Chairman Bachus, Chairman Pryce, and members of the Subcommittee on Financial Institutions and Consumer Credit and the Subcommittee on Domestic and International Monetary Policy, Trade, and Technology: It is a pleasure to join my colleagues from the other banking agencies to discuss the current status of Basel II in this country, as well as the Federal Reserve's views on H.R. 1226. The continued discussion among the Congress and the regulators--and, of course, the banking industry and other members of the public--is critical to the final implementation of the new capital accord.

The focus of recent attention has been the agencies' announcement that they will delay the notice of proposed rulemaking (NPR) for Basel II, originally scheduled for midyear 2005. In my remarks today, I will discuss the reasons for the delay and the Board's views regarding the timetable for implementation of Basel II in this country. But, first, I believe it may be useful to remind the members of the subcommittees why the agencies thought it wise to explore and then develop a modernization of the current capital accord; those factors have become, if anything, more important than they were when we began the process.

Our banking system is becoming more concentrated, with a number of very large entities operating across multiple business lines and national boundaries, each entity with positions and exposures that are both complicated and difficult for third parties to understand. These entities have outgrown the current regulatory capital regime, which is still adequate for most banks. But the current rules simply cannot keep up with the complex business of the global banking organizations toward which Basel II and its infrastructure prerequisites are directed. These organizations represent significant risks to the financial system should they develop substantial problems in a period of stress. Basel II offers the opportunity to work with these large entities to develop quantitative risk-measurement and risk-management systems that can both measure their risk more accurately and become the basis for more risk-focused capital requirements and prudential supervision. We would also require, as part of the Basel II approach, more public disclosures to improve market discipline and supplement supervisory efforts.

Many internationally active U.S. banks apparently agree that we should work toward Basel II. Indeed, we and the industry have already seen some benefits from work on Basel II implementation in that market participants and bank supervisors have developed a common language for inputs into risk-management processes.

Earlier this year, twenty-six banking organizations provided us with internal measures of credit risk as part of the fourth quantitative impact study, or QIS4. The agencies have now reviewed the risk parameter estimates provided and are discussing with individual participants their approaches to developing the required inputs. These discussions, which are ongoing, have significantly changed some of the data provided, and some modifications are still coming in.

Nonetheless, even with these revisions, two conclusions are already clear. First, the dispersion among the banks in their estimates of the key parameters that would be used to calculate Basel II capital requirements was quite wide--much wider than expected. Second, the implied reductions in minimum regulatory capital were often substantial--far more than previous quantitative impact studies, both here and abroad, had suggested. As responsible and prudent regulators, we believe it is appropriate to improve our understanding of these results and to see whether changes might be needed in our proposals.

From the outset of our participation in the development of Basel II, the U.S. agencies have clearly and consistently stated that the final adoption of the new capital rules in the United States would occur only after (1) we had reviewed all public comments and incorporated any needed adjustments to address legitimate concerns, and (2) we were satisfied that Basel II was consistent with safe and sound banking in this country. Throughout this process we have stressed that, should we become concerned about the level of overall capital in the banking system or the capital results for individual portfolios, we would seek to modify the framework, including possibly recalibrating the regulatory capital formulas that translate an individual bank's risk parameters into required capital. The agencies' current review and study is consistent with our historical position at Basel.

All of the agencies want to have a better understanding of QIS4 data and results. Does the dispersion reflect different risk profiles? Different model assumptions? Different estimates of risk for the same kind of asset? Different kinds of internal rating systems with some looking "through the cycle" and others being "point in time"? Different stages of institutions' implementation efforts? Limitations of current data bases? Some other factor? We hope that further analysis and discussion with respondents can provide some answers to such questions. All the agencies believed that the prudent approach was to delay the NPR to gain better understanding of the reasons for the unexpected results.

Still, this decision presents the U.S. banking agencies with a dilemma. There is good reason to delay the NPR and related supervisory guidance, but those very documents are needed to provide more complete blueprints for what banks will need in terms of the databases and systems to implement Basel II. We are not saying that these large entities today have inadequate risk-management systems. Rather, they do not yet have the systems for producing Basel II inputs that meet the standards set forth in the Basel II proposal. Until the banking organizations have the NPR, many just will not be able to provide us the inputs we need to assess how banks would operate under Basel II. The dilemma can be solved only by first issuing the NPR. But, we must then have a prudent and flexible way to make adjustments should the resultant data produce results that we, as supervisors, are not comfortable with. The plans developed before the delay in the NPR offer just such an opportunity.

Under those plans, institutions required or planning to move to Basel II would, after the adoption of a final rule, decide when to start their parallel run--with the first opportunity in January 2007. During the required parallel run, each bank would continue to calculate its required capital under the current capital regime and simultaneously calculate its Basel II capital statistics for review by its primary supervisor. When the supervisor believes that the bank has produced four quarters of credible Basel II estimates, the bank would be able to enter a minimum two-year transition run, the earliest in 2008 under these plans. During this transition run, the bank would be under Basel II capital rules, but it could not reduce its capital below 90 percent of what the current capital rules would require in the first year or below 80 percent in the second year. The length of either the parallel run or the transition run could be extended if the primary supervisor had doubts about the bank's Basel II system or the prudence of the resulting minimum regulatory capital level. Only after a minimum two-year transition run and only if its primary supervisor had no objection could a U.S. bank operate fully under Basel II capital rules.

This phase-in plan has been designed to ensure that bank inputs are reasonable and consistent with sound risk-management practices and that supervisors are comfortable with the safety and soundness of Basel II before it goes fully "live" in the United States. Please note that only when we get into the parallel run period will the agencies be able to accurately assess the aggregate capital effects as well as the effects on individual institutions from the new accord. Only then will banks' systems provide risk-parameter inputs that comport with the operational requirements of Basel II, and only then can the U.S. authorities be confident that the resultant capital calculations are reliable estimates of what will happen when Basel II is fully implemented. Such data would be far superior to those obtained through the four QIS exercises that we have conducted to date, which, as I have noted, have been carried out by banks on a best-efforts basis using systems that do not yet meet the standards required under Basel II. Once we have data from the parallel run period, the agencies can then consider the need, if any, for a recalibration of the Basel II parameters or other actions to ensure more-accurate risk sensitivity and a prudent level of overall capital.

This deliberate process provides multiple safeguards to help the agencies move to the final adoption of the new framework in the United States only when doing so is clearly appropriate. In other words, our implementation strategy has been designed to be both prudent and flexible enough to move banks from Basel I to Basel II as their own systems mature and they can provide reasonably accurate assessments of their credit and operational risks. The agencies' analysis of and reaction to QIS4 results show how those safeguards work: We saw results that gave us concern, and so we are investigating further before we go to the next stage. Additional, future safeguards--such as the NPR process and the minimum one-year parallel run and the minimum two-year transition period, with options to extend either--will also ensure ample opportunity to recalibrate or seek other adjustments if necessary.

But, for now, we believe that after a certain point, further analysis of QIS4 is likely to reap little or no additional benefit. We should, of course, try to learn what we can from these data and particularly look for indications of the need to modify the Basel II proposal where necessary. However, as soon as we have learned what we can, we should promptly return to the development of the joint NPR and related supervisory guidance. These documents are essential so that core and opt-in banks can continue to develop the databases and systems that they would need to operate under the Basel II capital rules and that would provide more-accurate risk parameter estimates than those in QIS4.

Recall that there have already been three rounds of U.S. public comments on the Basel II consultative papers between 1999 and 2004; an advance notice of proposed rulemaking (ANPR) in 2003; and numerous agency discussions with congressional committees, banking groups, and individual banks. All have resulted in significant modifications to the proposal. Once published, the NPR and the supervisory guidance will once more elicit comments that could result in further revisions. Particularly given its delayed issuance, the NPR must solicit feedback from core and potential opt-in banks as to whether the current timeline for implementing Basel II in this country needs to be delayed or can be retained.

Looking forward, I agree with my colleagues at this table that it is prudent to delay the NPR in order to see what we can learn from further review of QIS4 data, recognizing at the outset that final answers will not be forthcoming because the requisite databases and risk-management systems are not yet in place. I hope that we can return to the NPR before midyear, present it to the Office of Management and Budget for its review, as our OCC and OTS colleagues must do, and release the NPR in the fall. With such a schedule, one might hope that the parallel running period, currently scheduled for 2007, need not be delayed. But, as I noted earlier, it is important for the agencies to get feedback on this issue during the NPR comment period. The views of banking organizations will provide critical insights into the feasibility of the scheduled 2008 start date for the transition run. Once we have the views of the banking organizations, the agencies will be in a better position to reach a consensus on the timeline.

Basel II has the potential to be an important supervisory step forward. The Basel I framework is being arbitraged aggressively and provides us with less and less reliable measures on which to base a regulatory capital requirement for our largest and most complex banking organizations. Moreover, banks have spent tens of millions of dollars preparing for the U.S. implementation of Basel II and have contracts for further investment. They are awaiting the NPR, the guidance, and the final rules. Their global competitors are proceeding, and U.S. banks will be eager to avoid being placed at a competitive disadvantage. I might add that, as supervisors, we believe that the core risk-measurement and risk-management improvements contained in Basel II are appropriate, regardless of how the future accord is finally structured and calibrated. So it is, in our view, a good idea for banks to continue their current trajectory of making risk-management investments.

While the regulatory capital requirements ultimately produced by Basel II would be, we believe, considerably more risk sensitive than the current capital regime, importantly this is not the only capital regulation under which U.S. institutions would operate. Over a decade ago, the Congress, as part of the Federal Deposit Insurance Corporation Improvement Act's Prompt Corrective Action (PCA) regime, defined a critically undercapitalized insured depository institution by reference to a minimum tangible-equity-to-asset requirement, a leverage ratio. The agencies have also used other leverage ratios to define other PCA capital categories because experience has suggested that there is no substitute for an adequate equity-to-asset ratio, especially for entities that face the moral hazard that accompanies the safety net. The Federal Deposit Insurance Corporation (FDIC), responsible to the Congress for the management of the critical deposit insurance portion of the safety net, has underlined the importance of that minimum leverage ratio and PCA as part of a prudent supervisory regime.

The Federal Reserve concurs in the FDIC's view. We need, for reasons I have given, the risk-measurement and risk-management infrastructure and risk sensitivity of Basel II; but we also need the supplementary assurance of a minimum equity base. The market and the rating agencies will continue to require exactly that kind of base, and a regulatory minimum is prudentially desirable.

Even though the market and the rating agencies, not to mention bank management, will still require banking organizations to carry capital considerably above regulatory minimums, many of the thousands of depository institutions that will remain under the current capital rules are concerned about the impact of Basel II on their businesses. This concern is often voiced as a general disquiet about broad competitive feedbacks but also about competitive implications in specific markets. The Federal Reserve has published a series of research papers investigating such concerns voiced in public comments on the previous ANPR on Basel II. These studies have indeed suggested that there are potential effects that should be addressed in the small business and residential mortgage markets.

For this reason, as well as to continue to modernize the current capital regime, the agencies are developing, simultaneously with the Basel II proposal, a proposal to revise the current capital rules for non-Basel II banks to make those rules more risk sensitive and to blunt any unintended harm that Basel II might impose on non-adopters. Our intention is to keep these proposed changes simple to minimize any costs imposed on the many non-adopters. We plan to issue these proposals for public comment concurrently with or soon after the NPR on Basel II, to allow the banking community to comment on a combined package of proposed changes. However, these revised Basel I rules would not be an adequate substitute for the necessary capital reforms for the large, complex, global banks operating in this country because they would not provide the incentives for banks to adopt the more-sophisticated risk-measurement and risk-management techniques envisioned by the Basel II proposal.

Chairman Bachus and Chairman Pryce, I would also like to present the Federal Reserve's views on H.R. 1226, a bill setting up a committee of the four banking and thrift regulators to reach a common U.S. position on Basel issues and authorizing the Secretary of the Treasury, as its chairman, to determine a common position on any issue about which the regulators could not agree. The Federal Reserve believes that the bill does not fully reflect the existing process used by the four agencies to develop and modify Basel II and we would counsel that Congress not enact it.

Staff members of the four agencies have held frequent and comprehensive discussions about Basel II throughout the process. Certainly, the agencies have sometimes disagreed on specific issues, and we will sometimes disagree in the future. But we have in the past been able to find a common position that we can all support at Basel, and we will do so in the future. The salient fact is that any one of us has a veto over the entire proposal because we all realize that different rules cannot be applied to similarly situated insured depository institutions. That fact forces us to develop consensus positions on which all of us can agree. We have done so in the past because we understand that if the agencies cannot reach a collective agreement at Basel, the Basel II reforms will not be implemented in the United States while they go forward in the rest of the world. Communication, compromise, and comity are the prerequisites for agreement among the agencies.

Further, the ability of the U.S. agencies to negotiate effectively at Basel would be severely constrained if our foreign counterparties knew that we had to return to a committee before we could agree. The formalized "decision by committee" approach of H.R. 1226 would not advance U.S. interests in the complex and dynamic Basel negotiation process. The U.S. banking agencies need to preserve our current flexibility to respond to Basel issues if we are to develop a set of capital rules that are useful and productive for U.S. banks and thrifts. Moreover, it is possible that the agencies are more likely to implement effectively an agreement that they helped shape than they would be one that was imposed on them and for which they did not understand fully the rationale. While we urge the Congress not to move forward on this bill, we look forward to keeping Congress fully informed as the Basel process continues.

I will be pleased to answer your questions.

2005 Testimony