May 9, 2005

Honorable Richard C. Shelby
Chairman
Committee on Banking, Housing, and Urban Affairs
United States Senate
Washington, DC 20515

Dear Mr. Chairman:

In response to your request, the Congressional Budget Office (CBO) has prepared the attached analysis, which discusses the likely consequences of alternatives for modifying deposit insurance. In keeping with CBO's mandate to provide objective, impartial analysis, the document contains no recommendations.

If you have any questions about the analysis, please contact me or Judith Ruud.

Sincerely,

Douglas Holtz-Eakin
 

Attachment
 

cc:	Honorable Paul Sarbanes Ranking Member

 
 

Summary and Introduction

Currently, the Congress is considering legislation that would modify federal deposit insurance. Many of the proposals include increasing insurance coverage for various types of accounts. Others involve changing the premium structure, shifting the allocation of premiums among insured institutions, and modifying the link between premiums and the minimum required reserve ratio for the deposit insurance funds that are administered by the Federal Deposit Insurance Corporation (FDIC).

Increasing coverage would provide modest benefits to depositors (more than 98 percent of all accounts are below the current limit of $100,000). Higher coverage would also expand the government's liability for insured deposits and increase its expected payments for failed insured depositories. Those costs, which would probably be passed on to the banking system in the form of higher premiums and then to its customers, need to be weighed against the benefits from increases in coverage.

Advocates cite at least four reasons to increase deposit insurance coverage.⁽¹⁾ First, the coverage limit has not been increased since 1980, and inflation has eroded the value of coverage. Second, increasing the coverage limit would afford greater convenience to depositors who now divide their money into multiple accounts to get full insurance coverage of deposits in excess of $100,000. Third, boosting deposit insurance coverage could serve to attract more deposits to all FDIC-insured institutions. To the extent that small banks depend on deposits more than large banks do, increased deposit insurance could improve small banks' competitiveness. Fourth, raising the coverage limit for municipal deposits would provide more secure protection than currently occurs through the collateralization often required or by state municipal deposit insurance programs.

Opponents point out that because almost all domestic accounts are below the current limit, they are fully covered. In addition, the real value of coverage is about twice as high as it was during the 1934-1969 period and has proved sufficient to achieve the main objective of deposit insurance--preventing bank panics and disruptions to the financial markets. Moreover, because depositors who want more coverage can obtain it by dividing their funds into multiple accounts, the current ceiling is rarely binding. Finally, higher coverage for municipal deposits may be redundant because of the requirements for collateralization.

The incremental benefits and costs of the proposed increases in coverage are likely to be small relative to the total benefits and costs of the entire deposit insurance system. But increased coverage takes on added economic significance because deposit insurance appears to be underpriced for most institutions. Currently, 93 percent of FDIC-insured institutions, which hold 98 percent of insured deposits, pay nothing for deposit insurance.⁽²⁾

The Deposit Insurance Funds Act of 1996 stipulated that institutions in the best risk category get deposit insurance for free when the insurance fund reserves are above the designated reserve ratio (DRR) of 1.25 percent of insured deposits. As a result, those FDIC-insured institutions have not paid any deposit insurance premiums since 1997, even though they have received the benefits of insurance and some pose risks to the government.

In addition to setting premiums for most banks at zero for most of the time, linking premiums to the DRR increases their volatility by causing premiums to be procyclical (that is, high in bad times, low in good times). It also requires some losses to be financed after they have been incurred rather than funded prospectively.⁽³⁾ In a 2001 report on the structural deficiencies of the deposit insurance system, the FDIC identified the need to price insurance to reflect risk and to avoid raising premiums during bad economic times.⁽⁴⁾

The proposals currently being considered for increasing coverage and otherwise changing the federal deposit insurance system differ significantly in their effectiveness in addressing the weaknesses of the current system. This paper examines some of the policy alternatives.
 

The Objectives and Operation of Federal Deposit Insurance

Federal deposit insurance protects the payment system from disruption. That is, because most deposits are guaranteed by the federal government, depositors need not rush to the bank to withdraw funds if the bank's capacity to cover them is in question. Deposit insurance thus prevents the financial instability associated with runs on banks that can bring down even sound institutions, but it has costs. By reducing the market discipline on banks that results from the potential losses of uninsured depositors, insurance can weaken the incentives for banks to avoid risks, distort the efficient allocation of credit and investment, and increase the FDIC's and taxpayers' exposure to potential losses.⁽⁵⁾

Deposit insurance has two sources of financing: premiums paid by insured institutions and general tax revenues. Under current law, the costs of deposit insurance are paid from insurance premiums. Taxpayer funds are a backup source of financing for extraordinary losses such as those that occurred in the 1980s with the failure of the Federal Savings and Loan Insurance Corporation.

A key factor in the pricing of deposit insurance is the designated reserve ratio, which was established in the Financial Institutions Reform, Recovery, and Enforcement Act of 1989. The Federal Deposit Insurance Corporation Improvement Act of 1991 requires the FDIC to set assessment rates sufficient to restore its deposit insurance funds' reserve ratios to the DRR within a year if they fall below that level.

The DRR, originally a floor, also became, in effect, a cap under the Deposit Insurance Funds Act of 1996, when it prohibited the FDIC from collecting premiums from insured institutions in the best risk category as long as the reserve ratios exceeded the DRR. Because that category includes most banks during good economic times, that provision has exempted the vast majority of depositories from paying premiums.

The DRR of 1.25 percent of insured deposits is well below the historical average reserve ratio of the Bank Insurance Fund (the older and larger of the FDIC's two funds). From 1934 to 1969, over which the real (inflation-adjusted) level of coverage remained relatively constant, the average reserve ratio of the fund was 1.5 percent of insured deposits. Prior to the establishment of a minimum reserve, the ratio averaged 1.38 percent (see Figure 1).
 

Figure 1.

The Bank Insurance Fund Reserve Ratio, 1934 to 2004, and the Designated Reserve Ratio

(Percentage of insured deposits)

Source: Congressional Budget Office based on Federal Deposit Insurance Corporation, 2004 Annual Report (Washington, D.C.: Federal Deposit Insurance Corporation, February 14, 2005).

 

Assessing the Adequacy of Coverage

Lawmakers face a trade-off in determining the appropriate level of deposit insurance coverage: setting the ceiling high enough to protect most depositors and thus prevent bank runs but low enough to limit the distortion of banks' investment decisions and federal risks and costs. The insurance gives depositors in covered accounts the opportunity to hold liquid government-guaranteed assets in small denominations. At the same time, it increases the likelihood that banks--especially those experiencing financial difficulties--will use deposits to finance risky positions that they would not otherwise take because some of the losses would be borne by the government. Indeed, the FDIC has found that the increase in deposit insurance coverage from $40,000 to $100,000 in 1980 contributed to the losses suffered by the deposit insurance program.

The amount of insurance that a depositor can secure is not effectively limited to the current $100,000 ceiling. A depositor can obtain additional coverage at some additional cost by maintaining several accounts and by maintaining accounts with different types of legal ownership, such as individual, joint, and individual retirement accounts (IRAs). For example, a couple could have $600,000 in fully insured deposits at a single institution--two individual accounts of $100,000 each, two IRAs of $100,000 each, and one joint account of $200,000. Financial firms that own more than one bank sometimes offer their customers the service of dividing large deposits into multiple accounts of $100,000 so customers do not have to do it themselves. The Certificate of Deposit Account Registry Service, a new financial service created by Promontory Interfinancial Network, enables banks to offer customers FDIC coverage for multimillion-dollar deposits.

Core Coverage

As of June 2004, 98.3 percent of all domestic deposit accounts (including IRAs, Keogh accounts, and the accounts of businesses and state and local governments) in FDIC-insured depository institutions held $100,000 or less and were therefore fully covered. The typical household's deposit account was far below that limit. According to results from the Federal Reserve's triennial Survey of Consumer Finances, in 2001 the median balance in household checking accounts was about $4,000 and in certificate of deposit (CD) accounts, about $15,000.⁽⁶⁾ Even for families in the highest income bracket (those in the top 10 percent of the survey, with average annual income of over $300,000), the median checking account balance was $26,000, and the median CD account balance was $25,000. Thus, the current coverage level of deposit insurance is not binding for most households.⁽⁷⁾

Adjusted for inflation, the initial coverage limit of $5,000 in 1934 is now equivalent to about $60,000 (see Figure 2). The current limit of $100,000 that was established in 1980 would need to be roughly doubled to adjust for inflation since then.
 

Figure 2.

The Inflation-Adjusted Value of Deposit Insurance Coverage, 1934 to 2004

(2000 dollars)

Source: Congressional Budget Office based on data from the Federal Deposit Insurance Corporation and the Department of Commerce, Bureau of Economic Analysis.

Special Account Coverage

Depositors of amounts above the coverage limit have incentives to monitor banks' financial condition and operations and to withdraw their funds if they deem the banks to be too risky. So raising deposit insurance coverage would reduce the incentives for partially insured depositors--in particular, depositors of municipal funds--to constrain risk taking by insured institutions.

Further, deposits over the $100,000 coverage limit held by most local governments have alternative protection. Most states have municipal deposit insurance programs or require banks to pledge assets to ensure payment of any municipal deposits above the limit. However, federal deposit insurance provides more secure protection than that collateralization. When a bank fails because of fraud, the collateral intended to back municipal deposits may be missing. With federal deposit insurance, no risk of loss exists for insured deposits.⁽⁸⁾
 

Setting Premiums

Setting the appropriate premiums is a difficult, yet crucial, task of the deposit insurance system. For the period from 1934 to 1989, U.S. deposit insurance rates were based nominally on historical loss experience, even though effective rates were often reduced by rebates during periods of low losses. Specifically, during that period, the FDIC maintained a scheduled rate of 8.33 basis points (0.0833 percent), but beginning in 1950, it began providing rebates of premiums to insured banks. However, as the losses of the 1980s emerged, the agency reduced and ended the rebates.⁽⁹⁾ Over the entire period, the rate of 8.33 basis points would have been close to sufficient to cover costs (see Figure 3).⁽¹⁰⁾
 

Figure 3.

The Historical Actual Premium Rate and Required Overall Premium Rates, 1934 to 2004

(Basis points)

Source: Congressional Budget Office based on Federal Deposit Insurance Corporation, 2004 Annual Report (Washington, D.C.: Federal Deposit Insurance Corporation, February 14, 2005), Appendix A, and Options Paper (August 2000), Table 3.

Notes: A basis point is one-hundredth of a percentage point.

The assessment base for this figure is total domestic deposits.

The Congress responded to the failure of the Federal Savings and Loan Insurance Fund and the negative balance of the Bank Insurance Fund with legislation authorizing the FDIC to charge variable risk-based premiums. The initial premiums ranged from 4 cents to 27 cents per $100 of deposits, depending on the estimated risk posed by the bank. However, the linking of deposit insurance premiums with the designated reserve ratio weakened the connection between premium rates and historical losses. As a consequence, deposit insurance rates are now zero for most banks. So to the extent that historical costs are a reasonable guide to future costs, premium rates are likely to rise more in the future to pay for losses than they would have risen prior to the establishment of the DRR.

There is widespread agreement that volatile deposit insurance premium rates are undesirable. To smooth out premiums, designating a minimum annual premium rate, rather than a reserve ratio, is likely to be more efficient.⁽¹¹⁾ In particular, an aggregate premium rate based on the long-run average loss rate could stabilize premium rates and approximately cover losses prospectively.⁽¹²⁾ Annual premiums could be apportioned toward higher-risk banks, but all banks could pay something for deposit insurance.
 

Policy Alternatives

The Congress is considering a number of changes to federal deposit insurance, including proposals to increase the real level of insurance coverage, charge all banks and savings associations premiums for deposit insurance, authorize credits that some banks could use to pay premiums, change the designated reserve ratio, and specify a required level of premiums.⁽¹³⁾

While it may not be possible to assess every possible combination of features, each component can be evaluated on a stand-alone basis against a limited number of criteria. Relevant measures include the expected effect on the stability and prospective basis of premium rates, on the risk to the government and taxpayers, and on the budget deficit. Policies that would increase the stability of premium rates, set premiums that prospectively reflect differences in risk, lower costs to the FDIC and risk to taxpayers, and reduce the deficit appear to be superior to the alternatives.

Increasing Insurance Coverage

To justify increasing insurance coverage, advocates point to the erosion of the real value of coverage by inflation, greater convenience to depositors who now divide their money into multiple accounts, the possibility of attracting more deposits to all banks and thrifts, and more secure protection for municipal deposits. Opponents point out that increasing the coverage ceiling would expand the FDIC's liability for insured deposits in the event of bank failures, weaken market discipline on banks' investment decisions, and raise net costs to the FDIC and the risk to taxpayers.

Under current law, if higher coverage levels increased insured deposits sufficiently to reduce the reserve ratio below the DRR, the FDIC would be required to raise premium rates to restore the deposit insurance funds to the required level. Thus, in the near term, increasing coverage would probably increase collections. The effect would be short-lived, though. Once the DRR was met, premiums would fall to zero once again for most banks.

For the same reasons, an increase in coverage would not bring the deposit insurance system closer to a pricing structure that is prospective and actuarially sound. It could also necessitate a sharper increase in premiums during periods of losses than would have been required otherwise. At least some of the higher costs of increased coverage would be passed on to bank customers in the form of lower rates of return or higher fees on deposits.

Charging All Banks and Savings Associations Premiums for Deposit Insurance

Currently, 93 percent of FDIC-insured depositories (which hold 98 percent of insured deposits) pay nothing for deposit insurance. Charging those institutions some amount for deposit insurance would increase the FDIC's premium collections. Such a change would move in the direction of fair prospective pricing, thus reducing expected losses to the FDIC and to taxpayers. Clearly, proposals that increased premiums by a specified amount without raising insurance coverage would have a larger net effect on reducing the budget deficit than ones that also increased coverage. For example, the Congressional Budget Office has estimated that charging those institutions that currently pay nothing a premium rate of 2 basis points would increase federal premium receipts by more than $4 billion over the 2006-2010 period.⁽¹⁴⁾ Such a change in pricing would also produce more stable premium rates over long periods.

Awarding Premium Credits

Awarding premium credits based on the size of the deposit insurance funds has been proposed in conjunction with charging all institutions for deposit insurance as a way to address the "free rider" problem that results from charging nothing for deposit insurance. About 1,000 FDIC-insured institutions have been chartered since 1996 that have never paid any deposit insurance premiums, because they were in the best risk category. If institutions in that category were required to pay something for deposit insurance, awarding credits to those that existed before 1996 would aim to redress the problem.

Awarding credits for institutions to use to pay deposit insurance premiums in lieu of cash would reduce the government's collections. In general, the awarding of credits would tend to make premiums less prospective. If an institution that was using credits to pay deposit insurance premiums failed, the associated costs of deposit insurance claims would be borne by the surviving banks. However, credits could be used to mitigate the procyclical collection of premiums, if they were restricted to use in times of high losses.

Changing the Designated Reserve Ratio

Some proposals would allow the reserve ratio to vary within a range as a way to avoid sharp swings in premiums. However, specifying a range that extended lower than the current DRR would be effectively the same as reducing it, and doing so would not avoid large swings in premiums but rather postpone them until the fund was at a lower level.

Reducing the designated reserve ratio would lead to a reduction in premiums without reducing costs. It would, therefore, leave unaddressed the underpricing of insurance coverage and increase costs to the FDIC and the risk to taxpayers. It would also produce a sharper increase in premiums during periods with high levels of failures and insured losses, because it would reduce the extent to which premiums were collected for future losses.

Designating a Required Level of Premiums Rather Than a Reserve Ratio

Designating a minimum level of annual premiums that would be sufficient to cover expected average annual costs over the long run would improve the stability of premiums compared with that under the current policy of designating a required reserve ratio.
 

Conclusion

Policymakers need to weigh the benefits of potential changes in the deposit insurance system against their costs. Raising coverage levels for various types of larger accounts (municipal accounts and retirement accounts) would increase the safety of those deposits but could further weaken market discipline and increase the costs of federal deposit insurance, particularly in the event of large widespread failures, as occurred in the 1980s and 1990s.

In addition, the current pricing policy, based on targeting a designated reserve ratio, prevents pricing premiums so that they are prospective and actuarially fair and causes premiums to be volatile and procyclical. Addressing those problems may require revisiting the mechanism of the designated reserve ratio and, in setting premiums, emphasizing controlling and financing the long-run costs of deposit insurance.