[Senate Hearing 108-495]
[From the U.S. Government Publishing Office]



                                                        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) 512ï¿½091800  
Fax: (202) 512ï¿½092250 Mail: Stop SSOP, Washington, DC 20402ï¿½090001


            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. 




                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:

 Consideration of comments received by the Basel Committee on 
    CP-3. The comment period on this document concludes on July 31.
 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.
 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.
 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.
 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.
 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.
 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.
 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.
 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:

 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.
 A 1-year PD estimate for each grade must be provided as a 
    minimum input.
 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:

 Continue to promote safety and soundness and at least maintain 
    the current overall level of capital in the system.
 Continue to enhance competitive equality.
 Establish a more comprehensive approach to address risk.
 Contain approaches to capital adequacy that are appropriately 
    sensitive to risk.
 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:

 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).
 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.
 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:

 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.
 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.
 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.
 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.
