institutions—did not include a unique depositor ID or provisional hold requirement. The original purpose of the three-tiered approach was to reduce industry implementation costs. The software vendors indicated a less varied set of requirements would be easier and less costly to implement, hence the movement to a suggested two-tiered approach.

Appendix A—Primary FDIC Deposit Insurance Categories

Appendix B—Comment Summary

The FDIC received 28 comment letters in response to the 2005 ANPR. While most of the comment letters touched on multiple points, they generally focused on a common theme. The various themes of the letters are summarized in Table 3. Sixty-four percent of the comment letters indicated opposition due to the view that implementation costs of the options outweighed any potential benefits, high potential costs and regulatory burdens, or the options simply are not needed. In other words, these commenters expressed the general belief that the FDIC failed in the 2005 ANPR to make a compelling case in favor of any of the options in light of their perceptions of the costs.

The 2005 ANPR noted that the FDIC was considering expanding the definition ofa Covered Institution to include any institution with at least $20 billion in total assets, regardless of the total number of deposit accounts. Two institutions falling into this category commented that the definition of a Covered Institution should not be changed from the original definition of at least 250,000 deposit accounts and $2 billion in domestic deposits.

Some commenters were expressly supportive of one or more of the options, but in some cases indicated concern over costs. In particular, the letter from Dollar Bank stated it “understands and supports the need for the FDIC to have a rapid and effective process for determining insurance coverage. Not only does this benefit the FDIC directly, but effective performance by the FDIC also benefits the entire banking system by assuring the public of the reliability of federal insurance of deposits. The FDIC asked in this Proposal for suggestions on alternative approaches that might achieve approximately the same benefits for the FDIC at lower costs for banks. Because Dollar sees no reasonable alternative, it supports the general thrust of the Proposal.”

Two other commenters indicated support because the 2005 ANPR options were viewed as addressing the concept of too-big-to-fail (“TBTF”) and enhancing market discipline. Gary H. Stern, President of the Federal Reserve Bank of Minneapolis made the following five points.

• “To ensure effective use of society's resources, the FDIC must reform current insurance determination procedures which hinder its ability to carry out the least-cost resolution of a large bank.
• The FDIC's Board of Directors should focus on net benefits when evaluating the comments received on the 2005 ANPR and choosing which option to implement.

• The features of Option 2 are necessary but may not prove sufficient to correct weaknesses in the insurance determination process.
• The FDIC should give serious consideration to implementing Options 1 and 3.
• The reformed insurance determination regime should apply to all large banks for which the current regime could prevent a least cost resolution; the same insurance determination scheme need not apply to all covered institutions.”

One comment letter focused almost entirely on financial privacy issues. Numerous other commenters indicated financial privacy concerns as well, particularly as they may arise from any testing program implemented as part of the proposal.

The 2005 ANPR noted that “the FDIC solicits suggestions on alternative means of meeting the objective of conducting a timely insurance determination on Covered insured institutions.” No alternative suggestions were received.

Since such a large portion of the comment letters raised concerns about costs versus benefits, this topic will be discussed in the next section. This will be followed by a discussion of other issues raised in the comment letters.

Commenters' Views on Costs Versus Benefits

General arguments. Many commenters—including all responses from the trade organizations—argued that any option presented in the 2005 ANPR would impose high or significant costs on Covered Institutions. These costs would come in the form of dollar expenditures and the utilization of scarce technological resources. Some responders indicated this was the wrong time for a new technological initiative since “under both Basel II and Basel I-A as proposed, banks will be required to develop new and costly information technologies.”

Many commenters also argued that the likelihood of a Covered-Institution failure was remote. The Financial Institutions Reform, Recovery, and Enforcement Act of 1989 (“FIRREA”), the Federal Deposit Insurance Corporation Improvement Act of 1991 (“FIDICA”) and the Federal Deposit Insurance Reform Act of 2005 (“FDIRA”) were cited as containing provisions reducing the likelihood of large-institution failures. It was noted that the FDIC is undergoing the longest period in its history without a failure. Furthermore, responders pointed out that the most recent failures were of institutions not proposed to be covered by the regulation. It also was argued that the FDIC likely will have ample warning of a large-institution failure, thereby allowing for adequate preparation time. Several commenters recommended applying the 2005 ANPR options only in the event the Covered Institution reaches problem status. This suggestion is discussed in more detail below.

Failure preparation time. The joint trade association letter noted “failures that have occurred in the last few years were among financial institutions that would not be covered by this 2005 ANPR. Regulators frequently had knowledge of the problems undermining these institutions and had time to prepare for closure. Sudden failures were more likely to have been caused by fraud or other criminal activity. It is highly unlikely that such a series of similar events could cause a failure of covered financial institutions because of their size, capital strength and diversity of lines of business. Constructing, maintaining and periodically testing the programs proposed under this 2005 ANPR solely because of the remote chance of sudden failure resembles an expensive solution in search of a very low probability problem.”

The 2005 ANPR noted that Covered Institutions are more likely to be closed due to liquidity reasons, thus are prone to fail on any day of the week. Covered Institutions generally would be handled through a bridge bank structure, and to preserve franchise value the failed institution must open the day following failure. The provisional hold functionality included in Options 1 and 2 allows for a next-day opening of the bridge institution. The nightly processing cycle of Covered Institutions does not end until the early morning hours, often extending until 4 a.m. and, in some cases, until 7:30 a.m. Once the nightly processing schedule is complete a failed institution must generate deposit data to be used by the FDIC to make the deposit insurance determination. The 2005 ANPR options recognize that, even under the best of circumstances, it would be impossible for the FDIC to complete the steps necessary for a deposit insurance determination and have the results posted in time for the opening of the bridge bank the business day following failure. Therefore, it is the FDIC's view that one or more of the 2005 ANPR options appear necessary for a successful bridge bank opening, regardless of the advance warning or preparation time allotted.

Differentiation between options. While the majority of commenters opposed the FDIC moving forward, many clearly differentiated between the three options listed in the 2005 ANPR. The Clearing House stated, “we believe that Option 3 is so extraordinarily burdensome as to be unfeasible and that the burden of Option 1 is clearly excessive. Although Option 2 is less onerous and a possible solution to the FDIC's concerns, we believe that further study and dialogue between the Covered institutions and the FDIC are necessary to refine this option.”

Option 1 differs from Option 2 in that it would require the institution to supply a unique depositor ID and the insurance category of each account. Several commenters noted that—of the two—the insurance category requirement was significantly more burdensome. Wachovia Corporation noted that it “currently uses a unique customer identifier for each of [its] general bank customers. However, this identifier may not be available in all instances. An example of this is brokered CDs, in which the insurance is passed through to individuals who are the ultimate customers. We also do not have a unique way to identify insurance categories. Identifying and developing systemic ways to assess categories may be arduous and costly. Again, the development of this logic by multiple banks would be redundant and would shift responsibility to the bank that the bank should not have to bear.”

Capital One Financial Corporation noted that “we estimate the cost of complying with the FDIC's Option 1 as over $220,000. Most of that cost is attributable to the additional requirements of Option 1 as compared with Option 2—in particular, the requirement to identify the insurance ownership category of each deposit account.”

Estimated costs. No trade organization provided specific cost estimates on the 2005 ANPR options, other than to say the costs would be “high” or “very substantial.” Four of the 14 large-institution responders—Wachovia Corporation, Capital One Financial Corporation, First Tennessee and Dollar Bank—provided cost estimates for one or more of the options. These estimates generally were characterized as being “rough” and frequently contained caveats. The estimates provided are listed in Table 4, which also shows the assessable deposit base of the institution (indicating institution size) and the impact of a 1-basis point annual FDIC assessment (indicating a basis for relative cost comparison).

The paucity of data provided on Option 3 reflects the view among most commenters that it is unfeasible. Wachovia Corporation indicated, for example, that Option 3 was “wholly unacceptable,” which appears to be the reason why no cost estimate was provided for this option. First Tennessee was the only responder providing an estimate for Option 3 indicating it was roughly five times higher that that for Option 2.

For Options 1 and 2 the cost estimates provided in the table are fairly modest when matched against other potential deposit insurance costs. Compared to a 1-basis point annual FDIC assessment, the estimated implementation costs of Options 1 or 2 ranged from 5 to 44 percent. The FDIC expects that implementation costs will vary across institutions. The deposit systems at Covered Institutions are different. In particular, some institutions rely primarily on proprietary systems while others use software or servicing provided by an outside vendor.

The 2005 ANPR noted that many Covered Institutions use deposit software supplied by a common vendor or have their deposits serviced by a common servicer. The 2005 ANPR suggested this structure may help mitigate the implementation costs of the options. No deposit software vendor or servicer responded to the 2005 ANPR, nor did any commenter address the potential cost savings associated with the common use of software providers or servicers. The FDIC believes this common usage would mitigate implementation costs.

Too big to fail and market discipline. Several commenters raised the issue of TBTF, effectively expressing the concern that uninsured depositors of a large institution could be made whole in the event of failure, regardless of expected losses in the failed institution. Mr. Stern's letter noted that “[i]n the face of insufficient technology to segregate deposits or information to determine the insurance status of deposits, therefore, the FDIC would likely prefer to provide depositors with access to deposits even if they might be uninsured. This preference, even if understandable, undercuts least cost resolution and puts pressure on policymakers to invoke the systemic risk exception of [FDICIA]. Invoking the systemic risk exception due to limitations in the resolution process (as opposed to preventing a true systemic crisis) could contribute to substantial resource misallocation in the economy over time.” Mr. Stern noted that these costs are difficult to quantify, although they could be substantial.

FDIC's Views on the Cost/Benefit Tradeoff

Any option will impose industry costs, but benefits also will accrue. The FDIC must balance these costs and benefits.

Summary of costs. In its 2005 visitations to the four large deposit software vendors/servicers, two of the organizations indicated the cost of the provisional hold functionality was fairly modest. The 2005 ANPR specifically requested comment on the costs of implementing the three options. The limited data summarized above suggests fairly modest implementation costs for an Option 2 approach and, for some institutions, Option 1 as well. The consensus of comments was that Option 3 would be prohibitively expensive. While no commenters mentioned the potential cost savings that may arise from the use of common software vendors or servicers, they could be significant. The available data on costs currently is limited, although more information should result from this request for comments as well as other research conducted by the FDIC.

Many responders noted the low likelihood of a Covered-Institution failure. Historical evidence indicates this to be the case. The FDIC also agrees that the reforms implemented in FIRREA, FDICIA and FDIRA serve to reduce the probability of a Covered-institution failure. However, even if the likelihood of a failures among Covered Institutions is perceived to be low, it is not zero. The FDIC should have in place a credible plan for resolving the failure of an institution of any size with the least possible costs. The ability to determine the insurance status of depositors in a failed institution in a timely manner is a critical element for ensuring a least-costly resolution.

Meeting the FDIC's legal mandates. FDICIA was one of the most important pieces of legislation affecting the FDIC's failure resolution process. Its least-cost requirement effectively requires uninsured depositors to be exposed to losses. Also, FDICIA's legislative history and the nature of the systemic risk exception provide a clear message that uninsured depositors of large institutions are to be treated on par with those of any size. Meeting these mandates is an important benefit of the rules being proposed.

Enhancement of market discipline. The FDIC's legal mandates have direct implications for TBTF and market discipline. If financial markets perceive uninsured depositors in large institutions will be made whole in the event of failure, deposits will be directed toward these larger depository institutions. The result would be the misallocation of economic resources. Many market observers believe there are substantial benefits of improved market discipline that accrue even without serious industry distress or bank failures. The FDIC agrees with Mr. Stern's assessment that this resource misallocation could be significant.

Effective market discipline also limits the size of troubled institutions and results in a more rapid course toward failure. Both serve to mitigate overall resolution losses. Lower resolution losses benefit insured institutions through lower insurance assessments.

Equity in the treatment of depositors of insured institutions. In the absence of one or more of the options outlined in the 2005 ANPR, the FDIC is concerned that the resolution of a Covered Institution could be accomplished only through a significant departure from its normal claims procedures. This departure could involve leaving the bank closed until an insurance determination is made or the use of shortcuts to speed the opening of the bridge institution. The use of shortcuts or other mechanisms to facilitate depositor access to funds will imply disparate treatment among depositors within the failed institution and certainly different treatment relative to the closure of a non-Covered Institution. The FDIC places a high priority on the consistent implementation of its claims policies and procedures regardless of the size or complexity of the institution.

Preservation of franchise value in the event of failure. The sale of the franchise of a failed institution can provide significant value to mitigate failure costs and is a necessary ingredient to a least-cost resolution. Superior Bank, FSB, the largest failure over the past 10 years, generated a franchise premium of $52 million, or 17 percent of current estimated FDIC losses in the failure. An ineffective claims process—especially one deviating significantly from the FDIC's normal policies and procedures—risks reducing or destroying an important asset of the receivership. Preservation of franchise value in the event of failure of a Covered Institution will be an important benefit of the proposed options.

Suggested course of action. The strong industry opposition and high costs of Option 3 make it unlikely to be the most cost-effective option. In addition, the less costly options appear to meet the primary objective of the FDIC. Although the 2005 ANPR generated only limited data on the costs of Options 1 and 2, these costs are almost certainly low enough to merit moving forward—particularly given the substantial benefit to the FDIC in being able to meet its statutory mandate for least-cost resolutions and the uniform application of insurance limits, plus additional benefits associated with enhanced market discipline. Implementation costs may vary among Covered Institutions depending on conditions such as the number of deposit systems, the age of these systems and their architecture, and whether deposit operations are processed in-house or through a servicer. To some degree, the factors affecting costs also indicate a facet of operational risk which may influence failure potential.

Implementation of Options Upon Reaching Problem Status

Several commenters suggested delaying the implementation of any options until a Covered Institution reaches “problem bank status.” For supervisory purposes problem bank status refers to any insured depository institution with a composite CAMELS rating of “4” or “5”. None of the Covered Institutions currently are designated as problem institutions. The adoption of this exception likely would imply that no Covered Institutions would have to immediately comply with the new FDIC requirements.

Several commenters also provided insights into the potential time needed to implement the proposed rules. The Clearing House, for example, noted that “material information system changes take significant time. Our member banks have discussed the ANPR with their technical staffs and have determined that any of the requested changes could be made, but only over a significant period of time. Without more specific direction, they cannot put a specific timeframe on the project, but to make any substantial changes over multiple systems, and then fully test them, is likely to take more than a year.” Additional time would be needed for the FDIC to test the system changes.

The FDIC is concerned that a Covered Institution could fail prior to reaching problem status (with a CAMELS rating of “3”, for example), or relatively shortly after attaining problem status. If the one-year implementation time estimate is generally accurate, the FDIC risks not meeting its objectives should a Covered Institution fail more quickly than one year after being designated a problem institution. Further, a period of financial or operational stress is not the opportune time to make the proposed system enhancements.

Cost Reimbursement

Several responders to the 2005 ANPR suggested that the FDIC cover implementation costs, either through a direct payment or an assessment rebate. As shown in Table 4, the estimated costs of implementing Options 1 or 2 are fairly modest, ranging from 5 to 44 percent of a 1-basis point annual FDIC assessment. Implementation costs may be viewed as part of the overall cost of deposit insurance; therefore, not subject to reimbursement.

Extending Program to All Insured Institutions

Two commenters proposed extending the options to all insured institutions, and one commenter suggested the FDIC may apply the options to large institutions now but include small institutions at some future point. The 2005 ANPR specifically limited the scope of the options to the 145 insured institutions with at least 250,000 deposit accounts and more than $2 billion in domestic deposits. The 2005 ANPR noted that the FDIC was considering expanding the definition of a Covered Institution but only in a way that would include a handful of other institutions (for example, those with at least $20 billion in total assets, regardless of the number of accounts). The 2005 ANPR never suggested or mentioned in any way the possibility of extending coverage to all insured institutions.

As noted in the 2005 ANPR, the “FDIC is seeking to minimize [implementation] costs while at the same time ensuring that it can effectively carry out its mandates to make insured funds available quickly to depositors and provide a least-cost resolution for Covered institutions.” The FDIC's deposit insurance determination modernization initiative is directed at improving the process at the very largest institutions. The FDIC has never considered extending the options beyond the largest, most complex institutions. There simply is no business reason for doing so.

Financial Privacy

One comment letter focused primarily on financial privacy, but other letters mentioned the issue as well, especially in the context of any testing program. As noted in the 2005 ANPR, “[a]s part of its normal practice, the FDIC obtains depositor data only at the time an insured institution is in danger of failing. These data are received in the weeks or months prior to failure, and are obtained for the sole purpose of determining the insurance status of individual depositors and estimating the total amount of insured funds in the institution. The receipt of such depositor data is necessary for the FDIC to carry out its insurance function. The options provided in this [2005] ANPR do not alter the FDIC policy regarding the receipt of depositor information in preparation for the resolution of a failing insured institution. The FDIC is aware of the potential privacy issues surrounding the holding of depositor information and has in place strict safeguards to protect these data.” The 2005 ANPR also states “it is possible to conduct an effective testing process while on-site, without the need for sensitive depositor data to leave the institution's premises.”

The 2005 ANPR options would not change the treatment of depositor data in the event an institution is in danger of failing, nor have such changes been proposed. The FDIC still believes an effective testing program can be structured whereby sensitive depositor data never leaves the institution's premises. These testing safeguards eliminate privacy concerns.

Appendix C—Data Elements Included in the Standard Data Set

The Standard Data Request contains data structures which will be used by the FDIC to determine insurance categorization. This data structure may be divided into multiple Record Types/Formats. It is the FDIC's intent to work with the industry to define a standard data structure. If data or information are not maintained or do not apply, a null value in the appropriate field should be indicated.

XML may be the most beneficial format. XML has become a widely adopted standard for data interchange by enabling a common messaging format for the exchange of information between systems. XML will enable all the information listed below to be consolidated into one file and presented in plain text with hierarchical relationships providing a single source/file containing the required information.

Following is a list of the data fields that are to be included in the proposed data structure along with explanations of the data being requested. The fields are listed in the order they should appear in the file.

Representative Deposit Data Elements

The Deposit data elements provide information specific to deposit account balances and account data. The sequencing of these elements, their physical data structures and the mode or method of data transmission will be developed in cooperation with the Covered Institutions.

Note:

Fields 13-26 relate to the Account Name and Address information. Some systems provide for separate fields for Account Title/Name, Address, City, State, Zip, and Country, all of which are parsed out. Others systems may simply provide multiple lines for Name, Address, City, State, Zip, with no distinction. Please populate fields that best fit the system's data, either fields 13-20 or fields 21-26.

Representative Hold Data Elements

The Hold data elements provide information related to any holds for collateral placed on an account. If an account has more than one collateral hold, additional Hold elements may be provided to help the Covered Institutions or FDIC to process holds more efficiently.

Customer Record Held in Central Information File (“CIF”) or Central Information System (“CIS”)

The Customer Record provides information related to each customer of the financial institution. Customers may have more than one deposit account, or may be partial owners of more than one deposit account. Each of the customer's accounts are associated with a customer record. If there are multiple owners of an account, multiple customer records (CIF/CIS) will be associated to the deposit account and will be associated in the deposit record (pointed to or linked by a linking file). If a linking file is required to link customer records to deposit records, please provide the program along with instructions on how to link.