This report presents the results of our audit of the FDIC’s reserve ratio and assessment determination processes. The objective of this audit was to determine whether: (1) the Division of Insurance and Research (DIR) accurately determines the funds’ reserve ratios and (2) the Division of Finance (DOF) has adequate controls in place to ensure that the FDIC accurately calculates, collects, and processes assessments of financial institutions. During our review, we expanded our objective to include evaluating the FDIC’s communication of information relevant to the reserve ratios to the FDIC Board of Directors (Board). Appendix I contains additional details on our objective, scope, and methodology.
Historically, the FDIC maintained the Bank Insurance Fund (BIF) and the Savings Association Insurance Fund (SAIF) by assessing depository institutions a semiannual premium based on the institution’s total domestic deposits, with certain adjustments, and the degree of risk that the institution posed to its respective insurance fund. Section 302(a) of the FDIC Improvement Act of 1991 (FDICIA) required the FDIC’s Board of Directors (Board) to establish a risk-based assessment system. In September 1992, the Board amended its regulations on assessments to transition from a uniform rate to a risk-based insurance assessment system.[ 1 ]
The Federal Deposit Insurance Act (FDI Act) also required the Board to set assessments only to the extent necessary to maintain the insurance funds at the designated reserve ratio (DRR) of 1.25 percent of estimated insured deposits. The term “reserve ratio” was defined for each deposit insurance fund as the ratio of the fund’s net worth to the value of the aggregate estimated deposits insured by that fund. In setting the assessments, the Board considered expected operating expenses, case resolution expenses and income, the effect of assessments on member earnings and capital, and other factors the Board deemed appropriate. The Board set assessments higher for institutions that exhibited financial, operational, or compliance weaknesses ranging from moderately severe to unsatisfactory or were not well capitalized. The FDIC calculated the reserve ratio and set assessments separately for each deposit insurance fund.
DIR’s Financial Risk Management Branch is responsible for assessing the adequacy of the deposit insurance funds and implementing an effective and fair risk-based premium system. DOF’s Assessment Management Section (AMS) handles the assessment billing and collection function. AMS is responsible for establishing policy-related issues and assessing and invoicing FDIC member institutions on a quarterly cycle for FDIC risk-based assessment premiums. Table 1 presents information about the reserve ratio determination and assessment rate processes.
In February 2006, Congress passed The Federal Deposit Insurance Reform Act of 2005, which requires merging the BIF and SAIF insurance funds into a single Deposit Insurance Fund. The FDIC merged the funds at the end of March 2006. Accordingly, implementation of the legislation eliminated the need for the FDIC to allocate Oakar deposits between BIF and SAIF.
RESULTS OF AUDIT
The FDIC could improve internal controls over the reserve ratio and assessment determination processes. Although the FDIC accurately calculated fund reserve ratios and assessments due from financial institutions, a key underlying assumption supporting the reserve ratio calculations became outdated and was not representative of actual transactions. We concluded that the Corporation could improve its processes for estimating insured deposits and for communicating to the Board information relevant to assessment determinations and other corporate matters and activities.
The FDIC did not periodically validate key assumptions used to allocate estimated insured deposits at Oakar institutions to BIF and SAIF. In particular, the FDIC historically assumed that 100 percent of Oakar secondary fund deposits were insured. Over time, that assumption became less representative of actual transactions.
In October 2004, the FDIC established an interdivisional group to study the Corporation’s methodology for estimating and allocating insured deposits (Study). After considering various options, the FDIC devised a new method for allocating Oakar estimated insured deposits that the FDIC determined was more representative of actual transactions. DIR conducted an analysis simulating how estimated insured deposits would have been allocated had the new method been in effect since the beginning of 1997. DIR concluded that by September 2004, the new method would have resulted in just over $96 billion more in estimated insured deposits allocated to BIF and correspondingly fewer estimated insured deposits allocated to SAIF compared to the existing allocation method. Further, we recalculated BIF and SAIF reserve ratios using DIR’s analysis and concluded that reallocating estimated deposits insured by the funds to reflect the new method would have caused BIF to drop below the 1.25 percent DRR for a 6 quarter period starting in late 2001 and again in 2005. The Act generally required the Board to take action to return the reserve ratio to 1.25 percent within 1 year.
In February 2005, the FDIC formally adopted the new method and elected to apply the new method prospectively and not reallocate estimated deposits between BIF and SAIF. Because the nature, timing, and application of the new method could have had a significant impact on the reserve ratios, we concluded that the FDIC should have more fully: (1) informed the Board of the Study results and various alternatives that were considered and (2) involved the Board in the decision on whether, and how, to change the allocation method, including how to address the cumulative effect (estimated at $96 billion) of not applying the new, more representative allocation method earlier. For example, the Study results were not discussed in the 2005 BIF or SAIF semiannual assessment rate cases presented to the Board. FDIC officials involved in the Study indicated they had operated under Board delegations of authority in adopting the new method and that they had briefed the FDIC Chairman and deputies to the Board members before publicly announcing the change.
In addition to mandating the merger of the BIF and SAIF insurance funds, the recently enacted deposit insurance reform legislation also replaced the 1.25 DRR with a 1.15 to 1.5 percent range, which allows the Board more discretion in setting the DRR. This further supports our conclusion that the FDIC should improve controls over the assessments process by periodically validating assumptions and estimates and clarifying Board delegations of authority related to the assessments determination process. In addition, FDIC management and the Board should jointly review, and improve as appropriate, the Corporation’s procedures and practices for keeping the Board members informed of corporate activities and operations.
FINDINGS AND RECOMMENDATIONS
CALCULATION OF RESERVE RATIO AND ASSESSMENT DETERMINATIONS
The FDIC accurately calculated the BIF and SAIF reserve ratios and accurately calculated and collected assessments due from financial institutions. However, as discussed later in this report, the internal controls over the reserve ratio and assessment determination processes need to be strengthened. Specifically, the FDIC’s assumption used to allocate estimated insured deposits became outdated over time and less representative of actual Oakar institutions.
Calculation and Reporting of Reserve Ratio
The FDI Act required the FDIC to estimate insured deposits in setting assessments for each insurance fund and defined the DRR of each deposit insurance fund to be 1.25 percent of estimated insured deposits. Estimating insured deposits involves taking a snapshot of the depository institutions’ reported deposit holdings and estimating what portion of the deposits are insured. The FDIC was required to conduct this estimate to satisfy its obligation of maintaining the DRR at 1.25 percent of estimated insured deposits.
DIR was responsible for calculating the reserve ratios for the BIF and SAIF insurance funds. DIR calculated the reserve ratios by dividing the BIF and SAIF fund balance by the estimated deposits insured by each fund. To determine the reserve ratio, the FDIC first calculated the amount of estimated insured deposits. To calculate estimated insured deposits, the FDIC obtained deposit information from financial institutions’ Call Reports and Thrift Financial Reports and deducted estimated uninsured deposits from total deposits. The estimated insured deposits amount was the denominator in the reserve ratio calculation.
As a part of this audit, we recalculated the reserve ratio for the quarterly periods January 1997 through June 2005 based on the assumptions used by the FDIC in determining estimated insured deposits. We determined that the FDIC correctly calculated the BIF and SAIF reserve ratios based on those assumptions. Additionally, we verified that the FDIC accurately reported the reserve ratios calculated in this manner in the Quarterly Banking Profile. However, as discussed later, those assumptions became less representative of actual Oakar transactions.
Calculation and Collection of Assessments
The FDIC’s risk-based assessment regulations are set forth in Part 327 of the FDIC Rules and Regulations. The Board delegated authority to the Director, DOF, to administer regulations for the Corporation that governed the payment of assessments by insured depository institutions in accordance with the provisions of the FDI Act. DOF’s AMS handles the assessment billing and collection function. Additionally, AMS is responsible for establishing FDIC policy related to assessments and assessing and invoicing FDIC member institutions on a quarterly cycle for FDIC risk-based assessment premiums.
AMS maintains and uses AIMS II to calculate and collect FDIC insured institutions’ assessments on a quarterly basis. AIMS II has the capability to gather deposit and other data needed to calculate assessments. During the billing and collection process, AMS makes any necessary adjustments/amendments to financial institution demographic and financial data. AMS also produces and reviews assessment invoices and makes those invoices available to financial institutions through FDICconnect at least 15 days prior to the payment due date.[ 3 ] Additionally, AMS staff produce Automated Clearing House files; monitor financial institution changes (e.g., new institutions, terminated institutions, mergers, branch sales); and produce management reports.
To verify that the FDIC had adequate controls to ensure that assessments were properly calculated and collected, we selected a random sample of invoices from 43 of the 602 institutions that had paid assessments for the quarter ending December 31, 2004. Our audit tests included tests of controls to ensure that the assessment base amounts used for billing financial institutions agreed with Call Report and Thrift Financial Report data. In addition, we recalculated invoice totals to ensure that DOF accurately calculated assessment amounts. Further, we verified that the FDIC made assessment invoices available to financial institutions through FDICconnect at least 15 days prior to payment due dates and collected assessments accurately and in a timely manner. We identified no discrepancies during our audit testing.
THE FDIC’S PROCESS FOR ESTIMATING INSURED DEPOSITS
The FDIC could improve its process for estimating insured deposits. We found that the FDIC did not periodically validate assumptions used in estimating insured deposits for the purpose of calculating the reserve ratios. In particular, the FDIC historically assumed that 100 percent of Oakar secondary fund deposits were insured. Because the FDIC did not have a procedure or process for periodically validating the assumption, the Corporation was not aware for an extended period of time that the assumption became less representative of actual transactions. In early 2005, the FDIC adopted a new method for allocating Oakar estimated insured deposits that the FDIC determined was more representative of actual transactions. An FDIC analysis showed that had the new, more representative, methodology been implemented at the beginning of 1997, $96 billion more in estimated insured deposits would have been allocated to BIF and correspondingly fewer estimated insured deposits would have been allocated to SAIF compared to the existing allocation method.
Oakar Institutions
Oakar institutions resulted from transactions in which a member of one insurance fund acquired a member (or some deposits) of the other fund. The resulting Oakar institution’s “secondary” fund referred to the fund that insured the acquired institution. Most such acquired deposits, referred to as Oakar deposits, were SAIF deposits held by BIF members. Oakar deposits dated to the 1989 savings and loan clean-up legislation[ 4 ] wherein the Congress sought to prevent future thrift failures from burdening taxpayers by ensuring that the newly created SAIF would have a viable assessment base. While encouraging banks to buy troubled thrifts, the Congress provided that SAIF would assess some of the bank’s deposits after the acquisition. The amount assessed by SAIF would be derived from the acquired thrift’s deposits.
Oakar institutions were required to include, in their Call Reports, insured deposit information resulting from transactions involving Oakar institutions during the period in which the transaction occurred. However, Oakar institutions were not required to maintain or report separate accounting or allocation of deposits between BIF and SAIF during subsequent periods. Thus, to calculate quarterly reserve ratios, the FDIC had to allocate the estimated insured deposits of Oakar institutions between BIF and SAIF.
Adjusted Attributable Deposit Amount
The Oakar Amendment to the FDI Act introduced the concept of the Adjusted Attributable Deposit Amount (AADA). An AADA was an artificial construct: a number, expressed in dollars, that was generated in the course of an Oakar transaction and that pertained to the buyer. The AADA’s initial value was equal to the actual amount of the secondary fund deposits that the buyer acquired from the seller. Thereafter, the AADA increased or decreased at the same underlying rate as the buyer's overall deposit base, that is, at the rate of growth or shrinkage due to its ordinary business operations, not counting growth due to the acquisition of deposits from another institution (e.g., in a merger or a branch purchase). The FDIC used an Oakar institution's AADA for the following purposes.
Assessments: An Oakar institution paid two assessments: one for deposits in its secondary fund and the other for deposits in its primary fund. The secondary fund assessment was based on the portion of the assessment base that was equal to the AADA. The primary fund assessment was based on the remaining portion of the assessment base.
Insurance: The AADA fixed the amount of the institution's deposits to be treated as insured by an Oakar institution's secondary fund. The remaining portion of the institution's deposits was insured by the primary fund. If an Oakar institution failed and the failure caused a loss to the FDIC, the two insurance funds shared the loss in proportion to the amounts of deposits that they insured.
Estimated Insured Deposits
DIR also used the AADA to allocate estimated insured Oakar deposits between BIF and SAIF for the purpose of calculating the reserve ratios. The denominator in the reserve ratio was largely based on the financial institutions’ estimated insured deposit amounts. The FDI Act did not specifically define “estimated insured deposits.” However, since 1989, DIR assumed, for the purpose of determining the reserve ratios, that 100 percent of the AADA (also known as secondary fund deposits or Oakar deposits) was insured. According to DIR, for several years after FIRREA, this assumption appeared to closely mirror reality. Most of the Oakar deposits created in the early- and mid-1990s were SAIF deposits acquired by BIF members that were, on average, almost fully insured.
To maintain the secondary fund’s share of an Oakar institution’s total deposits (and thus its assessment base), the FDIC increased the Oakar deposit amount by the institution’s total deposit growth rate (excluding acquisitions). Because all Oakar deposits were assumed to be insured, the secondary fund’s estimated insured deposits also increased at the total deposit growth rate. According to DIR, in the early- to mid-1990s, industry insured deposit growth was, on average, close to total deposit growth. Thus, the secondary fund’s proportion of a typical Oakar institution’s estimated insured deposits changed very little during those years.
During 2004, DOF identified financial trends in SAIF that were inconsistent with SAIF reserve ratio trends. DOF determined that from 1998 through 2004, SAIF experienced favorable portfolio investment returns, few institution failures, proportionally smaller losses on failures than BIF, and low expenses but a declining reserve ratio.
One factor that DOF identified as possibly influencing these SAIF trends was the FDIC’s historical assumption that 100 percent of all secondary fund deposits were insured. To quantify the potential impact of this factor, DOF prepared a comparative analysis of the estimated insured deposit amounts allocated to BIF and SAIF for each Oakar institution using: (1) the existing method, and (2) the assessment base allocation method, which allocated an Oakar institution’s total estimated insured deposits on a pro rata basis to the primary and secondary funds. Table 3 shows how insured deposits estimates would have differed between the two approaches for a hypothetical Oakar institution.
Table 3: Insured Deposit Estimates for a Hypothetical Oakar Institution (in millions)
| Methodology for Estimating Oakar Insured Deposits |
Institution’s Total Deposits |
Percentage of Insured Deposits for the Total Institution |
Secondary Fund Deposits (AADA) |
Secondary
Fund Estimated
Insured
Deposits |
| Using the existing method that assumed 100% of the secondary fund was insured. |
$2,000,000 |
80% |
$500,000 |
$500,000 |
| Using the assessment base allocation method that allocated total estimated insured deposits to the primary and secondary funds on a pro rata basis. |
$2,000,000 |
80% |
$500,000 |
$400,000
($500,000 x .80) |
Source: OIG Analysis.
DOF calculated the net difference between the two methods for all Oakar institutions for the June 2004 quarterly reporting period and identified a net difference of $142 billion in the allocation of estimated deposits insured by BIF and SAIF. DOF concluded that the assessment base allocation method would have allocated $142 billion more to BIF and correspondingly less to SAIF. Most of the difference was attributable to approximately 40 BIF Oakar institutions with SAIF (secondary fund) deposits. DOF performed similar analyses for several other quarterly reporting dates. Table 4 presents information from DOF’s analysis.
Table 4: DOF Identified Differences in Oakar Insured Deposit Estimates (in billions)
| Reporting Date |
Total Deposits (Estimated Percentage Insured) |
SAIF Estimated Insured Deposits |
| As reported in the Quarterly Banking Profile(a) |
Based on the Assessment Base Allocation Method(b) |
Difference |
| As of 12/31/1996(c) |
$1,782.2 (75%) |
$338.7 |
$300.1 |
$38.6 |
| As of 6/30/2004 |
$3,727.3 (60%) |
$679.4 |
$537.6 |
$141.8 |
Source: DOF Analysis.
a Based on the premise that 100 percent of secondary deposits were insured.
b Based on the premise that secondary deposits were insured up to the percentage of total insured deposits.
c DIR’s simulation analysis (see page 9) applied the new method starting in 1997. Table 4 presents information as of December 31, 1996
and June 30, 2004 to allow for a more comparative analysis between the assessment base allocation method and DIR’s simulation
analysis of the new method for allocating estimated insured deposits.
FDIC Efforts to Address Estimated Insured Oakar Deposit Differences
In October 2004, the FDIC established an interdivisional group to study the FDIC’s methodology for estimating and allocating insured deposits (Study). The Study Group was tasked with:
- evaluating all of the options for allocating the insured portion of the AADA in Oakar institutions for the purposes of estimating BIF- and SAIF insured deposits and for allocating costs in the event of a failure of an Oakar institution;
- evaluating the merits of using total deposits in the denominator of the fund reserve ratio calculations instead of insured deposits; and
- making significant progress on the Study by February 2005 to ensure that the Chairman, Chief Operating Officer (COO), and Chief Financial Officer (CFO) could reference the Study in the annual management representations that they jointly made to the Government Accountability Office (GAO) as part of the FDIC’s financial statement audit.
Further, the CFO defined a project structure, milestones for project completion, and a Project Steering Committee composed of senior FDIC executive managers to oversee the project.
After the Study Group considered several alternatives for addressing the Oakar allocation issue, the Study Group issued an internal study report, Allocating Estimated Insured Deposits of Oakar Institutions Between the Insurance Funds (Oakar Study Report),[ 5 ] which concluded that unanticipated deposit trends at Oakar institutions made it appropriate to consider revising the estimated insured deposit allocation method for the reserve ratios.
The Oakar Study Report concluded that, over time, Oakar institutions began relying more heavily on uninsured deposit financing, and Oakar institutions’ total deposits (which included insured and uninsured deposits) grew at a greater rate than estimated insured deposits. Specifically, the Oakar Study Report stated that from 2000 through 2004, estimated insured deposits at the 10 largest BIF member Oakar institutions declined from 70 percent to 59 percent, and from the end of 1996 through 2003, BIF members’ total deposits grew at a compound annual rate of 6.2 percent compared to only 3.8 percent growth for estimated insured deposits. These uninsured deposit trends were not consistent with FDIC’s assumption that 100 percent of Oakar deposits were insured.
The Oakar Study Report recommended a new method for allocating estimated insured deposits at Oakar institutions that the Study Group determined was more representative of actual Oakar insured deposit trends. The FDIC elected to apply the new method prospectively beginning in the fourth quarter 2004. The new method differed from the existing method as follows.
- For future Oakar transactions, the insured share of Oakar deposits would correspond to the acquired institution’s ratio of insured-to-total deposits.[ 6 ]
- For existing and new Oakar institutions, the estimated insured growth rate of Oakar deposits would grow at the institution’s estimated insured deposit growth rate.
- Prospective application of the new method meant that, for existing Oakar institutions, at the start date of implementation (i.e., the start of the last quarter of 2004), insured Oakar deposits would equal total Oakar deposits.
The Oakar Study Report also included an analysis simulating how estimated insured deposits would have been allocated had the new method been in effect since the beginning of 1997. The Oakar Study concluded that by September 2004, the new method would have resulted in just over $90 billion more in estimated insured deposits allocated to BIF and correspondingly fewer estimated insured deposits allocated to SAIF compared to the existing method. The Oakar Study Report concluded that much of the difference appeared to have been concentrated in the past 2 to 3 years when total deposit growth was noticeably higher than insured deposit growth.
DIR and Legal Division representatives indicated that the CFO; COO; General Counsel; and Director, DIR, briefed the FDIC Chairman on the Oakar estimated insured deposit allocation issue, the results of the Study, and the proposed new method for allocating estimated insured deposits. DIR and Legal Division representatives also indicated that they briefed some information to the deputies to the FDIC Board members before publicly announcing the change in methodology. In February 2005, the FDIC formally adopted the new method. The FDIC included a disclosure discussing the change in methodology in the fourth quarter 2004, Quarterly Banking Profile, which was issued in late February 2005. Finally, the CFO briefed GAO on the Oakar issue in April 2005 and provided GAO a copy of the Oakar Study. We discuss the FDIC’s efforts to communicate the Oakar estimated insured deposit allocation issue later in this report.
Impact of New Allocation Method on Fund Reserve Ratios
We concluded that the nature, timing, and application of the new, more representative method could have had a significant impact on the reserve ratios. At our request, in August 2005, DIR recalculated its simulation analysis on a quarter-by-quarter basis. DIR also revised its analysis to conform with another DIR change in assumptions regarding branch purchases [ 7 ] and to reflect other technical changes to be consistent with the new method. These changes increased the cumulative net effect of not applying the new, more representative method between 1997 and 2004 from $90 billion to $96 billion. Figure 1 below presents the results of DIR’s revised analysis by semiannual period.
Figure 1: DIR Analysis of Applying the New Method Historically | |
 | |
| Source: DIR. | [ D ] |
We recalculated BIF and SAIF reserve ratios using DIR’s simulation analysis and concluded that the new more representative method would have caused the BIF to drop below the 1.25 percent DRR for a 6 quarter period starting in late 2001 and again in 2005. This was significant because the FDI Act generally required the Board to take action to return the reserve ratio to 1.25 percent within 1 year. We also determined that the new method would have increased the SAIF reserve ratio to as high as 1.53 percent in September 2003. Figure 2 presents our analysis of BIF and SAIF reserve ratios.
| Figure 2: OIG Analysis of Reserve Ratios | |
 | |
| Source: OIG Analysis of reserve ratio information and simulation analysis from DIR. | [ D ] |
| Note: For presentation purposes, Figure 2 presents SAIF and BIF reserve ratios for the period June 2001 through December 2005. However, we analyzed the simulation analysis impact on the reserve ratios for the period 1997 through June 2005. We determined that neither the reported reserve ratios nor the OIG recalculation of the reserve ratios dropped below the 1.25 percent DRR during the period 1997 through June 2001. | |
The outdated assumption may have also affected the allocation of FDIC administrative expenses between the two funds.[ 8 ] We did not attempt to quantify these amounts. However, it appears that application of the outdated assumption did not have an impact on the following processes or factors.
Actual Assessments Due From Oakar Institutions: Actual assessments due were based on total domestic deposits. Thus, the Oakar estimated insured deposit allocation issue would not have affected an Oakar institution’s assessment base, which the FDIC used to determine the assessment amount.
Allocation of Losses in the Event of an Oakar Failure: In the event of an institution failure, FDIC based liquidation of the failed institution on actual insured deposits. BIF and SAIF then shared any loss in proportion to assessable deposit shares (or the assessment base, which was based on total deposits). Again, the Oakar estimated insured deposit allocation issue would not have had an impact on the allocation of losses from an institution failure.
Credits or Dividends Due to SAIF Members: In certain cases, some financial institutions could be due credits or dividends for prior period assessments paid to an insurance fund. According to DIR and the Legal Division, no “1-A” Category SAIF members[ 9 ] had paid assessments to SAIF since 1997, thus no credits related to that period would have been due to SAIF institutions. Moreover, SAIF, unlike BIF, did not have a ceiling on the reserve ratio that could trigger credits or dividends to members.
We concluded that, while the new, more representative method will stem the growth of the disparity between the existing and new methods for allocating Oakar estimated insured deposits, prospective application of the new method will not resolve the cumulative effect ($96 billion) of not applying the new method earlier. In February 2006, Congress passed deposit insurance reform measures that included merging the BIF and SAIF insurance funds and eliminated the need for the FDIC to allocate Oakar estimated insured deposits between the funds.
DIR Process for Periodically Reviewing Key Assumptions and Estimates
In August 2004, DIR, DOF, and the Division of Information Technology, signed a memorandum of understanding (MOU), Allocating Estimated Insured Deposits between BIF and SAIF According to the AADA held by Oakar Institutions. The three divisions agreed to work more closely each quarter in order to ensure a more reliable process and data integrity in allocating estimated insured deposits and producing the FDIC’s calculations of BIF and SAIF reserve ratios. DIR and DOF also agreed to continue to examine the methods by which insured, domestic, and assessable deposits were calculated to ensure that these methods were applied consistently. We agree that the MOU should enhance cooperation between DIR and DOF and will help to ensure reliable data for the reserve ratio calculation.
However, we concluded that DIR could benefit from a more formal process (1) for periodically reviewing key assumptions and estimates related to the insurance funds and reserve ratio calculations, particularly assumptions related to estimated insured deposits, and (2) for documenting the results of such reviews, including the alternatives considered, the impact of alternatives on the insurance funds and reserve ratios, and the basis for decisions made to change assumptions or estimates. Without a periodic process that is clearly documented, DIR may not be able to make timely adjustments or revisions to assumptions and estimates, and the basis for DIR’s decisions may not be transparent.
Recommendation
We recommend that the Director, DIR:
- Establish policies and procedures for periodically validating key assumptions, estimates, or other components that factor into the calculation of the reserve ratio(s). Such procedures should address:
- a recommended schedule for validating key assumptions and estimates,
- discussion of disposition of alternatives considered but not accepted,
- documentation requirements to support decisions to change key assumptions or estimates,
- reporting the results of the validation process.
Corporation Comments and OIG Evaluation
The Director, DIR, provided a written response, dated April 13, 2006, for recommendation 1. The response is presented in its entirety in Appendix II. The Director, DIR, concurred with the recommendation and agreed to develop policies and procedures that will require DIR managers to periodically identify, validate, and approve key assumptions and estimates that support the calculation of the reserve ratio. The policies and procedures will require analysis of industry-wide issues and trends that could affect current or future estimates, as well as other factors such as economic trends, changes in institution accounting practices, etc. The policies and procedures will also focus on issues that affect the integrity of the application data that support reserve ratio calculations. For example, DIR will periodically validate the integrity of bank data systems, and discuss and approve the process and assumptions used to generate the data.
The policies and procedures will also include a schedule of periodic meetings to discuss concerns. Managers and staff will adequately document material discussions, decisions, testing, and validation results, including but not limited to, discussions of alternatives considered but not accepted, decisions to accept or change key assumptions and estimates, and the communication of recommendations and decisions to appropriate DIR management.
The Director, DIR, agreed to draft and submit policies and procedures to the Board for review by September 30, 2006. DIR’s planned actions are responsive to the recommendation, and we consider recommendation 1 resolved. However, the recommendation will remain open until we have determined that agreed to corrective actions have been completed and are effective.
In its response, DIR noted that there is no single correct method to estimate and allocate insured deposits and there is no scientific way to determine at what precise point in time an existing method becomes insufficiently representative so as to require a change. We acknowledge that there is no single correct method for estimating and allocating insured deposits. Estimates are expected to be reasonable and involve objective and subjective judgment. As recommended, a periodic validation of key assumptions and estimates should help to ensure that the Corporation’s method remains sufficiently representative and reasonable.
DIR also contended that previous changes in methodology impacting estimated insured deposits have been adopted prospectively and not applied retroactively to earlier periods, and that they were therefore following precedent. DIR offered two examples wherein changes in estimating methodologies were not applied retroactively. First, it should be noted that we did not suggest retroactive application. Rather, we stated that the decision on whether, and how, to address the cumulative net effect (estimated at $96 billion) of applying an updated methodology only prospectively, and not restating deposit funds to reflect the new method, was one on which the Board should have had more information and involvement.
In addition, in both cited cases, the Corporation elected to apply the new methodologies to all deposits at the date the methodologies were changed, which had an immediate effect on estimated insured deposit amounts, in one case for $47 billion and in another case for $56 billion. Conversely, in this case, FDIC elected to apply the new methodology only to newly generated Oakar deposits and growth occurring in the last quarter of 2004 (i.e., prospective application), rather than to all existing Oakar deposits. Thus, application of the new Oakar methodology did not resolve most of the $96 billion cumulative effect of not applying the new, more representative, allocation method earlier. Nevertheless, we are not concluding on whether the FDIC selected the correct methodology for estimating insured deposits or whether the FDIC properly applied the change in methodology. As discussed in the next section of this report, in our view, selection and application of the current methodology was a matter warranting deliberation by the Board based on timely, accurate, and complete information.
Finally, DIR states that even were the FDIC to retroactively apply the revised methodology, it could not legally impose assessments on banks for prior periods. As noted above, we did not suggest retroactive application of the methodology. We also did not suggest that the FDIC impose any assessments, and therefore did not evaluate related legal authorities.
BOARD MEMBER INVOLVEMENT IN, AND REPORTING OF, THE OAKAR ESTIMATED INSURED DEPOSIT ALLOCATION ISSUE
The FDIC could have done more to inform its Board of Directors of the Oakar estimated insured deposit allocation issue. FDIC Legal Division representatives indicated that FDIC officers operated under appropriate delegations of authority in adopting the new method and that they had briefed the FDIC Chairman on the issue. We concluded that the FDIC should review its delegations of authority related to the assessments determination process to determine whether the delegations need to be clarified or modified. Study Group representatives also stated that they had provided a briefing on the Oakar issue to the deputies to the Board members (Deputies), but this briefing did not clearly discuss the Study results, various alternatives considered, or the dollar and reserve ratio impact of the allocation methods being considered. Moreover, the timing of the briefing did not provide sufficient opportunity for Board members to evaluate the Oakar issue or raise questions before the change was publicly announced. Steps should be taken to clarify Board expectations regarding the types of issues that should be provided to the Board for review and approval and to ensure that existing mechanisms for providing the Board current, accurate, and complete information—such as the Deputies’ briefings—are adequate. Such actions should help to ensure that Board members are informed of, and have an opportunity to be involved in, matters that could impact their constituent financial institutions.
FDIC Board of Directors
The FDIC is managed by a five person Board of Directors whose members include the Comptroller of the Currency (Comptroller) and the Director of the Office of Thrift Supervision (OTS). The Board has all of the powers specifically granted by the provisions of the FDI Act. With respect to assessments, the FDI Act required the Board to set semiannual assessments for insured depository institutions when necessary, and only to the extent necessary, to maintain the reserve ratio of each deposit insurance fund at the DRR or to increase the reserve ratio to the DRR if the reserve ratio was less than the DRR.[ 10 ] The FDI Act required the Board to review and weigh the following factors when setting assessments:
- expected operating expenses;
- case resolution expenditures and income;
- the effect of assessments on fund members’ earnings and capital; and
- any other factors that the Board may deem appropriate.
Finally, the FDI Act also required the Board to set semiannual assessments for members of each deposit insurance fund independently from semiannual assessments for members of any other deposit insurance fund and to set the DRR of each deposit insurance fund independently from the DRR of any other deposit insurance fund.
Delegations of Authority
The FDI Act does not address whether the Board may delegate its rate-setting authority to FDIC officials. Further, the FDI Act does not address how estimates or assumptions impacting the reserve ratio should be treated. We confirmed with the Legal Division that the Board’s ratemaking authority had not been delegated to FDIC officers. However, Legal Division representatives asserted that the Board had delegated authority for implementing the assessment function to FDIC officers. Therefore, the FDIC General Counsel had concluded that these officers could act within their delegated authority to adopt a revised methodology for estimating Oakar insured deposits and decide how to apply that methodology, for example, prospectively without reallocating BIF or SAIF estimated insured deposit amounts.
Section 5 of the Bylaws of the Federal Deposit Insurance Corporation (Bylaws) addresses the powers of the Board and states that the management of the Corporation shall be vested in the Board, which shall have all powers specifically granted by the provisions of the FDI Act and other laws of the United States. Section 5 also states that, within limitations of the law, the Board may delegate any of its specific or incidental powers to any standing or special committee of the Corporation or to any officer or agent of the Corporation.
FDIC Legal Division representatives identified two areas of the Bylaws that it asserted provided FDIC officers with the authority to adopt and implement the revised estimation methodology. Article VI of the Bylaws identifies the following specific powers and duties:
The Deputy to the Chairperson and Chief Financial Officer shall implement programs consistent with the Chief Financial Officers Act of 1990, including establishing and maintaining sound financial management systems, accounting systems, corporate budgeting procedures, and cash management systems… .
The Director of the Division of Finance shall…receive, deposit, disburse, manage, safely keep, and account for all funds of the Corporation, including those funds payable to it in connection with its functions assigned to the Director of the Division of Resolutions and Receiverships; maintain all accounting records of the Corporation; prepare financial statements and reports therefrom; and administer regulations of the Corporation governing the payment of assessments by insured depository institutions in accordance with the provisions of the Federal Deposit Insurance Act.
As stated, the CFO’s delegations focus on establishing and maintaining financial systems and processes, while the Director, DOF’s, delegations focus on administering assessment payment regulations. The Board did not delegate assessment determination or rate setting authority to FDIC officers. Moreover, the delegations do not specifically address how changes in assumptions or estimates that could have a significant effect on the reserve ratios, and in turn, the need for assessments, should be handled or communicated. We concluded that an independent review of the delegations should be conducted to ensure that they clearly define the Board’s intentions in this regard. To facilitate this independent review, we consider the deputies to the Board members to be in the best position to make recommendations to the Board on this matter, in consultation with the FDIC’s Legal Division.
We also noted that the Bylaws include the following delegations to the Director, DIR:
The Director of the Division of Insurance and Research shall be responsible…for identifying and assessing existing and emerging risks to the deposit insurance funds; provide advice and assistance to the Board of Directors and the Corporation’s various organizational units on economic and financial matters of importance to the Corporation and to the depository institutions industry; conduct basic research on current and emerging major problems in areas of specific interest to the Corporation; analyze policy alternatives and make recommendations thereon; monitor current economic and financial developments, problems, and issues… .
In our opinion, the delegations to the Director, DIR, establish an expectation that the Director should communicate and advise the Board on financial matters of importance to the Corporation and the banking industry. The role of the Director, DIR, should be considered in the overall review of delegations.
Finally, Legal Division representatives noted that several years ago, the Board established a Master Resolution[ 11 ] to ensure that significant or potentially controversial issues reached the Board’s attention. Specifically, the Master Resolution stated that “…the Board hereby reserves to itself consideration of matters which would establish or change existing Corporation policy, could attract unusual attention or publicity, or would involve an issue of first impression.” The Board rescinded the Master Resolution in June 2002 because it was “overly subjective, vague, and unwieldy and therefore incapable of proper interpretation…” Legal Division representatives contended that the Board’s rescission of this resolution signified that the Board trusted FDIC officers to use discretion in determining what matters should be presented to the Board. As discussed later, we are recommending that steps be taken to clarify Board expectations regarding the types of issues that should be provided to the Board for review and approval.
FDIC Efforts to Communicate Issues to the Board
The Study Group representatives noted that during the initial stages of the Study, the FDIC General Counsel advised that the authority to make the change in methodology was clearly vested in the FDIC CFO under Board approved delegations of authority. Nevertheless, these representatives indicated that the CFO; COO; General Counsel; and Director, DIR, briefed the Chairman on numerous occasions about the Study’s status and the Study Group’s recommendations for adopting a revised methodology, including whether it should be adopted retroactively or prospectively and the implications thereof. The representatives noted that the Chairman chose to dispose of the Oakar issue at the staff level, but gave his assent to staff to brief the deputies to the other Board members as a courtesy.
We concluded that internal discussion among senior FDIC officials involved in the Oakar estimated insured deposit allocation issue was adequate; however, communications of the issue with the FDIC Board and Deputies were limited. Following passage of the Sarbanes Oxley Act of 2002, the Board issued the Pocket Guide for Directors to provide directors of financial institutions practical guidance in meeting their duties and responsibilities. The guide discusses the need for a strong and independent board of directors at financial institutions. Excerpts from the guide follow.
Maintain independence: Effective corporate governance requires a high level of cooperation between an institution’s board and its management. Nevertheless, a director’s duty to oversee the conduct of the institution’s business necessitates that each director exercise independent judgment in evaluating management’s actions and competence.
Keep informed: Directors must keep themselves informed of the activities and condition of their institution and of the environment in which it operates. They should attend board and assigned committee meetings regularly, and should be careful to review closely all meeting materials, auditor’s findings and recommendations, and supervisory communications ... Directors should work with management to develop a program to keep members informed. Periodic briefings by management, counsel, auditors or other consultants are helpful… .
Monitor implementation: The board’s policies should establish mechanisms for providing the board the information needed to monitor the institution’s operations. In most cases, these mechanisms will include management reports to the board. These reports should be carefully framed to present information in a form meaningful to the board. … Reports should be provided far enough in advance of board meetings to allow for meaningful review. Management should be asked to respond to any questions raised by the reports.
These guidelines are equally applicable to the relationship between FDIC management and the FDIC Board. We identified two opportunities wherein the Corporation could have better communicated the Oakar issue to the Board.
Semiannual Rate Cases: The Director, DIR, presents a semiannual written case to the Board that recommends maintaining or changing the fund assessment rate schedule for the ensuing 6 month period. We reviewed the 2005 BIF and SAIF Semiannual Rate Cases that had been presented in November 2004 and May 2005 and determined that the Oakar issue and revised estimation methodology were not discussed in the case memoranda. We also confirmed with DIR representatives that the Oakar issue and revised estimation methodology had not been discussed at the Board meetings approving the BIF and SAIF rate cases. These cases routinely discuss a range of estimates for income and expense items that affect BIF and SAIF balances. Table 5 presents excerpts of items discussed in the November 15, 2004 rate cases.
Table 5: Examples of Items Discussed in the November 15, 2004 BIF and SAIF
Rate Cases (in millions)
| Items Discussed in the Rate Case |
BIF |
SAIF |
| Projected Income From Assessments for the Semiannual Period |
$84 |
$7 |
| Projected Income From Assessments for the Semiannual Period |
$1,626 |
$537 |
| Projected FDIC Operating Expenses Allocated to the Funds |
($848) |
($136) |
| Provision for Losses (related to failures, litigation, and other contingent losses) |
($19) |
($56) |
| Unrealized Gain (Loss) on Available for Sale Securities |
($295) |
($93) |
| Comprehensive Income (Loss) |
$548 |
$259 |
Source: OIG Review of 2005 BIF and SAIF Rate Cases.
At the time of this rate case, the projected BIF and SAIF reserve ratios stood at 1.31 percent and 1.34 percent, respectively. A shift of $96 billion in estimated insured deposits from SAIF to BIF would have decreased the BIF reserve ratio by 5 basis points[ 12 ] to 1.26 percent and increased the SAIF reserve ratio by 15 basis points to 1.49 percent. The largest income and expense item (interest income) discussed in the rate case had the effect of increasing the BIF and SAIF reserve ratios by 6 basis points for each fund. Thus, in our view, the Oakar issue could potentially have had a greater impact on the reserve ratios than most of the other income and expense items that had been discussed in the rate cases. Further, given the timing of the November 2004 rate case, we believe that it would have been appropriate for the Study Group to brief the Board about the ongoing Study and alternatives being considered for addressing the Oakar estimated insured deposit issue.
The Study Group representatives indicated that the Oakar issue was not included in the May 2005 rate cases because the FDIC considered the Oakar issue to be a settled, fully briefed, and agreed upon change and noted:
There was no controversy with respect to the change internally—including with the Board, whose Deputies had been briefed (discussed below).
The FDIC had already publicly disclosed the new methodology for calculating estimated insured deposits in March 2005 with the publication of the fourth quarter 2004 Quarterly Banking Profile.
The prospective adoption of the change as of the fourth quarter of 2004 meant that it had an immaterial effect on the estimated insured deposit amounts presented in the two 2005 semiannual rate cases.
February 2005 Deputies’ Briefing: Although the revised methodology was not included in a formal board case, FDIC officials had communicated some information about the Oakar estimated insured deposit issue to Board member representatives. Specifically, on February 23, 2005, representatives from DIR and the Legal Division provided a briefing on the Oakar issue to the Deputies for the FDIC Vice Chairman, OTS Director, and the Comptroller of the Currency.[ 13 ] The Deputy for the FDIC Director was unable to attend the briefing. This briefing is significant because it occurred 1 day before the revised methodology was publicly announced in the Quarterly Banking Profile.
We interviewed DIR and Legal representatives and Deputies who attended the briefing to understand the nature and content of the briefing and whether the briefing sufficiently communicated the dollar impact of the Oakar issue and the potential impact on the reserve ratios. A DIR representative who presented the briefing provided a 2-page briefing document that the DIR representative believed had been provided to the Deputies to explain the Oakar issue. This briefing document did not contain any discussion of the dollar or reserve ratio impact. Further, other meeting attendees had differing recollections regarding the issues discussed at the briefing and whether the dollar value of the DIR simulation analysis had been discussed.
During our audit, senior FDIC officials involved in the Oakar issue contended that the Bylaws require that if at least two Board members specifically request that a matter that can be decided at the FDIC staff and/or Chairman level be referred to the Board, then the matter is placed before the Board for a final determination. This process is referred to as the “two member rule.” Those FDIC officials noted that the February 2005 Deputies’ briefing provided the Deputies an opportunity to invoke the “two member rule,” but at the conclusion of the briefing, there were no unresolved questions, and none of the Deputies requested to have the matter sent forward to the Board for a vote. However, we noted that the FDIC officials’ contention did not reflect input from the FDIC Acting Chairman or the other Board members.
In addition, we noted that the Bylaws state that:
Special meetings of the Board of Directors may be called by the Chairperson or, upon the written request of any two members of the Board of Directors, by the Executive Secretary. Reasonable notice of any such special meeting shall be given to all members of the Board of Directors who can be contacted after a reasonable effort and in sufficient time to permit their attendance or participation.
By its terms, this provision references only the calling of a special meeting of the Board, not the broader interpretation suggested above. Regardless, in our view, the timing of the Deputies’ briefing and limited nature of the information provided to the Deputies made exercising the “two member rule” concerning calling special meetings impractical.
We concluded that the FDIC needs to ensure that it has adequate mechanisms in place for providing the Board timely, accurate, and complete information about corporate matters and activities. With recent departures at the Board and FDIC senior management level, it is an opportune time to revisit processes and mechanisms for keeping Board members informed, including the Deputies’ briefings and the “two member rule.”
Communication of the Change in Methodology to the Industry and Public
The FDIC could have done more to publicly communicate the Oakar issue and change in estimating methodology to the banking industry and the public. The FDIC publicly announced the change in methodology in the Quarterly Banking Profile. However, the announcement did not clearly describe the reasons for the change, disparity between the old and new allocation methods, or the Corporation’s rationale for not addressing the cumulative effect of the prior allocation method. As a result, the FDIC’s actions lacked transparency. The Oakar change in methodology announcement was also inconsistent with other public communications involving estimated insured deposits.
On February 24, 2005, the day following the Deputies’ briefing, the FDIC published the Quarterly Banking Profile and included the following language in a footnote to communicate the Oakar revised estimation methodology:
BIF member institutions may acquire SAIF insured deposits, resulting in institutions with some deposits covered by both insurance funds. Also, SAIF members may acquire BIF insured deposits. Therefore, the BIF member and the SAIF member tables each include deposits from both insurance funds. Prior to the fourth quarter of 2004, all SAIF deposits held by BIF-member institutions and all BIF deposits held by SAIF members (Adjusted Attributable Deposit Amounts, or AADAs) were treated as fully insured. Beginning in the fourth quarter of 2004, the insured portions of newly acquired AADAs are based on the estimated insured share of deposits at the acquired institution.
The language does not provide sufficient detail about the reasons for the change in methodology or the financial impact of the change in estimates on the deposit insurance funds or reserve ratios. Further, the Quarterly Banking Profile notification was not consistent with the FDIC’s past practices for communicating changes in estimates or restatements of reserve ratio amounts.
In August 2002, the FDIC issued a press release, FDIC Reports Second Quarter 2002 Financial Results for Bank and Thrift Insurance Funds. The press release noted that the BIF reserve ratio fell from 1.26 percent at December 31, 2001 to 1.24 percent at March 31, 2002 as a result of an increase in estimated insured deposits of $75 billion, or 3.1 percent of the total estimated insured BIF deposits. The press release attributed the increase in estimated insured deposits primarily to a reporting change in quarterly Call Reports that provide the source data for estimating insured deposits. This reporting change required banks to report the amount of uninsured deposits so that a better estimate of insured deposits could be calculated.
In July 2004, the FDIC issued a press release, FDIC Revises March 31, 2004, Reserve Ratios for Deposit Insurance Funds. The press release announced revisions to the March 2004 reserve ratios as a result of an FDIC error in allocating estimated deposits between BIF and SAIF. Specifically, the FDIC reported that it had overstated BIF estimated insured deposits by $20.5 billion and correspondingly understated SAIF estimated insured deposits. The press release indicated that the FDIC properly reallocated the amounts insured by each fund. The press release included information on the revised estimated insured deposit amounts and the impact of the reallocation on the BIF and SAIF reserve ratios. The press release also stated that the revisions would not trigger adjustments to premiums previously paid by insured institutions and that the FDIC would review its processes and make appropriate changes.
Both of these press releases involved issues of a lesser magnitude than the Oakar estimated insured deposit issue. For example, the August 2002 press release involved an increase in BIF-estimated insured deposits of 3.1 percent. The Oakar issue involved a potential increase in BIF estimated insured deposits of 3.7 percent and a corresponding potential decrease in SAIF estimated insured deposits of 10.2 percent. The July 2004 BIF overstatement was $20.5 billion as opposed to the Oakar SAIF potential decrease of $96 billion. In our view, the two press releases establish a threshold for materiality and an example of the type of information and how that information should be communicated to the public. Although we are not making a recommendation, the FDIC should consider our conclusions regarding this aspect of the FDIC’s handling of the Oakar issue as the Corporation addresses the findings and recommendations contained in this report.
Recommendations
We recommend that the Acting Deputy to the Acting Chairman, in coordination with the deputies to the other Board members:
Review the existing Corporate Bylaws, specifically, the powers and duties delegated to the CFO and the Directors, DOF and DIR, to ensure that those delegations reflect the Board members’ intent and expectations with regard to the deposit insurance fund reserve ratio and assessment determination processes, and make recommendations deemed appropriate to the Board.
Work with FDIC management to evaluate procedures and practices for keeping Board members informed of Corporation matters and activities, and make recommendations deemed appropriate to the Board. Consideration should be given to:
the FDIC Board members’ expectations regarding communication and coordination with FDIC management, including the types of issues suitable for Board review and approval; and
the adequacy of existing mechanisms for providing the board current, accurate and complete information needed to monitor FDIC operations, such as the Deputies briefings and other informational reports and briefings.
Corporation Comments and OIG Evaluation
The Acting Deputy to the Acting Chairman provided a written response dated April 10, 2006, for recommendations 2 and 3. The response is presented in its entirety in Appendix II. The Acting Deputy concurred with both recommendations and agreed to form a working committee composed of deputies to the Board members and FDIC representatives to:
review the existing Bylaws, including a review of the powers and duties delegated to FDIC senior officials to ensure that those delegations reflect the Board members’ intent and expectations with regard to the deposit insurance fund reserve ratio and assessment determination processes and make recommendations concerning delegations of authority deemed appropriate to the Board, and
work with FDIC management to evaluate procedures and practices for keeping Board members informed of Corporation matters and activities and communicating matters of importance to the public. The working committee will prepare recommendations to the Board, giving consideration to Board members’ expectations regarding communication and coordination with FDIC management, including the types of issues suitable for Board review and approval, and the adequacy of existing mechanisms for providing the Board current, accurate and complete information needed to monitor FDIC operations, such as the Deputies briefings and other informational reports and briefings.
The Acting Deputy’s response anticipated that the working committee would draft and submit its recommendations to the Board for review by September 30, 2006.
FDIC’s planned actions are responsive to the recommendations, and we consider recommendations 2 and 3 resolved. However, these recommendations will remain open until we have determined that agreed to corrective actions have been completed and are effective. Appendix III presents a summary of the DIR Director’s and the Acting Deputy to the Acting Chairman’s responses to our recommendations.
