Remarks by Chairman Alan Greenspan
At the Conference of State Banking Supervisors, Traverse City, Michigan (via satellite)
May 18, 2001
Issues for Bank Regulators
Good afternoon. It is a pleasure, once again, to talk with you about banking supervision and regulation. The decade of the 1990s was one of dramatic change for the U.S. financial services industry, inducing us to rethink much of our approach to supervising and regulating banks. During the 1990s, the Congress adopted historic changes in banking laws that transformed the structure of, and outlook for, the U.S. banking system and laid the groundwork for significantly more change. Those events, in turn, have caused bank supervisors to rely more heavily on banks’ own internal risk management practices and to focus supervisory efforts to ensure that those practices are sound.
We entered the past decade dealing with an industry under stress--one that was saddled with high levels of nonperforming assets and that was still experiencing a high number of bank failures, all a result of misjudging risks. On the heels of the thrift industry debacle, these problems led the Congress to enact the Federal Deposit Insurance Corporation Improvement Act of 1991, imposing “prompt corrective action” and other requirements on bank supervisors.
By the decade’s end, we saw even more substantial legislation that has led to truly nationwide banking and the possible combination of banking, securities, and insurance underwriting activities under common ownership. In some respects, these legislative events contrast sharply with the action taken at the beginning of the 1990s to address weaknesses in banking and the bank supervisory process and reflect a much higher level of comfort with the industry’s financial condition and risk management techniques. As such, the industry’s new-found freedom speaks, in large part, to the progress it has made throughout the decade in preparing itself for the tests ahead. The different laws also reflect a greater understanding by the Congress of banking risks and how industry practices and financial markets have changed.
New technologies have altered business practices around the globe, spurring international trade and producing larger commercial firms with more efficient systems for information and control. This, in turn, has encouraged many banks to increase their own size and financial strength and expand the scope of their product lines. Confining banks to outdated rules was, in the circumstances, perceived as both unfair to them and their shareholders and harmful to the long-term economic interests of the United States. We need a banking system that is not only sound but also able to innovate and respond to customer needs. The question for us is how best to accommodate change while protecting the public interest.
In my comments today, I would like to draw upon some of our past experiences and outline a conceptual approach for addressing future issues. Even a casual observer of banking history will recognize that many of the matters we deal with--and perhaps the most important ones, in particular--are largely variations of older and more familiar themes. That may be because the fundamental nature of banking and its role in a modern economy have not changed materially for generations, despite the important operational changes taking place every day. However, the management and supervision of banking--especially bank risk-taking--is in the process of fundamental change.
The Role of Bank Supervision and Regulation
Reducing the broad objectives of bank supervisors and regulators to two sentences may be difficult, given the many demands placed on them in overseeing a business as dynamic and important as banking. Yet, in some respects, these statements could be simplified still more: To promote safety and soundness while allowing banking institutions to innovate and compete. There are clearly some tensions in these goals.
The hard part, of course, is in implementing these concepts on a daily basis. Can we always tell, for example, what is “unsafe and unsound” when we see it, and at what point do we interfere with management’s judgment when we disagree? What is “efficient competition” when banks have access to a subsidized safety net? And how do we judge the competitiveness of a bank’s returns? Reaching clear agreement on such questions is not easy, even among bank regulators with generally common goals. Resolving them becomes far more contentious when bankers weigh in with materially different views.
One approach is for regulators to be permissive, unless it is clear that established public interests are at risk. Banking in this country is, in most areas, highly competitive, and the industry has proven itself to be highly resilient. To survive and be effective, banks must be willing and able to take risk. Revenue, shareholder equity, and if necessary the Bank Insurance Fund are there to deal with mistakes. Put differently, while public policy needs to limit the financial and social costs of bank failures, we should not view every bank failure as a supervisory or regulatory failure. It is not our role to prevent all failures, let alone to guard against every earnings decline. Indeed, to do our jobs well, we should understand that the essential economic function of banks is to take risk, and that means mistakes will sometimes be made. A perfectly safe bank, holding a portfolio of Treasury bills, is not doing the economy or its shareholders any good.
The experience of the early 1990s illustrates the process of recovery after a period of excessive risk-taking. A decade ago, many regional and money center banks were reeling from excessive commercial real estate lending and still-substantial holdings of weak foreign country loans. Meanwhile, many agricultural banks were only then recovering from their own bout with problem credits that had proved too heavy for hundreds of other banks to bear. The FDIC was reporting that the Bank Insurance Fund was nearly drained, while the General Accounting Office was saying that, in substance, it already was. Banks nationwide were tightening their credit standards and cutting access to credit for many firms, producing a “credit crunch.”
It was clear that change was needed. The industry needed to resolve its problems, improve its risk management practices, and construct new strategies that would provide more stability in lending markets and more competitive, long-term returns to investors. For their part, bank supervisors needed to pay more attention to emerging weaknesses and deal with them more quickly.
Solving the immediate problem of weak asset quality and improving the industry’s overall profitability was, of course, the first priority, and fortunately that was accomplished sooner than anyone expected. By the mid-1990s, U.S. banks were enjoying stronger portfolios and higher earnings than they had seen in recent memory, attesting to the underlying strength and resiliency of our banking system. During the past decade, the industry’s profitability has arguably climbed to its highest level ever, as large banks finally joined small ones in achieving annual returns on average assets in excess of 1 percent.
To be sure, no one wants to repeat the experience of the early 1990s and, indeed, that period has set a standard for what we should seek to avoid. It is also clear, however, that the industry rebounded quickly when given the incentives to deal with its problems through market and regulatory pressures. The key was addressing asset quality problems and reducing the associated weight of bad loans. Getting there required some banks to reduce or eliminate dividends and raise new capital, often on dilutive terms. That did not always occur easily or without frank discussions between bankers and regulators. At times forceful intervention is required to ensure that banks don’t take risks that, at least in the judgment of their supervisors, they are not prepared to handle. We supervisors also need to resolve insolvent institutions while not trying to minimize every risk. But that does not mean we should avoid addressing the causes of banking crises, no matter how adept we may be at getting out of them.
Pro-Cyclicality of Bank Lending and Agency Supervision
Historically, as I have said, loan standards are softened during an expansion, when loan books are growing rapidly. As the economy expands, business and household financing needs increase and projections of future outcomes are optimistic. It is difficult for managers and supervisors to be critical of loan growth in such an environment, particularly one marked by intense competition. In addition, incentives for loan officers and managers traditionally have rewarded loan growth, market share, and nominal profits. In an effort to expand loans in an optimistic environment, potential returns are often not fully adjusted for risk, with the nominally high-yielding credit looking particularly attractive. Moreover, credit specialists at banks--a function often separated from the loan officers-- historically have had difficulty making their case about risk because of their inability to measure and quantify risk. At the same time, with debt service current and market risk premiums cyclically low, and with the same inability to quantify and measure risk, supervisory criticisms of standards are difficult to justify. When the economy begins to slow and the quality of some booked loans deteriorates, loan standards belatedly tighten and supervisors’ criticism of previous standards is taken more seriously, inducing an even greater tightening of credit. In part, the examiners’ observations have often provided a reality check by which some banks came to recognize a deterioration at first denied.
The last few years have had some of the traditional characteristics I have just described: the substantial easing of terms as the economy improves, the rapid expansion of the loan book, the deterioration of loan quality as the economy slows, and a cumulative tightening of loan standards. But this interval has had some interesting characteristics not observed in earlier expansions. First, in the mid-1990s, examiners began to focus on banks’ risk management systems and processes; at the same time, supervisors’ observations about softening loan standards came both unusually early in the expansion and were taken more seriously than had often been the case. The turmoil in financial markets in 1998 associated with both East Asia and the Russian default also focused bankers’ attention on loan quality during the continued expansion in this country. And there was a further induced tightening of standards last year, months before the aggregate economy weakened, in response to early indications of deteriorating loan quality.
All of this might have been the result of idiosyncratic events from which generalizations should not be made. Perhaps. But at the same time another, more profound development of critical importance had begun: the creation at the larger, more sophisticated banks of an operational loan process with a more or less formal procedure for recognizing, pricing, and managing risk. In these emerging systems, loans are classified by risk, internal profit centers are charged for equity allocations by risk category, and risk adjustments are explicitly made. In short, the formal measurement and quantification of risk has begun to occur and to be integrated into the loan-making process. This is a sea change--or at least the beginning of one. Formal risk management systems are designed to reduce the potential for the unintended acceptance of risk and hence should reduce the pro-cyclical behavior that has characterized banking history. But, again, the process has just begun.
The federal banking agencies are trying to generalize and institutionalize this process in the current efforts to reform the Basel Capital Accord. The proposed foundation approach, intended mainly for regional banks, and the advanced internal risk-based approach, intended mainly for large, complex banking organizations, would require banks to create and use internal risk classifications for their banking book. These classifications would not only be used to establish minimum regulatory capital, but the associated risk profiles would be disclosed by the bank for the review of its creditors and shareholders. How well this process works will depend on the supervisors’ skill and, for lack of a better word, toughness in validating the conceptual and empirical basis of the risk classifications and management system. A weak or misused classification system would destroy the process.
If applied seriously, and when operational near the middle of this decade, the revised accord, Basel II, promises to promote not only better risk management over a wider group of banks but also less-intrusive supervision once the risk management system is validated. It also promises less variability in loan policies over the cycle. To be sure, banks’ risk appetites can change as they decide to increase or decrease the riskiness of their loan portfolios. In addition, minimum regulatory capital might tend to be pro-cyclical as loans migrate from less- to more-risky classifications as the economy weakens, in the process increasing minimum capital requirements. This tendency must, however, be balanced against what should be the reduction in cyclical reserving and write-offs that traditionally have come with the late recognition of excess risk taken earlier. Moreover, if, as promised, better risk-management reduces the variance of charge-offs and earnings, bank equity values, to repeat, are likely to be rewarded with higher price-earnings ratios. Put somewhat differently, better risk management and supervision may well focus bank management on its real objective: maximizing shareholder wealth.
Building on Market Practice
We should encourage all banking organizations to continually strengthen their systems and controls as their operations and market practices evolve. But we need not expect--and surely not require--significant numbers of banks to develop risk management systems that are more complex than they need. Simplicity can work well, too, and for most banks that is likely to be the best approach. For the vast majority of U.S. banks, current practices, as I noted, seem fine.
It is, of course, our belief that, for the larger banks, improving the measurement and management of risk will provide bank supervisors with more reliable and efficient ways to evaluate a bank’s financial health. That thought underlies the concept of risk-based supervision and our increased focus on internal processes and controls. If we can gain greater confidence in a bank’s operating procedures and in its own evaluation of risk, we should be able to reduce our oversight role--certainly when compared with what would otherwise be required.
Innovative Banking System
Despite more than a decade of consolidation, the U.S. banking system remains highly competitive on virtually every front. With enhanced technology and deregulation we now have global and national markets for a growing number of banking products, while low barriers to entry encourage competition in more sectors than ever before. The pace of bank creation has actually increased in recent years, despite the contraction in the overall number of insured commercial banks. Interstate banking has also brought new competitors into previously settled markets, and Gramm-Leach-Bliley has the prospect of adding more competition, as nonbank firms seek entry into the banking business. Even when these events simply change the ownership of existing banks, one could still argue that, in most cases, they introduce stronger, more aggressive competitors into a market.
The role of bank regulators is, of course, to ensure that all of these activities do not reduce the safety and soundness of banks. In recent years, state bank regulators have worked well together and with their federal counterparts in accommodating interstate banking and minimizing oversight costs to both the supervisors and to state-chartered institutions. Now, Gramm-Leach-Bliley has introduced nonbank activities and functional regulators into the mix. In moving forward into new areas, the Federal Reserve will strive to build on the long tradition of coordination and cooperation with state bank supervisors in ensuring that the broader activities of financial holding companies remain adequately supervised and regulated.
On numerous occasions last year, Federal Reserve staff met with staff members of the Securities and Exchange Commission, various state insurance commissions, and other state and federal bank supervisors in a mutual effort to learn more about each other’s interests and how our role as umbrella supervisor can mesh best with their activities. It is important that we maintain this cooperative process and work to make supervision as efficient as it can be.
It seems clear that, building on bank practice, we are in the process of developing an improvement in both lending and supervisory policies that will not only foster better risk management but could also reduce the pro-cyclical pattern of easing and tightening of bank lending, and accordingly increase bank shareholder values. It is not an easy road, but it seems that we are well along it.