A reexamination of the costs and benefits of Federal deposit insurance.
FEDERAL DEPOSIT insurance in the U.S. was given birth in the midst of the Great Depression, with the creation of the Federal Deposit Insurance Corporation (FDIC) in the 1933 Glass-Steagall Banking Act. This federal government program was intended to accomplish two primary aims: (1) protect the "small" depositor from loss of wealth due to bank insolvency problems; and (2) prevent the contagious nature of isolated bank failures from turning into large scale bank runs, thus protecting the payments mechanism in the country. There is little doubt that these two goals have been achieved under our federal deposit insurance system. Since 1993, no federally insured depositor has lost funds as a result of a bank failure, and the only bank runs of any consequence in the U.S. have involved state insurance programs.
In light of such success, it is somewhat remarkable to see the extent of discussion both criticizing and calling for substantial reform in our deposit insurance system, especially in the academic community. The purpose of this paper is to provide an overall review of deposit insurance from the perspective of its costs and benefits.
The general cost/benefit framework developed in public finance provides a useful tool for the analysis of deposit insurance, because it requires a detailed analysis of the somewhat forgotten side of the equation -- the costs associated with the federal deposit insurance system. The general cost/benefit approach is also useful because it suggests a serious reexamination of the benefits of deposit insurance.
THE COSTS OF DEPOSIT INSURANCE
The financial failure of the Federal Saving and Loan Insurance Corporation (FSLIC), the resulting tax burden imposed on the American taxpayer, and the more recent insolvency of the Federal Deposit Insurance Corporation (FDIC) clearly represent explicit costs of the federal deposit insurance system. It was generally believed during the establishment of federal deposit insurance in the early 1930s that the program would be financially self-sufficient. Premiums paid by the depository institutions were to cover the handling of individual depository institution failure. For example, Representative Heurn Steagall (D-AL), May 1933, who was a co-sponsor of federal deposit insurance, stated in Congressional discussion, "I do not mean to be understood as favoring government guaranty of bank deposits. I do not. I have never favored such a plan ... Bankers should insure their own deposits."
This presumption concerning self-sufficiency was obviously incorrect. The American public has been called upon to supplement funds available to one federal deposit insurance agency (the old FSLIC) and to lend funds to the other (the FDIC). The Congressional Budget Office presently estimates that the FSLIC bailout will cost U.S. taxpayers about $215 billion in 1990 dollars. Taxpayers are being forced to pay for the savings and loan failure because the FSLIC insurance fund was insufficient to cover all insured deposits for the savings and loans, as their assets fell markedly in value relative to their liabilities. The fund was simply insufficient to serve it's purpose. Congress and the Bush Administration decided the American public would be better served by rescuing the fund, rather than dealing with the problems of losses to insured depositors.
Unfortunately, public discussion of deposit insurance in the U.S. has not proceeded much beyond the acknowledgement of such explicit costs, which generally have been understated by current practices. The public has been provided little in the way of explanation as to why the federal deposit insurance agencies have reached the point that they could no longer handle future failures without a public injection of funds. As such, the public has no way of gauging the likelihood of similar problems in the future. This is quite unfortunate, as academics have been warning that such losses are due in large part to the present structure of deposit insurance in the U.S., itself. Without substantial reform, then, these analysts argue such costs can be expected to reoccur at a future date.
It is definitionally true that the insolvency problems of the federal deposit insurance funds stem directly from either (or both) a large number of insured depository institution failures or failure of large institutions with a great difference between the value of their assets and liabilities. In the U.S., we have recently seen evidence of both problems. One would generally expect such losses to occur in the most difficult of economic times. Yet, it is quite difficult to document that the past few years have been substantially worse than earlier recessionary periods.
Most academics have turned their attention away from general macroeconomic conditions for explanations of the extent of recent depository institution failures. The significant losses of the deposit funds are not being blamed on tough economic times. Rather, many financial economists today point to the existence of deposit insurance as being responsible for the significance of depository institution losses. For example, some analysts point to the significant decline in bank capital as a percent of deposits following the establishment of the federal deposit insurance system. Carlstrom (1989), for example, argues that "Deposit insurance ... reduces the amount of capital a bank chooses to maintain." He then goes on to confirm such a decline in the U.S. In the 1930s the capital to asset ratio was about 15 percent. Today, this ratio is approximately 7 percent, more than a 50 percent decline.
Bank capital acts as a cushion against declining values of assets. If depository institutions have larger levels of capital, their financial losses will be less of a drain on the deposit insurance fund for a given economic shock. The deposit insurance funds and other regulations agency recognize this and indeed did (and continue to) impose capital requirements.(1)
However, with deposit insurance one of the incentives for a depository institution to acquire and hold more capital is substantially mitigated. In the absence of deposit insurance, the management of the institution has an important incentive to acquire more capital, as the institution with greater capital will be considered safer, or less likely to fail, by depositors. With deposit insurance, management has no similar incentive to signal to depositors its character concerning the potential of failure. Thus, it is no wonder that capital-to-asset ratios have declined in the U.S. following the establishment of deposit insurance.
Peltzman (1984) further suggests that not only is the existence of deposit insurance important in understanding the declining capital-to-asset ratios, but the extension of de facto insurance coverage to non-insured depositors has further exacerbated the problem. Peltzman, looking at more recent times, documents a significant decline in capital-to-asset ratios for the largest banks in the country between 1967 and 1980. He attributes this decline to the FDIC practice, established in the 1970s, of preventing substantial losses to uninsured deposits for very largest U.S. banks. Of course, this practice was further formalized in the 1980s with the establishment of the "Too-Big-to-Fail" doctrine (see Moyer in this issue, for further discussion).
In summary, capital is less of a protection against failure today then it once was. Depository institutions hold less capital (for a given size) today than they did earlier for at least two reasons. First, the existence of deposit insurance reduces depository institution's incentives to hold capital, as insured depositors are unconcerned about capital positions. Second, the de facto extension of deposit insurance coverage for uninsured deposits of the larger depository institutions even further reduces the incentives, as now even uninsured depositors are unconcerned about capital positions. Thus, we have the ironic result that the size of recent depository institutions losses today are likely to have been larger than those that would have resulted from an environment without deposit insurance.
Such adverse effects could be mitigated, of course, by increases in capital requirements for depository institutions. However, capital cannot be obtained without cost. Investors are not going to invest more capital in depository institutions without the prospect of increased future profits. These profits, however, are likely to be constrained by increased competition, as the financial world becomes more internationalized. Thus, increasing the capital position of U.S. depository institutions should not be viewed simply as something that can easily be legislated. Such legislation may indeed lead to the elimination of many depository institution in this country relative to the international community.
The Pricing Scheme
Academics have long emphasized the existence of the moral hazard problem as it relates to deposit insurance. Moral hazard simply refers to the fact that the existence of insurance may induce the insuree to take on more risk that would have been taken on without insurance. The existence of the moral hazard problem causes losses to the insurance agency are seen to be due, in part, to its own existence. The adverse capital developments just discussed are an example of moral hazard problems for federal deposit insurance. The moral hazard problem is further exacerbated, however, with a flat deposit insurance premium scheme such as the structure currently in place. With the present premium structure, higher risk-taking institutions do not bear higher insurance costs than their conservatively managed competitors.
A reform that could work to mitigate the moral hazard problem would be the introduction of a premium structure directly related to the risk of loss for the insuree. Those institutions that are perceived as riskier are charged a higher premium. Some discussion has favored a move to such a pricing scheme, although serious questions have been raised about regulators' abilities to design such a pricing system. The move to a risk-based capital requirement structure is viewed by many as a step in this direction.(2) It would seem, however, that a more direct approach of basing insurance premium on the estimate of risk exposure would be preferable to this indirect approach.(3)
Regulatory Reform and Excessive Regulation
One of the basic premises of federal deposit insurance in the U.S. is that the government can use regulations to prevent depository institutions from taking on excessive risk. It is believed that regulations can effectively substitute for market discipline that would have been present in the absence of deposit insurance. Unfortunately, the proper list of regulations was not handed down to Moses along with the Ten Commandments. The regulations that are put in place to "protect" the individual depository institution have many times been criticized as the source of risk, rather than a vehicle of risk prevention. More troubling than the existence of poor regulations is the fact that deposit insurance itself may work to reduce the political pressures for sensible regulatory reform.
Consider, for example, the existence of interstate branching restrictions in U.S. banking. This regulation was implemented in part to "protect" the insurance fund from being depleted as a result of "excessive" competition on the part of U.S. banks. Individual banks in the U.S. found much to like about this regulation, as it protected them from outside competition.
Without deposit insurance and the associated moral hazard problem, it is quite likely, however, that many individual banks would have seen gains from being allowed to branch across state lines. The potential adverse effects of increased competition would have been offset by the prospect of other gains. In particular, as O'Driscoll (1988) has emphasized, branching allows greater financial diversification. In fact, O'Driscoll details a sharp difference between U.S. and Canadian banks in terms of failures over the period 1920 to 1933, with U.S. banks suffering more numerous and larger bank failures. O'Driscoll attributes this disparate banking performance in the two countries primarily to the lack of geographical diversification in the U.S.
Consistent with O'Driscoll's view, many modern financial analysts, including major regulators of the day, now argue that branching restrictions are not beneficial to the overall health of U.S. banking. What was once characterized as a valuable regulation is now much discredited. Thus, a regulation that was held to protect the deposit insurance fund is now seen as flawed. More importantly, the existence of deposit insurance has probably prevented change in thinking on this matter from occurring earlier. Without deposit insurance, both bank shareholders and depositors would have been much more mindful of the gain from financial diversification. They would have understood that a diversified intermediary is less likely to fail than a nondiversified intermediary. It is quite likely that shareholders and depositors would have put pressure upon regulators to eliminate the prohibition that provided diversification gains from being actualized.(4) However, with a federal deposit insurance system in place, no such pressures were brought to bear. Depositors, especially the insured, had little or no reason to be concerned. Regulators have only recognized their mistake grudgingly and in a laggard fashion.
The Incentive Structure for Regulators
The existence of deposit insurance requires the substitution of regulation for market discipline. For example, as we have seen, capital requirements must be administered, rather than allowing bank management to decide the "proper" amount of capital to hold. Similarly, depository institutions must be examined periodically to assure that excessive risk is not being taken on by their managements. Rules must be provided that detail "excessive" risk taking activity. However, such rules may or may not be appropriate in managing the risk of a depository institution.
The creation of deposit insurance gives power to regulators that would have not occurred otherwise. Kane (1985), amongst others, has emphasized that regulators have their own goals and objectives and as such cannot be presumed always to have the best interests of the deposit insurance fund in mind. For example, in discussing the savings and loan association debacle, Kane is most critical of the extent of capital forbearance that was allowed. Regulators knew that many savings and loans were insolvent in the early 1980s, but regulators chose to be "forbearing" rather than close down the institution, either selling off the assets or allowing the institution to be acquired by another healthy institution.
It is argued that the regulators chose to be forbearing because it was in their own best interest, not because it was in the best interest of the deposit fund. Regulators wanted to avoid recognizing the extent of losses, which would have required the admission, in as early as 1982, that the FSLIC was insolvent. By being forbearing the regulators of the day allowed institutions that were effectively insolvent to remain in operation. The insurance fund purposedly avoided public recognition of the problem and its size, even when the problem was obvious to many outsiders. Regulators behaved this way because it was in their own best interest. By delaying recognition of the problem, the regulators could not be accused of being in charge at the outset of the crisis. The "not on my charge mentality," which serves each individual regulator, won out over protecting the interests of the fund.
The insolvent savings and loan associations of the early 1980s were allowed to stay in business. Regulators hoped that these institutions would be able to soon turn a profit and return to solvency. As Kane further details, however, these "zombies" of the day took on excessive risk in large part because depositors, management and owners of these institutions did not have anything to lose. Depositors were protected by the insurance fund; shareholders and management realized that their financial stake in the institution was essentially zero. If their risk-taking activities could return their institutions to solvency, they would win. But, if further losses occurred this would not harm them personally as the losses would be borne by the FSLIC.
In the process of sustaining their own existence, the zombies had significant adverse effects on the viable savings and loan association of the day. Zombie institutions were willing to pay more for deposits and accept less for loans than other viable institutions. This competitive pressure significantly narrowed the interest margin for solvent institutions. As a result, solvent institutions became less profitable. In this environment a smaller economic decline was sufficient to turn the viable institution insolvent.
The financial cost of the saving and loan debacle to the taxpayer today is held to be much larger, even on a net present value basis, because of early forbearance on the part of regulators. Early closure of insolvent institutions is obviously a reform that is much needed. It is encouraging that the Federal Deposit Insurance Corporation Improvement Act of 1991 makes a significant move in this direction. Still, the more general problem of regulators (including Congress and insurance fund employees) being more concerned with their own interest than that of the funds continues to exist. The Act should be perceived as only restricting the behavior of the regulator in one way. Whether or not this is sufficient remains an open question.
Summary of the Costs of Deposit Insurance
The significant financial losses to the FSLIC and the FDIC are widely recognized today. These losses can be viewed as explicit costs of deposit insurance over and above the deadweight cost of regulation, examination and oversight. There are two views concerning these losses. Under one view, such losses can be considered unique, representing a one-time event that is not likely to be repeated. In this case the future for federal deposit insurance, as presently structured, has great promise. Alternatively, such costs can be viewed as a direct consequence of the existence of deposit insurance itself. The future for federal deposit insurance holds less promise in this case. This is especially true if the argument that the existence of deposit insurance will cause these losses to grow is correct. Obviously, most of the academic literature on deposit insurance recently has been more supportive of the latter view of deposit insurance.
BENEFITS OF DEPOSIT INSURANCE
It is generally accepted that our federal deposit insurance was designed to: (1) protect the small depositor from loss due to failure of his or her depository institution; and (2) prevent the occurrence of systemic bank runs and protect the payments mechanism. At the time of the establishment of federal deposit insurance in the U.S., it was felt that problems in each of these areas had plagued the U.S. financial system in one way or another. It is generally accepted today that similar problems have not plagued the U.S. financial system since the establishment of federal deposit insurance in 1933. Thus, one could say that these essential goals of deposit insurance have been achieved.
Revisionist thought in this area, however, questions the value of such successes. In particular, revisionist thought suggests that these goals are not as important as they once were, and/or they could have been achieved without deposit insurance altogether. Revisionists question whether the costs of deposit insurance are worth the benefits, as the benefits might not be considered great.
Protecting the "Small" Depositor
With today's $100,000 protection of each individual account, federal deposit insurance has obviously moved beyond the protection of the "small" depositor. Thus, in terms of this one goal deposit insurance has expanded its objective quite markedly, even in constant purchasing power terms.(5) Moreover, the U.S. financial system has evolved sufficiently since 1933 so that one must question whether the small investor in the U.S. really needs protection of this type. For one thing, the U.S. Treasury has been kind enough to provide over $3 billion in Treasury securities to today's financial investor. Most of these securities are short-term to maturity and all are generally viewed as free of default risk.
In such a financial environment, do small investors really need additional federal guarantees? One growing financial intermediary in the U.S. -- the money-market mutual fund -- suggests the answer may be negative. Money-market mutual funds have been able to attract over $400 billion today without federal deposit insurance. This suggests that today's investors may not value federal deposit insurance highly. Of course, it should be acknowledged that money-market mutual funds rely heavily on the commercial banking industry for the payments mechanism. So eliminating deposit insurance altogether could significantly affect the business of money market funds, as well as depository institutions. Still, there remains an important question concerning the value associated with protecting the small investor in today's financial environment.
Preventing Bank Runs
Much of the current thought on bank runs questions the value added by deposit insurance in this area. To begin with, Kaufman (1988) has reexamined U.S. financial history prior to the establishment of deposit insurance and found very little evidence supporting the contagious nature of bank failures. Bank failures generally did not cause problems for otherwise solvent depository institutions. Most runs on banks in the U.S. were runs on banks that were insolvent. Such runs were only "rational" as investors tried to obtain their deposits before the institution was closed. These "rational" bank runs provided value to the financial environment. Such runs promoted prompt closure of insolvent institutions. No other investors were "suckered in" to depositing funds with such an institution. Moreover, the runs imposed market discipline on other bankers to avoid excessive risk taking, least they find themselves in a similar situation. Such runs generally were not problematic to the extent of inducing spillover effects on other viable institutions.
Similar evidence is provided by considering the experiences of countries without deposit insurance. For example, Pozdena (1992) recently examined the Danish banking system, which until very recently did not have a deposit insurance system. Pozdena did not find evidence of either significant depositor losses or contagious bank runs in Denmark prior to the establishment of deposit insurance. Pozdena attributes the lack of significant banking problems in Denmark to a market-value early closure policy, along with significant capital requirements, suggesting deposit insurance is not necessary for promoting financial stability. Bordo (1990) also concludes that deposit insurance is not necessary to prevent banking panics, based on his examination of other international experiences.
Finally, when it comes to preventing the systemic nature of bank runs from spillover and adversely affecting the solvent but liquid depository institution, recent thought questions the need of deposit insurance. For example, Smith and Wall (1992) argue that discount window lending is the most effective means of dealing with the systemic problem of banking panics. They argue implicitly that the central bank is better positioned to deal with such problems. The deposit insurance agencies attempts to deal with such problems may only exacerbate the problem by allowing bank managers to take on additional risk at no cost to them.
SUMMARY AND CONCLUSIONS
The public's current perception of deposit insurance appears to be that this federal program has served a real benefit to the financial environment in the U.S., far exceeding the costs of such a program. It appears that deposit insurance has achieved the laudable goal of assuring financial stability in the U.S. The banking community also appears to agree with such an overall evaluation, although they are more prone to recommend fine-tuning of the deposit insurance system to mitigate or eliminate perceived inequities, such as "too-big-to-fail problems."
Such an overall favorable review of deposit insurance by the public and the banking industry may be changing as the bill for deposit insurance comes due, but nevertheless recent academic work on the subject seems to be much more critical. The academic community has suggested that the overall costs of the present deposit insurance structure are not trivial. Such costs comprise not only the obvious financial costs of the bailout of the deposit funds, etc., but also the many large implicit costs too often ignored. Most disturbing, many in the academic community suggest that the present structure of the deposit insurance system perpetuates an ever-escalating cost of deposit insurance. The existence of deposit insurance itself is held to lay the foundation for increasing costs associated with the federal deposit insurance system. The recent thinking in the academic community also suggests that the overall benefits of the present deposit insurance structure are not as large as once believed.
Some in the academic community go so far as to question whether federal deposit insurance is actually a net social good for the economy. Unfortunately, much of the academic work to date, itself, has been quite piecemeal, looking only at one particular cost or benefit of federal deposit insurance. Such a treatment is surely as short-sighted as the general treatment of deposit insurance that only considers the benefits of the program. However one gauges the present costs versus benefits of deposit insurance, to the extent that the costs of deposit insurance are underestimated and/or the benefits exaggerated, one can anticipate that we have not heard the last regarding deposit insurance reform.
(1)However, many economists are also quite critical of the present book value accounting system, which many times overstates the capital position of a bank relative to current market conditions. See, for example, Keeley (1989). Pozdena (1992) also emphasizes the benefits of market-value accounting in examining the Danish banking system.
(2)For a pessimistic review of the move to risk-based capital requirements as it relates to deposit insurance see Bradley et. al.
(3)As part of legislation to recapitalize the FDIC, Congress has required the establishment of risk-based premiums. Such a move, while recognized as superior to today's flat premium structure, is not viewed as a complete resolution to the moral hazard problems by many analysts. For example, see O'Driscoll (1988) who compares administratively established risk premiums to those developed by competitive markets and finds the former lacking.
(4)Kane (1981) discusses the "regulatory dialectic" in which regulators establish regulations that limit the profitability of financial institutions and these institutions then skirt the regulation to improve profits. Regulators then must close such loopholes. I also believe that regulators listen to their constituents. As such, we need to be cautious when constituents lose the incentive to avoid their opinions.
(5)Mark Flood (Federal Reserve Bank of St. Louis Monetary Trends, November 1991) estimates that the original $2,500 per account coverage in 1934 translates to about $51,000 constant 1991 dollars. Thus, the real per account coverage today is nearly double that of 1934. Interestingly enough, the inflation of the 1980s has significantly eroded the real per account coverage, which peaked in 1980 at close to $165,000 constant 1991 dollars.
|Printer friendly Cite/link Email Feedback|
|Author:||Hein, Scott E.|
|Date:||Jul 1, 1992|
|Previous Article:||"Too big to fail": rationale, consequences, and alternatives.|
|Next Article:||The emerging bank structure of the 1990s.|