A quarterly magazine of the IMF

March 1999, Volume 36, Number 1

A deposit insurance system can contribute to financial stability, but only if it is adequately funded and if other safeguards—such as a strong bank supervision program—are also in place.

IN AESOP'S fable about the hare and the tortoise, the tortoise accepted the hare's challenge to a race. The hare was much faster than the tortoise, but, after sprinting ahead in the opening stretch, he lay down to take a nap, confident of his ability to outrun his opponent. When he awoke, the tortoise, who had plodded on slowly but steadily, was nearing the finish. The hare could not catch up, and the tortoise won the race.

A successful deposit insurance system is more like the tortoise than the hare. On the surface, it appears that a national deposit insurance system can be set up quickly and easily, by announcing a public guarantee of bank deposits. Some countries, hoping to prevent wholesale deposit withdrawals that could cause healthy banks to fail and to bring stability to a troubled banking system, have tried to create a deposit insurance system in just this way. Unfortunately, unless the system has both sufficient financing to ensure it will survive a serious financial crisis and a strong program of bank supervision, it is destined to fail.

Funding deposit insurance

Although the need for adequate funding is self-evident, it may be less obvious that the best approach is to fund the system through premiums paid by its member banks, even if an initial loan from the government for start-up capital is necessary. First, funding the system, rather than relying solely on a government guarantee, ensures that the agency in charge of its operations will manage it in a fiscally responsible manner. Second, the agency can be sure it will have the working capital needed to resolve bank failures quickly. Waiting for a legislature to appropriate the necessary working capital greatly increases the costs of resolving bank failures, as the United States discovered during the savings and loan crisis of the 1980s and as Japan is learning today. Third, paying premiums to fund the deposit insurance system will give its member banks an incentive to monitor the system's operations to ensure that they are sound and fiscally responsible.

According to an IMF survey carried out in 1996 by Gillian Garcia and Carl-Johan Lindgren, approximately 50 of the IMF's member countries had explicit deposit insurance systems at the time, but many did not conform to recognized best practices. In a subsequent study (Garcia, 1998), best practices were identified as providing legal and regulatory authority for the system, giving the national banking supervisor the power to take prompt remedial actions against failing depository institutions, resolving failed banks quickly, keeping the size of deposits covered by deposit insurance small, making bank membership in the system compulsory to avoid adverse selection, paying out on insured deposits quickly, charging risk-adjusted insurance premiums, and ensuring the independence of the deposit insurance agency.

The financial safety net

Although these standards are critical for effective deposit insurance, by themselves they are not sufficient to assure the stability of the financial system. Deposit insurance is only one element of the financial safety net that exists in many countries, particularly developed countries.

The safety net is intended to maintain the stability of the financial system by protecting the critical financial intermediation function of banks (making savings available for investment and economic development) and their role in the national payments system. Typically, in addition to deposit insurance, a safety net also includes short-term lending by a central bank to assure bank liquidity. In the United States, liquidity support takes two forms: short-term collateralized lending from the U.S. Federal Reserve System's discount window to provide solvent banks with cash when they experience short-term liquidity problems, and even shorter-term overdraft credit to assure the smooth functioning of the payments system. The role of deposit insurance is to stabilize the financial system in the event of bank failures by assuring depositors they will have immediate access to their insured funds even if their bank fails, thereby reducing their incentive to make a "run" on the bank. By discouraging bank runs, deposit insurance can prevent panic from spreading through a financial system. Such panic can threaten healthy banks as well as troubled banks.

Because banks intermediate deposits by turning them into illiquid loans, even the healthiest banks cannot survive unlimited, immediate demands to withdraw deposits. History has shown the importance of having a mechanism in place to reassure depositors. Between the start of the Great Depression in October 1929 and the creation of the U.S. Federal Deposit Insurance Corporation (FDIC) at the end of 1933, 4,000 banks failed in the United States, but only 9 banks failed in 1934.

There is a more complex story behind these numbers, however. Immediately after it was established, the FDIC sent 4,000 examiners into the field to qualify banks for membership in the bank insurance fund. Thus, it was recognized from the start that deposit insurance could not function effectively without adequate bank supervision.

In the absence of strong bank supervision, the central bank and the deposit insurance system might find themselves providing financial support for insolvent banks engaged in risky activities that could damage the health of the financial system. Prudential supervision, which consists of onsite surveillance of banks through examinations and offsite surveillance through regular financial reporting employing internationally recognized accounting standards, is the eyes and ears of central banks and deposit insurance systems. It makes it easier to determine whether institutions are insolvent or merely suffering from liquidity problems. Without effective prudential supervision, deposit insurance and the other elements that make up the safety net protect reckless banks from the losses they might otherwise suffer when they gamble with their assets in hopes of high returns.

A country needs the political will to set up an effective system of bank supervision, and the desire to gain credibility in the international financial markets is likely to be a strong incentive. It also needs professionals with training in banking supervision to implement the new system. Such training can be obtained from various organizations in developed countries as well as from international financial institutions. Even more important, the Basle Committee on Banking Supervision has developed guidelines—Core Principles for Effective Banking Supervision—that can be used by countries as a model for establishing an effective system of prudential supervision. In addition, beginning in 1988, the Basle Committee established risk-based capital-adequacy standards that set minimum capital standards for banks active in international financial markets.

Regulatory capital standards are the lifeblood of sound prudential supervision. Strong capital provides a cushion against problems and raises the costs of speculation and risky behavior because investors have more to lose if a bank fails. While strong capital is critical, however, it cannot, by itself, prevent a bank from failing. Problem assets can wipe out a bank's capital quickly. Moreover, newer banks that lack a track record and whose management is inexperienced are riskier than established banks; banks operating in a transition or developing economy without market experience are at even greater risk. The Basle Committee's standards apply to experienced, internationally active banks from developed countries, and the Committee has emphasized that capital levels should be significantly higher for banks in other circumstances.

Moral hazard

Regardless of how good a country's system of banking supervision is or how high its capital standards are, there is no question that the elements of a national safety net, including short-term discount-window lending and deposit insurance, present the danger that they will distort the marketplace by reducing the possibility of loss from bad business judgments. Thus, the safety net raises the specter of moral hazard.

Moral hazard is a term economists use to refer to anything—insurance or a government subsidy, for example—that encourages risky behavior by leading financial risk takers to believe that they will reap the benefits of the risky investments they make while being protected from the losses. Providing support for insolvent financial institutions clearly involves moral hazard, but even providing liquidity support for troubled but solvent institutions can involve moral hazard if it shields such institutions from the realities of the marketplace. In designing and operating a safety net, countries need to balance two competing goals—assuring stability in the financial system when liquidity and solvency problems arise while minimizing moral hazard.

The design of a safety net can be most effective at limiting moral hazard if the marketplace is permitted to discipline financial risk takers by letting insolvent financial institutions fail and by imposing costs on institutions that come close to failing. In the latter case, the solution could be as simple as charging higher interest rates for short-term liquidity support. In the former, however, countries worldwide have often tried to save institutions they felt were "too big to fail," in an effort to ward off systemic problems. A reasonable balance between moral hazard and a stable financial system would permit a very limited exception for failures that pose a systemic risk while letting the market discipline improvident behavior. Thus, insolvent banks should in general be allowed to fail and shareholders should lose their equity if a bank is assisted to stay open. The proper balance assures that, as a rule, the safety net will be a net through which insolvent institutions can fall and not a floor that prevents insolvent institutions from falling far enough to fail.

In the United States, concerns about moral hazard were an important part of the legislative response to the banking and thrift crises that racked the financial industry from 1982 to 1994. The banking crisis alone involved 1,617 banks that failed or were kept open with FDIC support. Congress questioned the approaches the financial regulatory agencies had taken to the crisis. The U.S. Federal Reserve System was criticized for lending from the discount window to too many banks that were, or became, insolvent. The FDIC was criticized for providing financial support to keep too many insolvent banks open. Other bank regulators were accused of not closing banks early enough, before their capital was depleted, thereby adding to the costs of resolving bank failures.

In response to those criticisms, the U.S. Congress enacted legislation to limit moral hazard. The Federal Deposit Insurance Corporation Improvement Act (known as FIDICIA), which took effect in 1991, reduced the Federal Reserve's discretion in discount-window lending by permitting lending only to problem banks likely to survive liquidity problems; it also required the FDIC to resolve bank failures using the method that presented the "least cost" to the deposit insurance fund. This reduced the FDIC's authority to keep failed banks open. The FDIC was also required to establish risk-based deposit insurance premiums for banks; it adopted such premiums in 1992.

In addition, the U.S. Congress greatly reduced the latitude that the FDIC and the Federal Reserve had for applying the "too big to fail" doctrine. Further, it required all federal bank and thrift regulators to use "prompt corrective action" in addressing the problems of insured troubled financial institutions by requiring closer supervision and more capital in institutions that did not meet graduated capital thresholds. To prevent losses to the deposit insurance fund and, potentially, to U.S. taxpayers, Congress also encouraged federal regulators to close financial institutions that were likely to fail—even if they had as much as 2 percent tangible capital.

FIDICIA sought to strike a balance between limiting moral hazard in the functioning of the banking system and ensuring stability during a financial crisis, but its main objective was to reduce moral hazard. The question today is whether this legislation has tipped the balance too much in favor of limiting moral hazard at the expense of financial stability. Based on current experience, the answer appears to be "no." For example, it is clear that with the "least cost" test, more banks will be closed, because the FDIC will usually be unable to take the costlier course—that is, compensating healthy banks to acquire failed banks, thereby protecting all, not just insured, depositors. While that may increase the likelihood that a recession will have a wider impact on small communities, it may also reduce the likelihood of bank failures by reducing moral hazard.

According to the FDIC's landmark 1997 study on the causes of bank failures during the most recent banking crisis in the United States, History of the Eighties: Lessons for the Future, between the end of 1992, when the "least cost" test went into effect, and 1995, uninsured depositors were not protected in 63 percent of all bank failures, compared with only 19 percent between 1986 and 1991. The "least cost" test is thus likely to reduce the moral hazard in deposit insurance by encouraging large, uninsured depositors and large creditors to pay more attention to the conduct and circumstances of the banks with which they deal. This is appropriate because they are, in fact, in a better position than small depositors to understand the risks they take in dealing with a bank that engages in imprudent behavior or operates in the midst of a recession. It may also mean that, over time, managers of financial institutions will be more judicious about the risks they take during economic downturns, knowing they may lose their jobs and any investments they have in their bank if it fails. Finally, wiping out shareholder equity in a failed bank, which the FDIC traditionally does, adds to the level of market discipline that the "least cost" test imposes.

The FDIC has found that the "prompt corrective action" standard may result in the closing of banks that could have been saved. Analyzing bank failures between 1980 and 1992, the FDIC found that 143 banks, with $11 billion in assets, that did not fail might have been closed after 1992 under the 2 percent tangible capital rule. Although keeping such banks open could increase financial system instability during a banking crisis, the $11 billion in bank assets represented a small percentage of the total assets (just over $206 billion) of all banks that failed during the period.

In contrast, the FDIC has also found that, under the 2 percent tangible capital standard, 343 banks that failed probably would have been closed earlier than they were during the bank crisis. Thus, the 2 percent tangible capital standard also serves to prevent losses by closing failed banks earlier. Moreover, permitting banks with some positive capital to be closed when they are near insolvency will, in all likelihood, increase market discipline in the banking system.

Lessons for developing countries

For countries considering the establishment or reform of a deposit insurance system, it is obvious that deposit insurance alone cannot increase financial system stability. Without a sound system of banking supervision that includes strong capital standards as well as mechanisms for enlisting help from the market in imposing discipline on system participants, deposit insurance and other elements of the financial safety net will be ineffective and will increase the costs and pain of resolving a financial crisis. Mechanisms for increasing market discipline include limiting deposit insurance protection in developing economies to small savers, putting uninsured depositors and large creditors at the end of the line for any recovery after resolution of a failed bank, and wiping out shareholder equity when a bank fails.

Unfortunately, these kinds of legislative reforms will be useless if governments in emerging markets continue to give broad guarantees of support to market participants on an ad hoc basis when the financial system begins to experience trouble, or if international financial institutions encourage such support in a misguided effort at ensuring stability, which did not work in the Asian crisis. Such actions result in greater moral hazard without increasing systemic stability. Although allowing insolvent firms to fail may increase instability in the near term, it holds out the prospect of medium-to-long-term market discipline that should enhance the strength and stability of the financial system over time.

For countries that are just beginning to think about setting up a deposit insurance system, this description of the limited role that deposit insurance can play in shoring up the stability of a financial system in the absence of strong bank supervision and mechanisms to limit moral hazard may be discouraging—and that is probably a good thing. Countries should move slowly, like the tortoise in Aesop's fable. If they move too quickly to provide deposit insurance protection, without the necessary supervisory and financial underpinnings and market discipline, they may find they have written a blank check for financial losses. That can only weaken their economies and create a strong environment of moral hazard that will increase risk taking and systemic problems while exacerbating the pain during the next crisis. This may seem like bleak news, but it is easier to digest than a bankrupt deposit insurance fund or a depleted national treasury.

U.S. Federal Deposit Insurance Corporation, 1997, History of the Eighties: Lessons for the Future (Washington).

Ricki Tigert Helfer, a Nonresident Senior Fellow of the Brookings Institution, Washington, D.C., was Chairman and Chief Executive Officer of the U.S. Federal Deposit Insurance Corporation from 1994 to 1997.