Each depositor insured to at least $250,000 per insured bank

The speakers and panelists at this forum have spent the day talking about what has happened in the five years since the passage of FDICIA and what may lie ahead. I am here to add another perspective: like the Ghost of Christmas Past, I will take us back to the way things were -- in this case, to the way things were in the banking crisis of the 1980s and early 1990s, the crisis that led to the passage of the law.

It has been said that experience is a tough teacher -- first you get the test, then you learn the lesson. Bank regulators were tested by the crisis, and learned lessons. Did we learn the correct lessons?

When I became FDIC Chairman, I initiated a project to find the answer to that question, an answer based on objective analysis. The result is a series of 14 papers that the FDIC will publish over the coming year. Next month, drafts of three papers -- an overview, as well as an analysis of bank examination and enforcement from 1980 through 1994 and an analysis of off-site surveillance systems during the same period -- will be presented at a symposium we are hosting. Although our studies cover many other issues, I will focus today on our findings relating to examination and supervision.

Effective bank supervision is critical to sound deposit insurance. Without it, the insurer is potentially faced with writing a blank check. It is also one of the important tools we have for containing the problem of moral hazard that arises from any form of insurance -- whether public or private. The failure of the federal savings and loan insurance fund was a direct result of the failure of supervision. It resulted in the taxpayer writing a blank check. Without strong supervision, deposit insurance simply becomes a public resource that risk takers exploit.

While economic, legislative and regulatory forces all contributed to a demanding environment for banking, the more immediate cause of bank failures in the 1980s and early 1990s was a series of severe sectoral and regional recessions. In agriculture, energy and commercial real estate -- and in the Southwest, the Northeast and California -- the recession followed periods of exuberant expansion often characterized by speculative activity. In all these cases, the conventional wisdom was that the boom would not end. Regulators, too, overreacted to the good times by becoming complacent.

Moreover, two decisions that were embraced by the Office of the Comptroller of the Currency and the Federal Deposit Insurance Corporation -- and to some extent by the Federal Reserve System -- to change examination policies during the late 1970s and early 1980s had an important negative impact on the outcome and severity of the crisis that was to follow. Those two decisions were (1) to place relatively more weight on off-site supervision and relatively less upon on-site examinations and (2) to concentrate examination resources on those institutions that posed the greatest risk to the insurance fund and the stability of the financial system. Both decisions ultimately resulted in fewer field examiners and reduced numbers of examinations for most of the 1980s, weakening the ability of bank supervisors to detect -- and respond to -- problems.

The total number of state and federal examiners declined by 13 percent from 1980 to 1984. The OCC and the FDIC experienced a greater decline of 17 percent. Even after hiring resumed, it was not until 1987 that the examiner force -- federal and state -- was restored to 1980 levels. In the meantime, the number of annual bank failures increased from 10 to 184 between 1980 and 1987, while the number of troubled banks increased from 217 to 1,575 over the same period.

The decline in the number of examiners led to marked changes in the frequency of examinations. In 1980, the average length of time between examinations was 15 months. By 1986, the average interval had increased to 20 months -- and in the most extreme cases, had increased to seven years. The greatest change was for CAMEL 1-rated banks, whose average interval increased from 15 to 28 months between 1980 and 1986.

With that background, today I will highlight six of the findings of our historical study -- findings based on evidence that, indeed, we regulators learned -- and are applying -- the correct lessons from our experience.

Lesson #1 -- There is no substitute for regular, on-site examinations of depository institutions for addressing specific problems at individual institutions. On-site examinations generate information on the condition of banks that is not available from any other source.

During the 1980s, examination ratings that were up-to-date generally identified most of the banks that required increased supervisory attention well before the bank actually failed. Examinations were generally effective in identifying problem banks in a two-to-three year window prior to failure. As we have seen, however, the problem was that far too many examinations were out-of-date, and could not, therefore, serve the function of identifying current difficulties in the industry. Of the 1,617 banks that failed in 1980 through 1994, 36 percent had CAMEL ratings of "1" or "2" two years prior to failure.

FDICIA, of course, requires annual full-scope examinations for all banks, except that an 18-month interval can be substituted for small banks with satisfactory ratings.

Lesson #2 -- Even though up-to-date CAMEL ratings were generally successful in identifying banks that required greater supervisory attention, they had limitations. Because CAMEL ratings are based on the internal operations of the bank, they do not take into account economic developments that may pose future problems. This partly explains why 1- or 2-rated institutions could fail only two years later.

We at the FDIC have created a Division of Insurance to monitor economic developments; to provide data to our supervisory staff, as well as to the staffs of the other regulatory agencies; and to make economic risk assessments available to the industry in order to bridge the gap between the individual institution and the economic environment in which it operates. We are also developing a model for projecting bank failures that will incorporate regional and macroeconomic information in the forecast, which up to now has been based solely on supervisory and historical information.

Lesson #3 -- Because CAMEL ratings are generally a measure of the current condition of the bank at the time it is examined, they do not systematically track risk factors that may produce future losses. In response to this lesson, all of the regulatory agencies today have programs aimed at tracking risk. At the FDIC, for example, we have developed a flow chart for our examiners to use in tracking interest rate risk. It reflects a graduated approach to determining the risk exposure of an institution -- the more risk the examiner finds, the more steps he or she must take.

We are now field testing 10 more flow charts that cover areas ranging from underwriting and credit administrative practices to loan review systems to insider transactions. The purpose of this structured risk-assessment approach is to look beyond the examination date to how a bank can respond to changing market conditions in the context of its individual risk profile.

Moreover, risk-based capital takes into account off-balance sheet risk.

Most recently, the CAMEL rating system has been updated to become CAMELS to emphasize risk assessment and the risk profile of the institution.

Lesson #4 -- Once troubled institutions were identified during the 1980-94 period, they were subjected to supervisory and enforcement actions that were by and large effective in reducing failures and losses to the insurance fund. About one-half of all banks rated "4" or "5 by the FDIC from 1980 through 1994 were the subject of formal enforcement actions; many of the remaining banks received informal enforcement actions.

About 75 percent of all problem banks recovered, while 25 percent failed.

As opposed to the thrift experience, bank supervisory actions led to lower asset growth, reduced dividend payments, and increased capital injections at troubled banks. This had the effect of limiting risk-taking by problem banks and limiting losses to the insurance fund when the banks failed.

Lesson #5 -- While capital is important as a cushion to protect banks from failure and the insurance funds from loss, even sizable capital will not save an institution with significant problem assets and a high risk profile. We looked at banks in 1982 and separated them into two groups. The first group survived the next five years. The second were the banks that failed in 1986 and 1987.

In 1982, the banks that did not fail had an average equity ratio of 8.84 percent, while failed banks had a ratio of 8.29 percent, only 55 basis points lower. Moreover, 8.29 percent -- the lower number -- is above the level needed to be considered well-capitalized under the risk-based system now in effect.

Capital is a lagging indicator of the health of an institution -- an important point in weighing the significance of the prompt corrective action requirements of FDICIA. Examiners analyze considerably more information than capital ratios to determine a bank's likelihood of failure.

The real value of prompt corrective action, therefore, may be that the regulators must maintain a staff of examiners sufficient to meet its demands and the demands of mandated regular examinations. In light of the experience in the early 1980s, that is valuable.

Lesson #6 -- Based on the experience of the 1980s, risk factors can be used to identify groups of banks that have a higher risk of failure.

For example, the banks that failed in the years 1982 through 1987 had distinctly higher risk profiles in 1982 than banks that did not fail. They had higher loan-to-asset ratios than survivors. They had substantially higher ratios of interest and fee income on their loan and lease portfolios, which suggests that their loans were riskier. They also had higher growth rates than the banks that did not fail, but these growth rates were sharply cut back as the banks approached failure, as FDIC enforcement actions took effect. This finding suggests that the focus on risk assessment in current supervisory thinking is on target.

Beyond these and other specific lessons that our studies confirm, the FDIC's history of the eighties and early nineties project reinforces the general lesson from that time: that balance is the key to success in both regulating banks and managing deposit insurance.

In banking regulation, balance means that we recognize that when things are going badly, the pendulum has a way of swinging back -- and when things are going well, the pendulum will someday swing the other way, too. We regulators can maintain this balance only if we follow the basic principles of bank supervision both in good times and in bad.

During good times, we must be alert to problems and do something about them before they result in severe problems for the banking system. We must be just as realistic when the cycle turns down as we are when the cycle is on the upswing. Banks are in the business of accepting risk as financial intermediaries and of making a profit. We should not fall into the mindset that problems lurk under every rock and in every loan file. We should justify the balance we maintain as regulators on the basis of fact and critical analysis.

Balance in managing deposit insurance means assuring stability in the financial system while addressing the problem of moral hazard that arises from public, or private, deposit insurance. By protecting depositors against loss, deposit insurance virtually eliminates the risk of bank runs and disruptive breakdowns in bank lending that damage the economy.

On the other hand, by assuming the risk of losses that would otherwise be borne by depositors, deposit insurance provides incentives for increased risk-taking by bank management, thereby exposing the insurance fund to greater losses. Moral hazard is a particularly serious concern if the institution is nearing insolvency. Then, the owners have strong incentives to make risky investments because profits accrue to the owners, while losses fall on the deposit insurance fund.

In the 1980s, the balance tipped in favor of stability. In assuring stability, the FDIC was eminently successful. Stability was achieved, however, at great cost -- and with respect to savings and loan failures, at great cost to the taxpayers. FDICIA was the Congress' call to us to restore the balance by giving more attention to the problem of moral hazard.

In carrying out the requirements of FDICIA -- and pursuing other initiatives -- we are doing so through risk-based and higher minimum capital standards, risk-related deposit insurance premiums, the least-cost test for resolving bank failures, and national depositor preference.

First, the development of internationally-accepted risk-based capital standards is one of the most significant innovations in the history of banking regulation. The Basle Committee on Banking Supervision has laid out a framework for assessing an institution's capital adequacy by weighing its assets and off-balance sheet exposures on the basis of counterparty risk. Moreover, recognizing that international banks have been actively involved in trading securities and derivative products, the Committee has developed progressive standards through the use of standardized and internal models to measure the unique market risks of specific portfolios.

Second, higher minimum capital standards are enforced through prompt corrective action. The principle embedded in prompt corrective action is gradation of risk and of appropriate regulatory response: The less capital a bank has, the smaller the cushion it has to absorb losses, and the greater the risk it poses to the insurance fund. The greater the risk, the more attention it should receive from regulators, but strong capital, as we have seen, is a necessary but not sufficient condition for safe and sound banking.

Third, the principle of gradation of risk and response is also reflected in our system of risk-related FDIC insurance premiums. The greater the risk, the higher the premiums the institutions pay. Risk-related premiums promote safety and soundness -- and help to address the issue of moral hazard -- by giving institutions an economic incentive -- through lower deposit insurance premiums -- to improve their conditions and maintain lower risk profiles.

The deposit insurance premium for an individual institution is now established on the basis of its capital and supervisory ratings -- with three categories of each and a nine-block grid. Currently, 94 percent of institutions insured by the Bank Insurance Fund and 89 percent of the institutions insured by the Savings Association Insurance Fund are in the FDIC's best category for deposit insurance premiums, which means these institutions are both well-capitalized and either 1- or 2-rated.

We are analyzing whether other factors are relevant to risk to the insurance funds -- and whether the nine-block grid for setting deposit insurance premiums should be expanded. We are also examining whether our current 27-basis point spread is sufficient to price the risks to the insurance funds posed by individual institutions. Those are questions that we will give a lot of attention to during the next year.

Fourth, in resolving bank failures, the FDIC is required by FDICIA to accept the proposal from a potential purchaser that is the least costly to the deposit insurance fund of all the proposals we receive. After the law took effect, in more than half of the failures in 1992 -- 66 out of 120 -- uninsured depositors received less than 100 cents on each dollar above $100,000. That was a significant increase in uninsured depositors experiencing losses from 1991, when fewer than 20 percent of the failures involved a loss for uninsured depositors. While the number of bank failures in 1992 was lower than in previous years, the number of uninsured depositors experiencing a loss was significantly greater.

Finally, the passage of a national depositor preference law in 1993 gave creditors of banks other than depositors an extra incentive to be concerned about the condition of their institutions. If a bank fails, anyone with a non-deposit claim gets nothing until all depositors, including the FDIC as insurer, have been made whole. It is still too early to assess the impact of this statutory change.

Conceptually, higher risk-based and minimum capital standards, risk-related deposit insurance premiums, and the least-cost test for resolving bank failures are direct and indirect surrogates for the discipline that depositors would logically impose if they had access to the economist's dream: perfect information in a purely competitive market.

In conclusion, we have been working to improve our system of banking regulation and supervision -- including the safety net -- for more than a decade. The banking crisis of the 1980s and early 1990s exposed weaknesses in the banking system -- and in the system of bank regulation. FDICIA was a reaction, but not the only one. Fortunately, regulators have continued to work beyond FDICIA's bounds to find better ways of responding to supervisory issues.

The Ghost of Christmas Past came with the message that the past was prelude to the future. In the euphoria of a year when the commercial banking industry is likely to make $50 billion in profits for the first time, perhaps we, too, can benefit from reflecting on that message. We have seen in our history of the 1980s and early 1990s project, it took years for problems at banks to surface.

In the end, the chief lesson of the 1980s is a clear one: there is a continuing, strong need for effective and balanced supervision.