<DOC> [108 Senate Hearings] [From the U.S. Government Printing Office via GPO Access] [DOCID: f:94514.wais] S. Hrg. 108-495 REVIEW OF THE NEW BASEL CAPITAL ACCORD ======================================================================= HEARING before the COMMITTEE ON BANKING,HOUSING,AND URBAN AFFAIRS UNITED STATES SENATE ONE HUNDRED EIGHTH CONGRESS FIRST SESSION ON THE NEW BASEL CAPITAL ACCORD PROPOSAL, WHICH WILL ADDRESS AN AREA WHICH IS IMPORTANT TO THE SAFE AND SOUND FUNCTION OF OUR BANKING SYSTEM __________ JUNE 18, 2003 __________ Printed for the use of the Committee on Banking, Housing, and Urban Affairs U.S. GOVERNMENT PRINTING OFFICE 94-514 WASHINGTON : DC ____________________________________________________________________________ For Sale by the Superintendent of Documents, U.S. Government Printing Office Internet: bookstore.gpo.gov Phone: toll free (866) 512-1800; (202) 512ÿ091800 Fax: (202) 512ÿ092250 Mail: Stop SSOP, Washington, DC 20402ÿ090001 COMMITTEE ON BANKING, HOUSING, AND URBAN AFFAIRS RICHARD C. SHELBY, Alabama, Chairman ROBERT F. BENNETT, Utah PAUL S. SARBANES, Maryland WAYNE ALLARD, Colorado CHRISTOPHER J. DODD, Connecticut MICHAEL B. ENZI, Wyoming TIM JOHNSON, South Dakota CHUCK HAGEL, Nebraska JACK REED, Rhode Island RICK SANTORUM, Pennsylvania CHARLES E. SCHUMER, New York JIM BUNNING, Kentucky EVAN BAYH, Indiana MIKE CRAPO, Idaho ZELL MILLER, Georgia JOHN E. SUNUNU, New Hampshire THOMAS R. CARPER, Delaware ELIZABETH DOLE, North Carolina DEBBIE STABENOW, Michigan LINCOLN D. CHAFEE, Rhode Island JON S. CORZINE, New Jersey Kathleen L. Casey, Staff Director and Counsel Steven B. Harris, Democratic Staff Director and Chief Counsel Mark F. Oesterle, Counsel Martin J. Gruenberg, Democratic Senior Counsel Vincent Meehan, Democratic Counsel Joseph R. Kolinski, Chief Clerk and Computer Systems Administrator George E. Whittle, Editor (ii) C O N T E N T S ---------- WEDNESDAY, JUNE 18, 2003 Page Opening statement of Chairman Shelby............................. 1 Opening statements, comments, or prepared statements of: Senator Sarbanes............................................. 2 Senator Allard............................................... 3 Prepared statement....................................... 53 Senator Reed................................................. 3 Senator Sununu............................................... 3 Senator Corzine.............................................. 21 WITNESSES Roger W. Ferguson, Jr., Vice Chairman, Board of Governors of the Federal Reserve System......................................... 3 Prepared statement........................................... 53 John D. Hawke, Jr., Comptroller of the Currency, U.S. Department of the Treasury................................................ 7 Prepared statement........................................... 71 Donald E. Powell, Chairman, Federal Deposit Insurance Corporation 10 Prepared statement........................................... 83 James E. Gilleran, Director, Office of Thrift Supervision........ 11 Prepared statement........................................... 86 Maurice H. Hartigan, II, President and CEO, RMA--The Risk Management Association......................................... 33 Prepared statement........................................... 90 Micah S. Green, President, The Bond Market Association........... 35 Prepared statement........................................... 92 Edward I. Altman, Max L. Heine Professor of Finance, Leonard N. Stern School of Business, New York University.................. 36 Prepared statement........................................... 96 Karen Shaw Petrou, Managing Partner, Federal Financial Analytics, Inc............................................................ 39 Prepared statement........................................... 98 D. Wilson Ervin, Managing Director, Credit Suisse First Boston, on Behalf of the Financial Services Roundtable................. 41 Prepared statement........................................... 103 Kevin M. Blakely, Executive Vice President and Chief Risk Officer, KeyCorp............................................... 44 Prepared statement........................................... 112 (iii) REVIEW OF THE NEW BASEL CAPITAL ACCORD ---------- WEDNESDAY, JUNE 18, 2003 U.S. Senate, Committee on Banking, Housing, and Urban Affairs, Washington, DC. The Committee met at 10:05 a.m. in room SD-538 of the Dirksen Senate Office Building, Senator Richard C. Shelby (Chairman of the Committee) presiding. OPENING STATEMENT OF CHAIRMAN RICHARD C. SHELBY Chairman Shelby. The hearing will come to order. I want to first thank the witnesses for coming here this morning. The purpose of the hearing today is to consider the New Basel Capital Accord Proposal. This important proposal addresses an area that is extremely important to the safe and sound function of our banking system. With only a quick glance at the economic history books, one can readily determine that thinly capitalized banks pose huge risks to depositors, the banking system, and to the overall economy, and perhaps to the taxpayers. To protect against such risks, we have employed minimum capital requirements as a means to ensure that banks possess the financial integrity necessary to carry on banking activities. What this really means is that bank capital, the bank owner's money, is ``on the line'' with the other bank resources used to conduct business. Thus, bank losses translate to bank owners' losses. By ensuring the sharing of losses amongst bank depositors, creditors, and owners, capital requirements properly align the interests of these groups, and alignment of these interests is crucial. Capital is a very valuable thing. There is tremendous competition for it. Those who provide it expect something in return for it. Those who obtain it must protect it and must make sure it produces. Thus, with their own capital in the breach, banks have developed very sophisticated risk identification, analysis, and management tools to achieve these ends. Ultimately, the combination of capital requirements and risk management techniques have served us well. Today, we are considering proposed changes to the current capital regime, changes which could have very serious effects on the amount of risk-based capital banks are required to hold, on the risk management techniques they employ, and even on the domestic and international competitive landscapes. Because of the significant nature of these issues, I believe this Committee has a responsibility to closely scrutinize the proposal and, at a minimum, become aware of its ramifications so that we can draw our own conclusions regarding its merit. I thank you for being here today, and I look forward to your testimony a little later. Senator Sarbanes. STATEMENT OF SENATOR PAUL S. SARBANES Senator Sarbanes. Thank you very much, Mr. Chairman. First, I want to welcome this distinguished panel of regulators, our line of defense, perhaps, for the Banking Committee this morning. And I want to commend Chairman Shelby for holding this oversight hearing on the Basel Capital Accord which is now being worked on. The New Basel Accord is a highly complex proposal with potentially significant consequences for both the United States and the international financial system, and I think it is appropriate for the Committee to review its likely impact, in fact, before the agreement is actually concluded and the process of implementing it begins. The whole dynamic changes at that point, and if there are concerns about it and what its implications are, I think it is important that they be heard earlier rather than later. Otherwise, we run the risk of examining what has been agreed upon, which is different from examining something that has not yet been agreed upon. The first Basel Accord, concluded in 1988 and fully implemented by 1992, was an effort to establish international standards for the measurement of bank capital in order to bring about greater uniformity of regulation and reduce risk in the international financial system. I think it is generally acknowledged to have been a significant step forward, but there now appears to be agreement that the system for measuring risk under Basel I may be inadequate today, particularly for large, complex financial institutions. In fact, former Federal Reserve Board Governor Laurence Meyer observed, ``Large, complex banking organizations now routinely structure their portfolios in ways that arbitrage around the current capital standard. These banks can often lower their capital requirements with little, if any, reduction in their actual risk taking. As a result, reported capital ratios may, and often do, overstate a bank's true financial strength.'' That is former Fed Board Governor Laurence Meyer. The proposed Basel II Accord is an effort to capture in a more sophisticated way the financial risks undertaken by banking institutions and to assign capital requirements appropriate to those risks. There appears to be agreement about the broad goals of the Basel II Accord, but press accounts report there are significant differences of view among the bank regulators about the agreement itself and, indeed, within the American banking industry. Most of the discussion is focused on whether the proposed Accord would raise or reduce capital; whether it would place one set of U.S. banking institutions at a competitive disadvantage to another; or whether U.S. institutions generally would be placed at a disadvantage relative to foreign institutions; and, finally, whether the agreement is too complex and difficult to implement. These are very important questions. It seems to me appropriate to have a public discussion about them. We need to have some sense of the domestic and international impact of the Accord. One of the key questions that will need to be addressed today is whether the regulators believe they have a sufficient grasp on the impact of the proposed agreement to conclude it internationally and implement it domestically. I must say, having sat on this Committee for a number of years, I would have concerns about any agreement that would significantly reduce the capital held by large financial institutions in terms of the safety and soundness of the system. Mr. Chairman, I look forward to the testimony. Chairman Shelby. Thank you. Senator Allard. COMMENTS OF SENATOR WAYNE ALLARD Senator Allard. Mr. Chairman, I would like to hear from our panel, and to expedite that, I would like to just submit my comments for the record. Chairman Shelby. Without objection, it will be entered into the record. Senator Allard. And I thank them personally for being here, and I look forward to their testimony. Chairman Shelby. Senator Reed. COMMENTS OF SENATOR JACK REED Senator Reed. Mr. Chairman, I think Senator Allard set the right direction. And I look forward to hearing the testimony. Chairman Shelby. Senator Sununu. COMMENTS OF SENATOR JOHN E. SUNUNU Senator Sununu. Thank you, Mr. Chairman. I have no opening statement. I am looking forward to hearing from the witnesses. Chairman Shelby. Our first panel today, we have with us the Honorable Roger W. Ferguson, Jr., who is the Vice Chairman of the Board of Governors of the Federal Reserve System. He is no stranger to this Committee. Welcome, Governor. The Honorable John Hawke, Comptroller of the Currency, who has also spent a lot of time with us. The Honorable Donald Powell, Chairman of the FDIC, who has been here many times. And Honorable James Gilleran, Director of the Office of Thrift Supervision. Gentlemen, we welcome all of you. All of your statements will be made part of the record in their entirety. You proceed as you wish. Governor Ferguson, we will start with you. STATEMENT OF ROGER W. FERGUSON, JR. VICE CHAIRMAN, BOARD OF GOVERNORS OF THE FEDERAL RESERVE SYSTEM Mr. Ferguson. Thank you, Mr. Chairman, and also Senator Sarbanes and Members of the Committee. It is certainly a pleasure to be here before you today on behalf of the Board of Governors to discuss Basel II, which, as you have already indicated, is the evolving New Capital Accord for internationally active banks. I appreciate the fact that the full statement will be made part of the record. Under Basel II, most of the banks in this country will remain under the current capital regime. The operations of the vast majority of U.S. banks do not require the full panoply of sophisticated risk management techniques involved in the advanced versions of Basel II. And the simpler versions of the new accord will not provide our banks, with their current supervisory and disclosure requirements, much additional benefit. We have an entirely different view of our largest and most complicated banking organizations, especially those with significant operations abroad. A very important objective of the proposed Basel II is to continue to promote consistency of capital requirements for banks that compete directly in global markets. And the supervisors of the industrial nations have agreed that banks operating across national boundaries should be under common capital standards to assure a competitive balance, and that standard will soon be Basel II. Supervisors also want to encourage the largest banking organizations in the world to continue to incorporate into their operations the most sophisticated techniques for both the measurement and the management of risk. Substantial difficulty at any one of those entities could have significant effects on global financial markets. In our view, prudential supervisors and central bankers would be remiss if we did not address the evolving complexity of our largest banks and ensure that modern techniques were being used to manage their risks. In the United States, the supervisory agencies will be proposing that to meet these objectives, banks with large foreign exposures and/or banks that are large and complex should be in the set of core banks that would be required to adopt Basel II. And the U.S. supervisors have concluded that these banks should be required to adopt the advanced versions of Basel II, the so-called Advanced Internal Ratings Based, or A-IRB, approach for measuring credit risk and the Advanced Measurement Approaches, or AMA, for measuring operational risk. These approaches, which are described in more detail in my written statement, are the most consistent with best practice risk measurement and risk management. Ten U.S. banks meet the proposed criteria to be core banks and, thus, would be required under our proposal to adopt the A- IRB and the AMA to measure their credit and operational risks, respectively. We would also permit any bank that meets the infrastructure requirements of the A-IRB and the AMA--that is, the ability to quantify and develop the necessary risk parameters on credit exposures and also to develop measurement systems for operational risk exposures--to choose Basel II. We anticipate that about 10 or so large banks now outside of that core group that I have discussed would choose to adopt Basel II in the near term. Thus, in total, we expect about 20 banks to adopt the advanced versions of Basel II before or shortly after implementation date. Now, let me turn to three issues that some have raised, and, in fact, these reflect some of the comments that you two raised, Mr. Chairman and Senator Sarbanes, in your opening comments. The first is competitive equity. While this concern takes several forms, the most frequently voiced is the view that competitive imbalance might result from the so-called ``bifurcated rules'' requiring Basel II for large banks while applying the current capital rules for all other U.S. banks. The fear in this regard is that banks that remain under the current capital rules, with capital charges that are not as risk-sensitive as those in Basel II, might be at a competitive disadvantage compared to the Basel II banks that would get lower capital charges on less risky assets. We take this concern seriously and will be exploring it through the upcoming Advanced Notice of Proposed Rulemaking. But without prejudging the issue, there are some reasons to believe that little, if any, competitive disadvantage will be brought to those banks remaining under the current capital regime. The basic question here is the role of minimum regulatory capital requirements in the determination of the price and the availability of credit. Our understanding of bank pricing is that it starts with the capital allocations that the banks themselves make internally within their own organizations, then factors in explicit recognition of the riskiness of credit, and is then further adjusted on the basis of market conditions and local competition from bank and nonbank sources. In some markets, some banks will be relatively passive price takers. In either case, regulatory capital, which is what Basel II deals with, in particular, regulatory minimum capital, is mostly irrelevant in the pricing decision and, therefore, unlikely to cause competitive disparities. Moreover, most banks, and especially the smaller ones, today hold capital far in excess of the regulatory minimums for a variety of reasons. Thus, changes in their own or their rival's minimum regulatory capital due to Basel II generally would not have much effect on the level of capital that they choose to hold and would, therefore, not necessarily affect internal capital allocations, which are the allocations that drive, in part, pricing decisions. Finally, the banks that most frequently express a fear of being disadvantaged by a bifurcated regulatory regime have for years faced capital arbitrage from larger rivals, who are able to reduce their capital charges by securitizing loans for which the regulatory charge was too high relative to either the market or economic capital charge. The A-IRB approach would provide, in effect, risk-sensitive capital charges for lower- risk assets that are similar to what the larger banks have for years already obtained through capital arbitrage. In short, competitive realities between banks might not change in many markets in which minimum regulatory capital charges would become more explicitly risk sensitive. Let me repeat that I do not mean to dismiss competitive equity concerns at all. Indeed, I hope that the comments on the Advanced Notice of Proposed Rulemaking, the ANPR, might bring forth insights and analyses that respond directly to the issues, particularly the observations I have just made. But, I really must say that, we need to see reasoned analysis and not just assertions. A second area of concern is the proposed Pillar 1 treatment of operational risk. Operational risk refers to losses from failures of systems, controls, or people. Capital charges for such risks have been implicit under Basel I for the last 15 years. These risks will, for the first time, be explicitly subject to capital charges under the Basel II proposal. Operational disruptions have caused banks to suffer huge losses and, in some cases, failure both here and abroad. In an increasingly technologically driven banking system, operational risks have become an even larger share of total risk. Frankly, at some banks, they are probably the dominant risk. To avoid addressing them would be imprudent and would leave a considerable gap in our regulatory system. The AMA for determining capital charges on operational risk is a principles-based approach that would obligate banks to evaluate their own operational risks in a structured but flexible way. Importantly, a bank could reduce its operational risk charge by adopting procedures, systems, and controls that reduce its risk or by shifting the risk to others through measures such as insurance. Some banks for which operational risk is the dominant risk oppose an explicit capital charge and would prefer that operational risk be handled case-by-case through the supervisory review of buffer capital rather than be subject to an explicit regulatory capital charge. The Federal Reserve believes that would be a mistake because it would greatly reduce the transparency of risk and capital that is such an important part of Basel II, and it would make it very difficult to treat risks comparably across banks. The third concern I would like to discuss is the fear that the combination of credit and operational risk capital charges for those U.S. banks that are under Basel II would decline too much for prudent supervisory purposes. Speaking for the Federal Reserve Board, let me underline that we could not support a final Basel II that we felt caused capital to decline to unsafe and unsound levels at the largest banks. There will be several stages before final implementation at which resulting capital levels can and will be evaluated. At any of those stages, if the evidence suggested that capital were declining too much, the Federal Reserve Board would insist that Basel II be adjusted or recalibrated, regardless of the difficulties with bankers here or abroad or with supervisors in other countries. But let us keep this in mind. Supervisors can maintain the same level of average capital in the banking industry, either by requiring each bank to maintain its current Basel I capital level or by recognizing that there will be divergent levels among banks dictated by different risk profiles. To go through the process of devising a more risk-sensitive capital framework, just to end, bank-by-bank with the same Basel I results, I think would be pointless. There will be some greater dispersion and greater dispersion in required capital ratios, if that reflects underlying risk, is an objective and not a problem to be overcome. Of course, I should add that capital ratios are not the sole consideration. The improved risk measurement and management and its integration into the supervisory system under Basel II are also critical to ensure the safety and soundness of the banking system. We are now in the middle of the comment period for the third Basel Consultative Paper, and next month we will begin our comment period on the agencies' Advanced Notice of Proposed Rulemaking. The comments on the domestic rulemaking as well as on the third Consultative Paper will be critical in developing the negotiating position of the U.S. agencies and highlighting the need for any potential modifications in the proposal. The U.S. agencies are committed to careful and considered review of the comments received. The record already underlines that comments and dialogue with bankers has had a substantive impact on the Basel II proposal, and that will continue. But at this stage of the proposal, comments that are based on evidence and analysis are most likely to be effective. In conclusion, the Basel II framework is the product of extensive multiyear dialogues with the banking industry regarding evolving ``best practice'' risk management techniques in every significant area of banking activity. Accordingly, by aligning supervision and regulation with these techniques, it provides a great step forward in protecting our financial system and that of other nations to the benefit of our citizens. We now face three choices: we can reject Basel II, we can delay Basel II as an indirect way of sidetracking it, or we can continue the domestic and international process using the public comment and implementation process to make whatever changes are necessary to make Basel II work effectively and efficiently. The first two options require staying with Basel I, which is not a viable option for our largest banks. The third option recognizes that an international capital framework is in our self-interest since our institutions are the major beneficiary of a sound international financial system. The Fed strongly supports that third option. I will be happy to respond to your questions. Chairman Shelby. Mr. Hawke. STATEMENT OF JOHN D. HAWKE, JR. COMPTROLLER OF THE CURRENCY U.S. DEPARTMENT OF THE TREASURY Mr. Hawke. Chairman Shelby, Senator Sarbanes, and Members of the Committee, thank you for inviting the Office of the Comptroller of the Currency to participate in this important hearing. I want to assure the Committee that the OCC, which has the sole statutory responsibility for promulgating capital regulations for national banks, will not sign off on a final Basel II framework for U.S. banks until we have determined, through our domestic rulemaking process, that any changes to our domestic capital regulations are practical, effective, and in the best interests of the U.S. public and our banking system. My written testimony provides a detailed discussion of the background and content of Basel II, and I appreciate, Mr. Chairman, that it will be included in the record in its entirety. It discusses the important issues with which this Committee is properly concerned. I would like to use this time before the Committee today to make four important points that may help to put today's testimony in proper focus. First, all of the U.S. banking agencies share a concern about the potential effect of Basel II on the capital levels of large U.S. banks. Our banking system has performed remarkably well in difficult economic conditions in recent years. I believe that is due in substantial part to the strong capital position our banks have maintained. While a more risk-sensitive system of capital calculation might be expected to have the effect of reducing the capital of some banks, we would not be comfortable if the consequence of Basel II were to bring about very large decreases in required minimum capital levels. By the same token, if Basel II were to threaten significant increases in the capital of some banks, it could undermine support for the proposal and might threaten the competitiveness of those banks. As things stand today, we simply do not have sufficiently reliable information on the effect of these proposals on individual institutions or on the banking industry as a whole. Before we can make a valid assessment of whether the results are appropriate and acceptable, we have to know, to a much greater degree of reliability than we now have, just what the results of Basel II will be. The OCC believes that significant additional quantitative impact analysis will be necessary. Ideally, this should take the form of another study by the Basel Committee itself. But even if the Basel Committee does not undertake such an additional study, I believe that it is absolutely essential that the U.S. agencies make such an assessment prior to the adoption of final implementing regulations. I strongly believe that we cannot responsibly adopt final rules implementing Basel II until we have not only determined with a high degree of reliability what the impact will be on the capital of our banks, but have made the judgment that the impact is acceptable and conducive to the maintenance of a safe and sound banking system in the United States. I believe all of the U.S. banking agencies share that objective, and we expect to work closely together to resolve any open issues. Second, some have perceived there to be significant differences among the U.S. banking agencies in their approach to Basel II and have suggested that some external mechanism is needed to resolve such differences. I do not think that is a correct conclusion. On the contrary, I believe the agencies have worked exceedingly well together on this project for the past 4 years and will continue to do so. To be sure, we have not always agreed on every one of the multitude of complex issues that Basel II has presented, but that is no more than one would reasonably expect when a group of experts have brought their individual perspectives to bear on difficult issues. Where there have been differences, we have worked our way through them in a highly professional and collaborative manner. The Advance Notice of Proposed Rulemaking (ANPR) for implementation of Basel II in the United States that the agencies will soon jointly issue is another example of a highly collegial and collaborative process. Our staffs have been laboring together diligently to get us prepared for this first round of rulemaking. In addition, we are now in the final stages of internal review on draft interagency guidance that we will jointly issue concurrently with the ANPR to clarify and elaborate on our expectations for those of our banks that will be subject to Basel II, and that guidance has been developed in a process in which every agency had substantial input. While reaching agreement on some of the proposed requirements was no small feat, I believe that every agency will concur with the outcome. Considerable consultation and deliberation still lie ahead before we can even consider final adoption of the implementing regulations. But I have every confidence that the agencies will continue to approach the issues in the same constructive and cooperative spirit that has prevailed up to now. Third, as I said earlier, I believe we are all committed to a process that has real integrity to it. The current Basel Committee timeline presents a daunting challenge to both the U.S. banking agencies and to the banking industry. While it is clearly necessary to address the acknowledged deficiencies in the current Basel Accord, the banking agencies must better understand the full range and scale of likely consequences before finalizing any proposal. We have identified in our written testimony the milestones that the agencies must meet under the current Basel II timeline. They include: Basel Committee consideration of comments received by it on its latest Consultative Paper; the issuance of an ANPR and draft supervisory guidance in the United States with a 90-day period for comments, which we expect to issue in mid-July; full consideration of those comments; the issuance of a definitive paper by the Basel Committee; the drafting and issuance for comment in the United States of a proposed regulation implementing the final Basel paper; the conduct of a further quantitative impact study, as I have just mentioned; consideration of the comments received on the NPR; and, finally, the issuance of a definitive U.S. implementing regulation. Each of these steps is critical in a prudential consideration of Basel II in the United States, and the agencies will be working closely together at every step. I anticipate that we will also be working in close communication with committees of the Congress. If we find that our current target implementation of January 1, 2007, is simply not doable--and my personal opinion is that realization of that target may be very difficult--we will take more time. But it is too early to draw that conclusion yet. The important point is that we will take great care not to let the time frame shape the debate. Equally important is that the time frame will be secondary to our responsibility to fully consider all comments received during our notice and comment process. If we determine through this process that changes to the proposal are necessary, we will make those views known to the Basel Committee, and we will not implement proposed revisions until those changes are made. Finally, some have viewed the New Basel II approach as leaving it up to the banks to determine their own minimum capital, or, as some have said, putting the fox in charge of the chicken coop. This is categorically not the case. While a bank's internal models and risk assessment systems will be the starting point for the calculation of capital, bank supervisors will be heavily involved at every stage of the process. We will publish extensive guidance and standards that the banks will have to observe. We will not only validate the models and systems but will assure that they are being applied with integrity. In my view, the bank supervisory system that we have in the United States is unsurpassed anywhere in the world in both its quality and in the intensity with which it is applied, and we are not going to allow Basel II to change that. In fact, if we do not believe at the end of the day that Basel II will enhance the quality and effectiveness of our supervision, we should have serious reservations about proceeding in this direction. Moreover, while Basel II has largely been designed by economists and mathematicians, and while these so-called ``quants'' will play an important role in our oversight of the implementation of Basel II, the role of our traditional bank examiners will continue to be of enormous importance. Such values as asset quality, credit culture, managerial competence, and the adequacy of internal controls cannot be determined by mathematical models or formulas. Nor can many of the risks that banks face be properly evaluated except by the application of seasoned and expert judgment. I can assure you that those national banks covered by Basel II will continue to be closely monitored and supervised by highly qualified and experienced national bank examiners who will continue to have a full-time, on-site presence. The new process will not replace them. It will simply give them even better tools to assess the true nature and measure of the risks confronting the banks for which they are responsible. I am pleased to have had this opportunity to provide our views, Mr. Chairman, on this important initiative, and I would be happy to answer any questions. Chairman Shelby. Mr. Powell. STATEMENT OF DONALD E. POWELL CHAIRMAN, FEDERAL DEPOSIT INSURANCE CORPORATION Chairman Powell. Thank you, Mr. Chairman and Members of the Committee. I appreciate the Committee's interest in the New Basel Capital Accord. I believe that Basel II ranks among the most important pieces of proposed banking regulation in our Nation's history. The FDIC supports the goal of lining up capital regulation with the economic substance of risks that banks take. Basel II encourages a disciplined approach to risk management, and it addresses important weaknesses in our current capital rules. We applaud the intense and prolonged efforts that have been made to address these important issues. We are approaching a crossroads where judgments will need to be made on some critical issues. We have an interagency process and a public comment period to help reach those judgments, and I am confident that our process will result in an appropriate outcome. My written testimony provides a broad overview of some of the critical judgments that will need to be made before the agencies commit to adopt Basel II in the United States. The first key issue is capital adequacy. The Basel II formulas allow, at least in principle, for significant capital reductions. The proposals issued by the Basel Committee specify that after a phase-in period, there would be no floor on the level of risk-based capital that banks would be required to hold. The level of risk-based capital that banks actually hold would depend upon their own internal estimates of risk--validated by their supervisors--and on the demands of the marketplace. It is very difficult to predict the ultimate effect of Basel II on overall bank capital, and we do know that the formulas are forceful tools for affecting risk-based capital requirements. There is no question that Basel formulas would help the regulators differentiate risk. The formulas cannot stand on their own. Banks face other risk besides credit risk and operational risk. Lending behavior can change over time, causing losses to escalate in activities perceived as low risk. The fact is that no one knows what the future holds. For these and other reasons, the FDIC believes that Basel II must be supplemented by the continued application of existing regulatory minimum leverage capital and prompt corrective action requirements. I am very gratified at the support our fellow bank regulators have expressed for this conclusion. We also understand that a leverage ratio alone cannot provide protection without the support of sound risk-based capital rules. It will be necessary to better understand the impact of the proposals on the capital required for specific activities. Maintaining capital adequacy under Basel II would be an ongoing task. Validating banks' internal risk estimates would be a challenge. Doing so consistently across agencies would be a greater challenge for which an interagency process would be needed. The other key issue is competitive equity. Basel II has been expected to provide some degree of regulatory capital relief. The banks that stand to be directly affected by Basel II have expressed strong support for such capital relief. They have expressed concerns where they believe Basel II capital was too high. The key policy question is: What economic benefits and costs would come with changes in regulatory capital requirements? Would the economic benefit of lower-risk-based capital requirements for large banks enhance their competitive posture or accelerate industry consolidation? We recognize there are differences of opinion about the importance of competitive equity issues. That is why we need to pay close attention to the comments we receive on this issue. The agencies received a number of comments on both sides of this issue at a recent industry outreach meeting and this dialogue will continue. In short, the ingredients of the success of Basel II continue to be: Appropriate minimum capital standards; a consistent approach to validating banks' risk estimates; an adequate vetting of competitive issues; and, time to address these and other policy issues as we finalize our views on this Accord. We will continue to work closely with our fellow regulators to work through these important issues and reach the right conclusions. We are committed to evaluating the costs and benefits of the Basel II proposals and their impact on the U.S. banking industry and the safety and soundness of the financial system. Thank you for the opportunity to present the views of the FDIC. Chairman Shelby. Mr. Gilleran. STATEMENT OF JAMES E. GILLERAN DIRECTOR, OFFICE OF THRIFT SUPERVISION Mr. Gilleran. Chairman Shelby, Ranking Member Sarbanes, and Members of the Committee. My fellow regulators have outlined very well, I believe, all of the risks that are inherent in Basel II, so just a couple of personal comments. I think that Basel II has moved the ball along very well in trying to think through the question of risk versus capital, and I think it has been a real contribution for all of us who are dealing with capital all the time. When you look at Basel I, you look at an idea that is very simple. Everyone understands it. It is applied to all, and it has yielded what everybody concludes is capital which seems to be sufficient. In fact, we have come off of just 2 years of almost the best results the financial services industry has ever had, and this year looks very well, so the capital levels do not seem to have impeded the results in the industry. Basel II, while having a step forward in thought process, concludes with a very complicated, detailed system that applies to only a few, and everybody believes will result in lower capital. Therefore, I think that your interest in this subject, and I believe the work that has been done by my fellow regulators, must continue so that we are all very sure that what we are doing in this next step will be safe and sound. Thank you very much. Chairman Shelby. Thank you. Governor Ferguson, first I am going to read an excerpt from the June 4, 2003 Wall Street Journal citing a recent Mercer Oliver Wyman study. ``The new accord will also change the amount of regulatory capital required from banks for different types of lending, meaning that certain types of products such as mortgages, lending to large corporations, and leasing will receive less back-up capital and could potentially be more profitable. Other types of lending, such as project financing and lending to small businesses will require more capital and so could become less profitable. Oliver Wyman expects this will lead banks to move away from products they cannot thrive in.'' Governor, looking at the results of this study as reported in The Wall Street Journal on June 4, I would conclude that the new accord will result in some winners and losers here, so to speak, in different sectors of the economy. In other words, this proposal could produce significant shifts in the economy. Inasmuch, it could be said that we are not merely talking about reform of banking regulation alone, we are talking about in a sense, some people believe, a proposal that involves policy decisions that will have significant macroeconomic effects. What is your take on that? Mr. Ferguson. I would not reach that conclusion based on my understanding of Basel II. Chairman Shelby. Do you believe that there will be winners and losers? There are always winners and---- Mr. Ferguson. By definition we are changing, if this goes through, we are changing the status quo, and there will be those who like that and those who do not. I do not think of any of them as winners and losers, because what we are trying to do under Basel II is to have regulatory capital that better reflects risk, and I see no losers in that process. Chairman Shelby. Do you see small business as a possible loser? This is what had turned this economy for years is the hiring of probably 75 to 80 percent of our people. And if they are not in the play like they have been, this could have a very significant impact on our economy in the future. Mr. Ferguson. I do not think small businesses will be a loser in the sense you are talking about. Chairman Shelby. Why? Mr. Ferguson. The reason is as follows. First, I am not sure that Oliver Wyman had it just right on what is going to happen in terms of risk-based capital for different portfolios. I would set that to one side. These things are still being calibrated, et cetera. Second, and more importantly, what we have seen is that small businesses have found a very large number of sources of capital or lending, some from large banks, some from small banks. None of that is being driven, I believe, by regulatory minimum capital. I think it is driven by a broader range of issues and regulatory minimum is relatively unimportant in that panoply of things that drive the decision of banks to lend to small businesses. Chairman Shelby. Governor, do you have some studies that you can share with the Committee that would back up what you are saying here with some analysis, not just your opinion but some other analysis? Mr. Ferguson. That is a fair question. I do not have studies at this stage. What we will be doing though in the ANPR is asking this question to get a greater sense of the information that you have just talked about. In fact, broadly speaking, the entire issue with respect to competition and different portfolios will be heavily examined under the ANPR, which will give us a better fact base, because I am looking at the same issue you are. As I said in my opening remarks, I do not think, based on my analysis and based on conversation with bankers, that anything other than economic capital and market structure are the kinds of things that drive these decisions. It is not, based on my current analysis, the regulatory minimum that determines pricing or availability. I would also observe that the thing that has been driving small business lending has been frankly more a question of technology, and to some degree, pricing. As I have been on the Board and observed what is happening over the past many years-- -- Chairman Shelby. In what way? Could you explain? Mr. Ferguson. Yes. I will explain both. What has happened in small business is that it has historically been, for many, many years the purview of local bankers, and those community bankers still have a very strong role that they are playing with respect to lending to small businesses. Chairman Shelby. They are the small business lender. Mr. Ferguson. Absolutely. That linkage still seems to hold true. Chairman Shelby. But will it hold true in the future, like my colleague--I know my time is nearly up and we will have another round, but Senator Sarbanes raised that question earlier, alluded to that question. Mr. Ferguson. Right, he alluded to it. Chairman Shelby. It is very important. Mr. Ferguson. It is very important. I believe---- Chairman Shelby. These are not accords that we should even look at in a cursory manner, as the Comptroller of the Currency said. The Basel II Accord has deep and broad ramifications. Mr. Ferguson. That is exactly why we are here today, and that is why we commend you for holding this hearing. I think it will continue to hold true going forward, because I do not believe that banks choose to get into or out of a business based on regulatory minimum capital. Indeed, we are trying to structure regulatory minimum capital so that it does not impact the strategic decisions of a bank to get into or out of a business. Chairman Shelby. My time is up, but I want to get this in, and I will turn to Senator Sarbanes. Governor, who is responsible here? In other words, this is a very basic question, but I think it is an important practical concern, at least of mine, perhaps other Members. I would like to know who is taking the sum total of this proposal, has digested it, and has a firm grasp of its entirety? I mean this is far reaching. Who has? Mr. Ferguson. Personally, I have been working with this for a year and a half. I have been heavily briefed. Chairman Shelby. I know you have. We have talked about it. Mr. Ferguson. I have worked through lots of the details. I believe I have a good grasp of the vast majority of the issues. There are a few issues here that I would want to understand a little better. I think in each one of our agencies, each would give perhaps a different answer. Chairman Shelby. We have some regulators here, the Comptroller, the FDIC Chairman, OTS Director, that voice other concerns, and they should. I would like to know who is ultimately responsible for this and who is accountable for a success or failure. Is it Chairman Greenspan? Is it the Fed? Is it the Comptroller? Is it the FDIC Chairman? Is it the OTS Director? Mr. Ferguson. You have before you a panel of the regulators of the major depository financial institutions. I think we have a collective responsibility, as both Comptroller Hawke and Chairman Powell have indicated, to work very closely together. Yes, these are tough issues and there are places where by definition---- Chairman Shelby. Profound issues, are they not? Mr. Ferguson. Absolutely profound. There are places where we will initially disagree. We will come to a middle ground that we all agree on, that we think is in the best interest of the United States, and we have, I believe, a collective responsibility to you and to the country to understand these implications of these profound changes, and to give you our best judgment on the impact that they will have. That is one of the reasons why I support what the Comptroller said with respect to the need to have ongoing quantitative impact studies so that we have a factual evaluation of the impact that this is likely to have in the U.S. banking industry. Chairman Shelby. We have this group, an ad hoc group of regulators, who have convened in Switzerland over the last few years. They put this together. Ultimately, I believe that this Committee, the Committee of jurisdiction in the Senate; as Senator Sarbanes said, capital is very important. A lot of us have been on this Committee a long time, where we have visited the taxpayer on things where there was not sufficient capital. This is going to lower some capital standards. It is going to concern me and other Members. Mr. Ferguson. As I have said, I would not jump to conclusions yet that it is going to lower capital overall in the industry. I think it will create greater dispersion of capital because there will be those banks that need to have higher regulatory minimum capital, and there will be some that will need to have lower regulatory minimum capital. When this Basel Committee got started in this process, the goal was to keep capital at about the same level as it is now, and thus far, first indications from early quantitative studies suggest that indeed if it lowers capital, it will be a very small amount, maybe about 6 percent or so. But as I said in my opening remarks, and I think all of us would say the same, if we believe, based on looking at the quantitative impact studies that we do going forward, that capital is lowered to levels that are unsafe and unsound, then we will have to go back and renegotiate, we will recalibrate, and I am sure your Committee would want to hear that we will do that. You have heard my commitment that the Fed believes that, and I think that is consistently shared across all the regulators before you. Chairman Shelby. Senator Sarbanes. Senator Sarbanes. Thank you, Mr. Chairman. Am I correct that the schedule for concluding the international agreement in Basel is the end of this year? Mr. Ferguson. That is the current schedule. Senator Sarbanes. Six months away? Mr. Ferguson. That is the current schedule, yes. Senator Sarbanes. We have the extent of concern or question, if you do not want to label it disagreement, amongst the legislators as reflected in their statements this morning, and we are only 6 months out; is that right? Mr. Ferguson. If I can continue to respond for myself, and allow them to speak. I would say, what you have heard from the regulators I would not describe as disagreement. I think we understand what the issues are. Senator Sarbanes. You may disagree on what should be done about them. Mr. Ferguson. I think we will reach an agreement on what should be done. This is certainly profound without question, but I think we will continue as we have to date to have an honest discussion among ourselves of what the options are and to develop a middle ground. I share some of the nature of your concern, and I share what Comptroller Hawke had to say. We have to give ourselves sufficient time in the comment period before we go back and do the final negotiations. Senator Sarbanes. Let me go to some of the basic concepts. The Economist, on March 29 of this year, in an article said, amongst other things, the following: ``American regulators intend to apply the new rules to fewer than a dozen other banks. This is a choker for European regulators who see Basel II like Basel I, as a global standard to be applied to all banks.'' That raises the first question. What is the rationale for applying Basel II to a limited number of banks? Mr. Ferguson. You are addressing that question to me, Senator? Senator Sarbanes. I do not know. Anyone on the panel who wants to answer it. Mr. Hawke. Let me give my colleague a rest. Mr. Ferguson. Thank you. [Laughter.] Mr. Hawke. I think the basic premise, Senator Sarbanes, in Basel I, as in Basel II, was that these accords were intended to apply to internationally active banks in order to achieve competitive equality on the international scene. The Basel Committee has played a valuable role in trying to establish a framework for capital that can be adopted by any banking system anyplace in the world. Senator Sarbanes. Did Basel I apply to all banks or a limited number of banks? Mr. Hawke. We voluntarily decided to apply Basel I to all of our banks. Senator Sarbanes. To all banks. Mr. Hawke. But the premise in Basel I, as in Basel II, was that it was intended to be applied to internationally active banks. And the 10 or 12 banks that we intend to apply it to constitute probably 95 percent or more of the international exposures of U.S. banks. So, by selecting that group, we have, I think, kept faith with the premise of the Basel Committee. We do not see any useful purpose to be served in applying Basel II to the thousands of smaller banks that we have in the United States which are already better capitalized than their larger counterparts, and, I think, significantly more intensively regulated than comparably sized banks anyplace else in the world. Senator Sarbanes. I gather one concern is that this will now enable their larger counterparts to have less capital vis- a-vis the smaller banks. Who determines which banks Basel II will apply, to whom it will apply; who determines that? Mr. Hawke. We have, through a joint interagency process, established guidelines that look both at the size of the bank and the extent of its international exposures, and it was through that process that we jointly selected the standard that resulted in 10 or 12 of our largest banks being included in the mandatory category. Senator Sarbanes. Is it also the case that another 10 or 12 banks can voluntarily subject to the Basel Accord? Mr. Hawke. We did not include any numerical limit. Any bank that can establish that it has the capacity to implement the systems that would be required can apply for regulatory approval to come under Basel II. Senator Sarbanes. So, you can calculate what advantage it will give you and decide to have that apply to you; is that correct? Mr. Hawke. That could be the case, yes. Senator Sarbanes. Well, what kind of regulatory system is that? Let me ask this question. Is it the case that the models to determine the risk will be internal models of the banks themselves? Mr. Hawke. As I mentioned in my statement, Senator Sarbanes, the banks will develop the models, but under very carefully defined guidelines that we will set forth for them, and under intense validation procedures that we will follow, and under continuing supervision by us with respect to the application of those models. Senator Sarbanes. But they will develop the models themselves. Mr. Hawke. The models would be developed by the banks subject to our validation and continuing oversight. Senator Sarbanes. That is a pretty tricky thing, is it not? Mr. Hawke. There are people who have expressed reservations about that, the fox-guarding-the-chicken-coop syndrome, but I do not think that is the case. First, as I said in my statement, the basic nature of bank supervision is not going to change. We are going to have full-time, on-site examiners in all of these large banks, just as we do today. Second, a new breed of people will be coming into the banks--mathematicians and economists who helped construct these rules--and they will be validating the models and helping our examiners to oversee the integrity of the application of the models. While intuitively many of us have had exactly that concern, I do not think this amounts to turning capital calculation over to the banks themselves. Senator Sarbanes. Another concern I have heard is the uncertainty of the impact of the proposed agreement on major U.S. financial institutions. Let me leave aside for a moment now the bifurcation, right, that it is only going to apply to a limited number of financial institutions, with others able to choose to have it apply if they decide it is to their advantage. I understand that three tests have been done to measure the impact of the new agreement on the largest U.S. financial institutions, and that no clear picture of the impact of the agreement on the capital held by the largest U.S. banks has emerged. In fact, and I want to just verify this factually, that the latest test shows the capital of some U.S. institutions rising by 40 percent and the capital at other U.S. institutions declining by 35 percent. Is that correct? Mr. Hawke. I think that is an accurate reflection of what the so-called QIS-3 reported, but I think QIS-3 was severely flawed. That is why we are insisting on a subsequent quantitative impact study that is overseen by the regulators and carefully monitored. That range would give me enormous concern. Senator Sarbanes. I would hope so. Mr. Hawke. At both ends of the spectrum. Mr. Ferguson. May I add one other point here, sir, on that? I think you should also recognize that what you are talking about I think is just the credit risk component of capital. There is also an operational risk component as well to this whole accord, which in most cases I think would end up reducing some of that range that you just talked about. Again, you also want us to ask the question of to what degree, if this is calibrated correctly, are we reflecting underlying risk? I agree with Comptroller Hawke that obviously going forward we will need to have more of these studies to try to continue to get a better handle on these issues that you are raising. Senator Sarbanes. My time is up. Before I close, Mr. Powell and Mr. Gilleran, you may want to add some observations on this discussion we have just been having here with Mr. Hawke and Mr. Ferguson. Mr. Powell. I think, Senator, your comments call to attention a strong view that the FDIC has, and that is, for minimum regulatory capital. That is very important to us at the FDIC. While I can understand that there are those in the marketplace who believe this is an outmoded ideal, I think it has served us well in good times and bad times over the years. I would also make the comment that economic capital and market capital for the last 13 years has exceeded regulatory capital. While these models are wonderful, I think the discussion that we have had here shows some of the inconsistency in them. Now, I believe in them, and we support that model, so we support that capital should be based upon risk, but we strongly support also minimum capital. Senator Sarbanes. Mr. Gilleran. Mr. Gilleran. Senator, I think that we also have probably misjudged the number of financial institutions that will be making application to have Basel II apply to them, and even though initially it is believed that just a few large organizations will apply because of the complexity and the need for computer programming and for a history to be able to demonstrate that the programs work, I believe that there are many consulting firms queuing up out there to be able to provide the data and the support to much smaller institutions so they can make application to use Basel II concepts. So, I believe the regulators will have many more organizations' plans to look at than some expect. Chairman Shelby. Senator Allard. Senator Allard. Thank you, Mr. Chairman. Basel II's third pillar has a requirement that banks not only calculate the risk positions in capital requirements, but it also ensures that the calculations are disclosed for review by the markets. I personally like to have disclosure for the markets because I think the more informed the consumers are, the better. I am curious if you have had any concerns brought to your attention about those disclosure requirements? I know that in a competitive environment banks sometimes have certain information that is proprietary in nature that they do not want their competitors to know. Have you received any comments in that regard? Mr. Ferguson. We have been working this for many years, and in fact this pillar you are referring to, Pillar 3, has evolved over time. It is clearly not intended to have any disclosure of competitively proprietary information. It is intended to have disclosure of information that the market should know. This disclosure, by the way, is part of the answer to some of the questions that have been raised before by Senator Sarbanes, because it is not just simply our judgment and validation of the inputs, but the market will be able to look at the inputs to some degree and make some comparisons under Pillar 3. So, we worked very hard to make sure that the disclosure regime under Pillar 3 avoids the kind of competitive issues that you have talked about, while being sufficient to give the market the information that it needs to help us do the work that we need to do with respect to validations. Mr. Hawke. If I could just add one quick point to that, Senator Allard. One important function of Pillar 3, in addition to providing the market with information, is to provide a basis for supervisors and banks in different countries to help assure that Basel II is being applied even-handedly by supervisors in different countries. That kind of transparency is very important to the integrity of the process. Senator Allard. I would think that would be very important in that regard. Let's say we have passed what you have recommended, and the regulations have become effective and everybody agrees on them. How do you make the transition from a Basel I to a Basel II bank? That is a hypothetical, but I think it is something that would very possibly happen in the future with the consolidation of the banks. Two or three banks could consolidate, and suddenly they are big enough to compete in the international market. How do you transition them into that arena once you have standards in place? What are your plans for doing that? Maybe the Fed is making appropriate plans, or the other members of the panel have some thoughts on that. I am curious as to how you see that happening. Mr. Hawke. Beyond that group of what we call the mandatory Basel banks, the 10 or 12 that we will explicitly subject to Basel, there may be 20 or more other banks that initially or over time will come to us and try to demonstrate that they have the systems capacity and the sophistication to opt in to Basel II. It will start with a judgment on the part of the bank as to whether they think it is to their advantage, either in terms of market perception or for other reasons, to come under the Basel II regime. We will have to determine that they are capable of developing the systems that are going to be necessary and that they have the kind of risk management and risk measurement capacity that will be necessary to deal with the Basel II requirements. We will make that judgment on a bank-by-bank basis. Senator Allard. Mr. Gilleran. Mr. Gilleran. Senator, if a bank is rated under Basel II and it has been deemed to have excess capital then, then it has to decide what to do with it. Do you either pay it out to your shareholders as a dividend? Do you take on additional risk in your portfolio to utilize the capital, or do you acquire other financial institutions? One of the things that may come out of the application of Basel II to only larger banks is the fact that it might provide a number of larger banks with excess capital to do many acquisitions within the industry, and therefore, one of the questions long range is whether or not there is a roll-up in the financial services industry even greater than there has been. This, Senator Shelby, I think has impact on small business lending because if you then have a roll-up of community banks, then you would probably have an impact on loans to small businesses. So that is an impact. On your question on disclosure, Senator Allard, the problem that I have is that the information in connection with Basel II provides a tremendous amount of information. I am unconvinced yet though that all of the disclosures will be really communicating the real true risk in the institution, and I believe that is something that we have to be careful about going forward, whether or not a lot of data is there, but not enough disclosure of the real risks. Senator Allard. I see my time is up. Chairman Shelby. You go ahead, proceed. Senator Allard. Well, just one other thought, and it was brought up by Mr. Hawke in his response. Basel II will require a significant investment in systems and training, particularly in examiners. All of you have admitted that Basel II is much more complicated than Basel I. What difficulties do you see in both getting your examiners trained and acquisition of equipment, if any? Mr. Hawke. That process has been ongoing now for quite a while. Our conventional safety and soundness examiners have been involved in the process. They are intimately involved, both within our agency and on an interagency basis, with the implementation process. So there is a great deal of ground laying that is going on. We are attempting to expand our staff of quants, if you will, to deal with Basel issues as well. We have all invested a lot in the preparations for Basel II. Senator Allard. Thank you, Mr. Chairman. Chairman Shelby. Senator Reed. Senator Reed. Thank you very much, Mr. Chairman, and thank you, gentleman. Let me raise a series of questions. First, one of the key aspects of the Basel II Accord is operational risk and it raises two questions in my mind. First, whether a quantitative approach, a certain level of capital, is the appropriate response or a qualitative approach of actually evaluating the systems that the institution has and making adjustments or directions to the system is the best approach. Mr. Ferguson, Mr. Hawke, might you comment on that, Mr. Powell and Mr. Gilleran. Mr. Ferguson. Well, I will start. We believe, and we have seen some recent evidence, that it is indeed more possible now than it was a few years ago to get a more quantified approach to operational risk. We had at the Federal Reserve Bank of New York, a week ago more or less, a conference that brought in banks from around the world, each one of which disclosed for the first time the approaches they have used, and we were actually quite impressed to see the degree to which progress has been made. What banks are doing these days--this is again, leading edge banks, not all banks but leading edge banks--are using both their internal operational loss information, external databases, scenario analysis, information from some insurance companies, for example, to come up with a very solid quantified approach to operational risk. I can name a few names. We have put the presentations on the website. So, I think indeed our degree of confidence in the ability to do more quantified approaches to operational risk has gone up, and it is a sign of how quickly things have changed in the industry. Obviously, one of the things that is done in that regard is to think about the impact of procedures, systems, backup sites, et cetera, on the probability of an operational problem having an impact on the ongoing day-to-day operations of the bank. So it does involve some management judgment as well. What Basel II calls for is that the approaches to doing the internal assessments of operational risk be things that we can replicate, that they be systematic and not purely judgmental, that they be based on data and facts that we can also evaluate. The good news is, as I have said, things have moved very much in that direction, and we have a pretty high degree--I think all of us, though it might vary somewhat--a reasonably high degree of confidence that indeed systems have moved in the right direction to make this measure of operational risk more quantifiable than it was, let us say, 5 years ago. Senator Reed. Mr. Hawke. Mr. Hawke. Senator, I have participated in the Basel Committee for 4\1/2\ years. Throughout that entire period I argued strenuously that operational risk should be left to supervisory assessment under Pillar 2, a qualitative assessment, because if you believe that operational risk inheres in the strength of an institution's internal controls, that is inherently a qualitative type of judgment. We, working together with the Fed and the other agencies, pursued the development of the Advanced Measurement Approach (AMA), which is a Pillar 1 approach, to the calculation of operational risk, and it has a very high degree of supervisory judgment involved in that process. I think whether we put operational risk under Pillar 1 or Pillar 2 is not a consequential issue any more. I think that even if it were under Pillar 2, we would still have to have a framework for determining what kind of operational risk charge we would apply. I think the AMA approach that we have developed is flexible. It has a lot of supervisory discretion and evaluation built into it, and I think that is where we need to continue to work on perfecting it. Senator Reed. Mr. Powell. Mr. Powell. I just say supervisory oversight--I think the Comptroller said it well--is the key to this whole operational issue. Mr. Gilleran. I believe that the Comptroller is correct, that it really does not matter whether it is in Pillar 1 or Pillar 2, as long as it is recognized. The interesting thing about Basel I is that Basel I implicitly recognizes operational risk, and that the Basel I system has worked pretty well, and has recognized that risk because we have gone through several years of tough economic times, and yet the banking system has held up. Therefore the implicit recognition has been a valid method. Basel II does require this explicit recognition, which is a whole different way of going about it, much more intellectual, but it remains to be seen whether it is better. Senator Reed. Thank you. Chairman Shelby. Senator Corzine. STATEMENT OF SENATOR JON S. CORZINE Senator Corzine. Thank you very much, Mr. Chairman. I welcome the witnesses. I will just say from my own personal experience that this is one of the most difficult topics that needs to be dealt with, and the complexity is almost mind-boggling. I want to understand what the process of the oversight of the self- derived models are. How is that foreseen to be certain? Are you going to certify different outside consultants or is it only going to be an internally driven responsibility of the regulators to look at the models that the organizations are putting together, and how do you see that process? The second question is, I think one of the most talked about and thought about risks are derivative risks. Is there any description on how that process, that you could share with us, that is tractable, that would be considered? And if we have time, I would love to hear what you think the international implications of this are, competitive implications for Americans, financial system on a relative basis, because there are some elements in here on assessment of risk with regard toward legal exposures and other things that are pretty significant and I would like to understand how those fit. But I will go back to the complexity issue. How are we going to assess the processes in a way that it is fair, consistent, in a way that people will feel like the system is not rigged for those that brightest quants? Mr. Hawke. Let me take the first question, Senator. First of all, I can assure you we will not be certifying consultants. That is not our intention at all. The validation process starts with the Basel documents, which have very detailed standards for risk measurement and risk management systems built into them, and for the kinds of models that will pass muster. Senator Corzine. How are those going to be tested? Mr. Hawke. The second step will be the validation of those models by our examiners and experts, who will come into the banks and review the models, review their compliance with the Basel standards and test them out. There will be an ongoing process beyond that where our examiners and experts will look at the application of the models to see how the models are actually working in practice. This is going to be an ongoing process that starts with the articulation of detailed standards by the Basel Committee itself. Let me defer to Governor Ferguson on the derivatives question. Mr. Ferguson. You asked a question on derivatives. You will see, when we come up to the ANPR, that we are going to ask some questions in that regard. The reality is, I think you will hear later, to broaden your question a bit beyond derivatives, to other products that are much more technically intensive, that drive securitization, for example. We have a number of very specific questions in the ANPR to try to get to some of the issues that you are talking about. The reality is that this is one of the areas that has been most complicated. We are continuing to work on it. You will hear from some people on the second panel, a range of issues and concerns. So at this stage I would say it is very much still being focused on, because it is, as you point out, very complicated. Senator Corzine. So with regard to those specific areas, it is still a work in progress. Mr. Ferguson. It is still a bit of a work in progress. We have a proposal, but all of these things are a work in progress, and we are looking for comment. We have gotten feedback from a number of the institutions. We have gotten feedback from the Bond Market Association, for example, with specific questions that we want to ask them to try to get some better input into these matters, because it is something that we still have to work through. This is, by the way, an area in which the United States is the world leader, as you know. We want to make sure that as we go down this path we do not undercut our ability to continue to innovate in this regard, and that we have a much better handle on the risk profiles that are starting to emerge. Mr. Hawke. If I could just add, Senator, we are about to put out in mid-July, jointly, substantial supervisory guidance on this whole internal ratings-based approach, and standards for the models and our expectations for management capacity to deal with those models. That is going out for comment in a month or so. Right now it runs about 130 pages. Senator Corzine. Is operational risk included in your derivative formulation of measurement of risk? Mr. Ferguson. It is not at this point. I think this is an interesting question that we need to raise. We have thought of derivative risk and securitization risk more down the path of credit risk. Obviously, there are some associated legal ramifications that might emerge that would fall into operational risk, but at this stage we are thinking of them separately. I think again it is something we can question. But the reality, as you may have sensed, with respect to operational risk in the areas of systems, regulations, et cetera, is that it is already sufficiently complicated. We have not then thrown in the derivative issues for a reason that I think you would sympathize with and understand. Mr. Powell. I was just going to comment on your earlier question about validating the banks' risk models. I think it is very important for consistency that that process should be interagency, like the same approach that we have with the Shared National Credits, because there obviously could be some potential inconsistencies, so it should be an interagency team effort. Senator Corzine. And is there every intent on having that? Mr. Powell. I think I have shared this with all of the gentleman here at the table--there is a spirit of cooperation here, and the intent would be that there is consistency, yes. Mr. Gilleran. I think we have demonstrated over the last several years the very good cooperative effort through the FFIEC Act and through other regulatory issues that we are capable of cooperatively dealing in most important areas. I think that will happen. I think the question you raise on the international impact, I think goes to the heart of whether or not the international regulation of banks is the same, and the United States clearly, in my view, is the leader in bank regulation. I think that on an ongoing basis, whether or not every country is looking at it the same way that we are, as an important issue. Senator Corzine. Certainly is in the long run a competitive issue. Capital is important actually on how you run your business. So if somebody gets a leg up, whether it is smaller banks versus larger banks or Swiss banks versus United States banks, whoever the primary supervisor is, how they look at those various models can end up potentially leading to disparities in how capital is applied to the various risks that are being taken. I am all for risk-based capital standards. It is just one heck of a complex issue. Mr. Ferguson. May I respond to the international question? We are well aware of the need to keep the regulatory community looking at these issues pretty much in the same way. The Basel Committee has put together a group called the Accord Implementation Group that brings together the regulators to share best practice and findings and try to create that sense of commonality across borders. By definition, no one can guarantee that any process is going to be foolproof, but we have not left this issue unexamined. We have created this entire process to get to some of the issues that you have raised, Senator. Senator Corzine. Thank you. Chairman Shelby. Thank you, Senator. The part of the proposal that will be applied to our banks, as I understand it, relies a great deal on banks' internal models. Do the regulators have the personnel with the requisite expertise to conduct a thorough review of these systems? How can we know that questionable practices or assumptions will be caught by the regulators before the models are in full operation? It is my understanding that modeling generally involves the use and analysis of historical value. Please help us understand the value provided by the use of modeling. Governor? Mr. Ferguson. I will start. I think we should take a step back here because I think there is a point that is being missed. The concepts in Basel II are not things that the regulators thought of independent of the market. I would say, in fact, just the opposite. What Basel II is trying to do at base is catch up with where leading edge banks already are in the way they run their banks. Not all banks are there. But we are trying to reinforce the decisions that banks have already made and to encourage others to adopt some of these techniques. So this is not brand new terra incognita to the banks. They are not starting at ground zero. The second point I make is on models. We have chosen not to have capital under Basel II that is purely model driven. It is true that the banks would provide some inputs into the formulas that we have derived for determining capital. So you should not, Senator, have the impression that somehow or another there is going to be a model that chunks away that no one understands and then a number---- Chairman Shelby. There is a model here. Excuse me. There is a model here though, is it not? Mr. Ferguson. There are approaches and models the banks use to determine some of the inputs, but the regulatory formulas take those inputs and then determine what the capital is going to be. Now, your question with respect to historical data. It is absolutely true that in order to evaluate your sense of a risk on probability of default or a loss given default, one of the inputs will be historical experience. That works extremely well in some portfolios for sure. There may be others in which it does not work as well. Banks will add to that important judgments, for example, stress testing. One of the things the regulators will look at is whether or not indeed a model or an approach to estimating these inputs looks across an entire cycle, so that banks are not just picking up the good times, but also, frankly, the bad times. Chairman Shelby. That is where capital is imported though. Mr. Ferguson. One of the reasons that one wants to have stress testing and a cross-cycle input is to get capital that deals with both the good times and the bad times. I think you should be aware that indeed one of the things we would be looking at is whether or not the data that are being used cut across a broad swath of time and not a short period that may or may not be representative, and also whether or not there is stress testing, for example, to figure out in the more extreme scenario what kind of loss a bank is likely to experience, what kind of exposures it is likely to have, et cetera. You asked an important question that Comptroller Hawke has answered, and I would endorse what he said. The regulators will have to make investments in training and staff. I would add two things that he did not say. One is that this entire process, at least from the standpoint of my agency, has allowed us already to start to build skills, just in the interaction over the last 4 years to do this. And also--again this is the Fed maybe more than the other agencies--we are blessed by already having, because of the nature of what we do, a number of economists already trained in these areas. Some of them have moved over to be the heads of supervision or senior members in supervision. We still have some way to go. The banks are not at ground zero. Frankly, nor are we. By definition, I do not want to overestimate or underplay the amount of investment that we are going to have to undertake. Chairman Shelby. Mr. Hawke, you are the Comptroller. I would like your views on this. This is important. Mr. Hawke. Let me say I envy their unlimited budget. [Laughter.] Chairman Shelby. At the Federal Reserve. Mr. Hawke. At the Federal Reserve. I think it is important to recognize that the use of models is not something new or something that is going to be initiated by this Basel process. Models have been used for years. Since 1997, banks have been using models to measure value-at-risk or market risk, as part of the Basel process. I think we at the OCC have been in the forefront of using economists in the examination process, working with banks and their models. The use of models as measurement tools in the supervision process is not brand new with Basel II. Chairman Shelby. But you cannot just use the model. You have got to have other means too, have you not, other than this model? Mr. Hawke. Oh, absolutely. As I said before, the role for conventional bank examiners will be undiminished. I view Basel II, if it really works, as providing our examiners with better tools to do the job that they already do. Chairman Shelby. We were told that the Long-Term Capital Management used models developed by Nobel prize winning economists and their models failed. So as a comptroller, any model, you will have to be looking at it closely to see if it does the job that you want, right? Mr. Hawke. I can assure you, we will not hire any Nobel prize winners at the OCC. [Laughter.] Mr. Hawke. But I think it is also important to recognize that the models in Long-Term Capital Management did not have the kind of supervisory triangulation that we are going to bring to bear in this process. Chairman Shelby. Mr. Powell. Mr. Powell. Senator, I think you bring a valid point. I think models are important. I think models are necessary, and I agree with what my fellow regulators have said. Models have been around for some time, and I think they are useful tools. But I am reminded of the old statement, junk-in, junk-out. I think models are wonderful, but the input data is extremely important. That is where the supervision, and where the oversight of the regulation will be critically important--to look at those estimates going forward. But I think models are very useful. Chairman Shelby. Mr. Gilleran. Mr. Gilleran. Regulators have faced sophistication all along the last decades because of data processing, and we are up to the task I think from a technical point of view, and we will hire others if we need them. But what regulators bring to the table is a gimlet eye, and it is that gimlet eye that is a very important one in the process, and I think that will continue to be the case. Chairman Shelby. From what I understand, this proposal involves taking the internal management practices of the banks, and saying such practices will now largely inform the regulatory standards. Typically--and I will direct this to you, Mr. Hawke--is it not Congress or at least the regulators who develop the policies which banks then implement? Does this proposal turn this equation around? Mr. Hawke. No, I do not think so, Mr. Chairman. We implement the basic policies that Congress sets. Congress has given us at the OCC the task of determining the rules under which the capital for our banks will be determined and evaluating capital adequacy on an ongoing basis. We are not turning over the policymaking function to the banks. The rules will be set in great detail, and there will be very careful oversight. Chairman Shelby. It can probably be said that there are some considerable differences between the markets of all the countries that are participating in the New Basel II Accord. What considerations were made regarding these differences? In this country our banks use very sophisticated methods of securitization. What impact will the proposal have on securitization in the United States? Mr. Hawke. That has been a great concern of mine, not only because U.S. banks do far more in the way of securitization than European banks, but also because national banks in particular are very heavy into the securitization markets. So, we have tried to make sure that the rules that emanate from Basel II on securitization do not discriminate against U.S. banks. There are other aspects of this whole structure that concern me in terms of international comparability and that relate to the nature of supervision. Our large banks, including the 10 to 12 large banks that will be covered by Basel II, have full-time, on-site resident examiners that are there day in and day out. Some of our counterparts on the Basel Committee examine their banks once every 2 years or once every 5 years. There is the potential for disparity here in application just because of the disparity in the nature of the supervisory functions from country to country. Chairman Shelby. Can we, just for the record, definitively establish the number and perhaps even the specific entities which will be required to comply with this proposal? Is the number you cited in your testimony, up to 20 banks definitive? I recall that Basel I was intended to apply only to internationally active banks, but now covers all U.S. banks. Do you want to take that, Governor Ferguson? Mr. Ferguson. Well, I will start and my colleagues will chime in and correct me if I am wrong. The plan, as we have indicated, is to have about 10 banks that will be mandatory banks. We think another 10 or so will want to opt in early on in this process for a variety of reasons. The question you ask puts us into a broader time frame though. The reality is, not in the short term but I would expect over the intermediate term, if indeed Basel II does what it is intended to do, which is to create more risk-sensitive capital which gives a better signal to management of what is going on, more banks will find it in their interest to opt in. That will also occur as the cost of building some of these systems and hiring some of these people comes down. So over time, not in the short term but over the intermediate term, almost regardless of what we as regulators do, I would expect, if this is indeed beneficial in helping banks to run themselves by giving them a better sense of risk, that we will have more banks gradually opt in. We will also, I think, as banks grow, find that we as regulators want more banks to come in because they will reach a larger scale of domestic operations or international operations, but I do not think that is going to be a short-term issue. I think it is more of an intermediate-term issue. Chairman Shelby. Mr. Hawke. Mr. Hawke. Mr. Chairman, the standard by which we would determine which banks are and which are not mandatory is one of the issues that will be addressed in the rulemaking. The standard will take into account both size and international exposure. Right now, our best estimate is that there will be about 10 banks that will be included. Those 10 banks probably account for, as I said, 95 percent of the foreign exposures of U.S. banks, and if another 10 banks were to opt in, we expect that we would cover probably 99 percent of the foreign exposures of U.S. banks. Chairman Shelby. Mr. Powell. Mr. Powell. Senator, may I say one thing about the comment on the role of Congress and the role of the regulator? Basel does not do away with prompt corrective action that Congress implemented during the crisis, and that still will be part of it. Chairman Shelby. Mr. Gilleran. Mr. Gilleran. I must say too that Basel is not going to take away the regulatory responsibility to be able to step up and tell any bank that it feels it is operating in an unsafe and unsound manner to have more capital. So therefore, this is a system for measurement, but it does not take away our regulatory powers or responsibilities. Chairman Shelby. Senator Sarbanes. Senator Sarbanes. That is interesting. I would like to follow up on that comment of Mr. Gilleran right there. Does that mean that if an agreement is concluded, our own regulators could then impose subsequently a higher capital requirement on our banks? Is that the point you were making? Mr. Gilleran. That would be my conclusion if a bank adopts it, but if we feel it has been adopted incorrectly or inappropriately, or we feel our estimation of the risks are such that more capital is required, then it is our responsibility to ask for it and to get it. Senator Sarbanes. Is that right? Mr. Ferguson. Yes, that is right. You also should recognize, as Chairman Powell has indicated, we already have some areas of difference from foreign countries. We have prompt corrective action, for example. We have the leverage ratio. We have the various legal requirements to become a financial holding company which require that subsidiary banks be well capitalized. There is also the market discipline. So there are a number of different tools that we have in the United States, some that are legislative, some that come from us as regulators, that are likely to increase the amount of capital we have far above or somewhat above the regulatory minimums. I should also mention the fact that banks themselves, because they recognize the cyclical ups and downs, already hold, and will continue to hold, a cushion of capital above the regulatory minimum that we are talking about here. Senator Sarbanes. Now I want to address what is required by the regulatory regime. If our regulators determine that what Basel II produces is inadequate, to what extent are they constrained by the agreement? Mr. Ferguson. Should I respond to that? The agreement already has a number of places that have what is called national discretion, where we can accept or reject some approaches. We also clearly have said over time, and will continue to say, that as we get a better sense of the impact of this on capital and as best practice changes over time, we will, if necessary, go back and we will renegotiate. We obviously, by definition, have the right and the authority to make, if you will, a step back from the agreement and to make some unilateral changes if we have to. That would not be the better approach. The better approach, for a variety of reasons, is to make changes in the international context because we want to make sure that we have some control over the way our banks that are operating overseas are treated, and therefore having an international agreement is the better approach. But we have not given up national approaches here if we find that we are having a great deal of difficulty with things that we think are important. Senator Sarbanes. All of this argues for making sure that we get it right in the first place, does it not? Mr. Ferguson. I think it argues for two things. One, obviously, making sure we get it right in the first place. Two, reinforcing the message that I have just given you and that we have also given to our negotiating partners, that as we see things evolve and change, we have the authority and we reserve the right to go back and renegotiate in places where we have to renegotiate, but obviously, we want to do the best we can to get it right the first time. But we should also be mindful, Senator, that we, I think, all believe that Basel I is not sending the correct signals to us, to the market, or to the banks about the capital that is being required. While we want to get this right the first time, recognize that delay is not necessarily holding onto a safe and sound system. Basel I is outdated for our largest institutions, and so it is important for us to continue to work for a common process, but to do it with a certain resolve so that we are not leaving large institutions on a capital framework that is not risk-sensitive and probably not fully reflective of what is going on in those banks. Senator Sarbanes. I guess my concern is that you seem to be very close to the concluding date, but not yet to have either worked out a number of problems that everyone concedes are problems, or to have developed, at least on our side, a full- scale consensus as to what should be done. Mr. Hawke. Let me say, Senator Sarbanes, that there is a lot of process to come here. We will be going out with an Advance Notice of Proposed Rulemaking, which will be the first occasion where we have officially solicited comments from the entire U.S. public, not just the banks that would be affected, on Basel II. The Basel Committee's schedule includes trying to finalize the Basel product by the end of this year. I think that may be too tight a schedule, given the need for us to analyze the comments that come in in response to the ANPR and to feed that back into the Basel process. Following the Basel Committee's decision, we are going to have another quantitative impact study, plus the potential for an economic analysis that might be dictated by an Executive Order that applies to rulemakings that have significant economic impact. We will have the NPR process, which will be another opportunity for public comment. It will not be until all of those things have run their course that we will be in a position to consider whether we are going to adopt the Basel proposal finally. So there is a lot of process, a lot of opportunity for input into this before the end of the day. Mr. Ferguson. May I add two other points to that? The final implementation for this is expected to be the end of 2006, very beginning of 2007. Before that we will have a year of parallel running between Basel II and Basel I, which will give us another chance to see if we are comfortable. Then in the first 2 years there are capital floors that are put in to make sure that if there are going to be reductions, we understand the sources of them and that we are comfortable with them. In some sense, Senator, what we are looking at ultimately, before we have an unfettered Basel II, is many years from now, not next year. So, 2004 will be when we hope we get the basics all run. We still have room to make adjustments. Senator Sarbanes. Once you close the agreement you are in a different framework than before you close the agreement. I made that point earlier, and I just want to repeat it. Mr. Chairman, could I ask one final question? Chairman Shelby. You may proceed. Senator Sarbanes. In her written testimony, which will be on the next panel, Karen Shaw Petrou, who has appeared before this Committee a number of times and given us some very perceptive and helpful testimony, poses a question, ``Whether the complexities in Basel II's advanced models are so daunting that supervisors at home and abroad will not be able to ensure that banks actually comply with the new capital rules.'' Before I put out my question I will just tell you a little story. John Biggs of TIAA-CREF testified before our Committee when we were working on the corporate and accountants responsibility legislation, and he told a story of one of his analysts who came in to see him, his financial analyst, who told him that he just could not see where Enron was making all this money, that he had been over their statements again and again, and he just could not figure it all out. And Biggs says to him, ``Well, if you cannot figure it out, we had better sell it.'' And they sold it. This was a couple of years before Enron took a nosedive and went into bankruptcy. So my question to you is whether you have the necessary expertise and resources to implement Basel II and effectively supervise the banks' internal risk measurements. I mean, everyone says right from the beginning, this is extremely complicated and complex. Only the big banks can handle it, and so forth and so on. That obviously raises a concern about whether the system can be gamed. Where are we on that rather important question? Mr. Hawke. I think that is a very important question, and I think it does present us with some challenges in terms of training and retention of people. I disagree with the suggestion that we are not up to the task of evaluating the models that our banks are going to be using. As I mentioned earlier, Senator Sarbanes---- Senator Sarbanes. Who will evaluate the models that other banks are using internationally, our competitors? Mr. Hawke. Their supervisors will be evaluating those models, and that is certainly a cause for concern because the intensity of the supervisory process in some of the Basel countries is quite different from ours. One of the purposes that the disclosure pillar, Pillar 3, is supposed to serve in this process is to provide as much transparency as we can with respect to this process, so that there can be some cross- checking from country to country, so that we can look at what is going on in other countries and determine the integrity with which the process is being applied. Mr. Powell. Senator, I hope this is the last time I will say this, but I think again you are raising lots of issues on the complexity and the validation issues and the competition. Again, I would just say that that is another reason for regulatory minimum capital. Mr. Ferguson. May I throw in a third if you will? A couple of points I want to make. You are also going to hear on the second panel from one of the bankers here who has a technical orientation by his background, who if I recall his testimony, talks a little bit about a set of models called VAR models, value-at-risk models, which 10 years ago were brand new and went through exactly the same degree of concern and uncertainty, and today we discover are well-established. We understand them. So, yes, this is in some sense relatively new, and I will come back to that point in a minute. But these things do evolve our own understanding and that of the banks does evolve. The second point to make is in some sense we have no choice here. We always can choose what we do, but what this is doing, again, is reflecting what many leading edge banks are already doing. It is reflecting what some of our regulators, some of our supervisors already have to come to grips with. It is making those changes more visible. I agree with you it is certainly important. We should not go into this being naive. We should not go into this thinking that it is easy, but we should not refuse to go into it because it is new, because indeed it is already happening in many of our leading edge banks. I will not underestimate the importance of the complexity and the degree of concern and the caution that you are raising, Senator, because they are extremely well thought out, obviously. But I do not want to leave the impression that we are stepping to terra completely incognita. We are, I think as regulators, doing what many of the leading edge banks are doing, we are reflecting where the well-run have already gone. You talked about the complexity allowing some gaming of Basel II. The reality is that one of the reasons we are going down this path is that--it is a quote that you heard from Larry Meyer--there is, if you will, gaming--I know ``gaming'' sounds much too pejorative--there is capital arbitrage already going on. One of the things we are trying to do is confront that directly by getting the capital to reflect the risk because we now have a system in which capital may well not be reflecting risk. Yes, this is complex for sure, but we are not going to be naive about going into this. We are not running into it with our eyes closed, if you will. But we also know that in some ways we are doing only what the market has already started to do. Mr. Gilleran. May I speak on this? Senator Sarbanes. Yes, sir. Mr. Gilleran. Senator, I believe that we will have the people necessary to do probably the finest regulatory job that can be done in connection with Basel II. In answer to your question, any time that you would make a system variable based upon the institution's own determinations of risk, that you are going to have a system where mistakes will be made, either intentionally or unintentionally, that has to be dealt with in the regulatory scheme. So any time that you would come up with a system like this, you are going to have to take into consideration that there will be mistakes made. Senator Sarbanes. Thank you. Chairman Shelby. Senator Allard. Senator Allard. I just have one question regarding the origin of needing to update Basel I. Did the Basel II proposal come about because we had financial institutions in this country concerned about the competitive environment, or did it arise from the regulatory community's concern with safety and soundness issues? How is it that the need to change the original Accord came up and was brought to your attention? Mr. Hawke. I think a little bit of both. There was a growing recognition, as Governor Ferguson has said, and as Larry Meyer said in the quote that was read earlier, that the old system was too coarse, that in defining several risk buckets, it was not really accurately reflecting risk. That was becoming evident as time went on. Senator Allard. That was a concern of yours. Mr. Hawke. Yes, Senator. And there was also a concern that we needed a better system of oversight internationally, that Basel I was not being applied in an even-handed way internationally. Senator Allard. Regulators internationally share some of those concerns? Mr. Hawke. Yes. Mr. Ferguson. I would like to add a third reason, which is a sense from the leadership of our largest banks. They were observing that the way they were managing and thinking about internal risk was diverging from what they had to do for capital purposes. While by definition those things may not always be perfectly aligned, I think all of us would recognize that there is probably a benefit to having banks building their basic regulatory capital under our guidance, supervision, validation, et cetera, in a way that is at least consistent with the way that they manage themselves because they should be getting the same signals from their own economic capital models and judgments, and also from what they are doing with respect to regulatory capital, what the minimum regulatory capital is telling them. Senator Allard. Are these large banks concerned about not being able to compete because of the capitalization requirements that we have here in the U.S. as compared to other countries? Mr. Hawke. Well, I have not heard that. I think that, as Governor Ferguson said earlier, regulatory capital requirements generally, frequently lag behind a bank's economic capital. Banks are maintaining higher regulatory capital than the regulatory minimums that are required. Senator Allard. In this country? Mr. Hawke. In this country. Senator Allard. How do U.S. banks view that requirement in regards to competing internationally with other banks that have not had a base in other countries? Mr. Ferguson. I do not think they have seen that recently as a disadvantage. To be fair, when the Basel process started back in 1988, there was some concern that perhaps some banks in some nations were letting their regulatory capital slip to levels that were really unreasonably low. I have not heard recently the kind of concern you are talking about, and the reason is that the banks that we are talking about, the largest internationally active banks around the world, compete with each other, but they also go to the markets for funding. The markets really are, in addition to their internal needs, demanding a certain level of capital. That is, for reasons Comptroller Hawke indicated and other reasons, higher often than the regulatory minimum. So, I had not heard from our banks a sense that they were at a competitive disadvantage because of regulatory differences of this type. That had been true 20 years ago more or less, but that is not one of the incentives that is driving this round of discussion with respect to Basel II. Mr. Hawke. Senator Allard, if I could just take a second to go back to your earlier point. Under Basel I with the very few risk buckets that cover all bank assets, it became clear that within a particular risk bucket, there could be included assets of widely different risk characteristics, and yet they had the same capital charge. One of the things that did was to encourage banks to invest in the riskier assets because there was no differentiation in the capital charge for those assets. So the current approach is to try to more closely match the capital allocation to the risk of a particular asset. Senator Allard. I see. Thank you, Mr. Chairman. Chairman Shelby. Thank you. Obviously, we have, and you have, a lot of unanswered questions. We expect answers here on the Committee, and we feel that is our obligation before things get finalized. In going forward, I hope that you will provide some of the means to the Committee that we will be kept abreast of changes in the proposal. All the Members have raised serious questions here today, and we do not have all the answers yet, and I do not think you do. But we are very interested in this. We appreciate, Governor Ferguson, you representing the Fed today, spent your morning here with us; Comptroller of the Currency, John Hawke; Don Powell, Chairman of the FDIC; James Gilleran, Director of the Office of Thrift Supervision. It has been a long morning. Thank you for your participation again before the Banking Committee. Mr. Powell. Thank you. Mr. Hawke. Thank you, Senator. Mr. Gilleran. Thank you. Mr. Ferguson. Thank you. Chairman Shelby. We will now, although we are in a late morning, going to call the second panel up, Mr. Maurice Hartigan, President and CEO, the Risk Management Association; Mr. Micah Green, President of the Bond Market Association; Professor Edward Altman, Max L. Heine Professor of Finance, Stern School of Business, New York University; Ms. Karen Shaw Petrou, Managing Partner, Federal Financial Analytics; Mr. Wilson Ervin, Managing Director, Strategic Risk Management, Credit Suisse First Boston, and he will be testifying on behalf of the Financial Services Roundtable; and Mr. Kevin Blakely, Executive Vice President and Chief Risk Officer, Key Corporation. All of your written testimony will be made part of the record. As I said, you have been here all morning. You have heard all the regulators. We would like to move this hearing on as quickly as possible. All of your written testimony will be made part of the record in its entirety. Mr. Hartigan, we will start with you. If you can sum up your testimony as brief as you can. Thank you so much. STATEMENT OF MAURICE H. HARTIGAN, II PRESIDENT AND CEO RMA--THE RISK MANAGEMENT ASSOCIATION Mr. Hartigan. Good morning, Mr. Chairman, Senator Sarbanes, and Members of the Committee. Thank you for inviting me to appear before the Committee. I am the President and CEO of RMA, the Risk Management Association. RMA is a member-driven professional association whose sole purpose is to advance the use of sound risk principles in the financial services industry. The main point I want to make to you today is that the New Basel Accord will be a step forward for the United States and world banking industries, provided it is modified as it is being finalized and provided it is implemented flexibly. It will be a step forward because it is directionally correct in improving the risk sensitivity of regulatory minimum capital adequacy standards. But it must be modified to ensure that it is not too conservative, that these are truly minimum and not maximum capital standards, and to ensure that it is not too prescriptive. The 1988 Accord relied solely on a regulatory minimum capital standard. In contrast, the new Accord will be grounded on three principles or ``pillars'' as they are called: Capital requirements, enhanced supervision, and greater disclosure. This alone represents a significant improvement. Nonetheless, we have specific concerns in this area. Pillar 1, which deals with the capital standard itself, must contain assurances that Basel will evolve toward a full models-based approach for credit risk, and it must avoid arbitrary specificity. Pillar 2, which deals with the implementation of the standard through the process of supervision, must allow regulators enough discretion to accommodate the diversity of best practices in risk management today. Pillar 3, which requires increased disclosure in order to provide greater market discipline, must ensure that comparability is meaningful across the varying international accounting regimes. Our research to date suggests that the new Accord, as proposed in the Third Consultative Paper, will require more overall capital than many banks' internal risk-rating systems require today, even though for some banks and some portfolios, the new overall requirement will be somewhat less than under the old Accord. This will often be inappropriate. The new Accord should represent a true minimum capital requirement. For well-run banks in normal times, this implies that regulatory capital levels should be set below a bank's economic capital based on best practice internal risk measurement procedures. Given the newness of the fields of study surrounding credit and operational risk management, it is natural that regulators should be prone to conservatism. But too much capital is just as bad as too little capital. Too much capital will drive down the risk-adjusted rates of return on a particular business line and cause bankers to lend less than they otherwise would and should. This is not a good thing either, for the shareholders of the bank, the loan customers, or the general economy. The next point I wish to make is that the process to reform the 1988 Accord has had a positive impact on the development of risk measurement and management procedures in the financial services industry. Moreover, the dialogue between the industry and its regulators surrounding Basel reform, while not without frustration on both sides, has been useful and productive. While outstanding issues clearly remain, some quite significant, continued discussion with the industry is ongoing, and I would expect this to be the case throughout the reform process, and into the implementation stage as well. Indeed, it may not be possible to resolve a number of specific issues without an active two-way dialogue between regulators and the industry as the implementation process takes place. Further discussion can only help promote innovation and investment in best practices throughout the industry. It is for this reason that the reform process must continue. However, it must be framed as a work in progress. There cannot be a prescribed end state for sound risk management practices. Otherwise, the ink on the new Accord would not be dry before it became obsolete, much like the 1988 Capital Accord. The quantitative analysis supporting sound credit risk measurement and management are still evolving. Many of these emerging practices were born out of the last economic downturn. The resilience of the financial services industry over the past 3 years should not go without comment. Many have credited the industry success to the better risk management practices established over the past decade. I clearly agree. Furthermore, in our own review of Basel II, we find that some of the new requirements are written in a very prescriptive fashion that does not lend itself to allowing individual banks to employ a diversity of best practices. Without such diversity we cannot have continued evolution of best practices, and without evolution we could not have had the improvements in risk measurement that have occurred over the past decade. In the interest of time, I redirect your attention to my writ- ten testimony for a fuller treatment of two more technical points which are of great importance, in essence, that the internal risk ratings-based approach must be followed with a full internal models approach to capital; and second, capital is required only for unexpected loss known as UL, not for expected loss, as the current version of the Accord argues. RMA has additional technical concerns specific to the Third Consultative Paper which we will address in the formal response. To conclude, I would like to reiterate RMA's belief that the reform process has helped advance the practice of sound risk measurement and management within the industry. RMA is hopeful that the New Capital Accord can be structured to encourage and enhance continued industry innovation and that it will recognize the benefit that diversity of practice within the industry provides. Thank you, and I would be happy to answer any questions that you may have. Chairman Shelby. Thank you. Mr. Green. STATEMENT OF MICAH S. GREEN, PRESIDENT THE BOND MARKET ASSOCIATION Mr. Green. Thank you, Chairman Shelby and Members of the Committee, for the opportunity to testify today on the Basel Committee on Banking Supervision's proposed New Capital Accords, or Basel II. My name is Micah Green, and I am President of the Bond Market Association, which represents securities firms and banks active in the United States and global bond markets. Association member firms account for at least 95 percent of all bond market activity in the United States, in addition to much of the bond underwriting and trading in the rest of the world. Together with our affiliates, the American and European Securitization Forums, we represent a majoirity of the participants in the growing securi- tization markets in the United States and Europe. The following comments focus only on those issues related to Basel II that are most important to our membership. First let me say the association supports the Basel Committee's overall goal of rationalizing the current risk- based capital regime and aligning regulatory capital requirements more closely with actual credit risk. We are grateful to the Federal Reserve Board and the other banking regulators, in particular, Vice Chairman Roger Ferguson, for working with us to address the issues presented by the proposed capital accord revisions that are important to our membership. We are still concerned, however, that if not amended, Basel II will diminish the economic benefits derived from large and growing sectors of the capital markets, benefits which accrue to consumers, as well as businesses. I will first make one general comment on the direction of Basel II and then focus on the two areas most important to us: Securitization and repurchase, or repos, and securities lending transactions. With regard to Basel II broadly, we believe it is important that this agreement not be viewed as the last word on regulatory capital. Risk management techniques are continually evolving, and the financial markets need a regulatory capital accord that evolves with them. Basel II must, therefore, be crafted in a way that ensures that it can better adapt to changing market products and developments. Ultimately, the global financial community will need to move toward a broader reliance on internal risk models to determine appropriate capital levels. Securitization is the process of converting illiquid financial assets, like loans and other receivables, into securities, which can then be traded in the capital markets. It is a large and growing market with tremendous economic benefits for consumers and businesses. Securitization lowers borrowing costs for consumers and others, improves risk management, and draws new sources of capital to the lending markets. Consumers benefit from these efficiencies with lower interest rates and lower prices. Just to summarize the size of the securitization market, in the United States over the last 5 years it has increased 5- fold, to $2.7 trillion. In Europe, it has increased 20-fold over that same period of time, to a much lower level overall but still $151 billion of outstanding securities. And in Asia, where the marketplace is just getting started in the last 7 years, the marketplace has increased 510-fold, to a level right now of $51 billion, and it is anticipated it will grow quite rapidly. Financial institutions participate in securitization as issuers and investors and as part of the risk management functions. For securitization generally under Basel II, the proposed risk weights for securitization positions held by banks are too high in light of the actual credit risk presented by these products. The proposed rules use unrealistically conservative assumptions that, cumulatively, would require financial institutions to set aside excessive levels of capital. And considering who ultimately benefits from a vibrant securitization market, consumers of homes, car buyers, or other people who need capital, this is very important. Repo and securities lending transactions, although little known outside the wholesale financial markets, are vital to our capital markets' liquidity and efficiency. Repo and securities lending transactions allow market participants to finance and hedge trading positions safely, cheaply, and efficiently. It is utilized by elements of the Government. The Federal Reserve, in fact, uses this marketplace to implement its monetary policy. Basel II may require banks to take capital charges inconsistent with the actual level of risks present in repo and securities lending transactions. Financial institutions should have greater flexibility to employ supervisory-approved internal risk models created to assess counterparty risk in order to accurately reflect risks present in these transactions. We agree completely that the current regulatory scheme for bank capital, Basel I, needs significant revision. The current regulations are outdated and inflexible, as you have heard before. Updating the regime can produce significant benefits, including the promotion of fair global competition, incentives for better internal risk management, and an economically efficient allocation of capital. Getting it wrong, however, and implementing capital regulations which do not reflect modern practices or true credit risks on balance sheets will diminish or eliminate the market efficiencies. The Basel Committee is on the right track. We need them to improve it even more, and we look forward to working with them in this process. Thank you again, Mr. Chairman. Chairman Shelby. Professor Altman. STATEMENT OF EDWARD I. ALTMAN MAX L. HEINE PROFESSOR OF FINANCE LEONARD N. STERN SCHOOL OF BUSINESS NEW YORK UNIVERSITY Dr. Altman. Thank you very much, Senator Shelby, for inviting me here today. I have followed the Basel II's consultative papers since the first one was issued in June 1999. A colleague and I have written several commentaries to them, particularly with respect to Pillar 1, the capital adequacy based on specific risk characteristics of bank counterparties. Our major comments were that capital requirements related to expected and unexpected losses from corporate and other loans should be based on actual historical experience of the Loss Given Default from the corporate bond and bank loan markets. The original 1999 suggestions bore absolutely no resemblance to real-world experience. However, they have made significant modifications since 1999, and I am pleased to say that the current revision does a much better job of relating the requirements to default experience, although in my opinion still too little capital is being required for the most risky categories, and probably too much capital for the least risky categories. A problem with the suggested regulations, however, is the complexity in determining capital requirements and the somewhat arbitrary choice of modifications to the standardized scale due to such items as the size of the counterparty and the existence, or not, of collateral on the loan or the bond. For example, Senator Shelby, you mentioned before small and medium- sized enterprises as a very important part of our system. Chairman Shelby. Do you agree with that? Mr. Altman. Absolutely. Chairman Shelby. Okay. Mr. Altman. Absolutely. The Basel Accord under the current recommendations will give lower capital requirements for comparable risk levels for as much as 25 to 50 percent less capital for SME's than larger counterparties. The argument that the correlation of default rates among these small counterparties is lower than for larger corporations may be valid, but I have seen little evidence that the haircut for SME's for these loans should be as much as 50 percent. In my opinion, this was a concession to those national banking systems of the world whereby SME's are the vast majority of borrowers; hence, lower capital requirements for banks in those countries. It is also true, however, that SME's make up the vast majority of loan assets of smaller banks in the United States and the same lower capital requirements would hold for U.S. SME's and the banks that make these loans, close to all but 100 of our Nation's 8,000 banks. But, as I will now discuss, almost all of U.S. banks will not be required to follow the recommendations of Basel II. So the reduced capital requirements for SME's will not be relevant and the old Basel I 8-percent rule will probably still be in effect for all but the very largest U.S. banks. In other words, SME's in other countries will be advantaged vis-a-vis the United States. As I indicated above and as you probably all are aware by now, and you heard the regulators this morning, the central banks of the world and other bank regulatory bodies set national regulatory policy based on Basel's recommendations, but they do not have to accept what Basel suggests. Indeed, it came as an enormous surprise to some observers, including this writer, that only the largest 10 U.S. banks, and perhaps the next 10 to 20 banks in terms of asset size and other requirements, would be required to conform and the next 10 to 20 having the option, to opt into Basel II, and for all other banks, Basel I still remains. In other words, the ``bad wine'' that Basel I might have been drinking in 1988 is still going to be drunk in the coming years by all but 10, maybe 20, banks in the United States. While it is true that as much as two-thirds of all bank assets are held by the top 30 U.S. banks and more than 95 percent of the foreign exposures of these banks will be covered under Basel II, it is likely that all the rest of our banks, almost 8,000, will not be asked to conform and will probably not do so for many of the reasons you heard this morning: Basel II is too complex and too costly; the U.S. banking system is presently more than adequately capitalized; the added Basel II capital required for operating risk is based on highly arbitrary and extremely difficult-to-measure variables; and, finally, the Federal Reserve System's and other regulatory bodies' maximum leverage ratios and prompt corrective action have worked very well. In other words, if it ain't broke, don't fix it. I believe that the choice of only the 10 largest commercial banks to conform to Basel II and the IRB approaches is unfortunate and should be reconsidered. Notwithstanding, the recent consolidation movement of many of our largest and most sophisticated banks, the possible exemption of number 11 to number 30, including such large banks as HSBC Bank, Citibank [West], Bank of New York, Key Bank, which you will hear later from, State Street Bank, number 11 to 15, and the very likely exemption of number 31 to 50, including such seemingly large banks as Charter One, Am South, Union Bank of California, Mellon Bank, and Northern Trust, to name just a few, seems arbitrary and belittles the possible sophistication and motivation of these banks which would be substan- tial institutions in most other countries of the world. For example, the 50th largest bank in the United States in terms of assets, Compass Bank, $24 billion in assets, or in terms of deposits Mellon Bank, would be huge institutions in most countries. They do not have to conform. The choice of a round number like 10 would seem to be insensitive to world opinion, as well as to the risk management motivation. Speaking from an economic standpoint, rather than a political one, I would prefer to see either no banks be required or some exemption level whereby the costs/benefits to our banking system would be more rationally presented and defended, not just the fact that they are internationally active or not. Certainly, a number like the top 50 to 100 banks would be much more in line with the number of banks conforming in other countries. Our largest banks are probably relatively happy to conform to Basel II even with its complexity and added costs to develop the systems and models we have been talking about. Some of those models, by the way, I helped develop a long time ago, and I am very happy to see that they are getting the play that they are. The reason is that they expect total capital required for credit assets will be less than what is required under the current regime. So we may have a new regulatory regime where everyone, large and small banks, as well as our bank regulators, are relatively pleased with the changes recommended. That does not necessarily mean that it is good legislation. I have always felt that despite its problems with complexity, too low capital requirements for risky counterparties, and the difficulty in managing against operating risks, Basel II had one extremely important by-product, at least one: The motivation for banks to develop or improve upon their existing credit-scoring models and systems to reduce total losses from nonperforming and eventually charged-off loans. These systems can be used to rate and set capital for all bank customers rather than using a ``one-size-fits-all,'' the 8-percent rule which is now in effect. I have observed the enormous strides achieved by banks throughout the world--mostly outside the United States, I might add--including ones of all size and location, as to developing risk management systems and training of personnel to prepare for Basel II. Indeed, from what I can surmise, banks in most countries, especially in the European Union, will all have to adhere to Basel II's Standardized, Foundation, or Advanced IRB approaches. Granted that regulators in these countries will need to sanction far fewer banks than U.S. regulators would have to do if all banks are mandated to conform, it must have come as a surprise, perhaps even a resentful shock, that the vast majority of U.S. banks will not adhere to Basel II. This is especially true since the United States and its representatives to the BIS were the early champions of the need to change the way banks allocate capital for credit risk of their clients. What is disappointing to me is that the Fed's decision to exempt all but the largest banks from building and implementing IRB approaches, et cetera, for risk management systems will demotivate the rest of the banks to do so. Under Basel I, they did not have to do so. They are not going to have to do so in the future, although some will opt to do so, as the regulators said. I am very sensitive to the problems in Basel II and what the regulators are faced with. But, in conclusion, what I would like to say is that our decision to exempt smaller banks from Basel II may backfire if many of the world's smaller, or even larger, banks in other countries decide also to opt out of the system because of what we do. This may cause an international problem in banking, which can affect us due to a contagion effect. I have other comments on the procyclicality issue, but in the interest of time, I would like to just conclude and thank you again. I am certainly happy to answer any questions. Chairman Shelby. Thank you. Ms. Petrou. STATEMENT OF KAREN SHAW PETROU MANAGING PARTNER FEDERAL FINANCIAL ANALYTICS, INC. Ms. Petrou. Thank you very much, Mr. Chairman, Senator Sarbanes, and the Committee as a whole. I am Karen Shaw Petrou, Managing Partner of Federal Financial Analytics, which is a firm that advises on the strategic impact of U.S. legislative, regulatory, and policy events like the Basel Accord. I also serve as Executive Director of a group called the Financial Guardian Group, which represents those U.S. banks most particularly concerned with the proposed new capital charge for operational risk-based capital. I thank Senator Sarbanes for mentioning that I have been here before. Through the 1980's, as a matter of fact, when this Committee spent a tremendous amount of time and ultimately had to allocate an awful lot of taxpayer money to rescue the FSLIC and address the savings and loan crisis. My firm then also advised a national commission on the causes of the S&L crisis chartered by Congress in 1989, and we concluded that the predicate cause of the savings and loan debacle was the failure in the early 1980's by the thrift regulators to set appropriate regulatory capital. I am a strong believer in regulatory capital because, at its most simple level, it means that shareholders put their money first. Their money is up before the deposit insurance fund, before the lender of last resort. It is a critical discipline. I must respectfully disagree with Mr. Ferguson in terms of the importance of regulatory capital in driving decisionmaking. It is a profound driver of profit decisions at the most senior level of every financial services firm, bank and nonbank. And as a result, it can have major policy impact. I would point to, for example, the question of Fannie Mae and Freddie Mac, an issue now before this Committee as a result of recent events. Congress and the markets have allowed Fannie and Freddie to operate with regulatory capital somewhere between a half or 20 percent of that which would be required on the same assets if they were held by an insured depository. If you want to look and ask yourself why have these two GSE's doubled in size every 5 years to now hold $3.3 trillion in obligations, are they smarter than everyone else? Maybe. Do they have advantages as GSE's? For sure. But, fundamentally, they can run a lot bigger and maintain tremendous profitability on much smaller amounts of regulatory capital. So these issues are very important, and they will have profound implications for the financial services industry. As Chairman Powell said, Basel II, for all its hundreds of pages and formulas, is a major strategic driver of the competitive direction of our financial services industry and its ability to serve a core customer base in a safe and sound fashion going forward. I would just like to emphasize four points in the interest of time this morning. First, I would suggest that, as Basel II concludes, hopefully quickly, and the U.S. regulators wrestle with their own implementing rules, we focus on first things first. Five years ago, Basel was set in motion to deal with the regulatory arbitrage issue, that is, banks holding high-risk assets when their regulatory capital ratios were too low, and low-risk ones left the banking system because the economic capital, regulatory capital numbers simply did not match. That is arbitrage, politely, or gamesmanship, not so politely. And it is risky. There is a lot in Basel II on which all agree. Some of it is in the more simple standardized models, and, sure, they are not perfect. But they are a lot better than what we have right now, and that part of Basel should move forward quickly because existing ongoing regulatory arbitrage undermines our banking system. I was surprised to see the regulators post on their website a couple of weeks ago the goals of Basel II, and suddenly they now are improving internal risk management, promoting market discipline, and, most mysterious of all, imposing a new operational risk-based capital charge. Two of those three goals make sense, but the underlying purpose of Basel II regulatory arbitrage termination has disappeared, and I would urge refocusing on that. Second, I do not believe that one-size-fits-all capital works. We have imposed the leveraged capital requirement. It is the most primitive of all of those on the table here in the United States because of concern about other sources of risk. But it is a piece of the puzzle that drives low-risk assets out of the banking system. Again, our minimum capital requirements are leverage standards. The way risk-based capital works now is one of the reasons why in recent years the highest-quality, lowest-risk mortgage assets have left our banks and savings associations for the secondary market and highest-risk, subprime loans are now being held by banks. We need a capital system that promotes safety and soundness not through an arbitrary cushion of capital but, rather, a system that rewards good risk-taking and imposes higher capital when higher risks are taken. The operational risk-based capital proposal, in my opinion, has a perverse incentive. It will encourage risk-taking not reduce it. And in the wake of September 11, we have all learned how really serious operational risk can be. And, most importantly, in those tragic days we also learned the value of effective operational risk mitigation, contingency planning, back-up facilities, and in the days thereafter, insurance. The crude capital charge proposed in Basel II will promote operational risk-taking because banks will have to hold an arbitrary amount of capital based on gross income, regardless of the way in which they invest in operational risk management or mitigation. I know the United States regulators have decided, rightly, that that approach is so flawed that we should not impose it here. But maintaining the operational risk capital charge in Basel means that it will still guide major EU and Japanese banks, and we cannot wall ourselves off from their operational risk. We need a good regulatory and supervisory system that rewards appropriate operational risk management and mitigation that can and should be done under good supervision. Finally, I think the challenge of Basel II is just that: Good supervision. We have it here. It does not exist uniformly elsewhere. Pillar 1 in the Basel new paper is about 200-some- odd pages, not counting footnotes. Pillar 2, which is supposed to improve supervision, is about 20 pages. I would rather see them reversed and put our efforts into ensuring that here and abroad, when banks threaten their safety and soundness or pose large economic risks, supervisors intervene quickly and meaningfully and, if necessary, shut the banks down. That does not exist elsewhere, and I think that is a fundamental challenge still left on the table. Thank you very much. Chairman Shelby. Mr. Ervin. STATEMENT OF D. WILSON ERVIN MANAGING DIRECTOR, STRATEGIC RISK MANAGEMENT CREDIT SUISSE FIRST BOSTON ON BEHALF OF THE FINANCIAL SERVICES ROUNDTABLE Mr. Ervin. Good afternoon, and thank you for inviting me here today. My name is Wilson Ervin. I am presenting testimony today on behalf of Credit Suisse First Boston and on behalf of our trade group, the Financial Services Roundtable. CSFB is a major participant in global capital markets employing approximately 20,000 people, mostly here in the United States. We are regulated as a U.S. broker-dealer, a U.S. financial holding company, and also as a Swiss bank. I head CSFB's risk management function. My job is to assess the risks of my bank and protect our capital, a goal that is similar to many of the goals of bank supervisors and at the heart of the Basel reforms. We agree with the importance of bringing the current regime up-to-date and fully support the objectives of Basel II. The regulators who have worked on Basel II have addressed a great many challenging issues with stamina and sophistication. I would like to thank Governor Ferguson, Comptroller Hawke, and FDIC Chairman Powell very much for their openness and willingness to listen during this process. Yet, while there is much to admire in the new rules, there are also many elements that still raise serious concern. On balance, we believe the advantages of the new rules now outweigh the drawbacks, but this balance remains close. This is a frustrating outcome for an initiative with so much potential. Today, I would like to highlight four macro issues that we believe are particularly important: Number one, the current proposal is unnecessarily complex and costly and suffers from an excessive reliance on detailed, prescriptive formulae. Number two, procy- clicality. The new accord could reduce liquidity in the credit markets during economic downturns and potentially deepen economic recessions. Number three, the operational risk charge is highly controversial. And, number four, the disclosure requirements require burdensome additional paperwork, raising costs but adding little information of value to users. The first topic I would like to address is the high cost and prescriptive nature of the new accord. Conceptually, the Committee has attempted to capture current industry best practice and then boil it down into a fixed formula. These new rules, while well-intentioned, will add burdensome qualification, testing, and reporting requirements and will be inconsistent with changing market reality and evolving best practice. The cost of implementing these sys- tems will be very high. We estimate approximately $70 to $100 million in start-up costs for our firm. This increased cost could run into the billions when added up across the whole banking sector globally and could tilt the playing field between banks and non- banks very significantly. The compliance costs of this accord will also be material and will be driven by how these rules are enforced. This is particularly true for international banks, who are regulated by multiple supervisors and subject to the risk of conflicting interpretations. This raises the risk of getting caught in a Catch-22 between regulators. While this problem does exist today to some extent, it will be a lot more important given the number of complex rules under Basel II. The Fed has recently indicated a constructive approach here which we hope represents the start of a broader international effort to resolve this problem. CSFB and other Roundtable members are also concerned about the cumulative effect of the numerous conservative choices that are built into the fabric of Basel II. A good example can be found in the rules for securitizations, which are a common method for financing housing and credit card loans in the United States. The rules proposed for this area include complex formulae and flowcharts with tough controls over business judgment. The combined effect of each of these individual items adds up to regulatory capital requirements that can depart significantly from the true economic capital needs that Basel II is aiming to emulate. The rules seem to be drafted primarily to avoid any possible circumvention, but are likely to deter good financial transactions as well. We believe the securitization rules will tend to raise capital requirements in markets where this technology is most advanced, notably here in the United States. Our suggested response to the problems of prescriptiveness and high cost is for the Basel Committee to place a much greater emphasis on a principles-based approach. Whereas Pillar 1 currently sets out capital calculations in a detailed, prescriptive way, the approach of Pillar 2 is to force the development of better internal models, based on evolving best practice, and then to scrutinize those results through the exam process. The new rules will change the capital requirements for bank lending which can have consequential effects on the state of the economy more broadly. Based on a review of the last 20 years of credit cycles, our calculations indicate that the new rules will require much more bank capital during an economic downturn when compared to the current system. My personal estimate is that my bank would have cut back its lending in these circumstances by perhaps 20 percent if the Basel II rules were in place during the recent recession. If all banks cut back in lending at the same time, as they will tend to do under a common regulatory regime, the potential adverse effect on the real economy could lengthen and deepen an economic slump. The proposed quantification of operational risk, the risk of breakdowns in systems and people, is highly controversial. All of the Roundtable's member companies agree that evaluating and controlling this risk is important and should be required. However, many banks, including CSFB, believe the current approach is deeply flawed because of the difficulties of measuring and predicting this type of risk using a quantitative model. Other banks take an opposite view and believe that the operational risk rules will lead to improved risk governance and better transparency and are, therefore, appropriately placed in the Pillar 1 category. One of the strengths of the Basel II proposals is that they go beyond capital calculations in Pillar 1. They look to improve market discipline via greater disclosure. While we appreciate that the Pillar 3 disclosure requirements have been reduced, they continue to be burdensome and potentially confusing. While we certainly support transparency, as Chairman Greenspan said recently, there is a large difference between more disclosure and more transparency. In sum, much hard work has been put into Basel II, but much also remains ahead. In the pressure to finalize and implement the accord, we hope enough time will be provided for everyone-- banks and supervisors alike--to consider the implications of this new regime. Streamlining the current proposal will require strong discipline in the final round of drafting and a return to the original philosophy underlying the project. I believe very much can be accomplished if we increase the emphasis on best practice prin- ciples rather than rigid formula, and if we increase the weight of Pillar 2. Pillars 2 and 3 have real people behind them: Regulators and the market. People can adapt to changes and new markets more easily than a rulebook can. This also puts the burden back where it should be--on the shoulders of bank management--to demonstrate to regulators and the public that they are doing a good job. That is in the spirit of the Sarbanes-Oxley reforms, and I think it is a smart and durable way to improve discipline. Thank you. Chairman Shelby. Mr. Blakely. STATEMENT OF KEVIN M. BLAKELY EXECUTIVE VICE PRESIDENT AND CHIEF RISK OFFICER KEYCORP Mr. Blakely. Thank you, Mr. Chairman, Senator Sarbanes, and other Members of the Committee. I am here today on behalf of KeyCorp, the 11th largest banking company in the United States KeyCorp has total assets of approximately $85 billion and spans the northern half of the United States from Maine to Alaska. While the vast majority of our business is domestically based, we do have a modest amount of international business activity. KeyCorp is not one of the companies included in the definition of ``the top 10 most internationally active institutions.'' Accordingly, under the present regulatory guidance, we will not be required to comply with Basel II when it becomes effective in 2006. Nonetheless, it is our intent to qualify as an advanced model institution. We simply believe that it is good banking practice to develop the risk management tools that are the foundation of Basel II. If that qualifies us as an advanced model company under the new accord, so much the better. KeyCorp believes that Basel I is broken and that a new accord needs to be implemented. Basel II is a major step forward, and we applaud its approach. It is not perfect now, nor will it be perfect when implemented, nor perfect 10 years after implementation. Regardless, it is light years ahead of Basel I, as well as any other proposal we have seen to date. We acknowledge that it is complex, but banking is a complex business. A simple solution to complex issues is probably not the right medicine. As an industry, we should not shy away from the remedy simply because it is complex. We should work collectively with the regulators to find the right solution, not the easy one. We have our doubts as to the high cost figures attributed to Basel II. Our own experience to date has proved to the contrary. We believe that many of the Basel II costs are simply expenditures we should otherwise be making as a matter of sound banking practice. Good risk management costs money, but it is intended to help avoid even bigger costs that arise from bad risk management. We do not believe the adoption of Basel II will trap the financial services industry in a time warp. It will not stifle the creation of new risk management tools, as some have alleged. Banks will continue to develop better methods of managing risks regardless of what Basel II requires. We believe there is substantial merit to including as much as we can in Pillar 1 versus Pillar 2. One of the greatest benefits that Basel II promises is that it will utilize the invisible hand of the market to discipline wayward institutions. In order to do that, investors must have adequate information to compare the risk of one institution against another on an apples-to-apples comparison basis. Pillar 1 is the best vehicle for ensuring that banks report on a consistent basis. Mr. Chairman, KeyCorp appreciates the opportunity to share our views on Basel II. We want to make sure our industry operates within a safe and sound environment. We know that this is the goal of the Committee, as well as our friends in the regulatory world. While Basel II is far from perfect, it certainly moves us further down the path. Chairman Shelby. Thank you. I think everyone and I hope everyone recognizes that risk management practices must constantly improve. You are in the business. That said, I am not sure there is complete agreement as to what actually leads to improved risk management practices. Is it market forces? Is it the use of detailed and specific capital rules such as those that have been proposed? What kind of impact can using these kinds of rules have on the development of risk management practices in the future? Mr. Ervin. Mr. Ervin. I think all those things are critically important. It is not just regulatory capital calculations that drive innovations in the market and drive innovations in best practice, although I think they can be helpful. Our concern here is that the complexity prescriptiveness of what they are trying to build in Pillar 1 may actually restrict that process, may restrict the evolution of market best practice and may trap us--I disagree with Mr. Blakely on this-- may trap us to some extent---- Chairman Shelby. You are trapped in the model, aren't you? Mr. Ervin. To some extent we are trapped in models today. We are living in an advanced world, and I do not think we can turn back the clock. The question is: Do we have the right market forces and the right ways to keep up with how the world is evolving? Or do we end up having to run two sets of books, one for a potentially outdated regulatory regime and another one for the world as it has come to be evolving? Chairman Shelby. Ms. Petrou, the proposal, as I understand it, requires banks to set aside capital for operational risk, assuming that the potential for these types can be quantified. Is it really possible to measure these risks? And don't we call some of those risks, you know, acts of God for a reason? You have heard that all your life. Do you want to comment? Ms. Petrou. I do not think it is possible at this point to quantify operational risk or to measure it in any way on which anybody agrees. In fact, Basel's own committee, the Risk Management Group, recently after reviewing the data that was gathered in an effort to come up with Basel II, said that the data needed to be used with caution. And another Basel Committee has said that it does not believe that a quantifiable operational risk capital charge is in sight. I believe the proposed ops risk capital charge now is, in essence, a way to top off the credit risk charge to protect against drops in credit risk capital. But when low-risk books of credit risk are there, then I think the capital should drop, and we should not be fudging the books, as the operational risk charge would do. Mr. Ervin. Senator Shelby, if I might expand on that just for a few moments? Chairman Shelby. Yes, you go ahead. Mr. Ervin. I am a quantitative person by nature. I came up through that part of the bank, and I would love to be able to-- -- Chairman Shelby. You are still there, too. [Laughter.] Mr. Ervin. But I would love to crack the operational risk management using quantitative people, hire a bunch of my people in my department and have the problem solved. I am just not convinced we can do that today. And I am concerned that we are building the system still on models that have not been validated, still have not been proven. Chairman Shelby. This is a work in progress in a sense? Mr. Ervin. Absolutely. Chairman Shelby. Professor. Mr. Altman. If I might piggyback on both comments. Chairman Shelby. Go ahead. Mr. Altman. I completely agree with them. I think it was a major mistake to add operating risk to Basel II. I mean, it almost was tacked on as an afterthought rather than the main problem, which was credit risk, which was what Basel II was supposed to handle. And then everyone plowed around as to how to measure this, and I agree completely with both Karen Petrou and Mr. Ervin that this is whistling in the wind. Most of the banks have no idea, and the regulators have less, about operating risk. I think it is plug figure, and it is unfortunate because it detracts from all of Basel II. Nobody likes it. I do not know anybody I have ever met who likes the operating risk part of Basel II. And yet it is in there. I think we should have the guts to say it does not belong in there and get rid of it. With respect to the first question you asked, it might add, one of my comments earlier was if we do not require some opting in by all but the largest banks, we are really going to demotivate risk management at these other institutions. They are going to throw up their hands and say it is too complex, I do not need to change what I have been doing all this time. And just the fact that we have not had any major bank failures lately--we have short memories. Things back in the 1980's were not so good. If we have a lot of stress to the system, I do believe that prompt corrective action on the part of our regulators, which is not part of Basel II, is a great thing we have, and Basel II does not have it. And I would like to see that be put in as well. Mr. Hartigan. Mr. Chairman. Chairman Shelby. Sorry, Mr. Hartigan. Mr. Hartigan. If I could just respond on operational risk, it is a nettlesome issue. It will continue to be. It is something which is not science. Chairman Shelby. While you are on that, would you please touch on the new capital charge for operational risk? That is all involved here. Mr. Hartigan. As the Comptroller of the Currency said, it really matters not whether it is 1 or 2. I think what we must do, what the industry must do, is continue to observe and analyze the data which surrounds operational risk, be more comfortable with it, acknowledge it more, and acknowledge that it does have a role in the capital charge. Chairman Shelby. Do any of you have any concerns about competitive issues associated with the new proposal? We know what the proposal basically is. Ms. Petrou. I would just like to say it has some unique U.S. issues. We have talked about international competitiveness. But it is also very important to get the rules right here because some big financial services firms and some little ones that elect to go in will be in, and some big ones will be out by virtue of their charter choice. And some of the ones that might think they are in can change that charter choice and take them out. In the EU, the proposal is to apply Basel to all financial services firms, but under U.S. law it can only apply to financial holding companies. If the regulatory capital incentives are not better aligned with the economic ones, some significant advantages to perhaps non- bank institutions will result. And that will push assets out of our good supervisory framework. So, I think that is a very troublesome issue that needs careful attention. Chairman Shelby. Professor Altman, I want to direct this question to you and also to Mr. Ervin. Could you elaborate on your statements regarding the procyclical aspects of the proposal? The proposal could increase booms and busts during the economic cycle, some people believe. Mr. Altman. Yes. I skipped over it in the interest of time. Thank you for giving me the opportunity. Procyclicality is considered by some as a very serious issue. Other people, including most of the regulators in the United States, seem to think that it is not that serious of an issue. I personally think that the New Basel II has all the ingredients to increase procyclicality, perhaps dramatically. It exists already. Banks cut back in difficult times, either through credit rationing or---- Chairman Shelby. What could that do to the economy? Mr. Altman. That is what I am getting to. Chairman Shelby. Okay. Mr. Altman. In times of stress, if the banks are motivated through Basel II higher capital requirements on losses and downgrades, then they will lend even less. That is exactly the time that we need it. One of the reasons we have had so many bankruptcies and defaults in the last couple of years is that banks were much too liberal in 1993 through 1998, and that is procyclicality. They were fat and happy. That is good. But they made too many bad loans as a result of being that. So that is the procyclicality. I recommend that we consider a smoothing of the capital requirements, that more capital be required in good times and less capital in bad times. You have got to prompt corrective action anyway to move into the banks if they have got too little capital. So exercise that, but reduce that procyclicality by much more proactive action on the part of the regulators. Chairman Shelby. The bottom line is capital is important. Let's face it. Mr. Altman. Absolutely. Chairman Shelby. Mr. Ervin, do you have a comment? Mr. Ervin. I would support what Professor Altman said. I do believe this could be potentially quite important. When we have looked at our own bank and looked at our own behavior, we think this could significantly affect the number of loans we choose to make. And if you pull liquidity out of the market in the tough times, that will have an effect in the economy. I am not smart enough to know exactly what effect that will have, but we do believe that could be a potentially substantial impact, and it is something that the U.S. regulators in particular need to be aware of and have contingency plans to make sure we do not fall into that trap. Chairman Shelby. Mr. Blakely. Mr. Blakely. Mr. Chairman, if I may, we disagree on the issue of procyclicality because the whole concept of procyclicality depends on the banks' operating at the absolute bare minimum level of capital. I do not think any bank worth its salt is going to be doing that. Most banks will maintain a buffer zone of capital above the bare minimum that should---- Chairman Shelby. Well, banks get nervous during stress time. Mr. Blakely. Banks do get nervous, but banks will not stop lending during a recessionary environment. Lending is the business that most banks are in. It is our lifeline of revenue. Chairman Shelby. Excuse me a minute. They might not quit lending, but they curtail some lending or tighten up on credit. We know that. And that can exacerbate an economic downturn, a cyclical thing, as Dr. Altman references. Is that correct? Mr. Altman. That is correct. Mr. Blakely. I will not deny that banks do curtail in some areas of lending. But, again, banks do need to continue lending in order to keep their revenue stream going. Also during a recessionary environment, you are afforded the opportunity to get better underwriting standards and to get better pricing. There is an old saying that says that the best of loans are made in the worst of times and the worst of loans are made in the best of times. And that is true right now. We do not stop lending. We may become a little bit more cautious, and as long as we have that buffer zone of capital, we will continue to lend when opportunities present themselves. Chairman Shelby. Thank you very much, Senator Sarbanes, for your indulgence. Senator Sarbanes. I will be very brief because it is getting late, and this panel has been with us quite a while. I am a little concerned or perhaps even confused by some of what I have been hearing from this panel. First of all, is there anyone on the panel who disagrees with the statements made by the previous panel that Basel I is deficient and does not constitute a framework within which we should continue to function? That is what I understood the previous panel to say. If you disagree with that, you should correct me. But is there anyone at the table who disagrees with that. Mr. Hartigan. No. Mr. Green. No. Mr. Altman. No. Ms. Petrou. No. Mr. Ervin. No. Mr. Blakely. No. Senator Sarbanes. I think the reporter should indicate that---- [Laughter.] All right. Now, my next question is: Is there anyone at the table who thinks the Basel II process should be terminated? Mr. Hartigan. No. Mr. Green. No Mr. Altman. No. Ms. Petrou. No. Mr. Ervin. No. Mr. Blakely. No. Senator Sarbanes. The question then is: How do we carry forward in the Basel II process to take into account some, if not all, of the concerns, to review the concerns that have been enunciated, and to try to weigh them and see what can be done? Would that be a fair statement? Mr. Hartigan. Could I start? I think that many of these issues that we have discussed today really have to continue to be broken down in a hierarchy of needs, and that some of the issues in the implementation are scalable. I think that the whole system of capital based on risk, and appropriate capital, is where the industry is moving and, indeed, where we are being encouraged to move, where the industry has been encouraged to move by the regulators. I think that the way to do it, Senator Sarbanes, is to continue to test the observations and to make recommendations based on valid tests. It may be that we do not meet the timetable that is in place now. Timetables can be moved. But at the end of the day, the system will be better off for a capital system based on risk. Mr. Green. Senator Sarbanes, I would say that this Committee deserves a good deal of credit for holding this hearing in a timely way before the comment period on the Consultative Paper 3 is complete. The fact is, what you heard here is unanimity in favor of the Basel process and support for what the regulators are trying to get done generally. And yet we all have issues that we would like to raise consistent with our support. By holding the hearing, you have advanced the dialogue of those constructive points of criticism. We came here today to put on your radar screen and to reiterate on the radar screens of the Basel Committee the effects on the securitization market and the repurchase agreement and securities lending market, and yet not be critical of the Basel process. It is not even a fine line. We can be wholly supportive of the Basel process and point out things that still need to be dealt with before it is finalized and fully implemented. So we thank you for what you have done. Senator Sarbanes. Does anyone else want to add anything on that point? Mr. Blakely. I would just like to say that since the initial draft of Basel II came out, there has been tremendous progress that has been made on it, and that has come about through work between the banking institutions themselves, as well as the regulators trying to come together on common ground. As I look back retrospectively and look at the progress that has been made, it is nothing short of phenomenal, and I think by the time Basel II eventually arrives, it will be better, much better than it is now. We do all agree that Basel II is the right direction to go. We just do not necessarily agree on certain aspects of it. But we will eventually get there. Senator Sarbanes. Well, Mr. Chairman, I--sorry, yes. Mr. Ervin. If I may just expand on that, I would agree that it has improved a lot, and I do agree that it is also through hearings like this one that the public airing and the public dialogue can be improved, and that has had a major impact on the quality of the work recently released in CP-3. I do think there is significantly more work to do in balancing between not sticking with an outmoded regime or going to Basel II as it exists today. I think as Professor Altman says, we need to have the guts to take out the parts that do not seem to be working, do not seem to be getting consensus, and focus on the parts that work, get that implemented, and take a more evolutionary approach. I think that will be a smart way to move forward here. Senator Sarbanes. Well, it is an interesting process, and I just wanted to clarify that. I once was the Executive Director of the Charter Revision Commission in Baltimore, and we had to take our proposed charter to the City Council. The director of finance, who was a very influential figure in the city government, came to testify. He was in support of what we were trying to do. But then he got in front of the City Council, and he had a number of pinpointed criticisms to make of our work. And by the end of his first session with the City Council, they were convinced he was against the charter changes. We had to go back, and he had to make very clear that he was basically in favor of the charter changes, and he was just trying to improve it here and there. And I wanted to be clear on that. This Economist article I quoted earlier, illustrates the difficult problem of this international relationship, said, ``For the past 5 years, the world's financial regulators have been working on a new set of rules for bank capital called Basel II. Getting this far has taken a lot of sweat and horse trading. American bankers and regulators have been at the forefront. American financial institutions have debated the rule changes as keenly as anybody. Imagine, therefore, the consternation of other committee members on learning how America plans to treat the new rulebook,'' and then they went on. You know, we have gone far enough down this path that we have, in a sense, that problem on our hands. On the other hand, I do not think that should lead us not to insist on shaping this thing in a way that people can look at it and say it did not make sense. And I do think the regulators need, one, to work to get themselves in alignment and, two, to continue to interact with the private sector to address some of these concerns so we develop a broader and deeper consensus about what we are seeking to do. And in that respect, I think, Mr. Chairman, that it was a very constructive contribution for you to schedule this hearing. Chairman Shelby. Thank you. A couple of quick observations. Mr. Green, has there been any work done on the effects of Basel II on securitization that considers the downstream effects on consumer financing? Mr. Green. Yes, that is our principal level of concern. As they have currently defined regulatory requirements, we believe that they are significantly too severe. And we think it could have an effect on the ability of financial institutions who are very active participants in that securitization market, not just of originators of loans and---- Chairman Shelby. And have an effect on our economy, right? Mr. Green. Absolutely, Mr. Chairman. All this high finance and complexity really does boil down to how does it affect real people. And the securitization market is one of those examples as to how real people would be affected, whether you are buying a home, refinancing a home, buying a car, have credit card debt, or just simply need capital. That $2.7 trillion, $151 billion, and $51 billion in the United States, Europe, and Asia of a market---- Chairman Shelby. That has to be addressed, doesn't it? Mr. Green. Absolutely. Chairman Shelby. If the Fed does not pull operational risk, should we address it here in Congress, Dr. Altman? Should we address operational risk here? Mr. Altman. I am not sure of the proper forum for that. I am not sure also that capital will--more capital will obviate the possibility of massive losses due to fraud or some other types of activity which is under operational risk. You put another 0.6 percent of capital for operating risk, but then you have got a rogue trader or some fraud action, and you have got an institution being brought down, regardless of that other 0.6 percent. Chairman Shelby. It has happened. Mr. Altman. Yes, exactly, and it is going to happen again. It has to happen again because that is the nature of fraud. I do not know when it is going to happen, and I do not think this extra 0.6 percent capital is going to do anything with respect to operating risk. Chairman Shelby. Ms. Petrou, you know a lot about capital. What do you think? Do you agree with him or disagree? Ms. Petrou. Yes, I do. Chairman Shelby. You agree? Ms. Petrou. Yes, I think it is a meaningless charge, and to your question of where Congress should go, I think what you and on the House side as well, these hearings are focusing attention back on the policy issues that Basel raises. When you have rules that are hundreds of pages long with the kinds of formulas as Mr. Ervin has said, all of us at this table and everywhere, we get into these ``how do we do it'' debates, and the ``should we do it'' issues get lost. And I think you are helping everyone focus back on the ``should it be done'' issue, which is of paramount importance. Chairman Shelby. Well, Senator Sarbanes and other Members of the Committee have thought this is a very important hearing. I know it is very technical in nature and probably boring to a lot of people, but it will have a tremendous effect on our economy down the road in our banking system. I want to thank all of you for being here and being patient, waiting through the first panel, but when we have these types of panels, yours and the other one before you, it is hard to leave. Thank you very much. The hearing is adjourned. [Whereupon, at 12:59 p.m., the hearing was adjourned.] [Prepared statements supplied for the record follow:] PREPARED STATEMENT OF SENATOR WAYNE ALLARD I would like to thank Chairman Shelby for holding this important hearing today to discuss the changing supervisory needs of some of the world's largest and most complex internationally active banks. The nature and activities of these banking institutions have evolved since the inception of the Basel I Capital Accords in 1988. It is necessary that the standards and requirements of internationally active U.S. banking organizations are subject to the appropriate standards and requirements to ensure that they remain competitive, healthy, and well- capitalized. The varying capital requirements for internationally active banks made the 1988 Basel I Accord necessary to address the problem of competitive inequality by establishing uniform capital requirements. In the last decade or so, securitization and the use of derivatives have prompted the Basel Committee to reconsider the relevance of the 1988 Accord standards. Since 1999, the Committee has been working to develop a more risk sensitive capital adequacy framework to replace Basel I. For most banks in the United States, Basel I is more than adequate in establishing a capital framework. For a few large, complex, and internationally active banks, the framework needs adjustment in order to maintain the health of banking organizations in the United States. I would like to thank the regulators for their ongoing work and attention to this critical issue. I also thank all of our witnesses for appearing before the Committee today. I look forward to your testimony. ---------- PREPARED STATEMENT OF ROGER W. FERGUSON, JR. Vice Chairman Board of Governors of the Federal Reserve System June 18, 2003 Chairman Shelby, Senator Sarbanes, Members of the Committee, it is a pleasure to appear before you this morning on behalf of the Board of Governors to discuss Basel II, the evolving New Capital Accord for internationally active banking organizations. After 5 years of discussion, the proposal is entering its final stage of public comment and review, although there still remains additional steps to the process. Why Is a New Capital Standard Necessary? The banking supervisors in this country believe that Basel I, the current capital regime adopted in 1988, must be replaced for the largest, most complex banks for three major reasons: (1) Basel I has serious shortcomings as it applies to these large entities, (2) the art of risk management has evolved at the largest banks, and (3) the banking system has become increasingly concentrated. Shortcomings in Basel I Basel I was a major step forward in capital regulation. For most banks in this country Basel I, as we in the United States have augmented it, is now--and for the foreseeable future will be--more than adequate as a capital framework. However, for the small number of large, complex, internationally active banking organizations, Basel I has serious shortcomings which are becoming more evident with time. Developing a replacement to apply to these banking organizations is imperative. Basel I is too simplistic to address the activities of our most complex banking institutions. The framework has only four risk categories, and most loans receive the same regulatory capital charge even though loans made by banks encompass the whole spectrum of credit quality. The limited differentiation among the degrees of risk means that the calculated capital ratios are too often uninformative and might well provide misleading information for banks with risky or problem credits or, for that matter, with portfolios dominated by very safe loans. Moreover, the limited number of risk categories creates incentives for banks to game the system through capital arbitrage. Capital arbitrage is the avoidance of certain minimum capital charges through the sale or securitization of bank assets for which the capital requirement that the market would impose is less than the current regulatory capital charge. For example, credit card loans and residential mortgages are securitized in volume, rather than held on banks' balance sheets, because the market requires less capital, in the form of bank credit enhancements, than Basel I requires in capital charges. This behavior by banks is perfectly understandable, even desirable in terms of economic efficiency. But it means that banks that engage in such arbitrage retain the higher-risk assets for which the regulatory capital charge--calibrated to assets of average quality--is on average too low. To be sure, through the examination process supervisors are still able to evaluate the true risk position of the bank, but the regulatory minimum capital ratios of the larger banks are becoming less and less meaningful, a trend that will only accelerate. Not only are creditors, counterparties, and investors less able to evaluate the capital strength of individual banks from what are supposed to be risk-based capital ratios, but also regulations and statutory requirements tied to capital ratios have less meaning as well. Basel I capital ratios neither adequately reflect risk nor measure bank strength at the larger banks. The Evolving State of the Art Risk measurement and management have improved significantly beyond the state of the art of 15 years ago, when Basel I was developed. Banks themselves have created some of the new techniques to improve their risk management and internal economic capital measures in order to be more effective competitors and to control and manage their credit losses. But clearly banks can go considerably further. One objective of Basel II is to speed adoption of these new techniques and to promote the further evolution of risk measurement and management by harnessing them to the regulatory process. Increased Heterogeneity and Concentration in Banking Market pressures have led to consolidation in banking around the world. Our own banking system has not been immune; it, too, has become increasingly concentrated with a small number of very large banks operating across a wide range of product and geographic markets. The operations of these large banks are tremendously complex and sophisticated, and they have markedly different product mixes. At the same time, significant weakness in one of these entities has the potential for severely adverse macroeconomic consequences. Although their insured liabilities have been declining over time as a share of their total funding, these organizations, with their scale and role in payment and settlement systems and in derivatives markets, have presented the authorities with an increasing moral hazard. It is imperative that the regulatory framework should encourage these banks to adopt the best possible risk-measurement and management techniques while allowing for the considerable differences in their business strategies. Basel II presents an opportunity for supervisors to encourage these and other large banks to push their management frontier forward. Basel II The proposed substitute for the current capital accord, Basel II, is more complex than its predecessor for very good reasons. First, the assessment of risk in an environment of a growing number of instruments and strategies with subtle differences in risk-reward characteristics is inevitably complicated Second, the Basel II reform has several objectives: U.S. supervisors are trying to improve risk measurement and management both domestically and internationally; to link to the extent that we can the amount of required capital to the amount of risk taken; to further focus the supervisor-bank dialog on the measurement and management of risk and the risk-capital nexus; and to make all of this transparent to the counterparties that ultimately fund--and hence share--these risk positions. To achieve all these objectives, the framework for Basel II contains three elements, called Pillars 1, 2, and 3. The most important pillar, Pillar 1, consists of minimum capital requirements--that is, the rules by which a bank calculates its capital ratio and by which its supervisor assesses whether it is in compliance with the minimum capital threshold. As under Basel I, a bank's risk-based capital ratio under Basel II would have a numerator representing the capital available to the bank and a denominator that would be a measure of the risks faced by the bank, referred to as ``risk-weighted assets.'' The definition of regulatory capital in the form of equity, reserves, and subordinated debt and the minimum required ratio, 8 percent, are not changing. What would be different is the definition of risk-weighted assets, that is, the methods used to measure the ``riskiness'' of the loans and investments held by the bank. It is this modified definition of risk-weighted assets, its greater risk- sensitivity, that is the hallmark of Basel II. The modified definition of risk-weighted assets would also include an explicit, rather than implicit, treatment of ``operational risk.'' Pillar 2 addresses supervisory oversight; it encompasses the concept that well-managed banks should seek to go beyond simple compliance with minimum capital requirements and perform for themselves a comprehensive assessment of whether they have sufficient capital to support their risks. In addition, on the basis of their knowledge of industry practices at a range of institutions, supervisors should provide constructive feedback to bank management on these internal assessments. Finally, Pillar 3 seeks to complement these activities with stronger market discipline by requiring banks publicly to disclose key measures related to their risk and capital positions. The concept of these three mutually reinforcing pillars has been central to the Basel II effort. Scope of Application in the United States The U.S. supervisory agencies will propose that most banking organizations in this country remain under the existing Basel I-type capital rules and would continue to have no explicit capital charge for operational risk. Earlier I emphasized that Basel I had outlived its usefulness for the larger banking organizations. How then did we conclude that most of our banks should remain under rules based on the old accord? Banks Remaining Under Current Capital Rules To begin with, most of our banks have relatively straightforward balance sheets and do not yet need the full panoply of sophisticated risk-management techniques required under the advanced versions of Basel II. In addition, for various reasons, most of our banks now hold considerable capital in excess of regulatory minimums: More than 93 percent have risk-weighted capital ratios in excess of 10 percent--an attained ratio that is 25 percent above the current regulatory minimum. No additional capital would likely have to be held if these institutions were required to adopt Basel II. Moreover, U.S. banks have long been subject to comprehensive and thorough supervision that is much less common in most other countries planning to implement Basel II. Indeed, U.S. supervisors will continue to be interested in reviewing and understanding the risk-measurement and management processes of all banks. Our banks also disclose considerable information through regulatory reports and under accounting rules and requirements of the Securities and Exchange Commission; they already provide significant disclosure--consistent with Pillar 3 of Basel II. Thus, when we balanced the costs of imposing a new capital regime on thousands of our banks against the benefits--slightly more risk sensitivity of capital requirements under, say, the standardized version of Basel II for credit risk, and somewhat more disclosure--it did not seem worthwhile to require most of our banks to take that step. Countries with an institutional structure different from ours might clearly find universal application of Basel II to benefit their banking system, but we do not think that imposing Basel II on most of our banks is either necessary or practical. Banks Moving to Basel II We have an entirely different view for our largest and most complicated bank- ing organizations, especially those with significant operations abroad. Among the most important objectives of both Basel I and the proposed Basel II is to promote competitive consistency of capital requirements for banks that compete directly in global markets. Another important objective has been to encourage the largest banking organizations of the world to continue to incorporate into their operations the most sophisticated techniques for the measurement and management of risk. As I have noted, these entities use financial instruments and procedures that are not adequately captured by the Basel I paradigm. They have already begun to use--or have the capability to adopt--the techniques of modern finance to measure and manage their exposures; and because substantial difficulty at one of the largest banking organizations could have significant effects on global financial markets, all of the largest banks should be using these procedures. In our view, prudential supervisors and central bankers would be remiss if we did not address the evolving complexity of our largest banks and ensure that modern techniques were being used to manage their risks. The U.S. supervisors have concluded that the advanced versions of Basel II--the Advanced Internal Ratings-Based (A- IRB) approach for measuring credit risk and the Advanced Measurement Approaches (AMA) for measuring operational risk--are best suited to achieve this last objective. Under the A-IRB approach, a banking organization would have to estimate, for each credit exposure, the probability that the borrower will default, the likely size of the loss that will be incurred in the event of default: And--where the lender has an undrawn line of credit or loan commitment to the borrower--an estimate of what the amount borrowed is likely to be at the time a default occurs. These three key inputs--probability of default (PD), loss given default (LGD), and exposure at default (EAD)--are inputs that would be used in formulas provided by supervisors to determine the minimum required capital for a given portfolio of exposure. While the organization would estimate these key inputs, the estimates would have to be rigorously based on empirical information, using procedures and controls validated by its supervisor, and the results would have to accurately measure risk. Those banks that are required, or choose, to adopt the A-IRB approach to measuring credit risk, would also be required to hold capital for operational risk, using a procedure known as the Advanced Management Approach (AMA) to establish the size of that charge. Under the AMA, banks themselves would bear the primary responsibility for developing their own methodology for assessing their own operational risk capital requirement. To be sure, supervisors would require that the procedures used are comprehensive, systematic, and consistent with certain broad outlines, and must review and validate each bank's process. In this way, a bank's ``op risk'' capital charge would reflect its own environment and controls. Importantly, the size of the charge could be reduced by actions that the bank takes to mitigate operational risk. This provides an important incentive for the bank to take actions to limit their potential losses from operational problems. Determining Basel II Banks To promote a more level global playing field, the banking agencies in the United States will be proposing in the forthcoming Advance Notice of Proposed Rulemaking (ANPR) that those U.S. banking organizations with foreign exposure above a specified amount would be in the core set of banks that would be required to adopt the advanced versions of Basel II. To improve risk management at those organizations whose disruption would have the largest effect on the global economy, we would also require the same of banks whose scale exceeds a specified amount. That is, banks meeting either the foreign exposure criterion or the asset size criterion would be required to adopt the advanced versions of Basel II, although most banks meeting one criterion also meet the other. Ten U.S. banks meet the proposed criteria to be core banks and thus would be required, under our proposal, to adopt A-IRB and AMA to measure their credit and operational risks, respectively. As they grow, other banks could very well meet the criteria and thus shift into the core group in the years ahead. We would also permit any bank that meets the infrastructure requirements of A-IRB and AMA--the ability to quantify and develop the necessary risk parameters on credit exposures and develop measurement systems for operational risk exposures--to choose Basel II. Banks that choose to use A-IRB and AMA would need to consider several factors, including the benefits of Basel II relative to its costs, the nature of their operations, the capital impact, and the message they want to send their counterparties about their risk- management techniques. We anticipate that after conducting such a review, about 10 or so large banks now outside the core group would choose to adopt Basel II in the near-term. Thus we expect about 20 banks to adopt the advanced version of Basel II before or shortly after the initial implementation date. Over time, other large banks, perhaps responding to market pressure and facing declining costs and wider understanding of the technology, may also choose this capital regime, but we do not think that the cost- benefit assessment would induce smaller banks to do so for a very long time. Our discussions with the rating agencies confirm they do not expect that regional banks would find adoption of Basel II to be cost effective in the initial implementation period. Preliminary surveys of the views of bank equity security analysts indicate that they are more focused on the disclosure aspects of Basel II rather than on the scope of application. To be clear, supervisors have no intention of pressuring any of the banks outside the core group to adopt Basel II. The 10 core banks that would be required to adopt Basel II, together with the approximately 10 self-selecting banks that we anticipate would adopt it before or shortly after the initial implementation date, today account for 99 percent of the foreign assets and two-thirds of all the assets of domestic U.S. banking organizations, a rate of coverage demonstrating the importance of these entities to the United States and global banking and financial markets. These data also underscore our commitment to international competitive equity and the adoption of best-practice policies at the organizations critical to our financial stability while minimizing cost and disruption at our purely domestic, less-complicated organizations. Issues Bankers have identified three key areas of concern: Cost, competitive equity, and Pillar 1 treatment of operational risk. Cost Implementing A-IRB and AMA in this country is going to be expensive for the small number of banks for which it will be required, for other banks choosing it, and for the supervisors. For the banks, the greatest expense would be establishing the mechanisms necessary for a bank to evaluate and control its risk exposures more formally. The A-IRB approach would not eliminate losses: Banks are in the business of taking risk, and where there are risks, there will be losses. But we believe that the better risk-management that is required for the A-IRB and AMA would better align risk and return and thereby provide benefits to bank stakeholders and the economy. And, more risk-sensitive capital requirements would assist in ensuring that banks would have sufficient capital to absorb losses when they do occur. The cost-benefit ratio looks right to the supervisors. This ratio is further enhanced because attributing to Basel II all the costs associated with the adoption of modern, formal risk- management systems is a logical fallacy. The large banks that would be required, or that would choose, to adopt A-IRB and AMA must compete for funding in a global marketplace and thus already have adopted many of these processes and would continue to develop them even without Basel II. The new accord may well appropriately speed up the adoption process, but overall, the costs of adopting these processes are being forced on these banks not by Basel II but by the requirements of doing business in an increasingly complex financial environment. In any event, the ANPR will include questions designed to quantify the cost of implementing Basel II. Competitive Equity A second key concern is competitive equity. Some are concerned that the U.S. supervisors would be more stringent in their application of Basel II rules than other countries and would thereby place U.S. banks at a competitive disadvantage. To address this concern, the Basel Agreement establishes an Accord Implementation Group (AIG), made up of senior supervisors from each Basel member country, which has already begun to meet. It is the AIG's task to work out common standards and procedures and act as a forum in which conflicts can be addressed. No doubt some differences in application would be unavoidable across banking systems with different institutional and supervisory structures, but all of the supervisors, and certainly the Federal Reserve, would remain alert to this issue and work to minimize it. I also emphasize that, as is the case today, U.S. bank subsidiaries of foreign banks would be operating under U.S. rules, just as foreign bank subsidiaries of U.S. banks would be operating under host-country rules. Another issue relates to the concern among U.S. Basel II banks of the potential competitive edge that might be given to any bank that would have its capital requirements lowered by more than that of another Basel II bank. The essence of Basel II is that it is designed to link the capital requirement to the risk of the exposures of each individual bank. A bank that holds mainly lower-risk assets, such as high-quality residential mortgages, would have no advantage over a rival that held mainly lower-quality, and therefore riskier, commercial loans just because the former had lower required capital charges. The capital requirements should be a function of risk taken, and, under Basel II, if the two banks had very similar loans, they both should have a very similar required capital charge. For this reason, competitive equity among Basel II banks in this country should not be a genuine issue because capital should reflect risk taken. Under the current capital regime, banks with different risk profiles have the same capital requirements, creating now a competitive inequity for the banks that have chosen lower-risk profiles. The most frequently voiced concern about possible competitive imbalance reflects the ``bifurcated'' rules implicit in the U.S. supervisors' proposed scope of application: That is, requiring Basel II through A-IRB and AMA for a small number of large banks while requiring the current capital rules for all other U.S. banks. The stated concern of some observers is that the banks that remained under the current capital rules, with capital charges that are not as risk sensitive, would be at a competitive disadvantage compared to Basel II banks that would get lower capital charges on less-risky assets. The same credit exposure might have a lower regulatory minimum capital charge at a Basel II bank than at a Basel I bank. Of course, Basel II banks would have higher capital charges on higher-risk assets and the cost of adopting a new infrastructure, neither of which Basel I banks would have. And any bank that might feel threatened could adopt Basel II if they would make the investment required to reach the qualifying criteria. But a concern remains about competitive equity in our proposed scope of application, one that could present some difficult trade-offs if the competitive issue is real and significant. On the one hand is the pressing need to reform the capital system for the largest banks and the practical arguments for retaining the present system for most U.S. banks. Against that is the concern that there might be an unintended consequence of disadvantaging those banks that would remain on the current capital regime. We take the latter concern seriously and will be exploring it through the ANPR. But, without prejudging the issue, there are reasons to believe that little if any competitive disadvantage would be brought to those banks remaining under the current capital regime. The basic question is the role of minimum regulatory capital requirements in the determination of the price and availability of credit. Economic analysis suggests that regulatory capital should be considerably less important than the capital allocations that banks make internally within their organization, so-called economic capital. Our understanding of bank pricing is that it starts with economic capital and the explicit recognition of the riskiness of the credit and is then adjusted on the basis of market conditions and local competition from bank and nonbank sources. In some markets, some banks will be relatively passive price takers. In either case, regulatory capital is mostly irrelevant in the pricing decision, and therefore unlikely to cause competitive disparities. Moreover, most banks, and especially the smaller ones, hold capital far in excess of regulatory minimums for various reasons. Thus, changes in their own or their rivals' minimum regulatory capital generally would not have much effect on the level of capital they choose to hold and would therefore not necessarily affect internal capital allocations for pricing purposes. In addition, the banks that most frequently express a fear of being disadvantaged by a bifurcated regulatory regime have for years faced capital arbitrage from larger rivals who were able to reduce their capital charges by securitizing loans for which the regulatory charge was too high relative to the market or economic capital charge. The more risk-sensitive A-IRB in fact would reduce the regulatory capital charge in just those areas where capital requirements are too high under the current regime. In those areas, capital arbitrage has already reduced the regulatory capital charge. The A-IRB would provide, in effect, risk-sensitive capital charges for lower-risk assets that are similar to what the larger banks have for years already obtained through capital arbitrage. In short, competitive realities between banks might not change in many markets in which minimum regulatory capital charges would become more explicitly risk sensitive. Concerns have also been raised about the effect of Basel II capital requirements on the competitive relationships between depository institutions and their nondepository rivals. Of course, the argument that economic capital is the driving force in pricing applies in this case, too. Its role is only reinforced by the fact that the cost of capital and funding is less at insured depositories than at their nondepository rivals because of the safety net. Insured deposits and access to the Federal Reserve discount window (and Federal Home Loan Bank advances) let insured depositories operate with far less capital or collateralization than the market would otherwise require of them and far less than it does require of nondepository rivals. Again, Basel II would not change those market realities. Let me repeat that I do not mean to dismiss competitive equity concerns. Indeed, I hope that the comments on the ANPR bring forth insights and analyses that respond directly to the issues, particularly the observations I have just made. But, I must say, we need to see reasoned analysis and not assertions. Operational Risk The third key area of concern is the proposed Pillar 1 treatment of operational risk. Operational risk refers to losses from failures of systems, controls, or people and will, for the first time, be explicitly subject to capital charges under the Basel II proposal. Neither operational risk nor capital to offset it are new concepts. Supervisors have been expecting banks to manage operational risk for some time, and banks have been holding capital against it. Under Basel I both operational and credit risks have been implicitly covered in one measure of risk and one capital charge. But Basel II, by designing a risk-based system for credit and operational risk, separates the two risks and would require capital to be held for each separately. Operational disruptions have caused banks to suffer huge losses and, in some cases, failure here and abroad. At times they have dominated the business news and even the front pages. Appendix 1 to this statement lists the 10 largest such events of recent years. In an increasingly technology-driven banking system, operational risks have become an even larger share of total risk; at some banks they are the dominant risk. To avoid addressing them would be imprudent and would leave a considerable gap in our regulatory system. A capital charge to cover operational risk would no more eliminate operational risk than a capital charge for credit risk eliminates credit risk. For both risks, capital is a measure of a bank's ability to absorb losses and survive without endangering the banking and financial system. The AMA for determining capital charges on operational risk is a principles-based approach that would obligate banks to evaluate their own operational risks in a structured but flexible way. Importantly, a bank could reduce its operational-risk charge by adopting procedures, systems, and controls that reduce its risk or by shifting the risk to others through measures such as insurance. This approach parallels that for credit risk, in which capital charges can be reduced by shifting to less-risky exposures or by making use of risk-mitigation techniques such as collateral or guarantees. Some banks for which operational risk is the dominant risk oppose an explicit capital charge on operational risk. Some of these organizations tend to have little credit exposure and hence very small required capital under the current regime, but would have significant required capital charges should operational risk be explicitly treated under Pillar 1 of Basel II. Such banks, and also some whose principal risks are credit-related, would prefer that operational risk be handled case by case through the supervisory review of buffer capital under Pillar 2 of the Basel proposal rather than be subject to an explicit regulatory capital charge under Pillar 1. The Federal Reserve believes that would be a mistake because it would greatly reduce the transparency of risk and capital that is such an important part of Basel II and would make it very difficult to treat risks comparably across banks because Pillar 2 is judgmentally based. Most of the banks to which Basel II would apply in the United States are well along in developing their AMA-based capital charge and believe that the process has already induced them to adopt risk- reducing innovations. Presentations at a conference held late last month illustrated the significant advances in operational-risk quantification being made by most internationally active banks. The presentations were made by representatives from most of the major banks in Europe, Asia, and North America, and many presenters enthusiastically supported the use of AMA-type techniques to incorporate operational risk in their formal modeling of economic capital. Many banks also acknowledged the important role played by the Basel process in encouraging them to develop improved operational risk management.\1\ --------------------------------------------------------------------------- \1\ Papers from that conference are available at http:// www.newyorkfed.org/pihome/news/speeches/2003/con052903.html. --------------------------------------------------------------------------- Overall Capital and An Evolving Basel II Before I move on to other issues, I would like to address the concern that the combination of credit and operational risk capital charges for those United States banks that are under Basel II would decline too much for prudent supervisory purposes. Speaking for the Federal Reserve Board, let me underline that we could not support a final Basel II that we felt caused capital to decline to unsafe and unsound levels at the largest banks. That is why we anticipate that the United States authorities would conduct a Quantitative Impact Study (QIS) in 2004 to supplement the one conducted late last year; I anticipate at least one or two more before final implementation. It is also why CP-3 calls for 1 year of parallel (Basel I and II) capital calculation and a 2-year phase-in with capital floors set at 90 and 80 percent, respectively, of the Basel I levels before full Basel II implementation. At any of those stages, if the evidence suggested that capital were declining too much the Federal Reserve Board would insist that Basel II be adjusted or recalibrated, regardless of the difficulties with bankers here and abroad or with supervisors in other countries. This is the stated position of the Board and our supervisors and has not changed during the process. Of course, capital ratios are not the sole consideration. The improved risk measurement and management, and its integration into the supervisory system, under Basel II, are also critical to ensuring the safety and soundness of the banking system. When coupled with the special U.S. features, such as prompt corrective action, minimum leverage ratios, statutory provisions that make capital a prerequisite to exercising additional powers, and market demands for buffer capital, some modest reduction in the minimum regulatory capital for sound, well-managed banks could be tolerable. I note that banks with lower risk profiles, as a matter of sound public policy, should have lower capital than banks with higher-risk profiles. Greater dispersion in required capital ratios, if reflective of underlying risk, is an objective, not a problem to be overcome. I should also underline that Basel II is designed to adapt to changing technology and procedures. I fully expect that in the years ahead banks and supervisors will develop better ways of estimating risk parameters as well as better functions that convert those parameters to capital requirements. When they do, these changes could be substituted directly into the Basel II framework, portfolio by portfolio if necessary. Basel II would not lock risk management into any particular structure; rather Basel II could evolve as best practice evolves and, as it were, be evergreen. The Schedule and Transparency I would like to say a few words about the schedule. In a few weeks, the agencies will be publishing their joint ANPR for a 90-day comment period, and will also issue early drafts of related supervisory guidance so that banks can have a fuller understanding of supervisory expectations and more carefully begin their planning process. The comments on the domestic rulemaking as well as on CP-3 will be critical in developing the negotiating position of the U.S. agencies, and highlighting the need for any potential modifications in the proposal. The U.S. agencies are committed to careful and considered review of the comments received. When the comments on CP-3 and the ANPR have been received, the agencies will review them and meet to discuss whether changes are required in the Basel II proposal. In November, we are scheduled to meet in Basel to negotiate our remaining differences. I fear this part of the schedule may be too tight because it may not provide U.S. negotiators with sufficient time to digest the comments on the ANPR and develop a national position to present to our negotiating partners. There may well be some slippage from the November target, but this slippage in the schedule is unlikely to be very great. In any event, implementation in this country of the final agreement on Basel II would require a Notice of Proposed Rulemaking (NPR) in 2004 and a review of comments followed by a final rule before the end of 2004. On a parallel track, core banks and potential opt in banks in the United States will be having preliminary discussions with their relevant supervisors in 2003 and 2004 to develop a work plan and schedule. As I noted, we intend to conduct more Quantitative Impact Studies, starting in 2004, so we can be more certain of the impact of the proposed changes on individual banks and the banking system. As it stands now, core and opt in banks will be asked by the fall of 2004 to develop an action plan leading up to final implementation. Implementation by the end of 2006 would be desirable, but each bank's plan will be based on a joint assessment by the individual bank and its relevant supervisors of a realistic schedule; for some banks the adoption date may be beyond the end of 2006 because of the complexity of the required changes in systems. It is our preference to have an institution ``do it right'' rather than ``do it quickly''. We do not plan to force any bank into a regime for which it is not ready, but supervisors do expect a formal plan and a reasonable implementation date. At any time during that period, we can slow down the schedule or revise the rules if there is a good reason to do so. The development of Basel II has been highly transparent from the beginning and will remain so. All of the consultative papers over the past 5 years have been supported by a large number of public papers and documents to provide background on the concepts, framework, and options. After each previous consultative paper, extensive public comment has been followed by significant refinement and improvement of the proposal. During the past 5 years, a number of meetings with bankers have been held in Basel and in other nations, including the United States. Over the past 18 months, I have chaired a series of meetings with bankers, often jointly with Comptroller Hawke. More than 20 U.S. banks late last year joined 365 others around the world in the third Quantitative Impact Survey (QIS-3), which was intended to estimate the effects of Basel II on their operations. The banking agencies last month held three regional meetings with the bankers that would not be required to adopt Basel II but might have an interest in choosing to adopt the A-IRB approach and the AMA. Our purpose was to ensure that these banks understand the proposal and the options it provides them.\2\ As I noted, in about 1 month the banking agencies in this country hope to release an ANPR that will outline and seek comment on specific proposals for the application of Basel II in this country. In the past week or so we have also released two White Papers to help commenters frame their views on commercial real estate and the capital implications of recognizing certain guarantees. These, too, are available at our web site. --------------------------------------------------------------------------- \2\ The documents used in these presentations are available at the Board's web site, http://www.Federalreserve.gov/banknreg.htm (Documents Relating to U.S. Implementation of Basel II). --------------------------------------------------------------------------- This dialog with bankers has had a substantive impact on the Basel II proposal. I have attached to my statement a comparison of some of the major provisions of Basel II as proposed in each of the three consultative documents published by the Basel Committee on Bank Supervision (Appendix 2). As you can see, commenters have significantly influenced the shape and detail of the proposal. For example, comments about the earlier proposed crude formulas for addressing operational risk led to a change in the way capital for operational risk may be calculated; banks may now use their own methods for assessing this form of risk, as long as these methods are sufficiently comprehensive and systematic and meet a set of principles-based qualifying criteria. That is the AMA. The mechanism for establishing capital for credit risk has also evolved significantly since the first Consultative Paper on the basis of industry comments and suggestions; as a result, a large number of exposure types are now treated separately. Similarly, disclosure rules have been simplified and streamlined in response to industry concerns. At this stage of the proposal, comments that are based on evidence and analysis are most likely to be effective. Perhaps an example of the importance of supporting evidence in causing a change in positions might be useful. As some Members of this Committee may know, the Federal Reserve had concluded earlier, on the basis of both supervisory judgment and the available evidence, that the risk associated with commercial real estate loans on certain existing or completed property required a capital charge higher than the capital charge on other commercial real estate and on commercial and industrial loans. In recent weeks, however, our analysis of additional data suggested that the evidence was contradictory. With such inconsistent empirical evidence, we concluded that, despite our supervisory judgment on the potential risk of these exposures, we could not support requiring a higher minimum capital charge on commercial real estate loans on any existing or completed property, and we will not do so. In the same vein, we also remain open minded about proposals that simplify the proposal but attain its objective. Both the modifications of the proposals in CP-3 and the changes in U.S. supervisory views, as evidenced by the commercial real estate proposal, testify to the willingness of the agencies, even at this late stage of the process, to entertain new ideas and to change previous views when warranted. Summary The existing capital regime must be replaced for the large, internationally active banks whose operations have outgrown the simple paradigm of Basel I and whose scale requires improved risk-management and supervisory techniques to minimize the risk of disruptions to world financial markets. Fortunately, the state of the art of risk measurement and risk management has improved dramatically since the first capital accord was adopted, and the new techniques are the basis for the proposed new accord. In my judgment, we have no alternative but to adopt, as soon as practical, these approaches for the supervision of our larger banks. The Basel II framework is the product of extensive multiyear dialogs with the banking industry regarding evolving best practice risk-management techniques in every significant area of banking activity. Accordingly, by aligning supervision and regulation with these techniques, it provides a great step forward in protecting our financial system and that of other nations to the benefit of our citizens. Basel II will provide strong incentives for banks to continue improving their internal risk-management capabilities as well as the tools for supervisors to focus on emerging problems and issues more rapidly than ever before. I am pleased to appear before you today to report on this effort as it nears completion. Open discussion of complex issues has been at the heart of the Basel II development process from the outset and will continue to characterize it as Basel II evolves further. <GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT> PREPARED STATEMENT OF JOHN D. HAWKE, JR. Comptroller of the Currency U.S. Department of the Treasury June 18, 2003 Introduction Chairman Shelby, Senator Sarbanes, and Members of the Committee, thank you for inviting the Office of the Comptroller of the Currency (OCC) to participate in this hearing on proposed revisions to the 1988 Capital Accord developed by the Basel Committee on Banking Supervision (Basel Committee). I welcome the efforts of the Committee to focus attention on these critical issues. The health of the U.S. commercial banking system is a critical element to a strong economy. Thus, it is essential that any regulatory changes that might affect the condition and competitiveness of our banking system be fully understood and carefully evaluated by the banking industry, the U.S. Congress and the American public. The 1988 Accord, referred to as Basel I, established the framework for the risk-based capital adequacy standards applicable to internationally active commercial banks in all of the G-10 countries, and it has been adopted by most other banking authorities around the world. U.S. banking and thrift agencies have applied the 1988 framework to all U.S. insured depository institutions. By the late 1990's, it became evident that Basel I had become outdated. The increased scope and complexity of the banking activities of our largest banking institutions over the last decade and the unintended consequences of various provisions of the regulations, severely undercut the utility of the Capital Accord. Basel I simply does not provide a meaningful measure of the risks faced by large, internationally active banks or the capital they should hold against those risks. Consequently, over the past several years, the Basel Committee has been developing a more detailed and risk sensitive capital adequacy framework to replace Basel I. The Committee's first draft document, Consultative Paper No. 1 (CP-1), was issued in June 1999. It laid the groundwork for the new capital adequacy framework (Basel II), but provided few details. The Committee provided additional details on the specifics of Basel II in its January 2001 issuance of Consultative Paper No. 2 (CP-2). Although more detailed, CP-2 still left a number of key issues unaddressed and unresolved. The Committee's most recent paper, Consultative Paper No. 3 (CP-3), which I will discuss today, was issued on April 29 of this year. As work on these consultative papers has progressed, the Basel Committee also has attempted to gauge the impact of its proposals on the required capital levels of banking institutions through a series of quantitative impact studies. In May, the Committee published the results of the most recent assessment, the third quantitative impact study (QIS-3). While the Committee concluded that the results were generally in line with the objectives of Basel II, the QIS-3 data still do not provide a sufficiently reliable estimate of the likely regulatory capital requirements for banks subject to Basel II. More work in this area is clearly warranted and I will discuss this later in my testimony. The Basel Committee has outlined an aggressive timeline for the remaining actions leading to the adoption of Basel II. As a consequence, the U.S. banking agencies, the agencies responsible for the maintenance of capital adequacy standards for U.S. financial institutions, are faced with a daunting task. While we will work earnestly in this effort, the timeline should be seen as a means to an end, not an end in itself. As will be highlighted in my testimony, basic principles of safety and soundness demand that the banking agencies have a more complete understanding of the consequences of this proposal on the overall capital levels of affected institutions, the competitive effects on our financial system, and associated compliance costs and burdens before moving forward to finalize this proposal. Our current primary focus in this effort is the development of U.S. implementing regulations and policies. As I will discuss later, the OCC and the other U.S. banking agencies will soon issue for comment proposed revisions to U.S. risk-based capital regulations to reflect the primary components of Basel II. Let me be absolutely clear about the integrity of this rulemaking process--the OCC, which has the sole statutory responsibility for promulgating capital regulations for national banks, will not begin implementing a final Basel II framework until we have conducted whatever cost-benefit and impact analyses that are required, and fully considered all comments received during our notice and comment process--as we would with any domestic rulemaking. If we determine through this process that changes to the proposal are necessary, we will not implement proposed revisions until appropriate changes are made. We made this point quite clearly to our Basel Committee colleagues before we agreed to go forward with CP- 3. Indeed, many of them will also have to go through their own internal domestic processes before they can adopt the Basel II framework. Current Basel Proposal The Basel Committee deserves considerable credit for its articulation of Basel II in CP-3. The proposal is still exceedingly complex, but CP-3 is a clearer presentation of inherently difficult material than its predecessors. This is an important step, since regardless of the complexity of the proposal, it is important the industry and other interested parties have a clear understanding of the proposed Accord. The attachment to this written statement provides a summary of the substantive provisions contained in CP-3. As before, this iteration of the proposed new Accord has three mutually reinforcing ``pillars'' that comprise the framework for assessing bank capital adequacy. The first pillar of the new Accord is the minimum regulatory capital requirement. The Pillar 1 capital requirement includes a credit risk charge, measured by either a standardized approach or one of the new internal ratings-based (IRB) approaches (foundation or advanced), an operational risk charge, and a market risk charge. Again, the attached document provides a more detailed description of the various components of the Pillar 1 charge. Pillar 2 addresses supervisory review. It is, ``intended to ensure not only that banks have adequate capital to support all the risks in their business, but also to encourage banks to develop and use better risk management techniques in monitoring and managing these risks.'' This pillar encourages supervisors to assess banks' internal approaches to capital allocation and internal assessments of capital adequacy, and, subject to national discretion, provides an opportunity for the supervisor to indicate where such approaches do not appear sufficient. Pillar 2 should also be seen as a way to focus supervisors on other means of addressing risks in a bank's portfolio, such as improving overall risk management techniques and internal controls. The third pillar recognizes that market discipline has the potential to reinforce capital regulation and other supervisory efforts to ensure the safety and soundness of the banking system. Thus, the Committee is proposing a wide range of disclosure initiatives, which are designed to make the risk and capital positions of a bank more transparent. As a bank begins to use the more advanced methodologies, such as the Advanced IRB approach, the new Accord will require a significant increase in the level of disclosure. In essence, the tradeoff for greater reliance on a bank's own assessment of the building blocks of capital adequacy is greater transparency. U.S. Implementation Actions It is important to recognize that the Basel Accord is not self- executing in the U.S. Even when adopted by the Basel Committee, Basel II will not apply to U.S. institutions unless and until the U.S. banking agencies adopt regulations to implement it. In accordance with the Administrative Procedure Act, 5 U.S.C. 551, et seq., the U.S. banking agencies must publish notice and seek comment from all interested persons on any such proposal, and must fully consider those comments, before adopting a new capital regulation in final form. Obviously, the OCC and the other Federal banking agencies intend to comply fully with these requirements. The importance of this rulemaking makes this comment process particularly critical to our success. Thus, we welcome this process as a means for positive contribution to this deliberative effort. We believe that the solicitation and assessment of comments is a critical step in determining the feasibility, effectiveness, and expected consequences of Basel II and related domestic capital regulations. Next month, the U.S. banking agencies expect to jointly issue an Advance Notice of Proposed Rulemaking (ANPR) soliciting comment on proposed revisions to the existing domestic capital adequacy regulations that would implement Basel II. The ANPR will be largely based on CP-3, and will provide a description of proposed revisions to current capital regulations, while seeking comment on outstanding or contentious issues associated with the proposal. The ANPR will also request information on the cost of implementing the proposal, and will seek comment on the competitive implications in both domestic and international markets for banks of all sizes. In conjunction with the ANPR, the banking agencies will also issue for comment draft supervisory guidance articulating general supervisory expectations for banks seeking to implement Basel II-compliant methodologies for the Advanced Measurement Approach (AMA) to operational risk and Advanced IRB for corporate credits. Recognizing that CP-3 is a complex document, we understand the importance of providing U.S. banks an opportunity to review and comment on U.S. implementing documents as soon as practicable. By describing these concepts within the context of our existing regulatory and supervisory regime, the ANPR and draft guidance will provide a meaningful forum for a full discussion of Basel II. After assessing comments generated during the ANPR process, the U.S. banking agencies will consider a complete cost analysis in accordance with applicable rulemaking requirements, including the standards of Executive Order 12866, discussed below, and will develop specific regulatory language for a joint Notice of Proposed Rulemaking (NPR). Again, the banking industry and other interested parties will have an opportunity to comment on this fully articulated proposal before any revisions to our capital regulations are finalized. Let me now focus on two important, unique features of the U.S. regulatory capital regime that will be highlighted in the ANPR and NPR--the scope of application of Basel II and the content and structure of the proposed revisions to the capital adequacy regulations. First, the United States expects to set forth in the ANPR proposed criteria for identifying which banks in the United States will be subject to the new Accord. Despite language in the 1988 Capital Accord that permitted a more limited application, U.S. banking and thrift agencies applied the Basel framework to all U.S. insured depository institutions. As we will highlight in the forthcoming ANPR, the U.S. agencies have determined to apply Basel II concepts more narrowly. Specifically, consistent with the focus of the Basel Capital Accord on banks that compete in the global marketplace, we will propose applying Basel II concepts on a mandatory basis only to large, internationally active institutions that compete on a significant global basis with other financial service providers. Other institutions will have the opportunity to voluntarily opt into the Basel framework upon application to, and approval by, their primary Federal supervisor. Preliminary analysis by the U.S. agencies suggests that under the narrow approach we are proposing, there are currently fewer than a dozen U.S. banks that would be mandatorily subject to Basel II-based regulatory capital requirements. Of course, the approach of requiring only a small population of banks to comply with Basel II will be subject to notice and comment in the ANPR and will be definitively resolved only after the U.S. rulemaking process has been completed. Second, in developing revisions to existing capital adequacy regulations, U.S. banking agencies recognize that the revised regulation, and interagency implementation policies, need not follow the literal structure and language of Basel II. While consistent with the objectives, general principles and core elements of the revised Basel Accord, the language, structure, and degree of detail of U.S. implementing documents may be very different from Basel II. These implementation differences are reflective of the particular statutory, regulatory and accounting structures and practices in place in the United States. It is important to note that U.S. implementation actions do not contemplate changes to many fundamental aspects of our regulatory/supervisory process, including a focus on regular on-site supervision, our prompt corrective action rules, and our minimum leverage ratio for capital adequacy. As described more fully in the attachment, the U.S. agencies will propose for notice and comment a Basel II-based regime incorporating only the Advanced IRB approach for credit risk, the AMA for operational risk, and the internal models approach for market risk. We are also very cognizant that in connection with this, or any rulemaking, existing requirements may compel preparation of detailed analysis of the costs, benefits, and other effects of our regulations, depending on threshold determinations of whether the rulemaking in question triggers the substantive requirements of particular statutes or Executive Orders. Relevant requirements are set forth in the Regulatory Flexibility Act (RFA), the Unfunded Mandates Reform Act of 1995 (UMRA) and Executive Order 12866 (E.O. 12866). Issuance of the ANPR will help us identify and determine costs, benefits, and other effects of the proposed rulemaking, for purposes of complying with these requirements. Timing As I noted early on in my testimony, the Basel Committee timeline presents a daunting task to both the U.S. banking agencies and the banking industry. While it is clearly necessary to move forward in addressing the acknowledged deficiencies in the current Basel Capital Accord, the banking agencies must better understand the full range and scale of likely consequences before finalizing any proposal. The list provided below identifies the milestones the OCC must meet under the current Basel II timeline. Each step is critical in a prudential consideration of Basel II in the United States: <bullet> Consideration of comments received by the Basel Committee on CP-3. The comment period on this document concludes on July 31. <bullet> Finalization, issuance, and consideration of comments on the U.S. ANPR. Based on the current estimates, the notice and comment period will run from July to October. <bullet> Finalization, issuance, and consideration of comments on supervisory guidance on Corporate IRB and AMA methodologies. Based on current estimates, the notice and comment period will run from July to October. <bullet> Development, issuance, and consideration of comments on supervisory guidance on other substantive aspects of Basel II-based regulations, especially including retail IRB. Based on current estimates, the agencies hope to commence solicitation of comment on this guidance by year-end 2003. <bullet> Participation in the Basel Committee's consideration of Basel II. Under the current timeline, the Committee is to consider approval of Basel II in December of this year. <bullet> Development, issuance, and analysis of results of additional agency efforts to evaluate the prospective effects of Basel II implementation. E.O. 12866 may compel the OCC and OTS to undertake such analysis prior to the issuance of an NPR. Even without regard to this requirement, however, it is essential that we have a reliable estimate of the impact of Basel II on the capital and competitive position of U.S. banks. <bullet> Development, issuance, and consideration of comments on the U.S. NPR. This document would only be issued after the Basel Committee finalizes its consideration of Basel II. If the existing timeline is maintained, solicitation of comment on the NPR would commence no earlier than the first quarter of 2004. <bullet> Development and issuance of a U.S. final rule and supervisory guidance. Again, assuming the present timeline is maintained, our best estimate for the issue date of a final rule implementing Basel II is the third or fourth quarter of 2004. <bullet> Completion of all necessary supervision-related steps to implement Basel II-based regulations in advance of the presently proposed December 2006 effective date. Most significantly, the agencies need to determine whether each bank subject to Basel II- based regulations has appropriate systems and procedures in place to qualify for using the A-IRB and AMA. Status of Basel Proposal--Outstanding Issues In commencing an objective assessment of the status of Basel II, it is important to reiterate and reaffirm the commendable work of the Basel Committee, and in particular, the strong and intelligent leadership of its former Chairman, William McDonough. The OCC firmly supports the objectives of Basel II. These objectives constitute a sound conceptual basis for the development of a new regulatory capital regime and should continue to serve as a useful benchmark to gauge our progress in this effort. Nonetheless, much of that conceptual basis has not been tested in practice in any manner approaching the magnitude of Basel II. We continue to be concerned about the potential for unintended or unanticipated consequences of the Basel II proposals. Implementation Challenges At its foundation, the Basel II proposals permit qualifying institutions to calculate their minimum risk-based capital requirements by reference to their own internal systems and methodologies. While it is the hallmark of Basel II, a greater alignment of internal risk assessment with minimum regulatory capital derived through internal models represents a radical departure from our existing regulatory capital framework. As we will highlight in the ANPR and accompanying guidance, this reliance on internal risk assessment systems mandates changes in the way we structure our capital regulations and, in certain important respects, how we conduct our supervisory activities. The fundamental question for the banking agencies in assessing Basel II is the issue the OCC has previously identified--whether the regime will work in practice, as well as theory, as the basis for a regulatory capital regime. For bank supervisors and other external stakeholders to be in a position to rely on a bank's internal process in the establishment of regulatory capital requirements, there must be a high degree of confidence that regulators can establish and enforce appropriate risk measurement and management standards consistently across the banks subject to a Basel II-based regime. The challenge for supervisors is to create a verifiably accurate system that appropriately balances the need for flexibility, to promote continued improvement in risk management practices, with the need for objective standards, to ensure consistency in application across institutions and supervisors, both foreign and domestic. The capital rule we implement must respect the evolutionary nature of risk management. As regulators, we must acknowledge that we are still in the relatively early days of model-based credit and operational risk measurement and management. We must recognize the inevitability of further innovation and improvements in this area. This respect for the evolutionary nature of this discipline must then be reconciled with the need for objective standards to ensure consistency in application. Much of the detail and complexity within Basel II derives from the need to establish more objective expectations for bank rating systems, control mechanisms, audit processes, data systems, and other internal determinations of risk by individual banks. In many cases, this has led to the establishment of supervisory standards in areas previously left to management discretion or supervisory judgment. Not surprisingly, the regulatory community has struggled with the establishment of these standards. Failing to achieve the proper balance for these often conflicting objectives while moving forward with the radically different Basel II-based regime can have dramatic consequences. If our regulation and supervisory process is overly flexible, bank internal calculations of capital adequacy may prove insufficient, noncomparable, or both. If we err on the other extreme, we establish an excessively prescriptive supervisory regime that stifles innovation, imposes undue regulatory burden, and inappropriately narrows the role of judgment. This need to carefully balance dramatically opposed objectives, together with the significant uncertainties that still exist about the practical feasibility of these proposed changes to the Capital Accord, raise doubts about the achievability of the timeframe established by the Basel Committee. Competitive Equality A stated goal of the Basel Committee in developing Basel II was that, ``the Accord should continue to enhance competitive equality.'' Realistically, we are not yet in a position to assess definitively the full range of consequences from the implementation of Basel II, including its effect on competitive equality in the global financial marketplace. There are risks that Basel II may create or exacerbate relative advantages between domestic banks and foreign banks; between banks and nonbanks; and between large domestic banks and mid-size/small domestic banks. It is imperative that the U.S. banking agencies remain sensitive to these concerns and assess, to the extent possible, any unintended consequences resulting from the implementation of Basel II. One of the primary objectives of the Basel Committee itself is the reduction of gaps and differences in international supervisory coverage by national supervisory agencies, especially as it relates to large internationally active banks that compete on a significant global basis with other financial service providers. This principle of competitive equality and a level playing field for international banks is an admirable one, and an appropriate goal of the Committee's efforts. Yet, the very complexity of the rules themselves calls this objective into question. Bank supervision varies significantly from one country to another in approach, intrusiveness, and quality. Is it realistic to think that an enormously complex set of rules will be applied in an evenhanded way across such a broad spectrum of supervisory regimes? For example, the OCC has as many as 30 to 40 full-time resident examiners in our largest banks. They are intimately involved as supervisors in assessing the banks' operations and judging the banks' compliance with a myriad of laws, rules, and guidelines. Some other countries may send examiners in once a year to a comparably sized institution, or may examine such an institution thoroughly only every 5 years, or may put heavy reliance on the oversight of outside auditors.\1\ --------------------------------------------------------------------------- \1\ See, Daniel E. Nolle, ``Bank Supervision in the United States and the G-10: Implications for Basel II,'' RMA Journal, June 2003. --------------------------------------------------------------------------- It is fair to ask, I think, in which type of supervisory regime detailed, prescriptive capital rules are more likely to be robustly and reliably enforced. The Basel Committee has not undertaken to set standards of supervision for member countries. Yet the attainment of competitive equity among internationally active banks is a bedrock principle of Basel II. Can we really achieve competitive equality without addressing disparities in supervision, particularly when we are operating on the assumption that the complex new rules we are writing will be applied in an evenhanded way throughout the world? Another principle source of competition for many banks is not other insured depository institutions, but nonbanks. This situation is especially pronounced in businesses such as asset management and payments processing. As you are aware, however, regulations implementing Basel II-based concepts in the United States will apply only to insured depository institutions and their holding companies. While differences in regulatory requirements for banks and nonbanks exist today, many institutions have voiced concern that implementation of Basel II may unduly exacerbate the current differences. These concerns have been mainly focused on the effects on competition from the application of the operational risk proposal and the enhanced disclosures required under Pillar 3. Finally, there is concern about the potential effect of Basel II on the competitive balance between large and small banks. As implemented in the United States, Basel II would result in a bifurcated regulatory capital regime, with large banks subject to Basel II-based requirements and small and mid-sized banks subject to the current capital regime. This structure is premised on the belief that, to the extent possible, regulations should reflect the size, structure, complexity, and risk profile of banking institutions. The Basel II framework was developed to address the unique risks of large internationally active institutions. Mandatory application of such a framework to small banks, with its associated costs, was deemed inappropriate. In fact, the banking agencies sought comment from the banking industry, especially smaller institutions, on the development of a simplified capital framework specifically for noncomplex institutions.\2\ Industry comments were overwhelming negative on the proposal--most institutions felt that the cost of adopting a new regulatory capital regime outweighed any potential benefits. Accordingly, the banking agencies tabled the proposal. --------------------------------------------------------------------------- \2\See Advance Notice of Proposed Rulemaking, Simplified Capital Framework for Non-Complex Institutions, 65 FR 66193 (November 3, 2000). --------------------------------------------------------------------------- With that said, the banking agencies need to continue to assess the competitive effects of a bifurcated regulatory capital regime, and it is one of the areas on which we will seek guidance in our ANPR. There are several concerns in this regard. First, banks using a Basel II- based regime may have a lower minimum capital requirement, allowing those banks to grow and compete more aggressively with smaller banks for both assets and liabilities. To be sure, banks subject to the New Basel II requirements will incur very significant systems and compliance costs in preparing for the new regime. These concerns are discussed in more detail in the ``Calibration'' section below. Moreover, banks using a Basel II-based regime may have significantly higher or lower marginal regulatory capital charges than non-Basel banks for some types of loan products, resulting in potential pricing differentials. While Basel II might enable larger banks to compete more effectively for high quality credits, it could also result in larger concentrations of lower quality credits in smaller institutions. Finally, the potential implications on industry consolidation are simply not known. The banking agencies must continue to assess this situation and, if warranted, take steps to mitigate adverse effects on the competitive balance between large and small banks. We would be seriously concerned if, as an unintended consequence of the implementation of Basel II, we significantly alter the structure of banking in the United States. Calibration The first objective of the Basel Committee in embarking on the Basel II effort was to calibrate minimum capital requirements to bring about a level of capital in the industry that, on average, is approximately equal to the global requirements of the present Basel Accord. That calibration was to be designed to provide an incentive to banks to develop and maintain sophisticated and risk-sensitive internal ratings-based systems. In order to gauge its success in meeting that objective, the Basel Committee attempted to measure the impact of its proposals on the required capital levels of banking institutions through several quantitative impact studies. On May 5, 2003, the Committee published an overview of the results of its most recent assessment, the third quantitative impact study (QIS-3). On the basis of QIS-3 results, the Committee concluded that the aggregate results were generally in line with the objectives established for Basel II. Unfortunately, the QIS-3 data do not provide a reliable estimate of the likely regulatory capital requirements for banks subject to Basel II. And banks encountered several practical impediments to providing accurate estimates of the effect of the proposals on their measured ratios; thus, the estimated risk-based capital ratios were subject to a substantial margin of error. For example, in many cases, existing bank systems were not able to produce the data requirements necessary for inputs required by the new Accord. In some areas, the QIS-3 instructions were not sufficiently clear or were misinterpreted, and in other cases, the proposals were still in flux as banks were completing the survey. Most important, QIS-3 was completed without the rigorous supervisory validation and oversight that would occur when the proposal actually takes effect. A key concern is that focusing on the overall results of the QIS-3 exercise masks the wide dispersion of results for individual institutions. In the U.S., measured against current risk-weighted assets, the use of advanced approaches yielded results that ranged from a decrease in regulatory capital requirements of 36 percent to an increase of 43 percent. Similarly broad dispersions are found in a great many of the underlying components that make up the total capital requirement. While some dispersion of results in a truly more risk- sensitive framework would be expected, we are not convinced that the wide ranges indicated by QIS-3 can be explained by relative differences in risk among institutions; it appears that comparability of QIS-3 results among different institutions may be severely lacking. Finally, the quantitative studies that have been done to date have been based on unilateral inputs from the participating banks. We and other supervisors have had only very limited ability to review the veracity of the results. I want to be clear that we have no reason to believe that U.S. banks did not make every effort to provide results as accurate as possible given the constraints they were operating under. Nonetheless, it is certainly conceivable--I would say highly likely-- that the results might change significantly, and not necessarily in any particular direction, when all the intricacies of real-world implementation come into play. It seems fair to assume that banks will have fewer incentives to take conservative stances and greater incentives to exploit any loopholes or gray areas in the final rules; the extent to which these effects might be offset, or exceeded by, greater supervisory oversight is unknown. Notwithstanding the significant uncertainties noted above, it presently appears that the required capital levels of some U.S. institutions could drop significantly, even taking into account the temporary minimum floor capital requirements, discussed in the attachment. The OCC does not believe that some reduction in minimum regulatory capital requirements for certain institutions is, in and of itself, an adverse feature of Basel II. Such a result is only acceptable, however, if the reduction is based on a regulatory capital regime that appropriately reflects the degree of risk in that bank's positions and activities. Given the fact that relevant bank systems and procedures are still in development, the OCC is not yet in a position to make that determination as it relates to Basel II. As such, the OCC is not yet comfortable allowing national banks to materially lower their current capital levels simply on the basis of the output of the currently proposed Basel II framework. The OCC expects that an additional quantitative study will be necessary after the Basel Committee's work on Basel II is completed. Ideally, this should take the form of another global study by the Basel Committee itself--that is, a QIS-4. However, even if the Basel Committee does not undertake such a study, I believe that it is absolutely essential that the U.S. agencies do so prior to the adoption of final implementing regulations. I strongly believe that we cannot responsibly adopt final rules implementing Basel II until we have not only determined with a high degree of reliability what the impact will be on the capital of our banks, but we have also made the judgment that the impact is acceptable and conducive to the maintenance of a safe and sound banking system in the United States. Conclusion As I have indicated, the OCC firmly supports the objectives of Basel II--a more risk-sensitive and accurate capital regime. However, in light of the issues that have been identified with the current iteration of Basel II, the U.S. banking agencies must now determine how best to proceed on this critically important issue. I believe the following are essential elements in the agencies' consideration of Basel II implementation within the United States. First, the agencies need to move forward with the solicitation of comments on a Basel II-related ANPR and associated guidance. That is the most effective mechanism to have full and complete consideration of the proposal from all interested parties. The solicitation of comments on a proposed regulatory and supervisory structure for Basel II implementation will also permit supervisors to tangibly assess the feasibility of the proposal. Second, the agencies need to undertake additional steps to evaluate the costs, benefits, and other effects of the proposal before moving forward with any final regulatory action. Frankly, we simply need additional information to reasonably address the numerous issues, concerns, and uncertainties associated with Basel II implementation. We must better understand the likely consequences of this proposal on overall capital levels of affected institutions, the competitive effects on our financial system, and associated compliance costs and burdens. In determining the appropriate additional steps, the agencies should consider the obligations imposed under E.O. 12866, the other statutory requirements for consideration of costs and impact, lessons learned from QIS-3, and perhaps, a U.S. version of QIS-4. Third, as I have consistently reiterated, if we determine through this process that changes to the Basel II proposal are necessary, the U.S. agencies must pursue those changes, both domestically and in the Basel Committee. In this regard, the U.S. agencies should not foreclose consideration of alternative proposals that address the acknowledged deficiencies of the 1988 Accord but that do not constitute such a radical departure from our existing regulatory capital framework. Fourth, the overarching consideration for supervisors in moving forward on Basel II is the need to act in accordance with our primary mission--to ensure the continued maintenance of a robust and safe and sound banking system. We need to incent banks to continue to better measure and manage the full panoply of risks they face and to make use of new and evolving risk management practices. We must also ensure that prudential consideration of safety and soundness principles remain paramount. As I said in the beginning of my statement, the OCC, the agency to which Congress has committed the authority to define capital requirements for national banks, will not sign off on implementation of a final Basel II framework until we have fully considered all comments received during our notice and comment process. Given the importance of this proposal, the significant issues that remain unresolved, and the prospect that whatever emerges from this process is likely to govern the financial landscape for years to come, we need to take whatever time is necessary to develop and implement a revised risk-based capital regime that achieves the stated objectives of the Basel Committee in both theory as well as practice. I am pleased to have had this opportunity to provide our views on this important initiative, and I would be happy to answer any questions you may have. * * * Attachment Summary of Basel II: The Proposed New Accord Office of the Comptroller of the Currency The Basel Committee (the Committee) has been developing the new Accord over the past 5 years. During that time, three full-scale consultative papers (June 1999, January 2001, and April 2003) and numerous working papers supporting various elements of the new Accord have been released to the industry for comment. This summary is intended to convey a general idea of the structure and substance of the proposed new Accord, and does not attempt to provide a complete analysis. It is based on the most recent publications from the Basel Committee, notably the New Basel Capital Accord (Consultative Document) which is out for comment until July 31; the document can be found on the Committee's website at http:// www.bis.org/bcbs/index.htm. The new Accord will include menus of approaches for measuring the capital required for credit risk, market risk, and operational risk. For credit risk and operational risk, each of the proposed approaches is described briefly below; capital charges for market risk are unchanged in the new Accord and are not discussed here. Some of the approaches described are unlikely to be implemented in the United States and have been noted as such. Moreover, based on preliminary analysis by the U.S. agencies, currently there are less than a dozen U.S. banks that would be mandatorily subject to Basel-based regulatory capital requirements. While other banks would be permitted to opt in to the Basel rules (subject to meeting prudential qualification requirements), the U.S. capital rules will remain in place for the vast majority of U.S. banks that either are not required to or do not opt to apply the Basel II framework. Of course, any issues regarding U.S. implementation of the new Accord will be definitively resolved only after the U.S. rulemaking process has been completed. The current structure of the Accord has been influenced by the results of several quantitative impact studies (QIS), the most recent of which was completed in December 2002. Approximately 20 U.S. banks participated in the QIS exercise in December and the results have been factored into the most recent version of the Accord. Changes were made in several areas including the treatment of retail credits, specialized lending, securitization, and operational risk. General Structure of the Proposed New Accord The new Accord has three mutually reinforcing ``pillars'' that make up the framework for assessing capital adequacy in a bank. The first pillar of the new Accord is the minimum regulatory capital charge. In order to calculate the capital charge under Pillar 1, banks will have to determine the individual charges for credit, market, and operational risk. The new Accord offers a series of options for calculating credit and operational risk. Market risk will remain unchanged from a 1996 amendment to the Accord. The new options for credit and operational risk were designed to be available to a wide range of banks, from relatively simple to very complex. For credit risk, the Pillar 1 capital requirement includes both the standardized approach, updated since the 1988 Accord, and the new Internal Ratings-Based (IRB) approaches (foundation and advanced). Pillar 1 has been the focal point of much of the discussion and comment from the industry on the new Accord. Pillar 2 covers supervisory review and banks' obligation to hold sufficient capital vis-a-vis their risk profile. The pillar is, ``intended to ensure not only that banks have adequate capital to support all the risks in their business, but also to encourage banks to develop and use better risk management techniques in monitoring and managing these risks.'' This pillar encourages supervisors to assess banks' internal approaches to capital allocation and internal assessments of capital adequacy. It provides an opportunity for the supervisor to indicate where such approaches do not appear sufficient. Pillar 2 is also a way to focus supervisors on other means of addressing risks in bank's portfolio, such as improving risk management techniques and internal controls. The third pillar recognizes that market discipline has the potential to reinforce capital regulation and other supervisory efforts to ensure the safety and soundness of the banking system. Thus, the new Accord proposes a wide range of disclosure initiatives, which are designed to make the risk and capital positions of a bank more transparent. As a bank begins to use the more advanced methodologies for market and operational risk, the new Accord will require a significant increase in the level of disclosure. In essence, the tradeoff for greater reliance on a bank's own assessment of capital adequacy is greater transparency. This pillar has been subject to numerous changes as the Committee has worked to balance the need for robust disclosure with a recognition of the proprietary and confidential nature of some of the information. Capital for Credit Risk Under Basel II, banks must select one of three approaches to determine their capital for credit risk. The three approaches, from simplest to most complex are: The standardized approach, the foundation IRB, and the advanced IRB. Standardized Approach The 1988 Accord introduced the standardized risk-bucketing approach for setting the minimum regulatory capital requirement, which is still used in the United States today. The approach has been subject to criticism that it lacks sufficient risk sensitivity. The revised standardized approach under Basel II enhances the 1988 Accord by providing greater, though still limited, risk sensitivity. Key changes to create a more risk-sensitive framework include the refinement and addition of risk buckets, the introduction of external credit ratings, and a wider recognition of credit risk mitigation techniques. Risk weights are still determined by category of the borrower--sovereign, bank, or corporate--but within each of these categories changes have been made to make the capital more reflective of the riskiness of the asset category. For example, the risk weight on mortgage loans has decreased from 50 percent to 35 percent and the risk weight on certain retail credits has moved from 100 percent to 75 percent. Risk weights for externally rated corporate credits, currently 100 percent, will range from 20 percent to 150 percent. Sovereign risk weights are no longer dependent upon whether a country is a member of the Organization for Economic Cooperation and Development (OECD), but rather on the external rating identified for the country. The standardized approach is not likely to be implemented in the United States. U.S. supervisors believe that credit risk measured under the standardized approach of Basel II would generally not be appreciably different than that measured under current rules for most U.S. banks, and the marginal changes in capital requirements would not justify the cost of implementation. Internal Ratings-Based Approach (Foundation and Advanced) The IRB approach represents a fundamental shift in the Committee's thinking on regulatory capital. It builds on internal credit risk rating practices used by some institutions to estimate the amount of capital they believe necessary to support their economic risks. In recent years, as a result of technological and financial innovations and the growth of the securities markets, leading banking institutions throughout the world have improved their measurement and management of credit risks. These developments have encouraged the supervisory authorities to devote greater attention to introducing more risk- sensitive regulatory capital requirements, particularly for large, complex banking organizations. Banks must meet an extensive set of eligibility standards or ``qualifying criteria'' in order to use the IRB approach. Because the requirements include both qualitative and quantitative measures, national supervisors will need to evaluate compliance with them to determine which banks may apply the new framework. The requirements vary by both the type of exposure and whether the bank intends to use the simpler foundation IRB framework or the more advanced IRB framework. The requirements are extensive and cover a number of different areas, including rating system design, risk rating system operations, corporate governance, and validation of internal estimates. A brief sample of actual criteria include: <bullet> The board of directors and senior management have a responsibility to oversee all material aspects of the IRB framework, including rating and probability of default (PD) estimation processes, frequency and content of risk rating management reports, documentation of risk rating determinations, and evaluation of control functions. <bullet> A 1-year PD estimate for each grade must be provided as a minimum input. <bullet> Banks must collect and store historical data on borrower defaults, rating decisions, rating histories, rating migration, information used to assign ratings, PD estimate histories, key borrower characteristics, and facility information. As mentioned above, the requirements that a bank must meet are partially dependent upon which of the two IRB approaches a bank will use. The first methodology, called the foundation approach, requires fewer direct inputs by banks and provides several supervisory parameters that, in many cases, carry over from those proposed for the standardized approach. For a variety of reasons, the United States does not plan to introduce the foundation approach in its regulations. The second approach, the advanced IRB approach, allows banks much greater use of their internal assessments in calculating the regulatory capital requirements. This flexibility is subject to the constraints of prudential regulation, current banking practices and capabilities, and the need for sufficiently compatible standards among countries to maintain competitive equality among banks worldwide. There are four key inputs that are needed under IRB, for both the foundation and advanced approaches. The first element is the PD of a borrower; the bank is required to provide the PD in both the foundation and the advanced approaches. The second input is the estimate of loss severity, known as the loss given default (LGD). The final two elements are the amount at risk in the event of default or exposure at default (EAD) and the facility's remaining maturity (M). LGD, EAD, and M are provided by supervisors in the foundation approach, but must be provided by banks operating under the advanced approach (subject to supervisory review and validation). For each exposure, the risk weight is a function of PD, LGD, and EAD. The IRB approach envisions internal rating systems that are two- dimensional. One dimension focuses on the borrower's financial capacity and PD estimates that quantify the likelihood of default by the borrower, independent of the structure of the facility. The other dimension takes into account transaction-specific factors such as terms, structure, and collateral. These characteristics would determine the second dimension, that is, the LGD. Implicit in this treatment is the assumption that when a borrower defaults on one obligation, it will generally default on all its obligations. (This assumption is relaxed with the IRB treatment of retail portfolios.) Calculating the capital charge under the IRB approach involves several steps. The first of these steps is the breakdown of the bank's portfolio into five categories: Corporate (including commercial real estate), retail, bank, sovereign, and equity. The IRB rules differ to varying degrees across these portfolios. As a result, the IRB capital charge is calculated by category, with the PD, LGD, and EAD inputs potentially differing across these categories. Supervisory approval is needed before banks can use the IRB approach for any of the five categories. The minimum requirements described above were written to apply across these five types of exposures. Another important step is the determination by the bank of the PD's for its loan grading categories. The PD of an exposure is the 1-year PD associated with the borrower grade, subject to a floor of 0.03 percent (excluding sovereigns). The determination of PD's for borrowers supported by guarantees or credit derivatives is more complex. Banks under the advanced approach would use their internal assessments of the degree of risk transfer within supervisory defined parameters, while those under the foundation approach would use the framework set forth in the new credit risk mitigation provisions. Overall, the PD must be, ``grounded in historical experience and empirical evidence,'' while being ``forward looking'' and ``conservative.'' A reference definition of default has been developed for use in PD estimation and internal data collection of realized defaults. Once the PD has been established, banks must then establish the dimensions of LGD based on collateral and M. Under the foundation approach, M is assumed to be 2.5 years. There are several options that may be selected for the advanced approach, but in general, M is defined as the greater of 1 year or the remaining effective maturity in years. After the bank determines the PD's and LGD's for all applicable exposures, these combinations can be mapped into regulatory risk weights. The risk weights, which are calibrated to include coverage for both expected and unexpected losses, are expressed as a continuous function. The minimum capital charge is then determined by multiplying the risk weight by the amount expected to be outstanding at the time of default (EAD), and by 8 percent. A final step in this process involves the ongoing review by the supervisors of the systems used to develop the IRB capital charge. Periodically, supervisors will need to validate these systems and review the internal controls that provide the foundation for the IRB approach. In addition, supervisors will also have to consider, under Pillar 2, whether the amount of capital generated by the IRB approach is commensurate with the bank's risk profile. Implementation of the IRB Approach In addition to the requirement that a bank meet the qualifying or eligibility criteria, the new Accord requires that banks using the IRB approach run parallel systems for 1 year before implementation. This means that a bank planning to implement the IRB approach in December 2006, will actually have to begin calculating results as of December 2005, while continuing to run its current systems. Adjustments to the Capital Charge for Credit Risk There are additional considerations that banks may have to factor in when determining the capital charge for credit risk. These additional considerations will further adjust required capital, outside of the requirements of the different approaches to credit risk. The two primary adjustments that might be made to the credit risk charge are for credit risk mitigation and asset securitization. Credit Risk Mitigation The new Accord provides a measure of capital relief for certain qualifying risk-mitigating techniques used by banks. However, it is important to note that most of the credit risk mitigation proposals in the new Accord are only directly relevant to the standardized or foundation IRB approaches, which are not likely to be used in the United States. In the advanced IRB approach, credit risk mitigation must meet certain qualitative requirements, such as legal certainty. In addition, specific proposals related to maturity mismatches and backtesting requirements of certain model results are applicable to the Advanced IRB approach. Otherwise, it is assumed that any credit risk mitigation efforts will be factored into the PD's and LGD's assigned by the bank. With that caveat in mind, the section on credit risk mitigation in the new Accord attempts to provide some rough approximations of the risk reduction attributable to various forms of collateralized credit exposures, guarantees, credit deriva- tives, and on-balance sheet netting arrangements. The Committee has proposed a conceptual approach to these risk mitigation techniques that, while recognizing their risk reduction benefits, attempts to capture the additional risks posed by such transactions. The credit risk mitigation proposal provides both a simple and a comprehensive approach to dealing with collateral. The proposal expands the range of eligible collateral from that recognized in Basel I. It also discusses the appropriate treatment for maturity mismatches between the credit risk mitigant and the underlying credit exposure. The proposal introduces ``haircuts,'' which the bank may estimate, to cover the market price and foreign exchange volatility that may be inherent in collateral. The proposal allows banks to greatly reduce the capital requirements for exposures with large amounts of high quality collateral. There are strict quantitative and qualitative factors that must be met in order for a bank to be permitted to use its own haircut estimates. The proposal encourages the use of credit risk mitigation by expanding the type of collateral, guarantors, and transaction structures that are recognized for capital reduction. Different types of credit risk mitigation techniques pose different levels of additional risk; the proposal incorporates flexibility that recognizes these differences and adjusts the capital treatment accordingly. Asset Securitization Asset securitization is clearly an important issue in the United States, as the securitization market is significantly greater than the securitization market of any other Basel-member country. The Committee believes that it is important to construct a more comprehensive framework to better reflect the risks inherent in the many forms of asset securitizations, including traditional and synthetic forms. The securitization framework in the New Basel Accord applies generally when there is a transaction that involves the stratification or tranching of credit risk. The Committee has developed securitization approaches for both standardized and IRB banks. The level of complexity is significantly higher for IRB banks. The framework tries to focus on the economic substance of the transaction, rather than on its legal form. Under the proposal for the treatment of securitizations by standardized banks, the capital charge is generally determined by multiplying the amount of the securiti- zation exposure by the risk weight mapped to the long- and short-term rating categories. Off-balance sheet exposures are subject to a conversion factor before the appropriate risk weight is applied. The proposal does allow for some recognition of credit risk mitigants provided on securitization exposures, but that recognition is permitted only when the bank meets a series of stringent criteria. Banks that adopt the IRB approach for credit risk are generally required to use one of two methods for determining capital requirements for securitization exposures. One method is the Supervisory Formula Approach (SFA), under which capital is calculated through the use of five bank-supplied inputs: The IRB capital charge on the underlying securitized exposures (as if held directly on the bank's balance sheet); the tranche's credit enhancement level and thickness; the pool's effective number of loans; and the pool's exposure weighted average loss given default (LGD). The second method is known as the Ratings-Based Approach (RBA). Under this approach, capital is determined by multiplying the amount of the exposure by the appropriate asset-backed security risk weights, which depend on external rating grades, short- or long-term. Granularity of the pool and the level of seniority of the position are also considered. The securitization proposal is one of the newest pieces of the Accord and its potential impact on the industry is still being assessed. In the December 2002 QIS exercise, banks were asked for the first time to provide data on the relative impact of the proposals. The QIS results did not provide entirely reliable results. However, the Committee has responded to some of the concerns raised during the QIS process by making changes to the securitization framework. One key change was the introduction of a simpler approach for liquidity facilities. Operational Risk One of the most significant changes in the new Accord is the proposal for an operational risk charge. It is expected to represent, on average, 10-15 percent of the total minimum regulatory capital charge. The framework is based upon the following operational risk definition: The risk of loss resulting from inadequate or failed internal processes, people, and systems or from external events. This includes legal risk, but excludes strategic and reputational risks. The Committee has proposed three approaches to calculate the operational risk charge, which represent a continuum of increasing sophistication and risk sensitivity. The Basic Indicator Approach (BIA) is the simplest of the three approaches; the capital charge is determined by taking an alpha factor decided by the Committee and multiplying it by an indicator, gross income. The next approach is known as the Standardized Approach and is similar to the BIA, but breaks out gross income into business lines. The Committee has introduced an Alternative Standardized Approach to address some of the concerns raised by the results of the December 2002 QIS exercise; this is not a separate approach, but rather a modification to the Standardized Approach. Because there is no compelling link between these measures and the level of operational risk, the United States does not plan to utilize the BIA or the Standardized Approach (including the Alternative Standardized Approach) to determine the capital charge for operational risk. The Committee has made the most significant changes to the advanced approach since it was originally introduced in January 2001. At that time, the Committee envisaged a single, very prescriptive advanced approach for operational risk, similar to credit risk. However, after numerous comments from the industry, the Committee made substantive changes in the proposal to reflect the evolutionary nature of the operational risk framework. The Committee recognized that, unlike credit risk, there are very little data and no internal systems specifically designed to target operational risk; instead, banks and supervisors rely primarily on internal controls to deal with a myriad of banking risks that cannot be as readily quantified as credit and market risks. The Committee considered the comments and analyzed the state of the art of operational risk and developed what is known as the Advanced Measurement Approaches (AMA). Rather than prescribing one methodology, the AMA will allow banks the option of designing the operational risk measurement framework that best suits their institution, subject to some broad criteria. The criteria will be the key to achieving a certain level of consistency and comparability among institutions, as well as providing a margin of comfort to supervisors who must assess these differing systems. The criteria currently identified in the new Accord include the need for internal and external data, scenario analysis, and consideration of business environment and internal control factors. Banks may also, under the AMA, consider the impact of risk mitigation (such as insurance), again subject to certain criteria set to ensure that the risk mitigants act as an effective capital- replacement tool. Temporary Capital Floors Two floors that have been established for the Basel II framework. In the first year of implementation, an institution's required minimum level of regulatory risk-based capital cannot be less than 90 percent of the minimum level of capital that would be required under the Agencies' general risk-based capital rules. In the following year, an institution's minimum level of regulatory risk-based capital cannot be less than 80 percent of the minimum amount required under the Agencies' general risk-based capital rules. ---------- PREPARED STATEMENT OF DONALD E. POWELL Chairman, Federal Deposit Insurance Corporation June 18, 2003 Thank you, Mr. Chairman and Members of the Committee. I welcome the opportunity to testify on behalf of the Federal Deposit Insurance Corporation on the New Basel Capital Accord (Basel II). The proposals contained in the Third Consultative Paper (CP-3) recently published by the Basel Committee on Banking Supervision, if adopted in the United States, would easily rank among the most important pieces of banking regulation in our Nation's history. Introduction Basel II would change bank capital regulation in the United States in at least three important ways. First, rather than emphasizing simple preset minimum numerical capital ratios, Basel II would allow qualifying banks to use their own internal risk estimates as inputs to regulator-supplied formulas with the supervisors providing oversight and evaluation of the banks' ability to measure risk. Second, the new framework would formally adopt a ``bifurcated'' capital system in the United States: One set of rules for the large, complex, and internationally active institutions, and another set for the balance of banks in the country. A third key change is that the total minimum regulatory capital charge under the new framework will include an explicit charge for operational risk. For those large institutions that qualify, the new framework may lead to reduced credit risk capital requirements for certain asset classes with additional capital held based on a flexible operational risk charge. The FDIC supports the overall goal of Basel II, which is to create regulatory capital standards that are more sensitive to the economic substance of risks taken by these large banks, to limit their opportunities for regulatory capital arbitrage and to encourage sound risk management. Over the years that Basel II has been under development, the Basel Committee and the U.S. Federal supervisors have reached out to the industry and the public for comment on how to more closely align the proposed new framework with the ways that large banks measure risk. There have been quantitative impact studies to assess the potential impact on capital levels. We have been engaged in roundtables and discussions. Over this time, various aspects of the new framework have been refined and changed. Today, these refinements are reflected in CP- 3, which the Basel Committee recently released for additional comment. The work in this country continues. The agencies intend to issue an Advance Notice of Proposed Rulemaking (ANPR) that will suggest how CP-3 will be proposed for adoption in the United States and will seek additional comments on all facets of Basel II. As in the past, it can be anticipated that further changes to the framework may be required. The FDIC is committed to an interagency process to achieve the overall goals of Basel II and to fully understand its possible impact on bank capital levels and competitiveness. The goal of more closely tying regulatory capital to banks' own internal assessment of risk is a good one. This goal is reached in part by using regulatory capital formulas that are based on ways of measuring credit risk and allocating internal capital that, to some degree, are already in place in large banks. The term ``economic capital'' is often used to refer to the amount of capital that should be allocated to an activity according to the results of a numerical loss analysis. Banks use models based on historical data and economic analysis to estimate future losses and the amount of income, reserves, and capital needed to ensure their portfolios conform to management's target level of risk. These calculations produce different results for different bank activities. For example, the measured risk on residential mortgages might be much less than the measured risk on construction loans. The bank might use the economic capital measures to compute its risk- adjusted returns on the two activities and to assist its pricing decisions. This is a disciplined approach to risk management, and Basel II establishes firm expectations for banks to be rigorous in this respect. Basel II expands these risk management expectations beyond the area of credit risk and into the realm of operational risk. Tying capital requirements closer to risk and increasing the incentives for disciplined risk management have the potential to improve the safety and soundness of the U.S. financial system. The FDIC supports enhancing the incentives for the largest banks in the United States to strengthen risk management processes. Tying regulatory capital closer to risk would reduce the incentives for banks to make uneconomic decisions designed to reduce regulatory capital. At the same time, the domestic impact of Basel II has not been determined. Given current analysis, it seems likely Basel II will confer some degree of regulatory capital benefits on the limited number of banks that qualify, in exchange for their substantial investments in systems and infrastructure intended to improve risk management. The critical issue for the safety and soundness of our financial system is whether the improvements in risk management systems, and the resulting bank risk profiles, would justify the level of capital reductions that banks might ultimately realize. It is virtually impossible to quantify at this time the potential changes in capital under Basel II. Basel II proposes floors by which risk-based capital would be allowed to decline by at most 10 percent the first year of implementation, and at most 20 percent the second year. After the second year, Basel II does not impose a floor on the minimum risk-based capital requirement. A quantitative study conducted in the fall of 2002 showed a wide range of changes in capital requirements for 19 large U.S. banks under the Advanced Internal Ratings-Based (A-IRB) approach, with an average reduction in capital requirements for credit risk of 17 percent. In this study, the reduction in capital was offset by the operational risk capital charge, which was substantial. However, the amount of this operational risk charge was by necessity estimated using an approach that will not be used in the United States. The agencies understand that the results to date of the impact studies do not provide a full picture of the possible impact of Basel II. There are many moving parts to the proposal and the banks' participation in the study was on a best efforts basis. Moreover, in the United States, leverage ratio floors and the demands of the marketplace would act as a constraint on the potential reduction in actual capital. Still, these initial estimated results show that the Basel II formulas are potent instruments for affecting risk-based capital requirements in the United States. This is a matter of great interest to the FDIC and we are committed to working with the other banking agencies as we move forward to more accurately assess the impact of the proposed new standards. A significant business challenge for the banking and thrift agencies would be how to achieve interagency consistency in the application of these complex rules. Required capital charges will depend heavily on the ongoing judgments of banks and regulators about a variety of specific risks. In addition to understanding the impact of Basel II on capital levels, we must also understand the significance of mandating two tiers of regulatory capital standards--a bifurcated framework that will offer competitors different regulatory capital charges for similar assets. The critical issues in terms of the competitive playing field are whether the direct competitors of a core group of about 10 large banks would feel forced to opt in to the new framework for competitive reasons, and whether banks in the tier below those able to opt in would be at substantial competitive risk. To resolve these fundamental issues satisfactorily, much hard work remains. Given the magnitude of the issues, we must proceed carefully. Capital Adequacy The U.S. banking system has weathered the last 10 years better than the banking systems of some other countries for a number of reasons. One significant reason is strong capital levels. Bank capital is subject to Federal legislation and regulation because of its critical importance to the health and well-being of the U.S. financial system. An adequate capital cushion enhances banks' financial flexibility and their ability to withstand periods of adversity. As insurer, the FDIC has a vital stake in the adequacy of bank capital--as do our fellow regulators and all U.S. taxpayers. Congress recognized this important principle when it established the Prompt Corrective Action (PCA) requirements in the Federal Deposit Insurance Corporation Improvement Act. A critical aspect of the existing PCA regulations is the minimum leverage capital requirement. To be considered well-capitalized, a bank must have a ratio of Tier 1 capital-to-total assets (the leverage ratio) of at least 5 percent. Banks with leverage ratios under 4 percent are considered undercapitalized. The agencies agree that maintaining the minimum regulatory capital standards as reflected in the current PCA legislation and existing implementing regulations is very important. Capital is not the only thing needed for safety-and-soundness. The strong risk management that Basel II promotes is also essential. There is no denying that banks with good risk management and a lower-risk profile should be able to operate with somewhat less capital than more risky banks. But there is also no denying that when the unexpected happens, the hard-earned benefits of risk management can evaporate overnight without adequate capital. The sophistication of the measurement of economic capital can make it easy to lose sight of the fact that, in reality, no one knows the range of potential future losses for a given activity, or the associated probabilities. Certain risk management practitioners express great faith in the calculation of economic capital, and believe that the regulatory capital standard should in all instances be less than the economic capital amount. The idea behind this philosophy is that banks tend to be forced out of low-risk activities where regulatory capital requirements exceed economic capital requirements. It is this belief that gives us concern about a clash of expectations about Basel II between a number of prominent risk management practitioners on the one hand, and the FDIC and our fellow bank regulatory agencies on the other. As the regulators move forward to finalize our views on Basel II, we need to proceed cautiously. Where a proposal seems to run counter to established U.S. supervisory practice, we need to ask whether the established practice should be reexamined in light of the proposed new rules, or whether the new rules need to be reexamined for U.S. purposes. Basel II is the object of intense scrutiny and comment. Changes have been and will be suggested by banks in many areas, including the treatment of commercial real estate, credit cards (and the related issue of future margin income), mortgages, securitizations, and capital recognition of certain risk-mitigating activities. The potential for many moving parts could make it difficult to evaluate the capital impact or the competitive impact of Basel II. Yet, we believe that we must achieve a better understanding of these issues before the bank regulatory agencies commit the United States to the new framework. Interagency Consistency Basel II would provide banks and supervisors some flexibility to determine what capital would be held on an ongoing basis. The degree of conservatism to apply to a particular situation would often be a judgment call. Is the loss given default on a secured commercial loan likely to be 20 percent or 40 percent? Capital for that loan would double, or be cut in half, depending on the answer--and the answer could well depend on a mix of historical data, the specific underwriting methods used by individual banks and the specific analytical techniques banks use to make their case. Supervisors would need to validate--uniformly and consistently across banks--the answers to such questions. In this new framework, regulators must be prepared to challenge the modeled outputs of sophisticated risk measurement systems of the largest U.S. financial institutions, a difficult and demanding task. It will require courage and discipline to respond to this new challenge. Much progress has been made by the regulators and the industry in deciding how this validation might be done. Interagency guidelines are being drafted and implementation approaches are being discussed. The FDIC has an active interest in the development of a sound approach to ensure the consistent and uniform review of bank risk measurement systems under Basel II. A Level Playing Field Capitalism, with its inevitable winners and losers, is about competition. It is the job of the regulators to make certain that the competition is fair. In our capitalist system, one of the key functions of regulation is to ensure the rules do not display favoritism and that the competitive struggle is carried out on equal terms. We need to evaluate Basel II against this standard before committing to implement it in the United States. The proposed agreement raises several very important questions. The fundamental question is what are the economic benefits of the regulatory capital relief some banks might realize under Basel II? Conversely, what are the costs of additional capital they might be required to hold for certain activities? Would small or mid-sized regional banks, unable to qualify for the new framework, become acquisition targets of Basel II banks whose reduced capital has boosted their returns on equity? Would a large credit card bank that must hold capital for unused credit card lines be at a disadvantage to a non- Basel bank that faces no such requirements? Would a securitizing regional bank that is forced to deduct most of its retained interests from capital be at a disadvantage to a Basel bank whose deductions from capital would now be capped? What would be the ramifications of significantly reduced capital requirements for Basel banks on specific assets held by banks of all sizes, such as mortgage-backed securities issued by the Federal Government sponsored enterprises? The Basel II formulas are designed to work for large diversified portfolios, and the capital requirements they produce might be too low for most small banks. The Basel framework also requires significant systems investments at a level likely beyond the reach of--and not essential for--small institutions. Therefore, it is not practical to think that any competitive concerns that may exist could be resolved simply by allowing all banks access to the Basel framework. To a large extent, the banking system in the United States is already a two-tier system, with large financial institutions possessing the vast majority of U.S. bank assets. Still, we must evaluate thoroughly whether Basel II will unnecessarily disturb this current, albeit divided, field of competition. Even though the industry may already be divided between the large and complex and the small and less complex, banking supervisors must understand fully whether Basel II adds significant additional competitive pressures or would trigger additional industry consolidation. The ANPR will seek input from all interested parties, including banks that believe they will be competitively harmed if they cannot embrace the Basel II framework. Conclusion An Advance Notice of Proposed Rulemaking will be issued this summer and will reflect the United States banking and thrift agencies' views on how Basel II would be adopted in the United States. More importantly, it will present issues and concerns, and raise questions to the industry and the public. The comments will pro- vide invaluable insight to many of the key concerns being raised by the agencies and by Congress. Given the importance of these issues, it is vital that we treat the implementation of Basel II in the United States as we would any other proposed regulation--with a dose of skepticism, a willingness to entertain the discussion of options, and a commitment to fully explore potential costs and benefits before reaching a final decision. We need to listen carefully to comments that will be received in the rulemaking process to ensure we address these threshold issues. It also is important that the financial services industry, the Congress, and the banking agencies have a full opportunity to review the response to the ANPR and achieve a better understanding of the impact of this proposed agreement before we commit the United States to the Basel II approach. The FDIC has no interest in delaying the agreement and its implementation beyond what is necessary to address the issues we have raised and to understand the impact of this new system of capital regulation. I have full confidence that this interagency process will work and will arrive at an appropriate outcome. The FDIC will continue to remain fully involved in this process and will work to ensure that the goals of Basel II and of Congress are being met as the process moves forward. Thank you for the opportunity to present the views of the FDIC. ---------- PREPARED STATEMENT OF JAMES E. GILLERAN Director, Office of Thrift Supervision June 18, 2003 Introduction Good morning, Chairman Shelby, Senator Sarbanes, and Members of the Committee. Thank you for the opportunity to discuss the proposed revisions to the 1988 Capital Accord (Basel I) developed by the Basel Committee on Banking Supervision (BSC). Although the Office of Thrift Supervision (OTS) has been involved in the Basel process for some time, we have only recently attempted to engage ourselves in the process internationally. While we are very supportive of the Basel process, there are numerous policy implications involved in the recently proposed international capital standards for banking organizations in the United States. These include issues that we all must strive to understand and address. I welcome your efforts to highlight these pending and important changes. The proposed change in capital standards currently under consideration arises from a third consultative paper, CP-3, recently issued for public comment by the BSC. CP-3 is expected to result in the New Basel Capital Accord, or Basel II. Basel II will directly affect the largest and most internationally active banking organizations around the world, including approximately 10 banking organizations in the United States. Basel II may also significantly impact, albeit indirectly, all other banking organizations around the world, including roughly 9,500 institutions in the United States. These institutions include large, medium, and small banks and thrifts that operate nationally, regionally, and at the community level, many of which compete domestically with our largest internationally active banking organizations. Development of Basel II Basel I, signed in 1988, addressed only the largest, internationally active banks in G-10 countries and encouraged countries outside the G-10 to adopt the framework for their banks that were operating internationally. The underlying principles of Basel I, however, were intended to apply to all banking organizations of any size and activity. Thus, while OTS did not sign Basel I, we applied it along with the other Federal banking agencies. Since Basel I, the four banking agencies have developed risk-based capital standards consistent with its underlying principles, but with modifications intended to enhance risk sensitivity. In connection with our involvement and experience with Basel I, OTS has been monitoring for many years the work leading up to Basel II. Because of the potential impact of Basel II on the institutions we regulate, we stepped up our involvement in the Basel process. In anticipation of the domestic application of Basel II, OTS is participating fully in preparation of an interagency Advanced Notice of Proposed Rulemaking (ANPR), with accompanying supervisory guidance, to be published in the Federal Register in the near future. The initiative will trigger the official kick-off of the national debate on the subject of new international capital standards, but, as you are aware, many of the issues raised by Basel II have already attracted significant attention. While OTS has not been directly involved in the international deliberations to date, our role on the domestic front-- particularly in the mortgage markets--provides us a unique and useful perspective for this discussion. In Basel I, the BSC identified two fundamental objectives at the heart of its work on regulatory convergence. As the Committee stated, first, ``the new framework should serve to strengthen the soundness and stability of the international banking system; and [second,] the framework should be fair and have a high degree of consistency in its application to banks in different countries with a view to diminishing an existing source of competitive inequality among international banks.'' Although the BSC developed a far more detailed and risk- sensitive capital adequacy framework in Basel II than in the original Accord, it does not stray from the objectives set 15 years earlier. In fact, the BSC expanded upon these objectives as a guide to its efforts in producing the current proposal. In particular, the Committee observed that Basel II should: <bullet> Continue to promote safety and soundness and at least maintain the current overall level of capital in the system. <bullet> Continue to enhance competitive equality. <bullet> Establish a more comprehensive approach to address risk. <bullet> Contain approaches to capital adequacy that are appropriately sensitive to risk. <bullet> Focus on internationally active banks, although its underlying principles should be suitable for application to all banking organizations. While the objectives for Basel II set forth by the BSC are important to ensure consistency and competitiveness among internationally active banking organizations, the impact of the proposed changes may affect many other banking entities domestically. It is important to encourage a thorough discussion among the regulators, Congress, and the thousands of banking organizations in the United States that may be affected, directly or indirectly, by Basel II. Hearings such as this and the upcoming ANPR will help stimulate this debate. Overview of Basel II Basel II contains three ``pillars'' that are intended to be mutually reinforcing. Pillar 1 is a minimum regulatory capital requirement; Pillar 2 addresses supervisory review; and Pillar 3 is intended to promote risk and capital transparency. Briefly, a description of these is as follows: <bullet> Pillar 1 includes a credit risk component that is measured by either a standardized approach or one of two internal ratings-based approaches. The two ratings-based approaches or models are the Advanced Internal Ratings-Based (A-IRB) approach and the ``Foundation'' approach. Pillar 1 also includes an operational risk component that has several optional approaches. The centerpiece of the operational risk component of Pillar 1 also permits use of an internal model, the Advanced Measurement Approach (AMA). <bullet> Pillar 2 is viewed by the BSC as a way for the banking supervisors to attain better overall risk management and internal controls at the banking organizations we regulate. <bullet> Pillar 3 includes a wide range of disclosure initiatives designed to make the risk and capital positions of banking organizations more transparent. Issues for Consideration As I noted at the outset, OTS has only recently sought to be involved internationally in the Basel process. While we are supportive of this process and encouraged by the work completed so far, both domestically and internationally, there are a number of issues that we have considered regarding the application of Basel II in the United States. In the following discussion, I highlight some of these issues. Competitive Equality Regardless of how we strive to explain Basel II, the extraordinary technical detail at its core is substantial. Our banking organizations will need to master the complexity of Basel II to provide effective feedback during the upcoming ANPR comment process on the balance of its burdens and benefits. As we proceed, we need their input to weigh changes to our existing capital rules, and to assure ourselves that our actions do not significantly alter the competitive landscape for all U.S. banking entities. We want to assure that U.S. banking organizations remain healthy, competitive, and well-capitalized. The key principle underlying Basel II, and the basis for the advancement from Basel I, is greater risk sensitivity. This principle has as much meaning for a small community banking organization as it does for a large internationally active institution. The challenge lies in how to address this issue simultaneously for both types of banking organizations, especially considering that under the proposed scope of application in the United States, all but the few largest banking organizations will not be ``Basel II banks.'' A significant issue in this debate is whether we maintain consistent capital standards for all banking organizations for lending activities that have the same risk characteristics. From our standpoint, maintaining competitive equality for community banks is important, particularly as our economy is showing encouraging signs of improvement. Community banking organizations play a significant role in small business lending, which feeds new job creation. ``Community banks are one of the key sources of credit and other financial services to small businesses--the most prolific job creating sector of our economy. Small businesses employ 60 percent of the Nation's workforce and have created two-thirds of all the net new jobs since 1970.'' \1\ --------------------------------------------------------------------------- \1\ Statement of Paul G. Merski, Chief Economist and Director of Federal Tax Policy, Independent Community Bankers of America, before the House Small Business Committee, March 1, 2002. --------------------------------------------------------------------------- Another aspect of this issue that we must consider is the extent to which we alter our existing capital rules, applicable to all banks, to accommodate changes proposed by Basel II. For example, under Basel I, the blunt-edged risk-based capital requirement for 1-4 family residential mortgages (a 50 percent risk-weight, or 4 percent capital requirement) is not commensurate with the historical risk associated with residential mortgage lending in the United States. For residential mortgage loans with relatively low loan-to-value ratios, a substantially lower risk--weight is more reflective of loss experience. By contrast, the Federal banking agencies have concluded that for some concentrations of subprime loans, a significantly higher risk weight than 100 percent--and therefore, a capital requirement higher than 8 percent might be more appropriate. While Basel II is intended to enhance the risk sensitivity of our capital rules, it is important that the proposed changes are truly reflective of actual risk, as measured over an appropriate historical timeframe. Supervisory Effectiveness Another important issue is the potential impact of Basel II on our supervisory effectiveness. The U.S. bank regulatory system is considered to be among the most comprehensive and admired in the world. Capital requirements are only part of our multifaceted supervisory response to ensure safety and soundness. Our supervisory system is grounded in a regular program of on-site examinations complemented by comprehensive and frequent reporting and off-site monitoring--a level of supervisory review that may be unparalleled. As we move forward with a relatively dramatic approach that places a tremendous emphasis on capital, we must be careful not to minimize or diminish the other supervisory tools and regulatory judgment that is integral to our supervisory system. In particular, we should focus on how Basel II fits within and improves our system, and how to strike the right balance between capital rules and effective supervisory oversight. In the end, sound regulatory judgment is the key to our supervisory effectiveness and cannot be compromised. Accountability in a Ratings-Based Capital Model A corollary to this issue is the role of examiners and our examination process in evaluating ratings-based models dictated in a Basel II supervisory world. The application of Basel II in the United States will include complex mathematical formulas and models used to measure regulatory capital levels for our largest financial institutions. While prior regulatory approval is required to use the models, once obtained, an institution would effectively set its own capital requirements. This would be based largely on inputs derived from credit assessments from the institution's own credit risk and operational risk models. The accuracy and consistency of ratings is extremely important in any ratings-based system. Numerous subjective decisions are made daily by bank personnel regarding model inputs. These inputs involve judgments made on items such as rating a loan's probability of default, an estimate of loss given default, and the probability of a major loss arising from an institution's operational risk. It is important to keep in mind that these are human inputs, and are not infallible. Of particular concern is how to account for the subjectivity of the ``human factor'' as we implement and apply Basel II. Equally important is that we take the steps necessary to support and train our examiners who will be expected to review the many subjective decisions made under, and evaluate the mathematical models of, Basel II. We must also consider how the Basel II models and mathematical formulas reconcile with our existing rules, such as with our asset risk classification and prompt corrective action rules. This includes whether any of our existing rules, in addition to risk-based capital, would have to be changed to accommodate Basel II. Operational Risk Another important issue is the operational risk capital charge in Basel II. The concerns include the difficulty of trying to measure something that cannot be readily modeled. Currently, the ability to measure and quantify operational risk is less advanced than the measurement and quantification of credit risk. In addition, the boundaries between credit risk and operational risk are not always clear. Another question is whether operational risk should receive a more qualitative Pillar 2 supervisory review as opposed to the quantitative Pillar 1 approach proposed in Basel II. This question is significant because assessment of operational risk inherently involves human judgment, which lies more squarely within Pillar 2. There are also questions about the availability of good data to measure operational risk. Under the AMA model of Pillar 1, the most sophisticated institutions would use available external data to measure risk and compute their own capital charge. While data may be readily available for ordinary risk events that can be budgeted, truly high- risk loss events occur infrequently. We must consider how to proceed where there is a lack of readily available data for precisely the type of risk for which capital may be most relevant to a particular institution or group of institutions. We will also want to consider the positive effect that an institution's internal systems and controls have on operational risk exposure. In computing their operational risk capital charge, it is important to understand whether and how different institutions would allocate capital appropriately for weaknesses in their internal systems and controls, as well as the disincentives in doing so. This is important to ensure both consistency and accuracy in the operational risk capital charge. Conclusion Thank you, Chairman Shelby, Senator Sarbanes, and Members of the Committee for the opportunity to testify on Basel II. As you are aware, Basel II raises very significant issues not only for our very largest banking organizations, but potentially for all our insured institutions. I urge all of the Members of the Committee to remain involved in this process going forward. PREPARED STATEMENT OF MAURICE H. HARTIGAN, II President and CEO RMA--The Risk Management Association June 18, 2003 Good morning, Mr. Chairman and Members of the Committee. Thank you for inviting me to appear before the Committee to discuss the important work under way to reform the 1988 Capital Accord, sometimes known as the Basel Accord. My name is Maurice Hartigan and I am the President and CEO of RMA--the Risk Management Association. RMA is a member-driven professional association whose sole purpose is to advance the use of sound risk principles in the financial services industry. RMA promotes an enterprise-wide approach to risk management that focuses on credit risk, market risk, and operational risk.\1\ --------------------------------------------------------------------------- \1\ Headquartered in Philadelphia, RMA has 3,000 institutional members that include banks of all sizes as well as nonbank institutions. They are represented in the Association by 16,000 commercial loan, credit, and risk management professionals in the 50 States, Puerto Rico, Canada, and numerous foreign cities, including Hong Kong, Singapore, and London. RMA was founded in 1914 and formerly known as Robert Morris Associates. --------------------------------------------------------------------------- RMA has been actively involved in the reform of the 1988 Accord. In 1999, we formed the RMA Capital Working Group, consisting of the chief economic capital officers of major banking institutions in North America. Our group conducted research to demonstrate how banks use their internal risk rating systems to assign economic capital. The RMA Capital Working Group has produced a substantial body of research, and has commented extensively on different drafts of the new Accord. It is currently formulating comments to the most recent Basel Committee Draft, the third consultative paper. This group also plans to comment on the forthcoming interagency advanced notice of proposed rulemaking that will deal with the U.S. implementation of the new Accord. The main point I want to make to you today is that the New Basel Accord will be a step forward for the U.S. and world banking industries, provided it is modified as it is being finalized and provided it is implemented flexibly. It will be a step forward because it is directionally correct in improving the risk sensitivity of regulatory minimum capital adequacy standards. But it must be modified to ensure that it is not too conservative--that these are truly minimum and not maximum capital standards--and to ensure that it is not too prescriptive. The purpose of the New Basel Accord is to make capital regulation truly risk sensitive. The 1988 Accord was called the Risk-Based Capital Accord, but it was that in name only. The new accord is designed to be much more risk sensitive. It will require additional capital for activities that are more risky and less capital for those that are not. The 1988 Accord relied solely on a regulatory minimum capital standard. In contrast, the new accord will be grounded on three principles or ``pillars'' as they are called: (1) Capital requirements, (2) enhanced supervision, and (3) greater disclosure. This alone represents a significant improvement. Nonetheless, we have specific concerns in this area. Pillar 1, which deals with the capital standard itself, must contain assurances that Basel will evolve toward a full models-based approach for credit risk, and it must avoid arbitrary specificity. Pillar 2, which deals with the implementation of the standard through the process of supervision, must allow regulators enough discretion to accommodate the diversity of best practices in risk management today. Pillar 3, which requires increased disclosure in order to provide greater market discipline, must ensure that comparability is meaningful across the varying international accounting regimes. Our research to date suggests that the new accord, as proposed in the third consultative paper, will require more overall capital than many banks' internal risk rating systems require today, even though for some banks and some portfolios, the new overall requirement will be somewhat less than under the old accord.\2\ This will often be inappropriate. --------------------------------------------------------------------------- \2\ All of RMA's research and our formal responses to the Consultatives Papers issued by the Basel Committee are available on our Web site at www.rmahq.org. RMA's Securities Lending Committee has also responded to the proposed treatment of securities lending activities, and the work of that Committee is available to the public on our Web site as well. Institutions participating in the research are listed on the Web site and may hold views different from those expressed in this testimony. --------------------------------------------------------------------------- The new accord should represent a true minimum capital requirement. For well run banks in normal times this implies that regulatory capital levels should be set below a bank's economic capital based on best- practice internal risk measurement procedures. RMA and many others within the industry have long argued that regulatory capital requirements should be more closely aligned with an institution's own internal risk rating systems. Best-practice institutions today assign internal capital to their portfolios and measure performance on a risk-adjusted basis. Doing so enables them to better price for risk and maximize shareholder value. Thus, good business practices are consistent with the economic capital principles underlying the proposed new accord. The old Capital Accord requires best-practice institutions to maintain two completely separate capital regimes: An internal system that mirrors their true risk profile, and a regulatory capital system that is a simple, flat capital charge. Advanced-practice institutions do not manage risk based on the current regulatory capital requirements. It would not be in their shareholders' or their customers' best interests to do so. This fact has certainly not gone unnoticed by the regulators. Indeed, that is why reform of the 1988 Accord is under way. For best-practice institutions, the possibility to align internal capital estimation processes and regulatory capital procedures represents a significant and meaningful improvement over the current system. Turning this promising possibility into reality is not an easy task, however. And that is why we are here before you today for a review of the New Basel Accord. The process to reform the 1988 Capital Accord has had a positive impact on the development of risk measurement and management procedures in the financial services industry. Moreover, the dialog between the industry and its regulators surrounding Basel reform, while not without frustration on both sides, has been useful and productive. While outstanding issues clearly remain, some quite significant, continued discussion with the industry is ongoing, and I would expect that to be the case throughout the reform process and into the implementation stage as well. Indeed, it may not be possible to resolve a number of specific issues without an active two-way dialog between regulators and the industry as the implementation process takes place. Further discussion can only help promote innovation and investment in best practices throughout the industry. It is for this reason that the reform process must continue. However, it must be framed as a work in progress. There cannot be a prescribed ``end state'' for sound risk management practices. Otherwise, the ink on the new accord would not be dry before it became obsolete. This is much like the 1988 Capital Accord. The quantitative analytics supporting sound credit risk measurement and management are still evolving. Many of these emerging practices were born out of the last economic downturn. The resilience of the financial services industry over the past 3 years should not go without comment. Many have credited the industry's success to the better risk management practices established over the past decade. I would have to agree. One way to look at the new accord is that it is aimed at bringing capital adequacy standards for credit and operational risk closer to those for market risk. For some time, market risk has had a well- established language among practitioners, strong analytics, and a robust disclosure framework to support it. It is for this reason that amendments to the 1988 Capital Accord were adopted in 1995 to acknowledge the industry's advancement in the field. Credit and operational risk management are still evolving to catch up with market risk management. The practice of credit risk measurement and management will no doubt benefit greatly over the next 2 years as new data become available to populate quantitative credit risk modeling systems. Operational risk measurement is a younger field, and it is making strides on the back of our achievements in credit and market risk. Given the newness of the fields of study surrounding credit and operational risk management, it is natural that regulators should be prone to conservatism. But too much capital is just as bad as too little capital. Too much capital will drive down the risk-adjusted rates of return on a particular business line and cause bankers to lend less than they otherwise would and should. This is not a good thing for the shareholders of the bank, the loan customers of the bank, or the general economy. Furthermore, in our own review of Basel II, we find that some of the new requirements are written in a very prescriptive fashion that does not lend itself to allowing individual banks to employ a diversity of best practices. Without such diversity we cannot have continued evolution of best practices, and without evolution we could not have had the improvements in risk measurement that have occurred over the past decade. I would now like to touch on two areas, which are somewhat more technical in nature, about which we have great concern at present. Foremost is the adoption by Basel of the same credit risk model as used by advanced banks. A key parameter of these models--the degree to which loan losses are correlated--is set by Basel, not by the empirical research of best-practice banks. In some cases, such as certain retail loan products, this critical parameter has been set too high by Basel, causing the regulatory capital minimums to be too high. This is why RMA has consistently stated in all our papers to the Basel Committee that, ``we believe strongly that the Internal Ratings-Based (IRB) approach must be followed with a full internal models approach to capital.'' Second, RMA also has repeatedly argued that the Basel definition of capital should be changed to conform to the definition used by the industry. Indeed, Basel II will run into problems to the extent that the Basel view of capital differs substantially from the view of economic capital held by the industry. In the industry view, economic capital is required only for unexpected loss (known as UL). The Basel Committee has proposed that both UL and expected loss (known as EL) be included in bank capital. RMA disagrees. For purposes of estimating economic capital and capital adequacy, EL is covered by earnings (spread and fees, net of expenses), and we believe that it is double counting to include expected losses in capital. Indeed, if EL is included in bank regulatory capital, it will clearly disadvantage banks with their nonbank competitors. RMA has additional technical concerns specific to the Third Consultative Paper that we will address in our formal response. To conclude, I would like to reiterate RMA's belief that the reform process has helped advance the practice of sound risk measurement and management within the industry. RMA is hopeful that the New Capital Accord can be structured to encourage and enhance continued industry innovation and that it will recognize the benefit that diversity of practice within the industry provides. Much good work has been done in conjunction with the new accord. It has helped foster valuable research that has contributed to industry innovation. It has also focused the industry and its regulators on the need for additional research. Data limitations remain in a number of key areas, and this is likely to be the case for some time. Again, this only reinforces the fact that development of the new accord must be an ongoing process. Regulatory capital standards must evolve over time as practices within the industry evolve. Otherwise, the industry and its regulators will continue to face the same limitations embedded in the current accord. The only way for this goal to be achieved is to allow for the development, over time, of a full Internal Models-Based approach to bank capital. The proposals contained within the third consultative paper, subject to the specific concerns we will be addressing shortly, can represent a necessary start to this process. Thank you, and I would be happy to answer any questions that you might have. ---------- PREPARED STATEMENT OF MICAH S. GREEN President, The Bond Market Association June 18, 2003 On the Basel II Capital Accord The Bond Market Association is grateful for the opportunity to testify on the Basel Committee on Banking Supervision's proposed New Capital Accords, or Basel II. The Bond Market Association represents securities firms and banks that underwrite, distribute and trade debt securities domestically and internationally. Association member firms account for in excess of 95 percent of all primary issuance and secondary market activity in the U.S. debt capital markets. Through our affiliate American and European Securitization Forums, we represent a majority of the participants in the growing securitization markets in the United States and Europe. The following comments focus on only those issues related to Basel II that are most important to our membership. TBMA Supports the Goals of Basel II The Association supports the Basel Committee's overall goal of rationalizing the current risk-based capital regime, and aligning regulatory capital requirements more closely with actual credit risk. This goal is critically important to the global financial market, in which capital flows are increasingly mobile and interdependent. Also, we are grateful to the Federal Reserve Board and other U.S. bank regulatory agencies for working with us to address the issues presented by the proposed capital accord revisions that affect the domestic bond market. While some of our concerns expressed previously were addressed in the Basel Committee's third consultative paper (CP-3) on Basel II, critical issues still remain. The Basel Committee has an important role in promoting a prudential but efficient allocation of capital throughout the banking system. An updated regulatory capital regime can produce significant benefits, including the promotion of fair global competition, the creation of incentives for better internal risk management, and an economically efficient allocation of capital to its most productive uses. Although we support the direction and goals embodied in Basel II, the revised Accord should not be viewed as the last word on regulatory capital. In attempting to promulgate a universal rules-based system that applies the same basic capital requirements to all regulated financial institutions, Basel II--like its predecessor--is overly rigid and prescriptive in certain critical respects. However, no such ``one- size-fits-all'' regulatory capital regime can fully accommodate the unique needs of these diverse institutions, or flexibly respond to rapid changes in the financial markets in which they operate, without suffering from this basic limitation. To overcome this deficiency, the global financial community will need to move toward a broader reliance on internal risk models, with supervisory review and approval, to determine appropriate regulatory capital levels, and we encourage financial market regulators to continue moving in this direction. In the meantime, our comments focus on aspects of the proposed Accord that we believe will, at least in the short-term, facilitate the goal of aligning regulatory capital requirements more closely with actual credit risk. The Association has principally focused on two areas of the proposed Basel Accord that significantly affect the bond markets-- securitizations and collateralized transactions, including securities repurchase (repo) and securities lending arrangements. By creating more risk-sensitive capital standards in these areas, Basel II can ensure these transactions continue to serve as useful funding, liquidity, and risk management tools. Securitizations allow banks and other entities to obtain efficient funding and to remove certain risks from their balance sheet so they can be borne by other parties who desire such an exposure. Repo and securities lending transactions also aid institutions in managing risk by allowing them to readily obtain securities in order to meet delivery obligations and to hedge exposures arising from separate transactions. Setting regulatory capital charges too high for these increasingly important and widely used arrangements threatens to distort economic decisionmaking on the part of a financial institution. This has the potential of eroding the significant benefits that consumers and businesses alike realize from securitization and collateralized transactions. Background on the Securitization and the Repo and Securities Lending Markets Market Size The past several years have seen phenomenal global growth of the securitization market. Since 1995, the United States, European, and Asian markets combined have grown from $497 billion to $2.9 trillion. The U.S. market by itself has accounted for about 95 percent of that volume. The repo market has also shown steady growth over the same period. Approximately $1.7 trillion in repo and securities lending transactions were outstanding on average in 1996 and today an average $3.7 trillion are outstanding. Hundreds of billions of dollars in repo transactions are conducted daily to fund the positions of bond market participants and allow the Federal Reserve Board to conduct open market operations. Benefits of Securitization and Securities Lending and Repo Agreements Securitization offers numerous benefits to consumers, investors, regulators, corporations, and financial institutions. Securitization has developed as a large market that provides an efficient funding mechanism for originators of receivables, loans, bonds, mortgages, and other financial assets. Securitization performs a crucial role for the entire U.S. economy by providing liquidity to nearly all major sectors including the residential and commercial real estate industry, the automobile industry, the consumer credit industry, the leasing industry, and the bank commercial lending and corporate credit markets. In addition, securitization has provided a means for banks to effectively disperse the risk of various positions they hold throughout the broader financial market. Securitization provides low-cost financing for banks and other companies, lowers borrowing costs for consumers and homebuyers, adds liquidity to banks' balance sheets, provides for efficient bank balance sheet and capital management, and draws nontraditional sources of capital to the consumer and corporate lending markets. The efficiencies introduced by securitization are passed on to consumers and businesses in the form of more widely available credit, lower interest rates, and lower prices. Securities Lending and Repurchase Transactions Securities lending and repo transactions are integral to maintaining liquidity in the capital markets. They are a secure and flexible method of obtaining funding and securities for market participants. For example, a market participant may purchase securities which are then sold in a repo transaction, with an agreement to repurchase such securities sometime in the future. The repo seller can use the proceeds of this transaction to fund their initial purchase. The repo buyer is able to invest funds for short periods in a safe and liquid product. By providing a ready source of funding, repos and securities lending transactions are critical to maintaining liquidity in the bond markets. In the Treasury markets in particular, this liquidity ensures that the Treasury's borrowing costs are kept low. In short, America's capital markets operate as efficiently as they do because wholesale market participants can use repos and securities lending transaction to finance and hedge positions. The liquidity and efficiency provided by the repo market lowers financing costs for the Federal Government, homebuyers, corporations, and consumers. Basel II's Impact on Securitization and the Repo and Securities Lending Market The Association applauds the goal of the Basel Accord to allow financial institutions the ability to more closely tailor risk-based capital requirements to the actual amount of risk present in financial transactions. The proposed Accord, however, does not currently meet this goal because under the proposal, institutions would be required to maintain a higher level of capital than is warranted by the practical risk of their positions. We have summarized below some of our principal concerns in connection with the proposed capital treatment of securitization exposures and repo and securities lending transactions. The Association is continuing to develop additional quantitative and analytical arguments to support these points, which will be submitted prior to the July 31 comment deadline in response to the CP-3. The Association will share our comments with Committee Members at that time. Securitization The Association is troubled by the treatment in Basel II of certain securitization products and positions. We are especially concerned that if Basel II is not amended, the onerous capital charges imposed on banks will discourage them from engaging in securitization transactions. As a result, the benefits conveyed by a robust and efficient securitization market would be diminished or lost. Securitization Risk Weights Are Too High The floor capital charge is too high for many types of securitization positions, given their actual risk profile. Subinvestment grade positions in particular attract too high a capital charge under the proposals, given the actual credit risk they present. Many of the key assumptions underlying securitization formulas and risk weights are too conservative, and lack a proper theoretical or empirical foundation. By setting the floor requirements at a higher level than the actual risk of a position, Basel II reduces incentives for banks to participate in securitizations. This would lower incentives to conduct transactions that actually lessen a bank's risk exposure and that allow banks effectively to disseminate the risk of a particular transaction throughout the marketplace. Conservative Rules Result in Inordinately High Charges In establishing rules governing the manner in which regulatory capital computations are to be made, Basel II defaults to the conservative alternative so often that--cumulatively--these rules result in an inappropriately high capital charge for securitizations. For example, given the general ability under Basel II to rely upon qualified external ratings to determine regulatory capital requirements, we believe that originators of securitized assets should be able to use such ratings to determine risk weights, even if this produces a lower capital charge than if the assets had not been securitized. Originators do not have this ability under the proposal as drafted. There are numerous other examples of excessively conservative rules that--in the aggregate--produce unduly high capital charges for securitizations. Synthetic Securitizations Should Not Be Discriminated Against Higher capital charges should not be levied against synthetic securitizations, in comparison to traditional asset securitizations. (Synthetic securitizations involve the bundling and securitization of credit exposures, rather than the underlying financial assets.) Synthetic securitizations are increasingly used by financial institutions to manage their balance sheets, and provide additional options and flexibility for risk management. Since the risk profile of a synthetic asset is the same as for a cash asset, the risk-based capital treatment should be equivalent. However, this would not be the outcome under the proposals as currently drafted and, in several respects, synthetic securitization positions attract inordinately high capital charges. Limited Credit Risk Inherent in Liquidity Facilities Should be Recognized In a number of important respects the Basel II proposals would require financial institutions to hold disproportionately high levels of capital against liquidity facilities they provide in connection with securitizations. Such liquidity facilities are extended by financial institutions to a variety of securitization issuance vehicles, including but not limited to asset-backed commercial paper conduits. Through the securitization market, these conduits provide competitive short-term financing for a wide range of asset originators. The performance history of liquidity facilities in this context demonstrates that the likelihood of draws are extremely low, and the incidence of credit losses negligible. We believe that internal modeling is the most appropriate method for determining regulatory capital for liquidity facilities. The key operational requirement for liquidity facilities is that there be an asset quality test that adjusts dynamically to preclude funding of defaulted assets. Such a dynamic test is one that is built into liquidity facilities that have been in the market for many years. This has led to historical performance data showing the relatively low risk of draws and of losses on such draws. Under Basel II, if a liquidity position is not rated, we believe that a bank should be able to look through to the risk weight assigned to the underlying transaction that the liquidity supports if that underlying transaction has been externally rated. Given that the underlying transaction reflects the ultimate risk of a liquidity position, we see no reason not to permit the reliance on the rating of that transaction if a liquidity position itself is not rated. Securities Lending and Repurchase Transactions The Association is concerned that Basel II, as proposed, falls short with regard to recognizing modern risk-management techniques as they relate to secured transactions such as securities lending and repurchase transactions. By failing to account for methods widely used to mitigate risk exposure, capital charges for banks would not reflect true balance sheet risk. The undue capital charges would ultimately result in less efficient and more costly markets. Encourage the Use of Cross-Product Netting as a Risk Management Technique The Association believes that the manner in which risk-based capital requirements for repo and securities lending transactions are calculated should be revisited along with the treatment of similar collateralized transactions. The Association strongly believes that transactions which present similar risks--and mitigate against similar risks--as repo and securities lending transactions should be treated in the same way for risk-based capital purposes. Many financial institutions currently manage risks for all collateralized transactions in a uniform manner. After conforming the manner in which risk is calculated for repo and securities lending transactions and other collateralized transactions, the Basel Accord should take the next logical step and allow for recognition of the netting of exposures across such transactions. Currently, the Basel Accord contemplates netting only between repo and securities lending transactions. It is widely recognized that netting exposures across different transactions helps financial institutions reduce their exposure to the risks such transactions present. Providing incentives in the Basel Accord through broader recognition of cross-product netting will provide added incentives for financial institutions to implement this risk-reducing practice. Encourage the Use of Internal Risk Models It is the Association's view that allowing financial institutions to utilize internal risk models--as Basel II would--to determine counterparty risk for collateralized transactions is a step in the right direction. Basel II should not, however, dictate rigid rules as to what models financial institutions must utilize in determining risk. The Accord should allow financial institutions to utilize their own risk models subject to the review and approval of national supervisors under Pillar 2 of the Basel Accord. Otherwise, financial institutions would likely devote resources to creating a model that may not accurately capture the risks present in collateralized transactions. In addition, the Association believes the Accord should not set out a rigid backtesting regime for such models. (In this case, backtesting refers to evaluating the performance of a model based on historical data.) In any event, the backtesting regime currently set out in the Basel Accord risks dissuading financial institutions from improving upon their existing risk management practices through the use of internal risk models by risking the imposition of significantly increased capital charges. As currently contemplated, should the results of the backtesting regime generate a number of mismatches or ``exceptions'' between estimated and actual data, an institution's risk-based capital charge would be significantly increased. Such backtesting regime--and its potentially punitive results--do not have any commercially reasonable basis in relation to the repo and securities lending markets. Conclusion The Association supports the overall goal of the Basel Committee to align capital requirements for financial institutions more closely to actual credit risk. While the revised Accord has the potential to move regulatory capital requirements in the right direction, the Association continues to have fundamental concerns with the proposal that must be addressed to uphold the Basel Committee's stated goals without causing economic distortions in the securitization, repo, and securities lending markets. The Association looks forward to continuing its dialog with the Federal Reserve Board and other U.S. regulators on the issues we have addressed above. We plan to offer formal comments on the third consultative paper this summer, and when the Board issues its advanced notice of proposed rulemaking describing the U.S. implementation of Basel II, the Association will provide further input. ---------- PREPARED STATEMENT OF EDWARD I. ALTMAN, PH.D. Max L. Heine Professor of Finance Leonard N. Stern School of Business New York University June 18, 2003 Thank you for inviting me to the Senate hearings on the B.I.S. recommended regulations on Capital Allocations for Bank Credit and Other Assets--the so-called ``Basel II'' Accord. I have followed Basel II's consultative papers since the first one was issued in June 1999. We have submitted two formal commentaries to the Basel Commission on Bank Supervision, primarily related to the first of the so-called pillars of the new recommendations--capital adequacy based on the specific risk characteristics of bank counterparties. Our major comments were that the capital requirements related to expected and unexpected losses from corporate and other loans should be based on the actual historical experience of Loss Given Default (LGD) from the corporate bond and bank loan markets. The original 1999 suggestions bore little resemblance to actual performance and we pointed this out fairly precisely. I am pleased to note that the latest version of Basel II's capital requirements based on the riskiness of bank portfolios does a much better job of relating the requirements to default experience, although too little capital is still being required for the most risky categories. A problem with the suggested regulations, however, is the complexity in determining capital requirements and the somewhat arbitrary choice of modifications to the standardized scale due to such items as the size of the counterparty and the existence, or not, of collateral on the loan/bond. For example, small and medium-sized enterprises (SME's) are given lower capital requirements for comparable risk levels of as much as 25-50 percent less capital. The argument that the correlation of default rates amongst these small counterparties is lower than for larger corporations may be valid, but I have seen little evidence that the ``haircut'' for these loans should be as much as 50 percent. In my opinion, this was a concession to those national banking systems of the world whereby SME's are the vast majority of borrowers-- hence lower capital requirements for banks in those countries. It is also true that SME's make up the vast majority of loan assets of the smaller banks in the United States and the same lower capital requirements would hold for U.S. SME's and the banks that make these loans--close to all but 100 of our Nation's 8,000 banks. But, as I will now discuss, almost all of U.S. banks will not be required to follow the recommendations of Basel II, so the reduced capital requirements on SME's will not be relevant and the old Basel I's 8 percent rate will probably still be in effect for all except the very largest U.S. banks. United States vs. Rest of the World and Basel II As I indicated above, and as you are probably all aware of, Basel II's recommendations on credit risk and operating risk dimensions of bank activity are just that--recommendations. The Central Banks of the world, and other bank regulatory bodies, who set national bank regulatory policy, may or may not choose to conform to all or any parts of Basel II's recommendations. Indeed, it came as an enormous surprise to some observers, including this writer, that only the largest 10 U.S. banks, and perhaps the next 10-20 banks in terms of asset size, would be required (top 10) or will have the option (next 10-20) to follow the advanced Internal Rate-Based (IRB) version of Basel II's Accord with respect to specifying the LGD dimensions of their portfolios and hence, set capital requirements based on portfolios risk characteristics. While it is true that as much as two-thirds of all bank assets are held by the top 30 U.S. banks and more than 95 percent of foreign bank assets operating in the United States will be covered by Basel II's most sophisticated guidelines, it is likely that all the rest of our banks (almost 8,000 smaller banks) will not be asked to conform and will probably not do so for the following reasons: (1) Basel II is too complex and costly to introduce and conform with. (2) The U.S. banking system is presently more than adequately capitalized and the recent decade's experience of very low numbers of bank failures makes change unnecessary. (3) The added Basel II capital required for operating risk is based on highly arbitrary and extremely difficult to measure variables. (4) The Federal Reserve System's, and other bank regulatory agencies, policy of ``prompt corrective action,'' and maximum leveraged ratios, when bank capital falls below a certain specified level has worked very well in the United States and is not specified as part of Basel II--even in Pillar 2's regulatory oversight. In other words, ``if it ain't broke, don't fix it!'' I believe that the choice of only the largest 10 commercial banks to conform to Basel II, and the IRB approaches, is unfortunate and should be reconsidered. Not withstanding the recent consolidation movement of many of our largest and most sophisticated banks, the possible exemption of #11 to #30 (including HSBC Bank USA, Citibank [West], Bank of New York, Key Bank, and State Street, #11-15) and the very likely exemption of #31 to #50 up to 8,000 (including such seemingly large banks as Charter One, Am South, Union Bank of California, Mellon Bank, and Northern Trust, to name just a few, seems arbitrary and belittles the possible sophistication and motivation of these banks which would be substantial institutions in most other countries of the world. For example, the 50th largest bank in the U.S. in terms of assets (Compass Bank with $24.3 billion) or in terms of deposits (Mellon Bank with $15.2 billion) would be huge institutions in most countries. The choice of a round number, like ten, would seem to be insensitive to world opinion as well as to risk management motivation. Speaking from an economic standpoint, rather than a political one, I would prefer to see either no banks be required to conform or some exemption level whereby the costs/benefits to our bank- ing system would be more rationally presented and defended. Certainly, a number like the top 50-100 banks would be much more in line with the number of banks conforming in other countries. This would help ensure a ``level playing field'' amongst banks. Our largest banks are probably relatively happy to conform to Basel II even with its complexity and added costs to develop information and credit scoring systems to conform to the requirements of the advanced Internal Rating-Based (IRB) systems mandated under Basel II. The reason is that they expect that the total capital required for credit assets will be less than what is required under the current regime (which will continue until 2007). So, we may have a new regulatory regime where everyone--large and small banks, as well as out bank regulators--are relatively pleased with the changes recommended under Basel II. A Related Disappointing Result of U.S. Policy I have always felt that despite its problems with: (1) complexity, (2) too low capital requirements on the more risky counterparty assets and (3) the difficulty of managing against operations risks, Basel II had one extremely important by-product--the motivation for banks to develop or improve upon their existing credit scoring models and systems to reduce total losses from nonperforming and eventually charged-off loans. These systems can be used to rate and set capital for all bank customers rather than setting a ``one-size-fits-all'' (8 percent) requirement on them. I have observed the enormous strides achieved by banks throughout the world, including ones of all size and location, as to developing risk management systems and training of personnel to prepare for Basel II. Indeed, from what I can surmise, banks in most countries, especially in the European Union, will all adhere to Basel II's Standardized, Foundation or Advance IRB approaches. Granted that regulators in these countries will need to sanction far fewer banks than U.S. regulators would have to do if all banks are mandated to conform, it must have come as a surprise, perhaps even a resentful shock, that the vast majority of U.S. banks will not adhere to Basel II. This is especially true since the United States and its representatives to the B.I.S. were early champions of the need to change the way banks allocate capital for credit risk of their clients. What is disappointing to me is that the Fed's decision to exempt all but the largest banks from building and implementing an IRB approach of some level of increased sophistication will reduce the motivation for most banks to move to a more risk sensitive lending policy. I recommend that our Federal regulators require some level of added due diligence on the part of banks with respect to economic capital decisionmaking and the use of credit scoring or rating systems even if they are not absolutely required under the old (and continuing) Basel Accords. One way to accomplish this on a cost effective basis is for smaller banks to combine resources (data and money) to accomplish these goals. Our decision to exempt smaller banks from Basel II may backfire if many of the world's smaller, or even larger, banks from developing and developed countries, also opt out of the process, leading to greater instability in these banking systems and perhaps to ours through contagion. A Note on Procyclicality One of the likely by-products of Basel II and its reliance on systems that require a careful assessment of credit ratings and loss given default reserves is the possible procyclical impact. That is, any system which requires more capital when defaults increase and banks' portfolios become more risky, such as what is likely to occur during periods of economic stress, will motivate banks to provide even less credit, for example, ration credit or a credit-crunch, thereby exacerbating economic downturns. The opposite will likely occur during periods of above average economic growth, thereby causing too much easy credit and subsequent higher levels of defaults and charge offs than would have been the case under the old system. Now I am aware that this problem called procyclicality already exists due to banks and other lending and capital providers having ``short memories'' of the last period of economic stress. Indeed, the procyclical problem resulting from the benign credit cycle of the mid-1990's (1993-1998) helped to cause the enormous level of defaults in 2000-2002. And, our research shows that when bond and loan defaults increase, we can expect a coincident reduction in recovery rates. Hence, the LGD result will be even greater due to the negative correlation between probabilities of default and recoveries given default. I would be surprised if bank regulators, and the banks themselves, have considered this double negative effect in times of economic stress. Fortunately, our banking system was very well capitalized prior to these problems and seems to have weathered the avalanche of large firm bankruptcies (77 in 2001/ 2002 with liabilities greater than $1 billion) without too much stress. Despite out seeming capital adequate condition and the fact that a great deal of procyclical behavior (for example, herding, over compensation for short-term loan losses) can be expected from current bank regulatory guidelines, I suggest that the Fed consider a more smoothed capital allocation system to even out the normal fluctuations in bank reserves, capital allocations, and lending behavior. This would require more capital set aside in good times and less during periods of stress. Conclusion Thank you for inviting me to attend today's hearings and express my views. On balance, Basel II has many positive recommendations but still may prove to be inadequate to overcome strong systemic problems that normally could be mitigated by a well-capitalized and prudent regulatory oversight policy. I look forward to observing the results of Basel II on a worldwide level as well as special concern for the U.S. banking system. For your information, I have provided my bio-sketch as an attachment to this document. ---------- PREPARED STATEMENT OF KAREN SHAW PETROU Managing Partner Federal Financial Analytics, Inc. June 18, 2003 It is an honor to appear today before this Committee to discuss the potential ramifications of the international risk-based capital rules under consideration in Basel for U.S. financial institutions and--even more important--for the economy that depends upon them. I am the Managing Partner of Federal Financial Analytics, a consulting firm that advises on U.S. legislative, regulatory, and policy issues affecting strategic planning. In this capacity, we advise a variety of companies on the implications of specific sections of the Basel proposal. We also advise the Financial Guardian Group, which represents those U.S. banks most concerned with the proposed operational risk-based capital charge. In my testimony I will focus on the most recent version of the Basel rules--the third consultative paper or CP-3, as well as on the advance notice of proposed rulemaking (ANPR) on which the U.S. regulators are now working. Although the effective date for the new version of the international capital rules-- Basel II--is December 31, 2006, its actual impact will be felt far more quickly. Indeed some markets have already begun to change in anticipation of the Basel standards. As the final shape of the rules becomes more clear, financial markets--and the larger economy--will change more noticeably. Congress' review of the rules thus comes in a timely fashion that ensures any policy concerns posed by the rules can be addressed well in advance. Much in CP-3 is very worthwhile. Overall, Basel II is a worthy and overdue effort to fix the problems in Basel I that have created all too many opportunities for banks to ``arbitrage'' the capital rules. When regulatory capital diverges from the ``economic'' capital dictated by the markets, banks change their portfolio, pricing, and risk decisions. This has profound impact on overall franchise value and on key lines of business, as well as affecting the cost and availability of credit to consumers and companies across the country and around the world. When regulatory capital is too low, banks can take undue risk--a problem observed since Basel I went into effect that needs a quick remedy. When regulatory capital is too high, banks cannot compete against nonbanks, and assets flee the banking system with possible adverse consequences for overall market stability. However, Basel II now has gone far from its initial clear goal of ending regulatory arbitrage. In fact, the most recent statement of the purposes of Basel II--posed 2 weeks ago by U.S. regulators--no longer even mentions this. Now, the goals of Basel II are said to be: Improvements to internal risk management and capital allocation, enhanced market discipline and--of all things--a new capital charge for operational risk. I shall have more to say about the operational risk capital charge later, but suffice it to say that this proposal worsens the relationship between regulatory capital and risk--absolutely the reverse of where Basel II initially intended to go. If Basel II cannot be brought back to its initial and important purpose, then the U.S. capital rules alone should change to do so. Based on our review of the third consultative paper and recent statements from U.S. regulators about our own implementing rules: <bullet> There is a ``first-things-first'' solution that fixes the Basel II's complexity problem. Much in the proposal can be quickly implemented at reasonable cost for all banks and savings associations here and abroad. U.S. regulators should act now on those sections of Basel II on which they can agree unanimously, and defer other sections until they can do so. <bullet> A ``one-size-fits-all'' approach won't work in the United States Capital should go up or down with risk, not be squeezed into the current requirements that were originally set with scant regard for actual credit losses. Unique factors in the U.S. market make it especially important that bank capital appropriately reflect risk. <bullet> The operational risk-based capital section of Basel II remains deeply flawed and should be dropped. Regulatory capital for operational risk will increase risk, not reduce it, and strong supervision with enforced standards is the right way to address operational risk. <bullet> Simple capital rules are essential for effective supervision. Agencies here and elsewhere cannot administer over-sophisticated rules. Further, laboring to do so will divert resources from emerging risks that often prove the undoing of individual institutions or serious risks to the financial system as a whole. Capital is not the only driver of safety and soundness. Banks have collapsed in the past and will fail in the future even as they hold more than the minimum amount of regulatory capital. Economists and financial analysts have spent literally thousands of hours working to revise the risk-based capital standards that govern interationally active banks around the world and all insured depositories in the United States. This effort is a very important one, as problems in Basel I have led to undue risk-taking and other concerns that warrant immediate attention. However, in the 5 years in which Basel II has been crafted, more and more attention has been devoted to the increasingly complex models that attempt to anticipate expected and unexpected losses in every line of business, under every scenario in each country for all time. The defense of this effort is that financial markets are now complex, so capital must be too. However, the universe is very complex, yet Einstein found a very simple formula that helped to explain it. Complexity is a weakness, not a strength, and Basel II should be difficult only when absolutely necessary to capture subtle risks with potentially severe consequences. Basel II rightly rests on three pillars: improved regulatory capital standards, better supervision and more disclosure. If Pillars 2 and 3 work well, then Pillar 1--the capital standards--need not be as formulaic and far-reaching as currently proposed because supervisors will have ample tools to tailor regulatory capital to individual circumstances and markets will know when this is not being done. One reason regulators rely so much on regulatory capital is the lack of effective supervision in many major financial markets. Here, though, supervisors have ample authority to discipline banks for problems that have nothing to do with capital standards. Companies must, for example, be ``well managed,'' as well as ``well capitalized'' to be financial holding companies and enjoy the privileges provided in Gramm-Leach-Bliley. Further, supervisors measure banks on a ``CAMELS'' scale in which capital--the C--is just one of a range of factors--all weighted equally--on which critical enforcement actions hinge. The other factors are asset quality, management, earnings, liquidity and sensitivity (to various risk factors). If non-U.S. regulators adopted a similarly wide-reaching supervisory regime--and backed it up with meaningful sanctions such as those deployed here--then much of the complexity in Basel II could fade away and the rules could focus on ending major sources of regulatory arbitrage that, on the one hand, threaten safety and soundness and, on the other, unnecessarily undermine bank profitability. First Things First Despite the intention of having a balanced international regulatory framework that emphasizes more than just regulatory capital, the vast majority of staff time has been spent on the regulatory capital charges. U.S. regulators, I think, could have done much for the global financial system and avoided many of the pitfalls in Basel II if more attention had been paid to exporting our strict supervisory standards and their effective enforcement. Japan, in particular, would benefit greatly from this--it is a clear case in which nominal adherence to regulatory capital has done nothing to prevent a grave banking crisis with serious macroeconomic impact. As Basel II advanced and the capital models grew ever more complex, U.S. regulators rightly became increasingly concerned about how this would work in our unique banking system. In sharp contrast to Japan and the European Union, we here have thousands of banks and savings associations; foreign banking systems are far more concentrated into a few nationwide banks. Regulators also--rightly--became concerned about several pieces of the simpler sections of Basel II that resulted from complex multilateral negotiations in which the end goal was often obscured. This is particularly true with the more simple versions of the operational risk-based capital standards, which are not only flawed, but could also actually increase--not reduce--banking risk. Further, even the relatively simple sections of Basel II grew ever more complex as negotiators sought to solve each problem and individual national political objectives as the rules worked their way along over the years. Based on these fears--some of them quite right--U.S. regulators have come up with a solution--mostly wrong. They now plan to impose only the most complex versions of Basel II and then to do so only for the Nation's largest banks. This may limit the pain, but it also undermines the gains close at hand in Basel II. Where Basel II drops regulatory capital--which it does dramatically in traditional lines of business like mortgages and small-business lending--banks left out of Basel II, which will still be required to comply with Basel I, will be at a serious competitive disadvantage to big ones in it. Where the models are overly complex or--worse--wrong, the fact that only big banks must comply with them does nothing to redress the adverse impact they might have. Further, writing off the most flawed sections of Basel II in the United States does nothing to address potential serious consequences in the global economy. U.S. banks--especially large ones--compete head-on with non-U.S. banks here and abroad. If differences in the simpler parts of Basel II--called the standardized approaches--give non-U.S. banks an excuse to rely on overlax rules and inadequate enforcement, then the major strength U.S. banks now have in the international financial services market will be undermined. Worse still, major financial services firms could operate under capital rules that do not actually address real risk. For all its flaws, much in the standardized proposal for credit risk reflects broad agreement on improvements to Basel I. U.S. regulators should turn this into clear language and propose it for smaller banks, while refining the advanced models and offering them to large ones. Where no agreement is in sight--on asset securitization, for example--regulators should act now on those areas where broad consensus exists and defer the others until it emerges or regulators are sure--absolutely sure--they are right and the industry is wrong. U.S. regulators now diverge on many key areas of Basel II, and they should act in unison on issues where they intend to contradict the best evidence and advocacy the industry can muster. One Size Won't Fit All As U.S. regulators have turned their attention from the international negotiations to implementation of Basel II at home, a major dispute has arisen over whether to follow the new rules where they lead. Under Basel II, capital could go down below current levels, especially for very large banks with major retail or mortgage operations. It is for that reason that the operational risk-based capital proposal has been superimposed on the credit risk reforms Basel II initially sought. It is also the reason why some U.S. regulators are now reasserting the importance of the most primitive of all capital charges--a simple leverage one--on banks and their parent holding companies. ``Topping off'' the right amount of credit risk capital with the operational charge and a surcharge for ``leverage'' will so undermine Basel II--especially in light of its high implementation cost--as to raise serious questions about whether the entire exercise is worthwhile. I shall have more to say about the operational risk charge below. With it, Basel II should not be implemented at all. Without it, a sound regulatory capital scheme is in sight. The leverage rule is a unique U.S. capital standard, and it is one that should be dropped as Basel II comes into force. Indeed, it is one that should have been dropped years ago. The leverage standard is a simple ratio of capital to on-balance sheet assets calculated without regard to risk. Under the leverage standard, a bank holds the same amount of capital if its book of business is solid gold or unsecured credit card loans to dubious borrowers. It is a capital standard that could not be more crude, but U.S. regulators clung to it in 1988 because they weren't sure they trusted Basel I. They wanted some form of insurance because they knew then--as now--that credit risk rules did not capture interest-rate risk. You will recall that this latter risk was the predicate cause of the collapse of our savings and loans--which cost taxpayers more than $250 billion and kept this Committee extremely busy for over a decade. So, ironically, Basel II still doesn't address interest-rate risk (IRR). Although the regulators think they know enough about operational risk to put it in the Pillar 1 mandated capital standards, they have decided to leave IRR in Pillar 2. In 1988, regulators were right about the problems measuring IRR; now, they are not. Markets price trillions of dollars of IRR each year in a fashion that Fed Chairman Greenspan has rightly praised. Why then keep the leverage rule? U.S. agencies appear to be clinging to it because they are afraid to follow Basel II's models where they lead. In some cases, the advanced models propose massive drops in regulatory capital. This is particularly true in mortgages and small-business loans--key lines of business for smaller banks that will face major competitive problems if big banks get to drop regulatory capital under Basel II while they are kept in the cold of Basel I. Of course, in other cases, Basel II will dramatically raise capital--for high-risk loans and certain equity holdings, for example. To adopt Basel II when it goes up and block it when it goes down is to create a regulatory capital regime that leaves arbitrage largely in place--again profoundly undermining why all this started in the first place. The best way to protect the deposit insurance funds from risk and small banks from competitive harm is to introduce Basel II's most advanced model-driven sections in an incremental way that-- essentially--hedges the model-builders' bets. How to do this? Despite the complexity of the advanced internal ratings-based approach to credit risk, it can be introduced in a remarkably easy way. Upon conclusion of Basel II's comment period and a review of all the analyses of the sophisticated models, regulators should make up their minds about the ``right'' amount of credit risk-based capital for specific assets. Where they cannot agree, as noted, they should defer action. Where they can, they should implement Basel II--but only in a phased-in fashion. If, for example, the ``right'' amount of capital is a dramatic drop, then set a schedule in which capital slides down year after year across the board for all banks that qualify to use the advanced models. Where it goes up a lot, capital should similarly be phased in. This incremental approach has two advantages. First, as noted, it hedges the regulators' bet on the skills of their model builders and the ability of supervisors to handle the complex new rules (on which more below). Second, it addresses concern that Basel II will exacerbate booms and worsen busts--``procyclicality'' in Basel speak. To be sure, phasing in Basel over time doesn't eliminate procyclicality, but it ensures that regulators are certain of their capital models when these come into full force, while giving them time also to assess the value of stress testing and other measures now under consideration. Eliminate the Pillar 1 Operational Risk Capital Charge Basel II's pending proposal and, we are told, the draft U.S. implementing rules will include a new regulatory capital charge for operational risk. Operational risk is that resulting from human or systems failures, natural disasters, and even terrorist attack. There is, though, no accepted definition of operational risk for supervisory purposes--for example, does it include reputation risk? Basel II says no--for now--but this risk has frequently proven the most serious of all in a business fundamentally founded on investor and depositor confidence. What about events like September 11--catastrophic operational risk? Basel II now has them in--although they were out at the end of last year--but who knows how to measure the likelihood of another attack and then to decide just how much capital is enough and whether capital the right antidote? It is particularly hard to understand why Basel II has a specific capital charge for this risk when one notes that many of its own documents agree that it cannot be well defined. The Basel Risk Management Group, for example, said its own data need to be used with ``caution'' and that a specific capital charge cannot now be based on them. A major Basel Committee on global financial safety also concluded earlier this year that there is now no way to determine a quantitative regulatory capital charge. Another major unanswered question: Is any amount of capital enough against catastrophic risk? I do not think so, and indeed imposing an operational risk-based capital (ORBC) requirement will create a serious and perverse incentive for banks to skimp on the forms of operational risk management and mitigation that proved their worth in the most recent and terrible case of catastrophic operational risk, the attack on the World Trade Center. What worked after the terrorist attack-- apart from undaunted heroism--were the backup systems and contingency plans that well-prepared financial firms had put in place. What worked in the terrible days thereafter--other than sheer courage and determination--was insurance. In Basel-speak, these are operational risk management and mitigation. Both are costly--indeed, the back up systems, which U.S. regulators have mandated since September 11, are very much so. Imposing a simple, arbitrary charge against operational risk will lead many banks to rely on this, not proven ways to protect themselves, their customers and the financial system more generally. Basel II now includes three variations on a regulatory ORBC requirement. Two of these--the ``basic indicator'' and ``standardized'' ones--rely on a simple percentage of gross income to calculate ORBC. This is, quite simply, nonsensical. There is no correlation between income and risk. In fact, operational risk also runs counter to gross income because banks that spend more on risk management and mitigation have less profits. Banks that generally have trouble making money also tend to be riskier--again, an inverse correlation between gross income and operational risk, not the positive, linear one on which Basel II relies. U.S. regulators have, apparently, realized that the two simple approaches to operational risk in Basel II do not work. As a result, they are planning only to impose the ``advanced measurement approach'' (AMA) here. This will, though, leave the other two methods in place in the EU and Japan, creating a perverse incentive for big banks there to run undue amounts of operational risk. We cannot wall ourselves off from the problems this will create, and U.S. regulators should thus push hard for meaningful supervisory standards for operational risk that bind all financial services firms, not compromise on a deeply flawed regulatory capital model. Further, the AMA does not solve the fundamental problems with an ORBC charge in Pillar 1. Some of these are unique to U.S. banks--which must compete with major nonbanks in lines of business like asset management and payments processing. Basel II in the United States will not cover nonbanks. Specialized banks will thus face major competitive pressures that may force them to review whether continuing to remain a bank is worthwhile. ORBC not only creates incentives for increased operational risk, but it also may create one for nonbank charters. This would drive assets outside our sound, proven system of bank supervision. The pending ORBC charge will also put U.S. banks at a competitive disadvantage against EU and Japanese ones because ``legal risk'' results in a regulatory capital charge. Our legal system is unique--no other nation has our plaintiffs' bar or our extensive array of laws designed to protect consumers, prevent discrimination, and promote workplace safety. There is no evidence that any of this legal risk has ever caused any U.S. bank to fail, and current law already requires reserves for material legal risk (and these must also be disclosed). The AMA epitomizes the problems in Basel II where reliance is placed on unproven models over which U.S. regulators rightly do not agree. Acceptance of these models now puts banks at undue and unnecessary risk--risk far better addressed through effective supervision with meaningful enforcement. Can Supervisors Supervise Under Basel II? Finally, I would like to turn to the question of whether the complexities in Basel II's advanced models are so daunting that supervisors at home and abroad will not be able to ensure that banks actually comply with the new capital rules. This is a major concern, and one the regulators are already trying to address through a major Basel Committee focused on supervisory implementation. In the United States, the agencies now think the best way to handle the complexity problem is to make Basel II apply only to the biggest banks, whose examiners tend to be those most familiar with complex financial arrangements. However, as noted, applying Basel II here only to the biggest banks will create a range of competitive and safety problems, while leaving the supervisory capability question largely unresolved. A recent survey of the cost of Basel implementation for the larger banks expected to use the advanced models indicates that it will reach $200 million per bank. One has to ask how it can cost so much for banks and not pose a comparable burden on supervisors who must assess these elaborate models. In point of fact, the rules must be as costly for the supervisors as for the supervised or undue reliance will be placed on untested models. If supervisors instead rely on ``benchmarks'' they will in effect superimpose standardized credit and operational standards that obviate the flexibility hoped for from the advanced approaches. These problems are not addressed by the proposed qualifying conditions for use of the advanced models--more board and senior management involvement, for example--because none of the proposed standards addresses the fundamental problem posed by complexity, let alone how top management can divert resources from their many other pressing investor protection and safety-and-soundness responsibilities. The right solution to the supervisory resource problem is the same as the right solution to the other challenges posed by Basel II: Impose a uniform system of improved rules across the board and then change them gradually over time as the rules are tested and we all learn how to work under them. Back up the more sophisticated models with meaningful supervision that binds banks in the EU and Japan, not just United States banks and give investors simple, clear disclosures to help them understand just how much capital banks have and whether the supervisors are concerned about it. ---------- PREPARED STATEMENT OF D. WILSON ERVIN Managing Director Credit Suisse First Boston on Behalf of the Financial Services Roundtable June 18, 2003 Introduction Good morning Mr. Chairman. I want to thank you for holding these hearings today and inviting me to appear before the Committee. My name is Wilson Ervin and I am a Managing Director of Credit Suisse First Boston (CSFB).\1\ I head our Strategic Risk Management (or SRM) department and also chair its risk committee. I am presenting testimony today on behalf of CSFB and on behalf of our trade group, the Financial Services Roundtable.\2\ CSFB employs approximately 20,000 people, primarily in the United States, and is a major participant in the capital markets. It ranks among the top firms in raising money for companies around the world and is a leading underwriter of mortgage and credit card financing. The firm is also among the largest managers of funds invested in private companies. --------------------------------------------------------------------------- \1\ Credit Suisse First Boston (CSFB) is a U.S. financial holding company and leading global investment bank serving institutional, corporate, government, and high net worth clients. CSFB's businesses include securities underwriting, sales and trading, investment banking, private equity, financial advisory services, investment research, venture capital, and asset management. CSFB operates in more than 89 locations across more than 37 countries on six continents. The Firm is a business unit of Zurich-based Credit Suisse Group, a leading global financial services company. \2\ The Financial Services Roundtable is a national association representing 100 of the largest integrated financial services companies in the United States providing banking, insurance, securities, and investment products and services to American consumers. --------------------------------------------------------------------------- My department is responsible for assessing the risk profile of CSFB on a global basis and for recommending corrective action where appropriate to protect our capital. This objective is very similar to many of the goals of bank supervisors, including the drafters of the proposed Basel Accord--to deter very large losses and protect bank solvency. The Basel II Capital proposals have been the topic of intense discussion and debate in the financial and regulatory community for the past several years. The industry supports the objectives of the Basel process: To better align regulatory capital to underlying economic risks, promote better risk management, and foster international consistency in regulatory standards. The proposed Accord is not a minor refinement to the bank regulatory process, but is, instead, a wholesale reform of bank regulation--a regime that covers roughly $2 trillion of capital and is a key economic engine for most developed markets. The impacts of these seemingly technical discussions will affect banks, the markets, and the economy in a deep way, and we would be wise to consider the effects carefully before implementation. Before I start, I would like to note that I have personally developed tremendous respect for the diligence and stamina of the regulators who have worked on Basel II. They have had to address a great many complex and challenging issues, and have been tenacious in trying to develop a ``best practice'' solution for each. Balancing all of this and applying it to very different financial markets around the world--with political sensitivities in each--does not make this an easy job. I wish to express appreciation for the efforts of Federal Reserve Board Vice Chairman Roger Ferguson, who has met with CSFB and Roundtable member companies several times in the past few weeks to listen to our concerns on the proposed Accord. Comptroller Hawke and FDIC Chairman Powell have also had open doors for discussion throughout the long process of developing the new Accord. We look forward to continuing this dialog as Basel II moves closer toward formal adoption and throughout the implementation period. CSFB and the Roundtable have worked hard to be constructive commentators on the new rules, particularly in respect to practical implementation issues. The recent revision of the proposals--called CP- 3--included significant improvements, and demonstrated a willingness by regulators to address specific issues raised by industry and academic critics. Just last week, the Federal Reserve announced that, in implementing Basel II in the United States, the regulators propose to reduce the capital charges on many types of commercial real estate loans, in response to comments and new data from the banking industry. We support the direction in which the Accord has been moving recently, and appreciate the regulators' willingness to reexamine earlier conclusions and consider further changes. However, in spite of the hard work of the Basel Committee and industry, we believe substantial areas for improvement still remain. Basel II has considerable momentum, and most people in the industry believe it will likely be implemented in the relatively near future. On balance, we believe that the advantages of the reform now outweigh the drawbacks, although that balance remains close, and in several areas, open issues remain. This is a frustrating outcome for an initiative with so much potential. We hope these hearings will help illuminate some of the important remaining issues that need to be addressed, so that the Basel II reforms can live up to their original, very worthy goals. Today, without getting too involved in the technical details of the Accord, I would like to highlight four ``macro'' issues which we believe are particularly important: 1. The current Basel proposal is unnecessarily complex and costly, and suffers from an excessive reliance on detailed, prescriptive rules. Under the rubric of comparability, these international rules could bring a more formulaic, inflexible style of regulation to the United States, which currently enjoys a much better balance between black- letter rules and supervisory consultations. 2. The new Accord and its sensitivity to credit ratings could reduce liquidity in the credit markets during economic downturns, potentially extending or deepening economic recessions (procyclicality). 3. The operational risk capital charge proposed by the Basel Committee remains highly controversial. Some Roundtable members support the proposed Pillar 1 operational risk charge; others believe operational risk should be addressed through Pillar 2 supervisory reviews instead. 4. The disclosures required under Pillar 3 of the new Accord are likely to add perhaps 20 pages of highly technical data to bank reporting requirements, raising costs and adding little information of value to the reader. While we appreciate that the Pillar 3 disclosure requirements have been reduced, they continue to be burdensome and potentially confusing. Prescriptiveness, Cost, and Adaptability The first topic I would like to address is the overall cost and prescriptive tone of the new capital rules, and the effect this will have on whether the rules remain relevant over time. The new rules shift the regulatory regime toward a highly complex, formula-based system, and will diminish the important role that is currently played by human judgment. Implementation of these rules will be high cost, but not highly cost effective. Moreover, we believe the very complexity of the new rules and the delicate political balance represented in them will make it challenging to update the rules over time. Most of this prescriptiveness is to be found in Pillar 1, which describes the ``recipe'' for calculating capital requirements. The most recent draft of the Pillar 1 calculations ran to nearly 200 pages, roughly 5 times the length of the original Basel Accord (not including technical papers and additional guidance that is expected to be issued). This is a common result from this kind of process. Once you start developing a system that attempts to capture the complexity of the real world in a series of mathematical rules, it is very hard to stop halfway. One issue or another will always be of major concern for some institution or country. Many of the Pillar 1 rules reflect a political compromise as much as the results of a scientific approach to risk management. The result is a very elaborate system that tries to address all circumstances by being ever more complex, and currently staggers under its own weight. The Basel Committee has done a commendable job in streamlining the earlier drafts in CP-3--the earlier drafts of Pillar 1 rules were even longer--but this remains a fundamental issue. Perhaps the underlying issue in this respect is the prescriptive nature of the new Accord. Conceptually, the Committee has attempted to capture current industry best practices and boil them down into fixed formulae, adding burdensome qualification, testing, and reporting requirements.\3\ These new regulatory requirements, while well- intentioned, will be unduly burdensome and inconsistent with changing market reality and evolving best practice.\4\ It is our recommendation that the Committee establish some basic requirements largely around the key input parameters and exposure calculations and publish best practices that provide guidance to banks and supervisors rather than a rigid rulebook. --------------------------------------------------------------------------- \3\ One editorial recently described this approach as ``prescribing and proscribing in equal measure . . . a monster that cannot clear the first hurdle: flexibility.'' Risk Magazine, editorial page, June 2003 edition. \4\ For example, the eligibility requirements for institutions to qualify to use the Accord's advanced IRB methods for credit risk capital charges are too detailed and burdensome. In general, we believe that these eligibility requirements should be scaled back and replaced with more general guidance. A specific example is the testing requirements for credit exposure in repurchase agreements, an area with historically very low losses. To its credit, the Basel Committee permits the use of internal market risk models to estimate potential collateral shortfalls under stress, which is in line with modern practice. However, the Committee requires substantial additional testing to use this technique, even though this calculation is based on the same model that governs overall market risk, which is a much bigger risk and already subject to comprehensive regulatory oversight. --------------------------------------------------------------------------- CSFB and other Roundtable members are also concerned about the cumulative effect of numerous conservative choices and assumptions that are built into this complex fabric. Each of these can be debated separately, and many are extremely technical. But the combined effect of each of these individual items adds up to regulatory capital requirements that can depart significantly from the true economic capital needs that Basel II was aiming to emulate.\5\ --------------------------------------------------------------------------- \5\ To mention a few examples: (i) The Accord significantly overstates the credit risk capital charges for exposures hedged by guarantees and credit derivatives, by failing to recognize the much lower risk of joint defaults by debtors and guarantors and by applying overly conservative rules on maturity mismatches. (ii) The proposed Accord requires capital against Expected Losses, even though these losses are already covered by loan loss reserves, and Future Margin Income is generally recognized only for credit card exposures. --------------------------------------------------------------------------- Home/Host Country Issues The complexity of the new rules poses particular challenges for an international bank that is regulated by supervisors in multiple countries. CSFB, for example, will be required to implement Basel II as both a Swiss bank and a U.S. financial holding company. Our implementation will be governed primarily by the Swiss Federal Banking Commission, in conjunction with the Federal Reserve in the United States and the Financial Services Authority in the United Kingdom, and also by other regulators around the world. Most international banks face a similar set of interlocking regulation in which both home and host countries interpret and enforce rules. This can give rise to conflicts, even under an international standard like the Basel Accord. At times, we have been given conflicting requirements by home and host regulators under Basel I, making compliance an impossible ``Catch-22''. While we have been able to resolve these issues to date, the potential tension between ``home and host'' regulators will become a bigger issue given the much wider and more detailed Basel II regime. If each country decides to require its own local rules and local data for each of the many calculations required under Basel II, the compliance burden will go from bad to worse. The Basel Committee has formed an Accord Implementation Group to deal with cross-border implementation issues, but experience shows that some differences between multiple supervisors are inevitable. We are pleased to note that, in a speech last week, Vice Chairman Ferguson indicated that the U.S. banking regulators expect to accept the Basel II approaches and calculations followed by a bank's home country supervisors, when evaluating an international bank with U.S. branches and for purposes of eligibility of Gramm-Leach-Bliley Act financial holding company status. This is reassuring to hear. We hope that other host countries adopt similar policies that defer to home country regulators, and that similar issues related to subsidiary banks also are addressed. We believe that stronger proposals should be developed to resolve home/host country conflicts in a timely and more predictable manner. Securitization A germane example of Basel II's complexity and excess prescriptiveness is its proposal for asset securitization. Asset securitizations are a cornerstone of how the U.S. markets finance residential mortgages, consumer credit card balances, automobile loans, and other receivables. The draft rules here are daunting, potentially quite burdensome, and often difficult to interpret. The result is that only a few experts in each area are likely to understand this and other specialized rules of the Accord. Yet, the interpretation of these experts on some technical points can have enormous impact on the capital calculation. These rules are written to deter possible arbitrages in the new rules, but risk throwing the ``baby out with the bath water.'' The industry and regulatory communities generally agree on the objective that capital should be similar before and after securitization, since the total economic risk is unchanged. However, apportioning the risks properly among the different securities poses a difficult challenge for any set of static rules. The Basel Committee's current proposal under CP-3 takes a conservative approach to this problem, focusing on avoiding improper capital arbitrage by building a technically complex system with a ``belts-and-suspenders'' philosophy. Unfortunately, this approach can also interfere with legitimate transactions and could undermine a widely accepted risk management tool used by many United States institutions. Several problems remain that should be reviewed by the regulators. First, the mere act of securitization and distribution will tend to increase the capital charge assigned to the same pool of assets.\6\ This increased capital is an important issue for the U.S. markets in particular, as foreign markets are much less reliant on securitization technology. This could raise costs for funding U.S. consumer loans and other asset classes where securitization techniques are important. Foreign regulators have much less at stake in their local markets. --------------------------------------------------------------------------- \6\ For example, the originating bank is charged the full risk of the pool if it retains a sufficiently large position in the junior securities. A second bank that purchases the senior securities also will be charged significant capital, meaning that the capital required of the banking system will be higher than if the assets had simply been held on an institution's balance sheet directly. --------------------------------------------------------------------------- Second, the calculations are subject to difficult interpretations, which can give rise to ``cliff edge'' uncertainties, where capital charges can change by a factor of 10 or more depending on whether a particular instrument can be fit into a specific regulatory box. For example, a credit line provided to support a credit card or receivables facility might attract a risk weighting of 100 percent if the bank can satisfy a number of technical tests about the structure of the credit facility.\7\ However, this charge can skyrocket to 1250 percent (that is, an outright deduction from capital) if a bank cannot meet one of these compliance requirements. This is a conservative approach \8\ that will certainly help deter arbitrage, but it may also deter good finance. It also will tend to restrict the evolution of new markets and new securities, since these future instruments might not fit easily into today's compartments. As with other areas of the Accord, we believe that moving to a more principles-based system that leaves more discretion to banks--subject to thorough supervisory oversight--will provide a more durable and flexible solution for the long term. It will be important to incorporate these changes into both the final text and in the practical implementation of the rules.\9\ --------------------------------------------------------------------------- \7\ In particular, questions remain regarding the proposed treatment of liquidity facilities for asset-backed commercial paper programs, which would face capital charges that seem disproportionately high relative to the level of risk. \8\ Some also have argued that the risk weights on such securitized assets are too high as a more general matter. Similarly rated corporate loans often attract a much lower capital charge. \9\ For example, banks that qualify for the Advanced IRB approach should be allowed to use internal ratings to determine risk weights, which is not allowed for securitizations under CP-3. Ratings based on rating agency methodologies or reasonably equivalent approaches, for example, should provide supervisors sufficient comfort that a market test has been met. Liquidity facilities and credit enhancements for asset-backed commercial paper conduits are prime examples where this approach could be easily adopted. --------------------------------------------------------------------------- Cost The monetary cost of complying with the Basel II rules will be significant. For Credit Suisse Group, our holding company, we estimate that our initial costs will be $70mm to $100mm just to implement the system, plus substantial ongoing costs. Multiply that by thousands of banks globally and this will amount to many billions of dollars of additional costs. Some of these costs will be passed on to consumers and corporations, and some of these costs may force banks to exit certain activities leaving these markets to unregulated entities. A major driver of the cost / benefit ratio of the new rules will depend on how they are applied. For example, there are more than 50 specific requirements that must each be met to use the so-called IRB advanced credit system. If each of them is interpreted and tested to rigorous audit standards, there will be enormous costs in compliance though the relevance to better risk management will be small. I would note that implementation costs will be substantial for regulators as well as for the banking community. Even more important, perhaps, than the direct monetary costs, are the indirect costs. These will depend on whether the new rules support the real risk management needs of the business, or whether they become an extra bureaucratic burden or even a diversion. CSFB's internal assessment is that most of the additional resources required will not be in the risk control departments. Instead, most of these new resources will be needed in the areas of financial reporting and IT support systems, in order to generate the volume of data and reports that Basel II requires to a reliable, audit quality standard. While further systems development provide some important benefits, this result suggests that the gains in risk management quality from the new proposal are likely to be relatively modest. Adaptability The proposed Basel rules are based on the financial markets as they work today, but are so complex and heavily negotiated that they will be difficult to update over time. Indeed, some commentators have suggested that the Accord will be outdated by the time of implementation.\10\ --------------------------------------------------------------------------- \10\ See Risk Magazine, footnote 3 above. --------------------------------------------------------------------------- The draft Accord also requires banks to use the Basel II processes in their internal management in many areas, regardless of whether they remain relevant for business practices. If bank management is required to compute and manage by the Basel II rules anyway, further improvements in internal practice could be seen as both costly and irrelevant. As a result, the Basel Accord could actually slow the progress of better private sector risk management techniques. Proposal Our suggested response to the problems of prescriptiveness and high cost is for the Basel Committee to place a much greater emphasis on the principles-based approach that underlies the ``Pillar 2'' section of the proposed Accord.\11\ Whereas Pillar 1 sets out regulatory capital calculations in a detailed, prescriptive way, the approach of Pillar 2 is to force firms to develop their own internal models, based on evolving best-practice, and then to scrutinize the results through the examination process and regulatory guidance. This ``principles-based'' approach, subject to some reasonable benchmarks and guidelines to maintain consistency, has some important natural advantages compared to the complex ``black-letter'' style rules currently prescribed by regulators under Pillar 1. Pillar 2 encourages banks and regulators to work together over time to improve risk management practice, rather than forcing compliance with a potentially dated rulebook. That approach permits steady, evolutionary improvement and should therefore be more durable and relevant than Pillar 1 rules that are designed with today's markets in mind. --------------------------------------------------------------------------- \11\ Our Pillar 2 comments here are strictly focused on the credit risk capital charges. As noted later in this testimony, Roundtable members have differing views on whether any operational risk charge should be addressed under Pillar 1 or Pillar 2. --------------------------------------------------------------------------- Addressing this issue will not be simple in the short time left before the rules are finalized. If these rules are all applied as black letter law and interpreted strictly, the new rules will be both costly and--since the risk management advances that lead in part to Basel II will not end in 2003--potentially irrelevant to ongoing best practice. We encourage an approach that emphasizes principles and simplicity as the rules are finalized, and a less onerous ``trust but verify'' approach to compliance. Specifically, we would support adding statements to the Accord to emphasize that compliance with the rules will be based not on ``box checking'' but with the spirit of the rules, based on economic content. Impact on Competition We believe that the cost and complex rules of Pillar 1 will have significant impacts on competition, and could tilt the current playing field significantly in various markets. This will be particularly important in the United States, where nonbank competitors like investment banks, finance companies, and insurance companies represent a large part of the financial system. The Basel rules do not apply to them. If the costs of Basel II are high, banks will earn a lower return on capital, will grow more slowly and may lose market share. There may even be some incentives to exit businesses or to de-bank altogether. We believe that the Basel Committee needs to do significantly more work in assessing the competitive impact of the rules across the financial marketplace. Procyclicality The new rules will change how banks calculate and manage their capital and the amount of business they choose to do. If banks all act in concert--as they will tend to do under a common regulatory regime-- this can significantly increase or decrease liquidity in the credit markets and ultimately affect the real economy. We have analyzed this effect over the last 20 years of credit cycles. Our calculations suggest that the impact on required bank capital will be substantial. In particular, the New Basel II calculations could require much more bank capital during economic recessions than the current system. The process by which these rules could widen economic swings is called ``procyclicality.'' \12\ This is, in effect, an implicit change in macroeconomic policy and it would be wise to consider that carefully. --------------------------------------------------------------------------- \12\ CSFB has taken particular interest in this issue among Roundtable members. See American Banker, ``Basel Capital Accord Must Leave Some Room for Human Judgment'' by Wilson Ervin and Joseph Seidel, August 30, 2002, and Risk Magazine, ``Procyclicality in the New Basel Accord'' by Wilson Ervin and Tom Wilde, October 2001. --------------------------------------------------------------------------- As a practical example, consider the credit environment of the last 2 years. We have seen a huge number of credit rating downgrades, which have increased the real risk of bank portfolios. The current system is relatively indifferent to this change in terms of required regulatory capital, but the proposed system will require significantly more capital when companies are downgraded. Banks will have to choose between raising more capital during recessions or reducing the amount of lending that they do. Some regulators have suggested that the fear of a capital shortfall will change banks' risk assessment and lending behavior so that this issue will disappear. Implicitly, they suggest that bank's risk assessment will improve so much that mistakes will be a thing of the past. While it would be wonderful if banks could always foresee the future, I do not think that is realistic. Bank management already makes risk assessment a top priority--it is perhaps the core judgment that determines whether a bank thrives or fails. Unfortunately, economies are likely to remain cyclical and predictions about the future will inevitably turn out to include their share of mistakes. Cutting lending during a downturn is probably smart, if your perspective is focused solely on bank solvency. However, it raises significant issues for the wider economy. My personal estimate is that my bank would have cut back its lending by perhaps an additional 20 percent to 30 percent if the Basel II rules were in place during 2002. If all banks cut back at the same time, the potential adverse impact on the real economy could lengthen and deepen the recession. We are currently working through an economic slowdown; it is difficult to think that adding pressure on bank capital during this period would be helpful to economic recovery. In fact, it defeats part of the reason for regulating banks in the first place--in order to have a stable supply of capital to support the underlying economy. We need to be particularly careful here because the new system is imposed across the whole banking system and everyone will have to operate at the same time on the same rules. Herd behavior can make smaller problems into bigger ones. The regulatory community has acknowledged this as a potentially serious issue, but we believe that further attention is warranted, because the consequences of getting this wrong are potentially quite important to the broader economy. When the first quantitative proposals in January 2001 revealed a significant potential problem, the regulators did react with a revised and somewhat ``flatter'' risk- weight curve.\13\ However, while this reduces the scale of the issue somewhat, it does not grasp the nettle. --------------------------------------------------------------------------- \13\ Basel Committee on Banking Supervision, Working Paper ``Potential Modifications to the Committee's Proposals'', Bank for International Settlements, November 2001. --------------------------------------------------------------------------- The current Pillar 2 proposals include a credit risk ``stress test'' which is directly linked to possible additional capital requirements.\14\ The exact design of this test remains unclear but the language suggests it amounts to an extra layer of buffer capital so that banks will not need to dig into their core capital in tough times. In effect, this is like creating a second fire department, because you want to always keep the first fire department in reserve. Creating two fire departments or requiring two pools of capital is unnecessarily expensive and doesn't seem to address the fundamental issue. That issue is that a risk sensitive system will inevitably lead to varying capital requirements through time, and that is a result that will require explicit management and thoughtful preparation. As with other areas of the Basel Accord, adding some flexibility to the rules is the simplest and most practical way of preventing these inevitable stresses from building up into major crises. --------------------------------------------------------------------------- \14\ Basel Committee on Banking Supervision, CP-3, April 2003, Paragraph 397-399 and 724. --------------------------------------------------------------------------- At a minimum, we suggest that the Basel Committee add to the proposed Accord an explicit acknowledgment that capital levels may fluctuate, and that Pillar 2 reviews and stress tests not become one- way ratchets that only increase regulatory capital requirements. If a stress test is to work properly, then when tough times arrive, banks should be permitted to live within their plans, and regulators should resist the temptation to continue to require the same untouched capital cushion. Otherwise, Basel II's stress test will not in fact reduce procyclicality, but will simply amount to an unpublished higher minimum capital standard. Operational Risk In addition to reforming capital charges for credit risk, Basel II establishes a new capital charge for operational risk--the risk of breakdowns in systems and people. This is the most controversial element of the proposed Accord. Financial Services Roundtable Comment It is important to distinguish between the concepts of managing operational risk and imposing a separate, quantitative capital requirement for it. All of the Roundtable's member companies agree that evaluating and controlling operational risk is important and should be required as a supervisory and business matter. Roundtable members do not agree on whether or how operational risk should be reflected in regulatory capital calculations. Many companies believe operational risk can best be addressed through case-by-case supervisory reviews under Pillar 2; others favor a quantitative and a publicly disclosed capital charge under Pillar 1. In several forums, the Roundtable itself has opposed a separate capital charge for operational risk and has argued for handling the issue through supervisory reviews under Pillar 2, much as interest rate and liquidity risk are handled. The Roundtable's senior management has expressed its concerns directly with Federal Reserve Board Chairman Greenspan and Vice Chairman Ferguson. Many Roundtable member companies strongly oppose any Pillar 1 operational risk capital charge. However, several Roundtable member companies just as firmly support Basel II's proposed Pillar 1 approach, following the development of the Accord's ``Advanced Measurement Approach'' (AMA), which gives banks flexibility to use their own internal methods for determining the regulatory capital needed for operational risk. Institutions that support a Pillar 1 operational risk charge believe it would improve transparency and comparability and bring regulatory capital requirements into closer alignment with the ``economic capital'' determinations used in these banks' internal management decisions. These institutions contend that any approach other than an explicit Pillar 1 charge for operational risk would impede progress toward a level playing field, by affecting the process of calibrating regulatory capital minimums. That is, these members believe that if an operational risk charge were not included in Pillar 1, the resulting capital charges on credit risk and market risk would remain higher to compensate, making it more difficult for international banks to compete with institutions that are not covered by the new Accord. The Roundtable continues to have concerns about the proposed operational risk capital charge, as well as several technical questions about its implementation. One problem that all of our members agree upon is that the proposed Accord fails to give enough recognition to the benefits of insurance in mitigating operational risk. CSFB Comment CSFB is concerned about the attempt to base an operational risk capital charge on new, unproven models, and believes this approach is problematic and possibly even counter-productive. We agree that operational risk is a critical risk to manage, and we set aside significant capital to cover potential surprises in our internal capital allocation process. However, we do not believe that operational risk can be modeled in the quantitative way proposed under the Basel II rules. Many efforts to measure operational risk have been proposed, often focusing on limited areas (for example, operations processing losses) that happen to be susceptible to statistical techniques. But these methods are not generally relevant to major risks, such as fraud, a changing legal environment or a major disaster, which are the risks that require capital. Operational risk capital is primarily to insure against the risk of being fundamentally surprised by a major event, but it is difficult to predict and measure what you do not expect. Basel II and other regulatory initiatives will push banks to devote significant resources toward operational risk systems and loss databases, but I personally feel that these resources could be better utilized elsewhere. Basel II's Advanced Measurement Approach to operational risk requires banks to attempt to verify their models statistically. Many are working hard on this, but we have yet to see any model that has actually been verified in a robust way. In fact, by emphasizing quantitative numbers for operational risk, we may be creating a real danger--creating a false sense of security that we have measured operational risk and hence controlled it. I am a model- oriented, technical person by training, but I do not want to rely on a model that is built on speculative assumptions. It is encouraging that the Basel Committee has sent signals suggesting an increased degree of flexibility in operational risk calculations. I am hopeful that this will bear out through the implementation. But we will still have a long way to go, and I am concerned that there will be a tendency to revert to prescriptive and unscientific requirements as regulators develop specific rules for approving operational risk models. Pillar 3--Disclosure Rules One of the strengths of the Basel II proposals is that they look beyond just calculating and maintaining capital levels. In designing Basel II, regulators realized that capital requirements--the so-called ``Pillar 1''--could never ensure the safety and soundness of the banking system alone. They understood that ultimately it is more important to encourage constructive relationships between financial institutions, their supervisors and the market to produce good risk management. This reasoning, which has the strong support of the banking industry, has lead to the creation of the two qualitative Pillars of the Basel Accord. Pillar 2 deals with the supervisory review process and, in particular, regulatory oversight of banks' internal economic risk assessments. Pillar 3 seeks to enhance market discipline through increased public disclosure requirements. The concepts behind the proposed rules for Pillar 2 and 3 are well accepted by the industry and regulators alike. However, many of the detailed proposals in the Pillar 3 market disclosures section are cause for concern in the industry. Unfortunately, the development of Pillar 3 is an area where consultation between the industry and the regulators came late in the process. Although CP-3 has improved the situation somewhat, we believe the proposals still are overly prescriptive, burdensome and subject to misinterpretation. The Pillar 3 requirements also reflect a somewhat narrow view of risk, focusing exclusively on a specific regulatory view of risk capital. We currently publish approximately 20 pages of risk information in our annual report, and we support transparency and disclosure as very worthwhile goals. The Pillar 3 proposals would add a large mass of additional disclosure which is highly technical in nature and which we believe will be of little benefit to the reader. Indeed, few people are able to digest all of the information that is already presented on risks, but now this information could be lost in a deeper, more technical pile of data. The additional requirements proposed under Pillar 3 are more likely to confuse than illuminate. As Chairman Greenspan has recently remarked, transparency is not the same as disclosure: ``Transparency challenges market participants not only to provide information, but also to place that information in a context that makes it meaningful.'' \15\ In this, we believe the prescriptive, volume oriented focus of Pillar 3 falls short. --------------------------------------------------------------------------- \15\ Remarks by Federal Reserve Board Chairman Alan Greenspan, Corporate Governance, at the 2003 Conference on Bank Structure and Competition, May 8, 2003. --------------------------------------------------------------------------- Of particular concern are the numerous required disclosures that relate directly to the capital calculations performed within Pillar 1. Instead of disclosing measures of risk used in internal risk management systems, these disclosures mandate an explicit regulatory capital view of risk. In the most complex areas, such as asset securitization, these disclosures will surely be mystifying to all but the most expert audiences. Moreover, given the likely longevity of the Basel II accord (the current accord is in its 14th year), there is a need to ensure risk management practice is able to mature beyond the concepts now embedded in the Basel II proposals. Just as the market has moved beyond the current accord, there will inevitably come a time when some Pillar 1 calculations are no longer regarded as good measures of risk for all products. In that case, it must be possible for banks to alter disclosures to represent emerging best practices. Under Pillar 3 as currently proposed, banks will likely find themselves constrained to disclosing risks under a system that is no longer wholly relevant. In designing the details of Pillar 3, the Basel Committee has placed too much emphasis on quantity, rather than quality, of disclosure. It is emphasizing consistency by prescription instead of consensus. In contrast, the demands of the market have produced broadly comparable and largely voluntary disclosures of market risk by banks. This is an example of how Pillar 3 should work. It would be more effective if Pillar 3 established a general set of principles, and then allowed the discipline of the market to produce continuous improvement in risk disclosure. This would produce information that the market actually desires, rather than seeking to impose today's ideas on future market participants by fiat. Summary We are at an important crossroads in the reform effort. A lot of good hard work on designing the framework and gaining political consensus has been accomplished. We have a high regard for the efforts of the Basel Committee and the regulators who have worked so hard to capture the best current practices in risk assessment. CSFB and the Roundtable have tried to contribute to the specifics of those discussions in a constructive manner. We believe that the current proposal should be streamlined significantly, reducing the level of prescriptiveness and cost, so that the advantages of this project are not tarnished by its current shortcomings. Simplifying the massive weight of detailed rules in Pillar 1 will require continued discipline in the final round of drafting. It will also require a new emphasis on the ``spirit'' of the rules, both as the rules are finalized and when they move to the implementation phase with national regulators. If, instead, these rules are written and interpreted as black-letter regulations, set at a highly technical audit standard, the cost of overall implementation will be high. Such an approach would mean the calculations could also become increasingly outdated and less relevant to risk management best practice over time. We can hope that all national regulators will avoid this pitfall, but international banks will tend to be driven by the standards set by the strictest and most literal of their major regulators. Much hard work has been put into Basel II, but much also remains ahead. The timetable for implementation is challenging, particularly since the Accord's require a minimum of 3 years of data for the advanced calculations--meaning that banks will need to revise systems to begin collecting the new information by early next year. In the pressure to finalize and implement the Accord, we hope that enough time will be provided for everyone--banks and supervisors alike--to digest and think about the implications of the new regime, and to develop appropriate transition rules. As a final comment, I believe that much more can be accomplished by increasing the emphasis on the concepts of Pillar 2 and Pillar 3, and a focus on the principles of evolving best practice rather than fixed formulae. This approach would not only help address ``prescriptiveness, cost and adaptability'', but could also help address the issues of operational risk and procyclicality. Pillars 2 and 3 have real people on the other side--regulators and the market. Human judgment can adapt to changes and new markets more easily than a rulebook can. This approach, properly applied, also puts the burden back where it should be--on the shoulders of bank management to demonstrate to the regulators and the public that they are doing a good job. That is in the spirit of the Sarbanes-Oxley reforms, and I think it is a smart, durable way to improve discipline and maintain best practice standards. Finally, it should also make the new system more responsive to change and therefore more relevant over time. Without adjustments to make Basel II more flexible and to allow it to evolve over time, I am afraid we might have to start work on a Basel 3 before the ink is dry on the current effort. Thank you. ---------- PREPARED STATEMENT OF KEVIN M. BLAKELY Executive Vice President and Chief Risk Officer, KeyCorp June 18, 2003 Introduction Thank you, Mr. Chairman. I am here today on behalf of KeyCorp, the 11th largest banking company in the United States. KeyCorp has total assets of approximately $85 billion, and spans the northern half of the United States from Maine to Alaska. While the vast majority of our business is domestically based, we do have a modest level of international business activity. KeyCorp is not one of the institutions included in the definition of ``top 10 most internationally active institutions.'' Accordingly, under the present regulatory guidance, we will not be required to comply with Basel II when it becomes effective in 2006. Nonetheless, it is our intent to qualify as an advanced practice institution. We simply believe that it is good banking practice to develop the risk management tools that are the foundation of Basel II: If that qualifies us as an advanced practice company under the new accord, so much the better. I believe my testimony today provides a rather unique perspective on the issue of whether or not Basel II is good for the banking industry. For the first 17 years of my professional career I was a bank regulator with the Office of the Comptroller of the Currency (OCC). Much of my time with the OCC was spent dealing with problem and failing institutions. During my last several years with the OCC, I was Deputy Comptroller for Special Supervision. That is a nice way of saying I was responsible for the department that dealt with severely troubled and failing financial institutions. My tenure in the Special Supervision Department ran from 1986 through 1990, a time when a significant number of banks failed in the United States. I was able to see first hand the myriad of reasons that caused banks to get into trouble. Not the least of these was the inability to appropriately identify and manage their risks. I left the OCC in 1990 to join the deeply troubled Ameritrust Corporation in Cleveland, Ohio. Ameritrust was a $12 billion company that had encountered difficulties arising from its loan portfolio. I was part of the new management team focused on turning the company around. Over an 18-month period, Ameritrust lurched from one crisis to another, but we eventually were able to stabilize the company. During the interim period I lived, first hand, through the effects of a firm that had little in the way of risk management practices and tools. My experience with the OCC's failing banks division and the Ameritrust debacle convinced me that there had to be a better way of managing risk in the banking industry. In 1992, Ameritrust was acquired by Society Corporation, the precursor of today's KeyCorp. I was placed in the position Executive Vice President of Credit Policy and Risk Management. In this capacity, I was given the opportunity to explore and experiment with new risk management tools that were beginning to bud in the industry. I was encouraged to do so by our CEO who expressed a desire to have a system whereby he could understand the totality of risk that our company faced on a daily basis. Our CEO envisioned a process that could tell him how much aggregate risk the company was taking, including the risks that emanated from our credit, market, and operational activities. He wanted a system that could allow us to increase, decrease, or maintain our risk position as circumstances warranted. Neither of us realized it at the time, but he was describing a process that is today commonly called ``enterprise- wide risk management.'' In 1993, I commenced the first step of his vision by installing a Value-at-Risk (VAR) system in our company's trading floor. VAR was a highly complex model designed to measure risk in the bond, equity, and foreign exchange trading we undertook on a daily basis. Due to the complexity of a VAR model, I had to engage several Ph.D.'s to help us implement it. During the course of their engagement, I happened to mention my frustration in finding an enterprise wide system that could aggregate the risk of each of our banking activities. One of the Ph.D.'s suggested that I look into the concept of economic capital allocation, now commonly known as ``risk-based capital.'' Once I investigated the premise of risk-based capital allocation, I concluded I had discovered a powerful risk management tool. Implementing such a model at KeyCorp would enable us to allocate capital to our lines of business based on the amount of risk they took. Each line of business would be charged for the amount of credit risk, market risk, and operational risk they encountered. Using the aggregate of that capital charge as the denominator, and the revenue they generated as the numerator, we could determine which lines of business were getting appropriately paid for the risk they took. For the first time, we would be able to put all our lines of business on an apples- to-apples comparison basis. Hence, the ability to know our level of risk and whether or not we would be paid for the risk being taken. Further, we would be able to aggregate the total amount of capital being allocated to all our lines of business to understand the totality of risk our company was taking. It was the enterprise-wide solution we had been looking for. KeyCorp commenced building an economic capital allocation program in the mid-1990's because we firmly believed that it was the right thing to do. It has taken us nearly a decade to build it, and we are still not finished with it. Nonetheless, even after nearly 10 years we remain convinced that it is the best way to run our company. No regulator has told us that we must do this. We are pleased to note that this powerful risk management tool, economic capital allocation, is now the underlying driver of Basel II. Our company was highly critical of the initial version of Basel II and publicly stated as much. We felt that it failed to address the sophistication and complexity that our industry routinely operated in. We felt it was inadequate and little better than the original Basel I. Put simply, it did not adequately address risk sensitivity. However, over the next several years we were pleasantly surprised to see how Basel II became a much better document. The regulators working on the new accord have been genuinely receptive to hearing the concerns that KeyCorp and others have raised. We haven't always gotten our way, but at least we have been heard. We believe that Basel II is now on the right track. Financial institutions will need to develop more sophisticated risk management tools to support the risk-based capital premise upon which it is built. This is a good thing. In today's world of complex financial markets, tools such as value-at-risk, two-dimensional loan grading systems, enterprise data warehouses, and operational loss databases are not a luxury; they are a necessity. In order to understand their risk positions, banks should be calculating risk-based capital and using these tools to do so. While models are no substitute for human judgment, they certainly create a more informed human with whom to make the decision. One of the benefits we see in the Basel II proposal is that we will finally be free to price our products and services commensurate with the risk they entail. As previously mentioned, Basel I provides very little in the way of risk sensitivity. One of the perversities of this shortcoming is that it has driven high quality borrowers away from the banking industry. These clients can access providers of credit not subject to the costly level of capital that banks are currently required to hold. In essence, banks are forced to overprice for this business, and they lose it to other cheaper, nonregulated providers. Conversely, Basel I's simplistic 8 percent capital requirement has allowed banks to hold less capital than they should against borrowers that are high risk. This has resulted in banks underpricing such credit. It should be no surprise, then, that Basel I has chased high- quality credits away from banks, while attracting low-quality credits to them. If banks are allowed to calculate the proper level of capital to be held based on a realistic stratification of credit risk, this serious problem will largely disappear. This is one of the tenets that Basel II is based upon: You hold the level of capital necessary to support the risk, and price for it accordingly. I would now like to address some of the criticisms that have been leveled against Basel II. These would include its cost, complexity, inflexibility, and propensity to foster procyclicality. I would also like to provide a few comments on the merits of Basel II's Pillar 1 versus Pillar 2. Cost Much has been said about the cost of building the models necessary to comply with Basel II. At KeyCorp, we wonder how anyone can afford not to build them. We, ourselves, have painfully learned the cost of not having them. In 1996, our risk-based capital process was still in its embryonic stage: In truth it did not begin to take hold until 2000. In 1996, we were still calculating profitability measures utilizing the primitive 8 percent capital standard stipulated by Basel I. On this basis, one of our loan portfolios, leveraged lending, was producing an eye-popping return on equity close to 30 percent. As a consequence, we unfortunately pursued expansion of leveraged lending over the next several years. At the end of 1998, the quality of this portfolio began to collapse and we have written-off many millions of dollars since. We have looked retrospectively on our experience with this portfolio. We believe if we had had our risk-based capital model in place (the kind proposed by Basel II) our anticipated return would have been in the single digit range. Such knowledge would have caused us to avoid this particular lending activity and to seek other opportunities that offered better risk/reward ratios. Through this experience, we have learned an important lesson from which others can benefit. The entire cost of the nearly 10-year effort to implement our economic capital model (the same kind proposed by Basel II) pales in comparison to the cost of not having it in place. We have read that others estimate the cost of compliance with Basel II to be staggeringly high. We are not convinced this is the case, and it certainly has not been so at KeyCorp. Yes, we have spent multiple millions of dollars over the years investing in risk management tools and models, but we have done so because we believe those tools are necessary to conduct our business in a safe and sound manner. Frankly, they will also make us a better competitor. The more we understand our risk, the better we will be at managing and pricing for it. Some have criticized the cost of auditing and back-testing the accuracy of the models that Basel II is based upon. We view such activities as nothing more than good common sense. Auditing and back- testing of outputs is critical to ensuring that the model is producing reasonable numbers. Auditing/back-testing serve as the tuning devices necessary to modify the models' calculations. For example, auditing and back-testing of VAR models is an accepted practice in the industry now: Everyone knows their benefit. We view auditing/back-testing as necessary investments needed to create a better model. Better models create better understanding of risk and the ability to better manage it. Better management of risk results in lower losses to banks. We believe the cost of auditing/back-testing is inconsequential compared to the losses that can occur due to inferior risk management processes. Before one accepts the large figures attributed to Basel II compliance, one must subtract the costs of building the risk management systems that a good financial institution would invest in, regardless. We do not believe the gap between the two is significant. Complexity We cannot deny that Basel II is a complex document. It is. Yet, it needs to be. Banking is a complex business that needs complex solutions to the issues it faces. We should not run from complexity but instead be willing to face it and manage our way through it. I have previously mentioned that KeyCorp installed a VAR system for its trading floors in the early 1990's. At that time, many were saying VAR systems were exceedingly complex, expensive, and too mathematically driven. Yet, today VAR systems are widely recognized as the standard by which to manage risk in their trading books. VAR is a superior risk management tool that never would have come to be had the financial services industry been intimidated by its complexity. I reiterate: When VAR first surfaced, it was accused of being too complex, costly, and mathematically driven, the same crimes Basel II stands accused of today. Yet, VAR has become the industry standard. Inflexibility Some fear Basel II will trap the industry with year 2000 era risk management tools and stifle creation of new ones. We believe this concern is overstated. The 1988 Basel Accord was a woefully inadequate document from the start. Its simplistic approach mandated a specific capital level and made no provisions to the contrary. Yet, over the past 15 years, the financial services industry has continued to develop new risk management tools never envisioned by the 1988 Accord. These would include: VAR models, two-dimensional loan grading systems, economic capital models, and enterprise-wide data warehouses. The fact that such tools were not contemplated by Basel I did not interfere with the industry's pursuit of them. We anticipate a similar situation with Basel II--banks will continue to pursue a better risk management mousetrap. We will acknowledge, however, that regulators must be willing to consider the new tools as they are developed, and work with the industry to accommodate them as their effectiveness is demonstrated. Procyclicality We have frequently heard that regulators are concerned that Basel II might allow substantial capital to escape from the banking system. We believe the whole premise of procyclicality is evidence that such concerns may be overstated. Basel II capital levels represent the minimum level of capital that an institution is to hold. The premise of procyclicality assumes that banks operate at or near the minimum capital level. We believe it is highly unlikely that any banking company worth its salt will allow their capital to sink to the lowest acceptable level. Some argue that under Basel II, economic downturns will cause financial institutions to become more reluctant to lend when liquidity is most needed. Banks would be placed in a position of making a difficult choice: Immediately raise new capital or stop lending. In truth, there is a third choice that most banks will probably follow: Retain a buffer level of capital to accommodate cyclical changes in risk that everyone knows will inevitably occur. We believe that even in times of economic stress, banks genuinely desire to make new loans to drive their own revenue streams. Our current economic situation is a prime example: Banks are anxious to lend money. The demand is not there. Pillar 1 versus Pillar 2 One of the basic principles of Basel II is to make risk transparent so that it is comparable from one institution to another. Pillar 1 encourages a formulaic based system that will enable this to occur. Consistency of methodology is critical to empower investors, regulators, and depositors with the information they need to gauge the risk of the institution with whom they are dealing. Without Pillar 1's consistency of approach, a Tower of Babel syndrome can occur. Pillar 2 relies more on flexible judgment as to how much capital is warranted at an institution. We acknowledge and accept that regulators must have the flexibility to invoke their authority to ignore the results of Pillar 1 when circumstances so dictate. However, completely abandoning Pillar 1 in favor of Pillar 2 yanks any comparability benefit away from investors and depositors. The invisible hand of the market will be impeded in its ability to quickly discipline a wayward institution. For example, much has been said about the need to place operational risk under a Pillar 2 approach. In this regard, the individual regulator that happened to be examining a particular bank would largely determine the adequacy of capital held for its operational risk. This lends itself to varying assessments, interpretations, methodologies, and enforcements. An investor attempting to compare the level of capital held for operational risk at multiple banks must assume that different examiners will utilize the exact same thinking in their operational risk assessments. That simply doesn't happen. A more formulaic approach, where all banks are using the same scorecard, lends itself much more to consistent comparability. The mere presence of a Basel II draft has caused many in the industry to start contemplating new ways of tracking operational risk. This would include KeyCorp. We have commenced building an operational risk database that will give us better information regarding the source, size, and amount of operational losses. This database will ultimately serve as the system that feeds our operational risk model. We believe it can be supplemented by exchanging information on operational risk losses with other financial institutions. This will help us build the critical mass necessary to create reliable, predictive loss forecasting models. I will readily admit that we have a way to go in this particular area, but the presence of Basel II over our heads encouraged KeyCorp and others in the industry to get moving on building the databases sooner. Conclusion In conclusion, KeyCorp believes that Basel I is hopelessly broken and that a new accord needs to be implemented. Basel II is a major step forward and we applaud its approach. It is not perfect now, nor will it be perfect when implemented, nor perfect 10 years after implementation. Regardless, it is light years ahead of Basel I, as well as any other proposal we have seen to date. It should be supported. We acknowledge it is complex, but banking is a complex business. A simple solution to complex issues is probably not the right medicine. As an industry, we should not shy away from the remedy simply because it is complex. Instead, we should work collectively with the regulators to find the right solution, not the easy one. We have our doubts as to the high cost figures attributed to Basel II. Our own experience to date has proven to the contrary. Further, we believe many Basel II costs are simply expenditures we should otherwise be making as a matter of sound banking practice. Good risk management costs money, but it is intended to help avoid even bigger costs that arise from bad risk management. We do not believe adoption of Basel II will trap the financial services industry in a time warp. Banks will continue to develop better methods of managing risk regardless of what Basel II requires. However, regulators must be open and responsive as these new tools are developed. We believe there is substantial merit to including as much as we can in Pillar 1 versus Pillar 2. One of the greatest benefits that Basel II promises is that it will utilize the invisible hand of the market to discipline wayward institutions. In order to do that, investors must have adequate information to compare the risk of one institution against another on an apples-to-apples basis. Pillar 1 is the best vehicle for ensuring that banks report on a consistent basis. Mr. Chairman, KeyCorp appreciates the opportunity to share our views on Basel II. We want to make sure our industry operates within a safe and sound environment. We know this is the goal of the Committee as well as our friends in the regulatory world. While Basel II is far from perfect, it certainly moves us further down the path.