An official website of the United States government
News Release 1997-111
December 12, 1997
Share This Page:
Good morning. As the current chairman of the Federal Financial Institutions Examination Council, it is my pleasure to welcome you all to our conference on regulatory capital. The subject before us today is one whose importance and complexity are truly worthy of the outstanding representatives of the financial, academic, and government gathered here to discuss it. Let me extend a special welcome to our featured speakers, Governor Phillips and Mr. Medlin.
Almost ten years ago, the Basle Committee on Banking Supervision adopted the accord on international convergence of capital measurement and capital standards. The Basle accord was immediately—and rightly—hailed as a landmark event. "Never before," one commentator wrote in 1988, "have regulators from so many countries reached agreement on a basic issue affecting the operation of financial institutions." Basle was the breakthrough that formed the basis for rationalizing the international crazy quilt of regulatory capital standards. It helped to assuage a building protectionist backlash against non-U.S. financial service providers, some of whom had been accused of taking advantage of unequal regulation to leapfrog their American competition. With the Basle accord, the rationale for retaliation—which had real potential for disrupting the global flow of financial products and services — lost urgency.
And that was not all. The Basle accord highlighted and ultimately helped reverse the slippage in bank capital levels worldwide. It focused attention on the whole concept of risk, as a tool both for bank managers and bank supervisors. It gave official recognition to the growing importance of off-balance sheet activities in bank operations. Finally, Basle pointed the way to the future — a future full of possibilities for continued cooperation and international harmonization of bank regulation — cooperation which has continue to bear important fruit in such important areas as the regulation of derivatives, Year 2000 compliance, money laundering, and more. Unquestionably, the Basle accord has significantly advanced the effectiveness of bank supervision worldwide.
Certainly no one expected that the Basle risk-based capital formula would endure for all time. But few of us imagined that we would have cause to revisit these issues so soon. Since 1988, the financial marketplace has evolved far faster than most of us anticipated — and in new, often uncharted directions. The explosive growth of asset securitization, derivatives and other hedging instruments, and new internal risk management strategies and technologies have transformed the traditional banking business — and introduced new risks of their own. Changes in the statutory and regulatory framework have paved the way for interstate banking, new opportunities for affiliations, additions to the permissible product mix, and corporate consolidation. Globalization has proceeded apace, bringing new competitive pressures and opportunities in its train. We have become a nation of investors instead of savers, with all that implies for bank funding and liquidity. Money itself is in the process of a historic transformation from a tangible commodity to a series of electronic impulses embedded on microchips.
But even if the banking world had stood still over the past decade, the current risk-based framework would nevertheless have required attention. As it is, many outside analysts believe that the time has arrived for major modifications of our current capital framework. They point both to its technical limitations and the practical problems that have flowed from them. Certainly Basle reflected the temporal wisdom of its time, when there was less focus on risk as the fulcrum for safety and soundness supervision and our tools for measuring risk were not as refined as they are today. The original accord of 1988 primarily addressed the issue of credit risk and did not address many other kinds of risk that affect banks' need for capital. And its approach to credit risk presupposed a limited ability to distinguish between different levels of risk.
One important test of any regulatory regime is whether it promotes rational or irrational economic behavior. Does it allocate investment capital fairly and efficiently? Or does it divert resources unproductively and serve primarily to make more work for lawyers and other financial engineers whose job it is to identify and exploit loopholes?
Our implementation of the risk-based capital standard, some critics conclude, is problematic in this regard. For example, the current system, these critics point out, has given rise to a whole cottage industry of consultants and advisers, producing ream upon ream of ponderous interpretations designed to help banks calculate and manage regulatory capital. What's more, the capital accord applies only to banks and their subsidiaries. Other types of financial institutions are not held to the same level of rigor. So while the Basle accord has clearly helped level the playing field between U.S. and foreign banks, the absence of uniform international standards in other sectors of the financial marketplace continues to create competitive inequalities that can make it difficult for banks to compete effectively with nonbanks here at home and abroad.
So where do we go from here? Can we draw on the experiences of the past decade to reform the current risk-based capital framework, altering it to bring it into line with the current realities of the financial marketplace? Or is a more dramatic overhaul needed? As a member of the Basle Committee, I am certainly raising these questions with other members of the international supervisory community.
Some reforms of the risk-based framework are already underway. For example, supervisors in several Basle Committee countries are currently looking at ways to extend the capital accord framework to credit derivatives and other novel financial instruments. There are those who believe that a transaction-based form of capital adequacy could eventually capture risk of credit concentrations and other types of risk such as operational risk and settlement risk.
But there is also a school of thought which says that because the more subjective — but no less critical — risk factors can never be quantified for inclusion in any risk-based capital formula, we should throw in the towel altogether and lay the risk-based approach to rest, honoring it as a regime that accomplished much in its heyday but one that has outlived its usefulness.
Some bankers and regulators might be prepared to do just that, if we were only able to agree on a better substitute. That is where the difficulty arises—and that is why we have our work cut out for us today. Not that we can realistically expect to achieve a consensus on such a complicated subject in the space of a few hours. The options we will be discussing range across a wide gamut, and each one has distinct pluses and minuses. But we can at least expect to learn more about the possibilities from the varied perspectives of the distinguished representatives of the financial, legal, and academic communities here with us today.
As our discussion proceeds, it seems important to me that we make a special effort not to confuse means and ends in two very fundamental respects. First, there is a danger that, in dealing with the technical challenges involved in measuring regulatory capital, we lose sight of the function of capital itself. For the truth is, that even though the financial world has seen massive changes in recent years, the logic behind regulatory capital is pretty much the same as it has always been: bank owners are most likely to operate prudently when they have their own funds at risk. Capital provides a buffer against losses and thus protects the interests of depositors—and deposit insurance.
Second, it is important that we not lose sight of the fact that while capital is only one, albeit important, indicator of an institution's overall health, it is also only one, albeit important, tool in our overall supervisory arsenal. Most regulators would hesitate to say that capital, even a mountain of capital, will guarantee a bank's stability or future solvency. Indeed, excessive capital can be almost as detrimental as inadequate capital, if it compels bankers to take greater risks to earn the hurdle rate of the return that the markets require. Some have argued that no amount of capital will salvage a bank that is grossly mismanaged. Catastrophic events do occur, and when they do, all bets are off. On the eve of the Great Depression, commercial bank capital was well in excess of regulatory minimums, sometimes by a factor of two or three. But it was all swept away by a flood of unanticipated losses.
That is why some have argued that, as my friend and distinguished predecessor John Heimann once put it, capital adequacy is "situational"—just one factor to consider in the context of the caliber of the bank's management, the level of its earnings, and a host of other factors. At the OCC, we have always tried to look beyond the raw numbers to interpret what those numbers mean for a particular institution, with its own peculiar risk characteristics. As a matter of policy, capital measurements should be determined objectively, consistently, and uniformly, but the interpretation of those measures is necessarily subjective and should be an adaptable component of overall supervision.
Notwithstanding that fact, I for one firmly believe that we need solid, substantial, and tangible capital — regulatory capital with healthy, uniformly applicable minimum standards.
Historically, bank supervision has always involved compromises and tradeoffs and trying to strike the optimum balance of intervention and detachment. Over time, the pendulum has swung between those two poles. It was not that long ago that examiners conducted what amounted to intensive audits of loans and passbook accounts, and counted all the cash in the vault. But the marginal benefits of this exhaustive approach led to its abandonment in favor of a framework in which examiners made only rare and fleeting appearances inside the banks they were responsible for. Because that approach also proved unsatisfactory, for obvious reasons, the search resumed for the right middle ground. We may not yet have found the perfect system, but our risk-based approach to supervision takes us a giant step closer to supervisory prudence.
The history of regulatory capital has included similar give and take. Over the course of the last century, until the implementation of the Basle accord in the late 1980s, bank capital levels fell almost steadily. More than a hundred years ago, the first Comptroller of the Currency discouraged bankers from increasing capital because he worried that increased capital would lead to too-rapid asset expansion — a story which illustrates yet another way in which regulatory capital has been used to promote supervisory goals. But this view of capital as a public policy liability changed, as concern shifted to the safety of deposits in the pre-insurance era, when confidence in the banking system often hung by a thread. In 1914, the OCC adopted its first minimum capital ratios — a flat ten percent of deposits. Banks typically posted the most current levels of capital and surplus in gold-leaf letters as a way of reassuring depositors.
With the advent of deposit insurance, however, the focus shifted away from depositor confidence to the asset side of the balance sheet, as federal regulators grew increasingly concerned over risk-taking with insured deposits. By 1948, the OCC had abandoned the capital-deposit ratio and was placing emphasis on the ratio of capital to assets. But which assets? Banks had emerged from World War II with huge quantities of government securities—essentially riskless from the credit point of view. Thus, the OCC embraced the then-novel concept of risk assets — a concept generally interpreted as meaning assets less cash, physical plant, and both direct and indirect government obligations.
This seemed simple, but it proved anything but that. After conclusion of the 1951 Treasury-Federal Reserve accord, which exposed longer term securities to market forces, these government obligations no longer looked so riskless. And, just as obviously, commercial loans varied dramatically in their risk characteristics. So the regulators tried to fine-tune their formulae. In the early 1950s, the Federal Reserve adopted a scheme based on "adjusted capital," which assigned varying percentage capital requirements according to their presumed relative riskiness. The Federal Reserve Bank of New York refined this approach to produce a formula that was even more complex.
The system soon began to buckle under its own weight, and the regulators themselves were among the first to acknowledge it. The OCC backed away from its risk-asset emphasis in the 1960s and, by the early 1970s, the Federal Reserve had followed suit. Straight leverage ratios were adopted and refined — until the 1980s, when the rise of off-balance sheet activities led risk-based formulae to come into vogue once again.
If one were to take this history literally, one might predict that flat ratios are due for a comeback. Indeed, there are those who advocate just such an approach, as a kind of regulatory backstop. For those whom history teaches that the regulators will never get it right at acceptable cost, the argument is that, in this age of transparency, the markets are fully capable of determining how much capital a given institution should hold.
The odds are, however, that neither straight leverage nor a pure laissez faire approach will hold sway at the end of the day. For one thing, FDICIA institutionalized regulatory capital—using a hybrid of Basle-defined risk weights plus leverage — as a matter of law. Until Congress decides otherwise, regulatory capital with a risk-based component will be an integral part of our overall supervisory strategy. And secondly, as long as that strategy is genuinely harnessed to the concept of risk, it is hard to imagine that we will dispense altogether with risk-weighted capital standards.
So the issue, for all practical purposes, reverts back to the traditional one of supervisory balance — in this case, setting capital standards that accurately reflect the risk they insure against without incurring the kind of burden that is ultimately counterproductive. Some have proposed that we adopt a simplified approach for community banks, in the same way that we have modified our overall examination procedures to make them less burdensome for smaller, healthy banks. Another approach suggests that regulators should rely upon the banks' own internal risk capital allocation models — sometimes known as RAROC — or "risk-adjusted return on capital"— that tells bankers how much risk there is in a particular line of business and how much capital is needed in that business. From such calculations, they can determine how much capital is needed overall. Even so, most experts agree that these models will never capture all unanticipated risks. The problem, of course, is that if capital is to protect against what cannot be anticipated, then standards for its need are necessarily vague because it is impossible to measure them. And some regulators, with indelible memories of the bank and thrift failures of the 1980s and early 1990s, have misgivings about letting financial institutions in effect set their own capital levels. Certainly, before we embrace risk allocation models too enthusiastically, they will have to have met the test of several down cycles in the economy.
What I have tried to do in these brief remarks is to provide some perspective for our discussions today — perspective that points to the importance of maintaining a reliable, substantial and tangible capital base for all institutions, but one which also reflects the genuine risks financial institutions face. Striking the right balance is the challenge we face. With the help of the experts gathered here today, I know it is challenge we will meet.
Public Affairs (202) 649-6870