<DOC>
[109 Senate Hearings]
[From the U.S. Government Printing Office via GPO Access]
[DOCID: f:34133.wais]
S. Hrg. 109-890
THE DEVELOPMENT OF
NEW BASEL CAPITAL ACCORDS
=======================================================================
HEARING
before the
COMMITTEE ON
BANKING,HOUSING,AND URBAN AFFAIRS
UNITED STATES SENATE
ONE HUNDRED NINTH CONGRESS
FIRST SESSION
ON
PLANS OF THE U.S. BANKING AGENCIES TO UPDATE AND ENHANCE THE REGULATORY
CAPITAL PROGRAM THROUGH IMPLEMENTATION OF THE INTERNATIONAL CONVERGENCE
OF CAPITAL MEASUREMENT AND CAPITAL STANDARDS, AND REVISIONS TO THE
EXISTING DOMESTIC RISK-BASED CAPITAL FRAMEWORK FOR BANKS
__________
NOVEMBER 10, 2005
__________
Printed for the use of the Committee on Banking, Housing, and Urban
Affairs
Available at: http: //www.access.gpo.gov /congress /senate/
senate05sh.html
______
U.S. GOVERNMENT PRINTING OFFICE
34-133 WASHINGTON : 2007
_____________________________________________________________________________
For Sale by the Superintendent of Documents, U.S. Government Printing Office
Internet: bookstore.gpo.gov Phone: toll free (866) 512-1800; (202) 512ÿ091800
Fax: (202) 512ÿ092250 Mail: Stop SSOP, Washington, DC 20402ÿ090001
COMMITTEE ON BANKING, HOUSING, AND URBAN AFFAIRS
RICHARD C. SHELBY, Alabama, Chairman
ROBERT F. BENNETT, Utah PAUL S. SARBANES, Maryland
WAYNE ALLARD, Colorado CHRISTOPHER J. DODD, Connecticut
MICHAEL B. ENZI, Wyoming TIM JOHNSON, South Dakota
CHUCK HAGEL, Nebraska JACK REED, Rhode Island
RICK SANTORUM, Pennsylvania CHARLES E. SCHUMER, New York
JIM BUNNING, Kentucky EVAN BAYH, Indiana
MIKE CRAPO, Idaho THOMAS R. CARPER, Delaware
JOHN E. SUNUNU, New Hampshire DEBBIE STABENOW, Michigan
ELIZABETH DOLE, North Carolina ROBERT MENENDEZ, New Jersey
MEL MARTINEZ, Florida
Kathleen L. Casey, Staff Director and Counsel
Steven B. Harris, Democratic Staff Director and Chief Counsel
Mark Oesterle, Counsel
Andrew Olmem, Counsel
Aaron D. Klein, Democratic Economist
Joseph R. Kolinski, Chief Clerk and Computer Systems Administrator
George E. Whittle, Editor
(ii)
?
C O N T E N T S
----------
THURSDAY, NOVEMBER 10, 2005
Page
Opening statement of Chairman Shelby............................. 1
Opening statements, comments, or prepared statements of:
Senator Sarbanes............................................. 12
Senator Hagel................................................ 13
WITNESSES
John C. Dugan, Comptroller of the Currency, U.S. Office of the
Comptroller of the Currency.................................... 2
Prepared statement........................................... 39
Susan Schmidt Bies, Governor, Board of Governors of the Federal
Reserve System................................................. 4
Prepared statement........................................... 45
Response to written questions of Senator Shelby.............. 95
Donald E. Powell, Chairman, Federal Deposit Insurance Corporation 6
Prepared statement........................................... 50
Response to written questions of Senator Shelby.............. 119
John M. Reich, Director, Office of Thrift Supervision............ 7
Prepared statement........................................... 66
Response to written questions of Senator Shelby.............. 133
L. William Seidman, Former Chairman, Federal Deposit Insurance
Corporation and Former Chairman, Resolution Trust Corporation.. 21
Prepared statement........................................... 72
William M. Isaac, Former Chairman, Federal Deposit Insurance
Corporation and Chairman, The Secura Group..................... 22
Prepared statement........................................... 73
George G. Kaufman, Co-Chair, U.S. Shadow Financial Regulatory
Committee and John F. Smith Professor of Finance and Economics
Loyola University Chicago...................................... 25
Prepared statement........................................... 76
Daniel K. Tarullo, Professor, Georgetown University Law Center... 27
Prepared statement........................................... 85
Katherine G. Wyatt, Head, Financial Services Research Unit, New
York State Banking Department.................................. 29
Prepared statement........................................... 91
(iii)
THE DEVELOPMENT OF
NEW BASEL CAPITAL ACCORDS
----------
THURSDAY, NOVEMBER 10, 2005
U.S. Senate,
Committee on Banking, Housing, and Urban Affairs,
Washington, DC.
The Committee met at 9:30 a.m., in room SD-538, 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. You can
tell we generally start around 10:00, but some people will have
to leave here today, and I want to put you on notice that we
have stacked votes starting about 11:30, so maybe we can
accelerate the hearing and not lose any substance.
Today's hearing will examine recent developments regarding
the new Basel Capital Accord known as Basel II and certainly
proposed revisions to existing U.S. capital requirements known
as Basel I-A. Federal banking regulators have recently made two
important announcements with regard to Basel II. First, on
September 30, Federal banking regulators announced their
revised plan for implementing Basel II. Under the revised plan,
the implementation will begin in January 2008 with a parallel
trial run followed by a 3-year transition period from 2009 to
2011.
Because Basel II will require banks to maintain complex and
costly internal risk assessment processes, Federal banking
regulators have said that Basel II will apply to only the
largest, most sophisticated banks. Just a few days after
banking regulators announced the revised Basel II timetable,
they announced proposing revisions to existing capital
requirements, which have been referred to as Basel I-A.
The purpose of Basel I-A, as I understand it, is to
modernize the capital requirements for the roughly 7,000 or so
banks that do not qualify for Basel II. Additionally, Basel I-A
aims to address concerns that the application of Basel II to
only the largest banks would create a bifurcated capital
requirements system that could put small and regional banks at
a competitive disadvantage.
It is my hope that today's hearing will shed additional
light on the implications of adopting Basel II and the new
Basel I-A. The changes to U.S. capital requirements proposed by
Basel II and Basel I-A will have far-reaching implications on
not only the safety and soundness of the U.S. banking system
but also on both the competitiveness of the U.S. banking
industry and the effectiveness of Federal banking regulators.
Before we implement any changes, I believe we need to make
sure that we fully understand these implications.
Today's hearing will have two panels. Our first panel
includes John Dugan, Comptroller of the Currency; Governor
Susan Schmidt Bies of the Federal Reserve Board of Governors;
Chairman Donald Powell of the Federal Deposit Insurance
Corporation; and Director John Reich of the Office of Thrift
Supervision.
Our second panel will follow them and will include Mr.
William M. Isaac, former Chairman of the FDIC and Chairman of
Secura Group LLC; Professor George Kaufman, Co-Chair of the
Shadow Financial Regulatory Committee and the John F. Smith,
Jr. Professor of Finance and Economics at Loyola University
Chicago School of Business Administration; Mr. William Seidman,
no stranger to this Committee, former Chairman of the FDIC,
former Chairman of the Resolution Trust Corporation and CNBC
Chief Commentator; Professor Daniel Tarullo, Professor of Law,
Georgetown University Law Center; and Ms. Katherine Wyatt, Head
of the Financial Services Research Unit, New York State
Banking.
We welcome all of the witnesses to today's hearing. Without
objection, your written testimony will be made part of the
record in its entirety, and Mr. Dugan, we will start with you.
Sum up your testimony.
STATEMENT OF JOHN C. DUGAN
COMPTROLLER OF THE CURRENCY,
U.S. OFFICE OF THE COMPTROLLER OF THE CURRENCY
Comptroller Dugan. Thank you, Chairman Shelby. I appreciate
this opportunity to discuss the plans of the U.S. banking
agencies to strengthen regulatory capital requirements for our
banking system. We intend to do this in two fundamental ways:
First, through implementation for our largest banks of the
framework generally known as Basel II, and second, for banks
not adopting Basel II, through revisions to our capital rules
in an initiative generally known as Basel I-A.
In both efforts, our primary goal is to substantially
strengthen the long-term safety and soundness of our banking
system. Our largest banks require a more risk-sensitive
regulatory capital system to address their complex operations
and activities. For banks not adopting Basel II, we need to
increase the risk sensitivity of risk-based capital without
unduly increasing regulatory burden.
To ensure that Basel II would be consistent with continued
safety and soundness, the U.S. agencies conducted a
quantitative impact study earlier this year known as QIS-4. As
you will hear in more detail today, the QIS-4 results, which
were based on crude approximations of Basel II requirements,
nevertheless raised real concerns among the agencies because
they forecast substantial reductions in capital for Basel II
banks and substantial differences in capital requirements for
very similar credits.
Because these preliminary results would be unacceptable if
produced by the final Basel II framework, the agencies
conducted in-depth discussions about QIS-4 and what that should
mean for the future of the Basel II process. The result of
these discussions was an agreement by all the agencies to move
forward but only with substantial new safeguards to address the
QIS-4 concerns.
This agreement, which was released as a joint statement on
September 30, was based on several key premises. First and
foremost, despite the preliminary forecasts of QIS-4, the Basel
II framework provides necessary improvements to address
recognized flaws in the existing risk-based capital regime for
our largest, most complex banks. Basel II will promote
significant advances in risk management that will benefit
supervisors and banks alike and substantially enhance safety
and soundness.
Second, apart from the notice and comment process, further
study of the Basel II framework itself will do little to
resolve concerns raised by QIS-4, which by necessity was based
only on preliminary approximations of a completed Basel II
system. Instead, we need to observe live systems in operation--
and subject them to rigorous supervisory scrutiny--before we
will be able to rely on Basel II for regulatory capital
purposes.
And third, we must proceed deliberately, gaining a better
understanding of the effects of Basel II on bank risk
management practices and capital levels. That means a
meaningful transition period during which we can scrutinize
Basel II systems while strictly limiting potential reductions
in capital requirements through a system of simple and
conservative capital floors. Based on the experience we gained
through supervisory oversight in the transition period, we will
incorporate any necessary revisions to Basel II before the
transition period ends. I believe that once the Basel II
framework is implemented completely and rigorously supervised
in the controlled environment of the transition period, and
once we have had the opportunity to make necessary changes to
the framework based on the knowledge we gain during that
period, the concerns raised by QIS-4 will be addressed.
While the comprehensive Basel II framework is necessary and
appropriate to address the complex risks of our largest banks,
it would be far too burdensome and expensive to impose on our
other banks. Instead, we need meaningful but simpler
improvements to our capital rules for these banks that would,
first, make capital more sensitive to risk, and second, address
competitive disparities raised by the Basel II changes for our
largest banks. That is the purpose of our Basel I-A initiative,
in which the modifications we are considering would increase
the number of risk-weight categories, expand the use of
external credit ratings, and employ other techniques to
increase the risk sensitivity of capital requirements.
The banking agencies agreed that it is critical to have
overlapping comment periods next year for the rulemakings on
both Basel II and Basel I-A. This will allow the industry to
compare the proposals as they prepare their comments, and will
allow us, as regulators, to take both sets of comments into
account in finalizing each proposal--a process that will allow
a better assessment of the potential competitive effects of
these proposals on the U.S. financial services industry.
To summarize, doing nothing to revise our capital rules
would, over time, threaten the safety and soundness of the
banking system, especially with regard to our largest banks
that engage in increasingly complex transactions and hold
increasingly complex
assets. Basel II provides a conceptually sound and prudent way
forward for these largest banks by more closely aligning
regulatory capital and risk management systems with actual
risk. Likewise, Basel I-A will provide a more risk sensitive
framework for non-Basel II institutions.
Although both processes will take time and will inevitably
change to address supervisory concerns, I believe they both
will substantially enhance safety and soundness. It is for this
essential reason, safety and soundness, that I believe we
should support the approach embodied in proposals for Basel II
and Basel I-A.
Thank you very much.
Chairman Shelby. Thank you very much, Mr. Dugan.
Governor Bies on behalf of the Fed.
STATEMENT OF SUSAN SCHMIDT BIES, GOVERNOR,
BOARD OF GOVERNORS OF THE FEDERAL RESERVE SYSTEM
Ms. Bies. Thank you, Mr. Chairman.
I appreciate the opportunity to join my colleagues from the
other Federal banking agencies to discuss the current status of
Basel II in this country as well as the status of our proposed
amendments to our existing Basel I risk-based capital rules.
The Federal Reserve considers the maintenance of strong and
stable financial markets to be an integral part of our
responsibility and criticality
related to the safety and soundness of the participants in
those markets.
Part of maintaining a strong financial system is to ensure
that banking organizations operate in a safe and sound manner
with adequate capital cushions that appropriately support the
risks taken. There have been two major developments in the past
6 weeks regarding U.S. regulatory capital requirements that
apply to banking institutions: One relating to the Basel II
framework and another to the proposed Basel I amendments.
The Federal Reserve considers the ongoing discussion
between the Congress and the U.S. banking agencies--and of
course with the banking industry and members of the public--to
be critical to the success of both sets of proposals. While the
current Basel I-based rules have served us well for nearly two
decades, they are no longer appropriate for identifying and
measuring the risks of our largest, most complex banking
organizations.
The Basel II framework is designed to improve supervisors'
ability to understand and monitor the risk-taking and capital
adequacy of these institutions. It should also provide market
participants with a better ability to evaluate the risk
positions at those institutions. And perhaps most importantly,
Basel II should enhance risk management at institutions
adopting it. The advanced approaches under Basel II offer
particularly good improvements in terms of risk sensitivity,
since those approaches incorporate risk management processes
already used by certain banks today.
In April of this year, the agencies announced jointly their
reaction to initial results of a fourth quantitative impact
study pertaining to Basel II known as QIS-4. The initial QIS-4
results prompted the agencies to delay issuance of a notice of
proposed rulemaking (NPR) for Basel II in order to conduct
further analysis of those results and their potential impact.
During the summer, the U.S. agencies jointly conducted
additional analysis of the information reported in QIS-4 and
collectively decided to move ahead with an NPR but adjusted the
plan for U.S. implementation of Basel II by extending the
timeline and by augmenting the transitional floors.
Probably the most important thing we learned from the QIS-4
analysis is that progress is being made toward developing a
risk-sensitive capital system. From QIS-4, we also learned that
the favorable point in the business cycle from which the
measurement was conducted had an effect on minimum regulatory
capital produced; that the dispersion was largely due to
varying risk parameters used by different institutions; and
some of the data submitted by individual institutions, as
expected, was not complete.
Based on the results of QIS-4, the Federal Reserve
recognizes that all institutions have additional work to do.
Indeed, QIS-4 should not be considered a complete forecast of
Basel II's ultimate effects, since it was a point-in-time look
at how the U.S. implementation was progressing. We certainly
expect that as we move closer to implementation, supervisory
oversight of the Basel II implementation methodologies would
increase. In fact, we would not allow any bank to move to
transitional floors until we are satisfied that their Basel II
process met our standards.
The proposal for the Basel I amendments, an advance notice
of proposed rulemaking, relates, in part, to some longstanding
issues in our current capital rules. It is also intended to
mitigate certain competitive inequalities that may arise from
the implementation of Basel II. In considering these possible
revisions, the U.S. agencies are seeking to enhance the
relationship between inherent risks in bank portfolios and
minimum regulatory capital without undue complexity or
regulatory burden on the Basel I banks.
For both Basel II and Basel I amendments, the Federal
Reserve fully supports the retention of the existing prompt
corrective action regime, which the Congress put in place more
than a decade ago, as well as supporting existing leverage
requirements. These regulations help to ensure a minimum level
of capital both at individual institutions and in the
aggregate, which we consider to be absolutely vital to the
health of our financial system and economy.
We look forward to continuing to engage the industry, the
Congress, fellow supervisors, and others in discussion about
what effects the Basel II framework and the Basel I revisions
might have on our banking system. We will remain vigilant about
potential unintended and undesired consequences. We will make
adjustments as necessary to ensure that minimum regulatory
capital reflects risk exposures at individual banks and is
adequate for the safety and soundness of the banking system.
Finally, I would like to emphasize that from my
perspective, the U.S. agencies have worked well with one
another on these regulatory capital proposals, and worked in a
general environment of cooperation and good will, and we expect
that will continue. I thank my colleagues at the other agencies
for their efforts in developing the U.S. proposals for Basel II
and Basel I amendments.
Thank you, Mr. Chairman.
Chairman Shelby. Chairman Powell.
STATEMENT OF DONALD E. POWELL
CHAIRMAN, FEDERAL DEPOSIT INSURANCE CORPORATION
Chairman Powell. Good morning, Mr. Chairman. Thank you for
holding this important hearing.
Recently, the agencies described a plan for moving forward
with Basel II in the United States. Basel II has the potential
to bring about positive change for U.S. capital regulation and
that is why I participated in, and support, the agency plan.
Basel II will require better risk management for adopting
banks and give the agencies better data to identify emerging
risks. Basel II will give banks more flexibility on risk-based
capital. For Basel II banks whose risk-based capital
requirements exceed leverage ratio requirements, there even is
a potential for a moderate reduction in overall capital
requirements. The FDIC can and does support these things.
I would like to focus on the interaction of the Basel II
framework and our U.S. leverage ratio for bank regulatory
capital. There are other important issues, of course. Should
capital requirements for mortgages be reduced 90 percent or
more as Basel II permits? Do capital requirements for credit
card lending need to be increased by 66 percent on average? How
can we get consistency in capital requirements under Basel II?
What about the competitive impact on smaller institutions?
Those questions are important, but none are so fundamental
to the long-term cost of our Federal banking safety net as how
Basel II interacts with the leverage ratio.
QIS-4 supports earlier FDIC analysis showing that Basel II
is a much lower numerical capital standard than we have in
place today. For most insured banks, QIS-4 would require tier
one capital-to-asset ratios that are far less than what would
be allowed under current prompt corrective action regulations.
As we ask what to make of these QIS-4 results, it is
important to remember that the current Basel II framework is
not a scientifically pure method for measuring bank capital
needs but an imperfect framework crafted to reach consensus
among Basel Committee members. The U.S. banking system is among
the most well-capitalized in the world, and it is no surprise
that an internationally acceptable capital standard might not
be consistent with U.S. capitalization levels.
Putting aside these international comparisons, let me be
clear that the FDIC views the extremely low capital numbers
coming out of Basel II formulas as evidence that changes must
be necessary going forward. We view the QIS-4 results as
examples of why, under Basel II, the leverage ratio will play a
more important role than ever in ensuring the soundness of our
banking system.
I believe my colleagues at the table share our discomfort
with the low levels of capital indicated by QIS-4 results. They
have noted that the leverage ratio would rightly prevent banks
from lowering their capital that much. They also share our hope
that experience in the Basel II transition years would help us
pinpoint and correct the aspects of the framework that give
rise to troubling QIS-4 results.
Yet, beyond those hopes for the future, there is a great
conflict of expectations that is going to have to be better
understood and resolved. The agencies' plan allows banks to
reduce their risk-based capital requirements over 3 years by 5
percent, then 10 percent, then 15 percent, after which there is
no floor or risk-based capital requirements.
This phase-out of floors dangles the prospect before banks
of much lower overall capital requirements. Yet, most banks in
the QIS-4 would continue to have overall capital requirements
determined by the 5 percent leverage ratio. Therefore, any bank
that thinks Basel II will not be worth the effort without the
elimination or significant reduction in the leverage ratio
needs to make its views known.
With the PCA standards in place since the early 1990's,
U.S. banks have faced more conservative capital regulations
than most of their international competitors. This same time
period has been one of historic profitability for large U.S.
banks, and I do not believe they are suffering under a
disadvantage.
There can be little doubt, however, that with Basel II
requiring half or less the core capital as does our current 5
percent leverage standard for many insured banks, Members of
Congress will hear from our industry about overly conservative
regulators and the competitiveness of large U.S. banks.
Those are fundamental questions that cannot be answered by
looking at the output of a formula. They will require a
different kind of policy debate than we have had so far about
the scope of the Federal safety net and the role of bank
capital in international competition. Since Basel II was never
advertised as an effort to reduce capital systemwide,
discussions of the new framework have never revolved around
those issues.
The regulators today appear generally of one mind about the
future of bank capital regulation in the United States and on
this basis have agreed to proceed with Basel II. As I have said
today, there is more work to be done. I hope and believe the
future leaders of the agencies will continue to insist on the
best of both worlds: Sophisticated bank risk management and a
clear-cut minimum of regulatory capital to protect against
future banking crises. Thank you.
Chairman Shelby. Thank you, Chairman Powell.
Mr. Reich.
STATEMENT OF JOHN M. REICH
DIRECTOR, OFFICE OF THRIFT SUPERVISION
Director Reich. Thank you, Chairman Shelby and Senator
Hagel. I appreciate the opportunity to be here today to present
views similar to that which you have heard expressed, the views
of the Office of Thrift Supervision on the development of the
Basel II capital framework in the United States along with the
parallel modernization of current risk-based capital standards
under Basel I.
We are more than 2 years from the start of a proposed 4-
year phase-in of Basel II. There are significant hurdles to
overcome before we can represent to you, Mr. Chairman, that it
will be ready to be implemented on a permanent basis. Some
suggest that Basel II is not necessary; that we simply need to
update our existing capital rules to accommodate advances and
changes in the banking system since the 1988 Basel I Capital
Accord. We are reviewing our current rules, but it is important
for us to work with all of our institutions to determine what
is most important and appropriate for them.
For the largest and most sophisticated banks and especially
those that operate internationally, Basel II may be the most
appropriate capital framework. For other institutions, either
the current Basel I rules or a modified Basel I framework may
be most appropriate.
In an advance notice of proposed rulemaking issued last
month, we raised this issue, and we welcomed comments on the
ANPR to guide us on how best to update our existing Basel I
rules, including whether the status quo should be retained as
an option. OTS supports the goals and objectives of Basel II,
and we are committed to implementing a more risk sensitive
capital framework for all of our regulated institutions.
While it is important that the United States continue to
move forward on Basel II, we should proceed in a cautious,
well-studied and deliberate manner. The revised timeframe for
Basel II is consistent with this goal. However, it is critical
that all interested parties, including the industry, the
Congress, and the regulators, that we continue an active, open,
and thorough dialogue regarding Basel II. We will continue to
work with Members of this Committee, the other Federal banking
agencies and with our international colleagues on these issues.
Perhaps the most significant challenge that we face in this
process is maintaining competitiveness. While Basel II provides
an opportunity for our largest U.S. institutions to move to a
more logical, risk-based capital framework and to remain
competitive internationally, it is equally important to
identify ways to improve the risk sensitivity of existing Basel
I rules.
These objectives are not mutually exclusive but rather
mutually dependent in order to prevent potential competitive
inequities between Basel II and Basel I institutions. OTS is
pleased that Basel I modernization, an initiative that we have
advocated, has evolved into a commitment by all of the Federal
banking agencies. The goal of this initiative is to achieve
greater risk sensitivity without undue complexity.
We strongly support finding ways to improve existing Basel
I rules in conjunction with the implementation of Basel II if
not sooner. The results of the QIS-4 study suggested that Basel
II remains a work in progress in the United States. Consistent
with this message, the Federal banking agencies recently
announced a 1-year delay in the start of Basel II and an
additional phase-in year for its implementation. We also
provide for greater supervisory control over individual
institutions at each step of Basel II implementation and
graduated capital floors for several years prior to full
implementation in 2012.
Throughout implementation, institutions will be subject to
close supervisory scrutiny and a strict leverage ratio
requirement. Basel II does not include explicit capital
requirements for interest rate risk, an important issue for all
institutions but particularly for OTS-supervised institutions.
The banking industry currently has almost 40 percent of its
assets in residential mortgages and mortgage-related assets,
and this number is significantly higher for most OTS-regulated
institutions. Interest rate risk is especially important for
mortgage products, and one of the most important risks
confronting mortgage lenders must be addressed uniformly with
guidance from our Federal banking agencies on how to measure
and manage this risk.
Any discussion of Basel is incomplete without a discussion
of the interrelationship between leverage and risk-based
requirements. This issue has spawned a substantial amount of
dialogue about whether there should be a leverage requirement
at all which has, unfortunately, created more heat than light.
OTS supports a leverage ratio, which we view as a mainstay
of our regulatory system. As a final note, I would urge a point
previously made by my predecessors at the Office of Thrift
Supervision; that is, Mr. Chairman, to consider legislation
supporting OTS' representation on the Basel committee. I think
it is important that OTS' international role be formalized for
a number of reasons, not the least of which is the potential
impact of Basel II on the institutions and the holding
companies that we regulate.
Thank you.
Chairman Shelby. I thank all of you.
I touched on this in my opening remarks. One of the key
concerns about Basel II is its impact on the competitiveness of
small and medium-sized banks and how that could relate to our
economy, because as we all know, 75 to 80 percent of our job
creation in this country comes from small businesses, medium-
sized businesses, which deal with small banks in communities
from coast to coast, and I think it would be an unfortunate
result if instead of improving the capitalization of the U.S.
banking industry, Basel II instead created a regulatory
framework biased toward large financial institutions.
Basically, we have the most dynamic and healthy banking
systems here in the world, because we have a competitive
banking system, all the banks in this country. Since I believe
it is critical that Basel II does not dampen the
competitiveness in any way of the U.S. banking industry,
Governor Bies, I pose this to you: What type of analysis have
you conducted, and when I say you, the Fed, or do you plan to
conduct on the likely competitive effects of Basel II? In
particular, has any of your research indicated that Basel II
may hasten the trend of consolidation in the banking industry?
Is that one of the goals here?
Ms. Bies. Mr. Chairman, it is not one of our goals to
foster consolidation of the banking industry. We have completed
several studies of particular loan portfolios and mergers and
acquisitions, and let me talk about the loan portfolio piece
first.
Chairman Shelby. Okay.
Ms. Bies. When we looked at mortgages and small business
loans, we found there might be a small impact, and we propose,
in the Basel I amendments, to address these potential impacts.
We will get those comments to the advanced proposed rulemaking
that we just put out, and try to make sure that we follow
through with that in the final NPR in Basel I.
We have already had quite a bit of dialogue with the
community and regional banks, all of the agencies, in order to
come up with the ANPR, and we will address it. We have just
completed a study that was led by the Federal Reserve Bank in
Philadelphia around the credit card competitiveness, that is
being vetted by all the agencies now, and we hope to make that
public within the next few weeks.
In terms of mergers and acquisitions, one of the challenges
that I think of when in terms of capital and something that
people need to realize is that we have had a dramatic change in
the accounting for mergers in the past couple of years.
In the days of pooling, when you just added two banks
together, if the bigger bank had a higher amount of leverage,
you could carry it through. Now, we are in a world where every
acquisition is a purchase accounting transaction. What that
means is that when a large bank buys a small bank, on day two,
the large bank has no new capital, has all of the assets and in
our methodology risk exposures of the smaller bank and a
probably added good will, which reduces its regulatory capital.
Chairman Shelby. It is what you bought.
Ms. Bies. It is what you bought, and the purchase
accounting reflects that.
The only way the bank can even get to even or ahead is the
old fashioned way: Issue stock. And if they issue stock to the
selling shareholders, there will be an impact on capital. Or
they may go to the market, or they may use retained earnings.
But the only way any merger creates capital above what is used
in the merger is by issuing equity, and we think anytime you
issue equity, whether there is a merger or not, that is
capital.
Chairman Shelby. The Basel I-A release indicated that there
were competitiveness concerns that prompted the proposal to
revise existing capital requirements. Governor Bies, is there a
plan to conduct any analysis on how Basel I-A is likely to
impact the competitiveness of small and medium-sized banks
against large banks operating under Basel II?
Ms. Bies. Once we get closer to formal proposals in the
NPR, we will try to look at that more intensely. Now, one of
the challenges we are going to have is to differentiate between
minimum regulatory capital, which is at the heart of both of
these proposals, Basel I and II, versus what is held by the
banks for business purposes.
And we know for the vast majority of banks of all sizes
that they hold real capital way above what they need for
minimum regulatory purposes to demonstrate that they are a
highly rated institution by the financial markets. And so, we
will have to differentiate between those in our analysis, but
as we get closer to the proposal, we will be looking at that.
Chairman Shelby. By charging banks with the responsibility
for calculating their capital requirements, that is a little
different from what we do today, is it not, Mr. Reich?
Basel II effectively delegates regulatory authority in some
areas from bank regulators like yourselves to the banks
themselves. While bank regulators must approve a bank's method
for calculating its capital requirements, Basel II, as I
understand it, nevertheless transfers a lot of responsibility
for the safety and soundness of the banking system to the banks
themselves.
Chairman Powell, you are the Chairman of the FDIC. I want
to pose this question to you. Does Basel II rely too much on
the judgment of banks rather than regulators? Do banks have the
proper incentives to accurately calculate their own risks? And
was Basel II premised on the belief that there were competitive
pressures to lower capital requirements below what was required
for safety and soundness? In other words, how do we expect
banks to be able to ignore those pressures? Is Basel II founded
on a realistic incentive structure?
That is a lot of stuff, but you have spent a lot of time on
this, and you have a big portfolio on safety and soundness, you
all do.
Chairman Powell. I think, Mr. Chairman, that clearly, as we
have all said, because the nature of the balance sheets of
these large institutions has changed the risk profile has
changed. So, the way that we measure capital against those
risks needs to be modified. I support a more rigid risk-based
capital regime.
Having said that, you are correct, judgements will be based
upon the formulas and the input that institution management
will give to the regulators. But, remember that the regulators
have oversight and will test those formulas and will look at
not only the formulas themselves but will also look at the
overall capacity of management to manage their risk. But, that
will depend upon the will, very frankly, of the regulators to
enforce capital standards they believe properly reflect risk.
Going forward, I think clearly, there will be times that
the regulators will have to step in and say the formulas are
flawed. If the formulas do not produce the necessary capital
that we as regulators believe institutions should maintain, we
need the will and the courage of the regulators to step in.
Comptroller Dugan. Mr. Chairman.
Chairman Shelby. Yes, go ahead, Mr. Dugan.
Comptroller Dugan. I just wanted to add one point to that.
It is not the individual bank's model. It is our model. It is a
model that we have designed. The differences come in the inputs
to the model and what the institutions do with them. And, we do
have a responsibility to supervise that process, as we do now
with other types of inputs.
Chairman Shelby. And to supervise that process, you are
going to have very sophisticated personnel, are you not?
Because they are going to have them.
Comptroller Dugan. With regard to, the models themselves,
you do have to have very sophisticated quantitative experts,
which we do, and which we will expand over time. The
supervisory process to make sure that the inputs into that are
based on sound judgments that are valid does not take the same
kind of quantitative skills and plays more to the basic
fundamental, practical judgment of, in our case, national bank
examiners. We think we do have the skills to do that and
supervise that process.
Chairman Shelby. As the Comptroller of the Currency, you
will have the primary responsibility to supervise them.
Comptroller Dugan. That is right. Most of the Basel II
banks are national banks, and we will be on the front lines of
that supervisory process.
Chairman Shelby. Senator Sarbanes, I think he has an
opening statement.
STATEMENT OF SENATOR PAUL S. SARBANES
Senator Sarbanes. Have you finished your--
Chairman Shelby. Yes, I will come back. We will go another
round.
Senator Sarbanes. Mr. Chairman, first of all, I apologize
to the panel and to my colleagues for not being able to be here
at the outset. I will defer my questioning to Senator Hagel,
but I would like to briefly make an opening statement.
Chairman Shelby. Go ahead. You are recognized.
Senator Sarbanes. First of all, Mr. Chairman, I want to
commend you for calling this timely and important hearing on
the development of the new Basel Capital Accords. In fact, you
have had a continued commitment to the Committee's oversight
function with a focus, of course, on preserving the safety and
soundness of our banking system. There is a distinguished group
of witnesses, the regulators, who are here now and in the
second panel that is going to follow, and we are looking
forward to hearing from them. We have some very able people on
that panel as well.
Let me say right at the outset, I want to congratulate FDIC
Chairman Powell for undertaking the responsibility of serving
as coordinator of Federal support for the recovery and the
rebuilding of the Gulf Coast region. First of all, Chairman
Powell, we thank you for your tenure as the Chairman of the
FDIC. I think you have rendered fine service to the banking
system and to the country in that capacity.
There is even a deposit insurance bill moving its way
through the Congress. It went through this Committee quite
easily in the end after a lot of back toing and froing over the
years. But you are taking on a very critical and challenging
responsibility now in dealing with the aftermaths of Hurricanes
Katrina and Wilma, and I know that you have the best wishes of
all of us here on this panel, and if there is any way that we
can be helpful to you in that effort, we certainly want to do
it.
Chairman Powell. Thank you.
Senator Sarbanes. But thank you very much for your service.
With respect to the subject at hand, I must say, Mr.
Chairman, this is an important and timely hearing. The last
quantitative impact study, QIS-4, indicated the possibility
that Basel II will result in a large scale reduction in bank
capital. Now, Basel I resulted in an increase in bank capital.
Half the institutions which participated in the study had a
reduction of 25 percent or more. One institution had a 50
percent capital reduction. I do not see how one can avoid the
conclusion that a reduction of capital on this scale would
represent a major change in the safety and soundness status of
our Nation's banking system.
Second, the Basel II framework is extremely complex. It
will require banks and regulators to create and operate many
highly technical financial models. The sheer complexity of
these models raises concerns of how many people will be able to
operate, understand, and critique the models and their results.
Furthermore, if the largest banks, which are to be eligible
for the Basel II framework, receive large capital reductions,
it will create enormous pressure to reduce capital for all of
the other banks for reasons of competitive equity. In addition,
if risk-based capital is reduced significantly below the
leverage ratio for the largest banks as a result of Basel II,
it could create great pressure to weaken or eliminate the
leverage ratio.
We must recall that our banks are currently very well-
capitalized. They are very well-capitalized; they are very
profitable. It has really been a good performance for banks, in
the general. As a result, it has been a long time since we have
faced any significant financial crisis. So you base these
models on recent experience. They may not be sufficient for a
period of stress in the banking system.
Moreover, Basel II is premised on allowing the banks
themselves to determine their own capital requirements. I mean,
this premise raises a host of regulatory concerns. I mean, it
is like someone said; I was a teacher; if the students could
set the exam, they all would have gotten A's.
So, this is an important hearing. I have a very real
concern that this thing is just moving down the path, that
there is a lot of pressure from the other countries that are a
part of Basel II to get this thing into place and so forth and
so on. But generally speaking, the pressure for strong safety
and soundness standards and everything has come from the United
States, I think it is fair to say.
And I just think--I know now you had an October 3 meeting,
I think, of the Basel II, and before the October 3 meeting, you
put out--did you go for--I will ask it in the question period,
but my sense of it is this thing is just moving along. I know
the Fed has had a long, vested interest in this Basel II, and
presumably, they go to meetings, and all their fellow central
bankers say, when are you going to do this thing and so forth
and so on?
But we cannot let that type of dynamic determine what the
result is going to be, and if we just keep moving along without
solving some of these problems that I raised, someday,
everything may well come to a stop. And then, your problem of
your relationships will be much more severe than I think they
are right now. So, I just put that as a--and I will come back
to some of these in the question period, and I thank the
indulgence of my colleague.
Chairman Shelby. Thank you, Senator Sarbanes. I think your
observations are well taken right here.
Senator Hagel, thank you for your indulgence.
STATEMENT OF SENATOR CHUCK HAGEL
Senator Hagel. Mr. Chairman, thank you.
Welcome. If I could, I would like to pick up where Senator
Sarbanes ended and without trespassing on his set of concerns
and questions, but this will warm the panel up when you get to
Senator Sarbanes.
[Laughter.]
And by the way, Chairman Powell, congratulations on your
new gainful employment.
Safety and soundness, what are your concerns about what
Senator Sarbanes talked about? Do you have any? This is
somewhat of a departure, and I would like to hear from each of
you about the issue.
Mr. Dugan.
Comptroller Dugan. Senator, as I mentioned in my opening
statement, I think safety and soundness is why we should go
forward. I recognize very much the concerns that Senator
Sarbanes raised about the results of QIS-4. Frankly, if those
were the results that popped out at the end of the process,
they would be unacceptable. They are unacceptable. The drop in
capital is too great. The dispersion in results among similarly
situated institutions is too wide.
It was staring at those results that forced the regulators
to get together and say, what does this mean? Should we go
forward, or should we readjust? Should we stop now? What we
came up with is a system that (A) recognized that those results
were based on a very preliminary read of QIS-4, which, by
definition, could not be accurate because the system had not
been built and (B) that we needed to put the system in place
and supervise the system and see it in operation. Only then
would we know exactly what the outputs would be from the
capital process.
It is our belief that that process, where we have a better
specified system reflecting the final rule that banks have to
comply with and banks are subject to rigorous supervision, will
address some of those issues. But it probably won't address all
of those issues, and we are going to need to adjust the rules
once we see the process in operation.
And so, what we agreed on was the need for a longer and
controlled transition period where banks cannot drop their
capital to impermissibly low levels. During that period, we
need to rigorously supervise the system and see what it does,
and only when we are comfortable during that carefully
controlled environment, and after we make changes to address
problems that we see, do we allow the system to go forward. It
is our belief that that will address the problems raised by
QIS-4.
But to come back to your fundamental point, we believe that
we need to go in a direction to get our arms around the risks
that our largest, most complex banks are likely to take in the
coming years. We at the OCC have three institutions with assets
that exceed $1 trillion each. They take risks that are
different and more complicated than the risks taken by smaller
institutions. They have very complex risk management systems
already in place to address those risks, which we believe is
appropriate. And we believe that we will gain a great deal by
having a common framework across these banks to encourage them
to develop state-of-the-art risk management systems that we can
supervise, and it will increase the safety and soundness of the
system.
As Albert Einstein said, everything should be made as
simple as possible--but not simpler. We believe that you have
to address the complexity of the risk with a system that is
adequate to do the task. We do not have that now. We need to
make changes to move institutions in that direction, and we
think this is the right way forward.
Senator Hagel. That is the Shelby formula of success, the
Einstein.
[Laughter.]
Chairman Shelby. Absolutely. We try to follow it, but we
get off track, do we not?
Senator Hagel. Occasionally. That is why we get up in the
morning.
Chairman Shelby. But these people here probably understand
the Einstein method.
Senator Hagel. They do.
Chairman Shelby. We hope they do.
Senator Hagel. Governor Bies.
Ms. Bies. Thank you, Senator Hagel. Let me just add a
couple more thoughts to what the Comptroller just said, and I
agree with his comments.
One of the things I think really came out in QIS-4 to
support why we need to move ahead for safety and soundness
reasons on the new framework of Basel II are the issues we
raised; for example, similar credits being graded differently
as input into the models by different banks have been there for
ages.
What happened is that the transparency of having the banks
look at their models and risks in a similar way, having us look
at them in a horizontal way comparing them brought more
transparency to light about the differences in the ways the
banks actually manage credits and grade individual facilities.
So that additional transparency is also going to be an
important new tool to help us do our job better, to be able to
identify the outliers or the banks that are not following sound
practice.
The other thing the effort has done and one aspect that
came out in July on an international basis and that will be in
our NPR is that as we have said, there are a lot more
sophisticated financial instruments out there that the banks
have exposure to. In July, we adopted a new framework
internationally, and we will tailor it in the United States
that really recognizes that the nature of risk in the trading
book of banks as opposed to the book of loans, has changed
significantly from what is in the Basel I market risk
framework.
And it is grossly inadequate for measuring the risk of the
kinds of positions that are sitting in trading accounts today.
If we look at what is sitting in the trading accounts, we are
seeing not the traditional trading activity where bankers buy
and sell securities and derivatives all during the day. Instead
of that, we are seeing about 15 percent on average of the
assets being comprised of illiquid, high risk positions in
sophisticated financial instruments.
Many of these are engineered financial instruments, where
they take something that starts out as a mortgage loan, goes
into a mortgage-backed security, gets tranched into CDO's, and
they are holding equity type risk positions sitting in a
trading account and not getting the capital treatment for that
high risk.
That kind of sophisticated approach, we cannot put within a
Basel I framework, because the normal small community bank or
regional bank just does not engage in those types of
activities. But as we know from derivatives and complex
transactions, they can blow up very quickly. We need more
capital. We need a better measure to look at that.
And then, the other thing that Basel II does, which is
increasingly important for these banks, and if you look at the
banks' earnings and shocks to their earnings in the last few
years, bankers in these large organizations earn more and more
of the revenue through activities that are related to doing a
transaction, processing various forms of payments and
transactions for customers, and none of that is reflected on
the balance sheet.
The balance sheet has a very small asset for fixed assets
or what happens to be there at the end of the day, but there
are millions of transactions running through there. And when
there is an operational glitch, all of a sudden there is a big
earnings hit potential, either through the form of a model
failure or compliance risk that entails large penalties.
Operational risk in QIS-4, and this is still being
developed in the databases, added 10 percent to capital, above
what the banks would have if they stayed under Basel I. So
those kinds of changes in these big institutions are why we
need more effective tools to make sure we have capital to
address these changing risks.
Senator Hagel. Thank you.
Chairman Powell.
Chairman Powell. Senator, Basel II is a good thing if we
keep the leverage ratio. I believe it will bring better risk
management processes and better data for supervisors to
identify emerging risks as you just identified. But without the
leverage ratio, Basel II would not be a good thing.
Senator Hagel. Thank you.
Mr. Reich.
Director Reich. Senator, I think safety and soundness is at
the top of each of our lists of priorities. It certainly is at
the top of my list, and I believe it is for each of the four
people at this table.
Our first guiding principle should be to do no harm to the
industry today. And I think that the process which currently
exists, calling for a parallel run in 2008, with capital floors
graduated in 2009, 2010, 2011, full implementation delayed
until 2012, and then, only then, with the approval of the
primary supervisor of each of the Basel II banks, plus the
continuation of the leverage ratio should provide us with
essentially a 6-year period of time to get it right and to make
any changes that are needed to be made during that period of
time.
We expect to come out with an NPR early next year. That
will not be the final word, in my opinion, on Basel II. As we
go through a 6-year period of testing and observation, I think
we are committed to make whatever changes are necessary to make
sure we get it right.
Senator Hagel. Thank you.
Chairman Shelby. Senator Sarbanes.
Senator Sarbanes. Thank you, Mr. Chairman. I have a very
deep concern that we are just marching along even though, you
know, the QIS had terrible results.
Everyone says, well, we are going to fix it. Just leave it
to us; we are going to fix it. And you keep moving down the
path. Then, we are going to have a transition period. But of
course, at the end of that transition period, you are going
down what? Ninety-five, 90 and 85 percent, I think, so, we are
still moving along the same path.
I think you need to tell your international partners there
are serious problems here. You probably should say to them
look: The Congress may nix this thing altogether. We have to go
back and do some careful rethinking. These consequences that
were shown in the QIS are unacceptable. Let me ask you: Does
anyone at the table today think the U.S. banking system is
overcapitalized? I see everyone shaking their head no. Ms.
Bies, you did not shake your head one way or the other, so I am
kind of interested.
Ms. Bies. I am trying to decide if you are focusing on
minimum regulatory capital or the real capital that banks hold,
which is determined by the marketplace, and I think that is
appropriate. The total capital that the banks hold in the
system is appropriate for the risks they hold today, yes.
Senator Sarbanes. During this process of negotiating and
implementing the Basel Capital Accords, one of the concerns
regularly raised in the Congress is that the minimum capital
requirements on federally insured banks should be preserved.
The regulators, including the Fed, which I understand is the
lead agency in this negotiation; is that correct?
Ms. Bies. We have tried to work all of this through in a
joint way for the United States, and we are working jointly
toward an NPR.
Senator Sarbanes. What do the regulators in the Fed think
about all of this, not the economists but the regulators within
the Fed system? What do they think about all of this? Are they
concerned by this lowering of capital?
Ms. Bies. I think all of us were concerned about the
results of the QIS-4, and that is why we all, field examiners,
economists, as well as the regulatory staff all felt we needed
to do a time out and find out why we got the results that we
did.
We have engaged in the whole process through the Fed, both
the participation of people through the various Reserve Banks
who are the examiners on these complex institutions and have
had very good participation and really look through and analyze
the data, and we have done this with the other agencies. So we
are concerned about it, and some of the things that show up in
the NPR based on today's information we hope will address some
of those areas.
But as has been said by some of my colleagues, one of the
challenges here is we have to give the banks enough definition
of where they need to enhance their models so they can get on
with it and deliver the information so that we can recalibrate.
We will revise this. Basel I has been amended 25 times already
before this proposal. We are not going to go 6 years on Basel
II without an amendment to whatever we put out in this NPR
coming out early next year.
Senator Sarbanes. Yes, but once you move to a NPR, then,
you have crossed an important threshold. Let me ask you this
question: We have been assured by the regulators that minimum
capital requirements would not be affected by Basel at various
hearings
before this Committee, correct? Then, Ms. Bies, I read in the
American Banker, which, of course, we follow closely, Ms. Bies
said the leverage ratio down the road has to disappear.
I would say to the industry if you work with us and be
patient, we understand the concerns about leverage ratios, and
as we get more confidence in the new risk-based approach, it
will be easier for us to move away from the leverage ratios. Is
it your view that leverage ratios have to disappear?
Ms. Bies. Senator, I did not choose my words well in that
meeting. I have tried to make amends to that Q and A that I
handled. In my written testimony here, my oral testimony, my
testimony in the House in May and every speech I have given on
capital since then, I have clearly stated that the leverage
ratio needs to stay in place. And in the context in which that
was answered, I wish I had answered it in a much more effective
manner. We are committed to keeping the leverage ratio in
place.
Senator Sarbanes. Was there a suggestion in that exchange
that the cost of implementing--in exchange for the costs of
implementing Basel II, the Fed would work to eliminate the
leverage ratios?
Ms. Bies. No, we are not working on eliminating the
leverage ratio. We have no staff working on that at all.
Senator Sarbanes. Where are the banks looking to get the
recompense for the costs involved in instituting these complex
models to do the risk-based capital calculations?
Ms. Bies. Senator, I think as prudential supervisors, what
Basel II effectively does is sends a message to the industry
that if they are going to be operating large organizations that
are difficult to manage in an enterprise-wide basis and deal
with sophisticated transactions, we expect a quantum leap in
the quality of their risk management, and this is part of it.
We want models that are used by the bankers as well as by the
regulators.
And so, this is part of the cost of doing business as part
of a large, complex organization. We feel it is necessary for
safety and soundness supervision purposes.
Senator Sarbanes. Would you say that the large financial
institutions are operating on the premise that there will be a
significant reduction in the amount of capital they have to
carry as a consequence of Basel II?
Ms. Bies. I think there may be a few, because they are
saying that today's capital framework, given their mix of
business that they choose to be in in the aggregate requires
more capital than they feel is appropriate. What we are trying
to do in this Basel II framework is make it sensitive to the
business mix of each of the firms, and the proof will be in the
results of these models.
Whether their expectations are met or not, we want to make
sure there is enough capital for their particular
organization's risk that is inherent in that bank. If we do not
feel that their risks are really low, their capital is not
going to drop at all. And so, what we have here is trying to
get better sensitivity, because the mix of the businesses are
very, very different across these large organizations.
So we are going to likely have more variability in minimum
regulatory capital than we have today, but for each
institution, it should be appropriate for the risks, and that
is why we are making them invest in these models, to
demonstrate how well they manage these risks.
Senator Sarbanes. I understand that under the QIS-4, a
significant number of the institutions that are currently
adequately or well-capitalized would have their status changed.
Their safety and soundness would lessen; that none of those
institutions are critically undercapitalized today, but some of
them would be under the application of the Basel II; is that
correct?
Ms. Bies. That is incorrect, Senator. QIS-4IV was a test
along the way of a process that is going to stretch forward for
6 more years. None of us, and we are unanimous, would have
qualified any one of the 26 banks who participated in QIS-4 to
move forward with Basel II. Their models, their data, are not
ready for it.
And so, right now, it is a work in progress. We are meeting
with each of the bankers this month so that we can talk to them
about what we observed in their submission in the exercise both
quantitatively as well as process-wise and emphasizing what
they need to do to try to qualify. None of them would have, and
that is why the results in and of themselves are really not
indicative of what will happen once the models and databases
are complete.
Senator Sarbanes. Well, you are talking about a changed
analysis. But on the existing analysis, what I stated is
correct, is it not? I mean, the answer you are giving me is not
that what I stated is incorrect but that, well, we are going to
change the criteria or the standards, and we are going to get a
different result. But I am looking at Chairman Powell's table
here. Have you seen this table?
Ms. Bies. Yes, sir.
Senator Sarbanes. What do you think about that?
Ms. Bies. What I am telling you is that I interpret the
information very differently. We had some banks who basically,
for some of the elements, did not have an estimate for an
extreme loss and plugged a zero because they did not have a
number.
Well, to put in a zero for the extreme losses, which is
where you need capital, we would not allow that to be used. But
in the model that they submitted, it is there as a zero, and
that is part of the reason these numbers went down. We would
not accept that. So to hypothetically say this is what the
results show, surely, I mean, mathematically, they may. But
what I am saying is we would never accept it. These banks would
never move to Basel II, any of them, based on the quality of
the information they gave us. We would not let anybody start
the transition based on what they gave us.
Senator Sarbanes. Well, my time is up. That answer seems to
me is like a ship passing in the night. It does not really come
head to head with what this problem is. It seems to me if you
are going to put out this thing, it then has to show a result
that is acceptable, and obviously, this does not come anywhere
close to doing it. But nevertheless, you are going ahead.
Ms. Bies. But the banks need more specific guidance from us
on what these expectations are so that they can move ahead to
complete their work. They all are still working very hard to
complete their models, and they need it so we can do a more
rigorous test.
Senator Sarbanes. My time has expired. I guess in the end
the question that should be put is, suppose you did not do
Basel II? What would be the bad result of that if you did not
do Basel II? We have good capital standards right now. The
banks are extremely profitable. We do not have this problem
with a competitive disadvantage. Anyhow, my time is up and----
Comptroller Dugan. Could I respond to that?
Chairman Shelby. Mr. Dugan wants to answer that.
Comptroller Dugan. That last question, Senator.
I will tell you I sat in this hearing room behind you
during the period in the late 1980's and through the changes
that caused banks to have incentives to raise their capital.
Chairman Shelby. The crisis.
Comptroller Dugan. That is right.
Senator Sarbanes. Scars you for life.
Comptroller Dugan. Yes, and we believe that banks have
raised capital appropriately, and it is a fundamental part of
our system. The results from QIS-4 were frankly totally
unacceptable. We cannot have a system that produces such
results.
If you ask, on the other hand, what would happen if we do
nothing, we think, our supervisors think, not just me, and we
supervise the largest of these institutions, that it is
absolutely critical that we have a sophisticated risk
management system and model conceptually like Basel II, a model
that we build, not the banks, to get our arms around the most
complex risks of these institutions, and that we have a process
that will allow us to address what came out of QIS-4 so that we
can fix it before these institutions can go live.
That is why we stretched out the transition. That is why we
required a much more carefully controlled environment where
cannot drop their capital, no matter what Basel II says, during
that period, and why we have anticipated we are going to have
to change this rule at some point in the future when we see
what it looks like, when we have built it to specifications,
when we have supervised it, and when we see what the results
are at that time.
We think that approach will address part of the problem,
but in the end, we are going to have to make some changes. But
for safety and soundness reasons, we think it is a
fundamentally good idea to go forward.
Senator Sarbanes. My time has expired, and I know the
Chairman has other questions. We have a very good panel coming.
I have looked at some of their statements. I suggest to this
panel that you give careful study to the submissions that are
coming to the Committee from the next panel we are going to
hear from, because they have raised some pointed questions
about the situation.
Thank you, Mr. Chairman.
Chairman Shelby. Thank you, Senator Sarbanes, for your
experience and your service here and your long-term memory from
many, many crises; we were both here together, probably the
only two left, maybe; Senator Dodd, from the crises. Mr. Dugan
was here and so forth. But I do not believe there is any
substitute for capital. I mean, you have to do it, and as
Senator Sarbanes said, our banks are doing very well. Our
economy is good.
But things will happen, and they will be challenged. They
always are. And I hope, as regulators, that when the system is
challenged that these models, whatever they are, will withstand
the challenge they sometimes will be in and sometimes they will
break. I hope whatever you do, you do it the right way. And we
have this responsibility to question this, as you well know.
We thank all of you for your service, and we thank you for
your participation here today.
Chairman Shelby. We will go to our second panel: Mr.
William Seidman, no stranger to the Banking Committee, who has
served this country very well; Mr. William Isaac, I would say
the same thing, former Chairman, now Chairman of Secura Group,
LLC; George Kaufman, Professor of Finance and Economics, Loyola
University, Chicago; Mr. Daniel Tarullo, Professor of Law,
Georgetown University Law Center; Ms. Katherine Wyatt, head of
Financial Service Research Unit, New York State Banking
Department.
As I said earlier, your written testimony of this panel
will be made part of the hearing record, and Mr. Seidman, we
will start with you, if you are ready to sum up any points that
you want to make. Your written testimony, all of you, it will
be made a part of the record. I know you have a tight schedule,
but we have stacked votes, too, later on. We are not going to
push you too hard, but we think this is a very important
follow-up panel from the first.
Proceed, and welcome again to the Committee, sir.
STATEMENT OF L. WILLIAM SEIDMAN, FORMER CHAIRMAN,
FEDERAL DEPOSIT INSURANCE CORPORATION;
FORMER CHAIRMAN, RESOLUTION TRUST CORPORATION
Mr. Seidman. Thank you very much, Chairman Shelby and
Senator Sarbanes. It is a great pleasure to be back here. I
hope you will realize it has been a long time, so I am probably
a little rusty and, you know, not quite as well-informed as I
used to be. So, I will just try to hit some high points based
primarily on my experience. I mean, you have heard some people
who are all up-to-date.
Senator Sarbanes. Seidman rusty is better than most people
well-oiled.
[Laughter.]
Mr. Seidman. Thank you, Senator. That is very kind of you.
First, I was there when the Basel I standards were created,
I was part of putting that together. And the idea was that
there were certain banks around the world that were
undercapitalized that were part of a global system, and the
whole effort was to get a minimum capital in the global system
so that if something happened, we would not have weak links.
So the idea was to have a uniform, simple standard which
could be applied worldwide, and I think essentially that was
accomplished. And it was a simple standard that people could
understand.
Basel II, of course, is not simple. Here are the regs for
Basel II right here. I can just about lift them up. That is the
regs for Basel II. They have been working on for 7 years, and
as you have heard----
Chairman Shelby. Does anybody understand them?
Mr. Seidman. Not me, I guarantee you that.
Chairman Shelby. If you do not understand them, I would be
nervous----
Mr. Seidman. I do not think anybody would be willing to
take a test on that right now.
Chairman Shelby. Go ahead.
Mr. Seidman. So our experience with minimum capital
standards and with capital standards in general was when there
is stress, they are too low. I can tell you that if capital
standards had been any lower than they were back in the 1980's,
our two largest banks would have failed, without question. They
were right on the edge.
So the idea that we would like to have less capital, and
let me just comment that do you remember Walter Wreston's
famous statement that big banks do not need capital? And that
is true until there is a crisis, because as long as you are
making money, they do not need capital. What they need it for
is the unexpected, and that is what I think we are all trying
to work on.
Our experience with actually using these models, the only
experience I am aware of in the United States is the GSE's,
where, as you know, it took 4 or 5 years to create a model, and
it was absolutely useless when it came to the problems they
had, because how are you going to model the improper
accounting? So if you had been using their model obviously did
not provide protection.
I thought it was interesting: Chairman Greenspan, who, as I
last remembered, was a part of the Fed. They asked him the last
thing, the last hearing what do you think about using models
for setting regulatory standards? And he said, well, I think it
is a good idea. And then, he said are you ready to do that
today? And he said no, the models are not ready. And then, they
said, well, when can we use them? He said it will be a very
long time before we--so even the Fed Chairman is pretty clearly
in doubts.
And of course, you have heard about the differences between
large banks and small banks, and I will not dwell on that. Let
me make clear: I think the models that are being made by these
banks are very useful. They are useful for current operations.
That is what they are designed to do. They are useful for
pricing. They are useful for a lot of things. They are a factor
in setting regulatory standards. But in no way that I can think
of can you build a model that would be the basis for absolute
standards of required capital.
I think there are some adjustments to Basel I necessary,
and we have I-A, and I think we have people well-evaluated, who
can evaluate that.
Finally, I would just like to quote from one of the learned
professors on the Berkeley faculty, and his basic quote is:
``Everyone knows when to discard his models. Of course, we
understand that even in normal times, the best model is just a
guide. If something extraordinary happens, like Russia's
problems or the stock market goes down 20 percent, anyone with
a modicum of common sense knows that models are not going to be
reliable guides.'' That's when you need the leverage ratios.
Thank you.
Chairman Shelby. Thank you, Mr. Seidman.
Mr. Isaac, welcome again to the Committee.
STATEMENT OF WILLIAM M. ISAAC, FORMER CHAIRMAN,
FEDERAL DEPOSIT INSURANCE CORPORATION;
CHAIRMAN, SECURA GROUP, LLC
Mr. Isaac. Mr. Chairman, Mr. Sarbanes, I appreciate the
opportunity to be here. This is a very important hearing. The
Basel II debate is probably one of the most profound debates we
are facing in the banking system right now.
As I explained in my written statement I have grave
reservations about Basel II, and I believe it carries the
potential to do enormous harm to our banking system, which is
the strongest, the most profitable, and the most innovative in
the world. No regulator, legislator, or banker who lived
through the banking crisis of the 1980's, which those of us who
are in this hearing right now did, can ever forget the lessons
of that period, and one of the most important is that capital
matters.
And I agree with Chairman Seidman, that we would have had a
lot more banks fail if we had had less capital in the system. I
do not believe that the U.S. banking system has too much
capital. You asked that question to the earlier panel. I would
state that in my opinion, the U.S. banking system does not have
too much capital. The rest of the world has too little capital,
and the United States should not go down to world standards.
That is not the problem. The problem is we need to try to find
ways to bring the rest of the world up to U.S. standards, and I
do not think we should play any games with that.
We have spent 25 years trying to build a great banking
system and increase the capital levels in this country. We have
done it, and we have the best system, and it is not time to try
to fix it. Basel I was developed in the 1980's to bring
uniformity around the world and also to try to calibrate the
capital ratios to the risks in the banks.
Despite its objectives, Basel II has done nothing to bring
about parity in capital standards around the world. The 20
largest banks in the world outside the United States at the end
of 2004 had a median capital to assets ratio of 3 percent.
That, I might add, is below the minimum standards required for
U.S. banks. That is the median. The median among the 20 largest
banks in the United States was 6 percent, so twice as high, and
yet, the U.S. banks had much higher returns on assets. The
argument that U.S. banks are disadvantaged in the international
marketplace is fallacious. We have the best banks. We have the
strongest banks. Our banks are taking the fewest risks, and
they are making the most money. And so, I do not see the
competitive inequity that exists.
When Basel I came into place, most of the career bank
regulators were skeptical about it, because when you said we
are going to give different risk weightings to different asset
classifications, they did not know what would happen to
capital. So they quite wisely put in place the leverage ratio
to make sure that capital standards did not go too low and they
also gave themselves the ability to override Basel I at any
point that they felt that they needed to. For example, in
subprime lending, the regulators are requiring a lot more
capital, risk weights above 100 percent on high risk lending,
something that they reserve the power to do despite what Basel
I may have said.
But almost since Basel I was adopted, the Committee at
Basel, at the Bank for International Settlements, started
agitating for a more sophisticated risk system for the more
complex banks. My major concerns about Basel II are first of
all that it will be based on inadequate and unreliable data. No
bank in this country has detailed data on losses that goes back
as much as 10 years. How can we build models when we do not
have data that takes us through various business cycles?
The past 10 years is the most profitable period in U.S.
banking history, and we cannot base models on that period.
Moreover, the banks that we are trying to get to build these
models did not exist 15 years ago for the most part, at least
not in close to their current form. Their names may have
existed, but those banks do not resemble anything like what
they did 15 years ago. They are amalgamations or mergers of
very disparate cultures, data systems, and business practices.
Even if we could go back 15 or 20 years with these banks, it is
not the same bank it used to be. And so, I am not sure how they
build models from that.
Another concern I have with Basel II is trying to get the
banks to model their operational risks. I only know of two
kinds of operational risks: Those that you can identify,
predict, price, and insure against--those have never been a
problem in our banking system, because you can predict them,
you can price them, and you can insure against them. Then there
are operational risks you cannot predict. I do not know how you
build a model around something you cannot predict. So, I think
the whole notion of creating a model for operational risks and
basing capital requirements on the model is flawed.
I think one of the things we need to be concerned about
with Basel II if it goes forward in anything approaching its
current form is that it might well-foster a complacency and a
false sense of security on the part of bank management, and on
the part of bank boards of directors, bank regulators, and
analysts. I am concerned that people get a false sense of
security from the models. They will have more confidence in
them than they deserve.
And you do not have to go back very far in history to find
out what happens when you do that. Long Term Capital in 1998
was a company that allowed, by all accounts, brilliant
mathematicians and Nobel Prize-winning economists to create
models and build a firm based on those models. They made some
huge bets, and they lost. It resulted in the Federal Reserve
forcing the banking system to come in and bail out the company.
I think that the argument that U.S. banks are at a
disadvantage vis-a-vis foreign banks is simply wrong. I heard
this argument in the 1970's and 1980's about the Japanese
banks--that they were going to rule the world. We hear very
little about the Japanese banks today except that they are a
real drag on the Japanese economy. They pursued growth with
reckless abandon. They did not have capital. They did not have
earnings. Their business model was not sustainable, and it
failed.
I do not know of a single professional bank supervisor,
someone who does it for a living, not part-time, I do not know
of a single professional bank supervisor who is enthusiastic
about Basel II, and I really do not know any bank CEO's who are
enthusiastic about it. Capital regulation, in my judgment,
should be simple and easily understood. It is foolhardy to
accept a capital regime that will be virtually impossible for
senior management, boards of directors, regulators, and market
participants to understand.
There is nothing wrong with the current capital regime in
this country that some relatively simple fixes to Basel I could
not cure. I think Basel II is a good exercise. I agree with Mr.
Dugan, for example, on the previous panel when he said it is
important for bank regulators and bank managements to get their
arms around different kinds of sophisticated risks. I agree 100
percent, but Basel II it should be a management tool and a
regulatory tool; we should not base our capital standards on
it.
The last thing in the world that U.S. regulators should do
is engage in a competition in laxity with supervisors in other
countries by lowering U.S. standards to international norms. We
have the best and strongest system. We should work hard to keep
it. Thank you.
Chairman Shelby. Thank you, Mr. Isaac.
Professor Kaufman.
STATEMENT OF GEORGE G. KAUFMAN, CO-CHAIR,
U.S. SHADOW FINANCIAL REGULATORY COMMITTEE;
AND JOHN F. SMITH, JR., PROFESSOR OF FINANCE
AND ECONOMICS, LOYOLA UNIVERSITY CHICAGO
Mr. Kaufman. Thank you, Mr. Chairman, Senator Sarbanes. It
is a pleasure to testify before this Committee on the public
policy implications for the health and safety of the banking
system and the U.S. macroeconomy of the proposed Basel II
capital standards. I will summarize my longer written
statement.
Basel and structured early intervention and prompt
corrective action in this country have different histories. The
capital standards constructed in Basel I in 1988 effectively
resembled guidelines at the time for best practices in bank
capital management; in particular, with respect to
incorporating credit risk exposure. In contrast, in the United
States, emphasis was not on developing best practices schemes
but on developing public policy measures to prevent a
reoccurrence of the large scale failures of thrift institutions
and commercial banks in the 1980's, which imposed high costs on
the insurance agencies and for thrift institutions also on the
taxpayers.
The structure was designed to turn troubled institutions
around before insolvency, primarily through recapitalization or
merger with healthier institutions, and failing that, as a last
resort, to legally close and resolve them at lowest cost to the
insurance agency and potentially also to taxpayers.
With its emphasis on risk-based capital, pillar one in
Basel II basically remains a best practices guide for internal
bank management and not a public policy instrument. While there
is substantial empirical evidence of an inverse relationship
between leverage ratios, which use total assets and not risk-
rated assets, and bank insolvency, there is no such evidence
between risk-based capital ratios and bank insolvencies.
In other industries, analysts do not compute risk-weighted
assets or risk-based capital ratios for individual firms. But
they do compute, and investors use, leverage ratios. Risk-
weighted assets are an inferior scaler to the total assets to
gauge how much capital is available to a bank before the value
of its total assets declines below the value of its liabilities
and it becomes insolvent.
Unfortunately, Basel II may be on the verge of causing
major mischief in the United States that could weaken financial
stability. It appears that the 4 percent risk-based tier one
capital requirement ratio can be achieved under Basel II for
many large banks intending to use the advance internal ratings
approach with lower capital than is currently required both
under Basel I and to be classified as an adequately capitalized
bank according to the 4 percent tier one leverage ratio.
Consequently, the leverage ratio is likely to become the
binding constraint for these banks and prevent a reduction in
required regulatory capital. The FDIC has concluded that,
``U.S. policymakers will be confronted with a choice between
ignoring the results of Basel II or substantially weakening the
PCA requirements.''
Although almost all U.S. banks currently maintain capital
ratios at well above the regulatory requirements--indeed, the
FDIC reported that bank equity capital ratios at mid-year 2005
climb to the highest levels since 1938, more than twice the
ratio required to be adequately capitalized--some large banks
appear to be lobbying U.S. regulators to lower the numerical
threshold capital leverage ratio to qualify as adequately
capitalized to below 4 percent, say, to 3.5 percent or lower.
Congress in FDICIA delegated to the appropriate Federal
regulators the setting of the numerical thresholds for all
tranches but the minimum critically undercapitalized closure
trigger of 2 percent of equity capital. Some large banks are
also arguing that the current leverage ratio requirements put
them at a disadvantage with their competitors in the rest of
the world, who are not subject to these ratios.
For U.S. regulators to cave in to such pressures would be a
big and costly mistake. Considerable evidence suggests that
even a 4 percent equity leverage ratio is lower than that
maintained by almost all domestic nonbank competitors of banks
who are not similarly regulated nor covered by a safety net.
When industry leverage ratios fall below 6 percent capital,
bank failures increase, particularly in a recession.
Indeed, on the whole, there is a negative relationship
between leverage ratio and defaults in all industries. With
respect to individual large banks, there is no evidence either
that equity capital increases the bank's overall cost of funds
or that there is an inverse relationship between bank capital
ratios and bank returns on either assets or equity.
A time series analysis for U.S. banks shows a weak positive
relationship between bank capital and profitability. A cursory
cross-section analysis of the world's largest banks also shows
a positive relationship between capital and profitability since
the 1980's. Recently, United States, United Kingdom,
Australian, and Spanish banks have both high capital ratios and
high profitability, while German, Swiss, and Japanese banks
have both low capital ratios and low profitability, although
some adjustment need be made for the possibility of
simultaneity in the direction of causation.
Nevertheless, this helps to explain why capital ratios
actually maintained by U.S. banks are considerably higher than
the regulatory requirements. They are signalling strength to
their customers. Likewise, among the largest 1,000 banks in the
world in 2003, U.S. banks accounted for only 15 percent of
aggregate assets but 22 percent of aggregate tier one capital
and fully 37 percent of aggregate pretax profits.
To summarize, adoption of Basel II for large banks in the
United States is likely to have little effect on the banks and
the economy if the current numerical threshold values for the
leverage ratios for adequately and well-capitalized banks are
not reduced. It would do little damage. If, however, because
the new Basel II risk-based requirements can be met, on
average, with lower leverage ratio numbers, the numerical
definitions for adequately and well-capitalized banks and PCA
were reduced, there are likely to be longer-term adverse
consequences. The integrity of prompt corrective action should
not be compromised for the sake of harmonizing bank capital
standards across countries.
But the Basel process has not been totally negative. It has
greatly improved the measurement and management of risk by both
bankers and regulators and thus enhanced financial stability
worldwide. Basel should be maintained as an ongoing process to
develop ever-better bank best practice schemes for internal
management purposes. But it should not be halted and put in
place in the United States. It is the process, not the end
result, that will provide the major benefits.
Thank you.
Chairman Shelby. Thank you, Professor.
Mr. Seidman. Mr. Chairman, I am sorry; I apologize.
Chairman Shelby. Mr. Seidman.
Mr. Seidman. I have to leave. But I agree totally with my
predecessor, Mr. Isaac, so anything that he says, I subscribe
to.
Chairman Shelby. You are going to associate with him. Well,
we appreciate you. We appreciate your appearance. We know your
time constraint.
Mr. Seidman. Sir, thank you very much.
Senator Sarbanes. Thanks very much for coming. It is nice
to have you back before the Committee.
Mr. Seidman. Thank you.
Chairman Shelby. Mr. Tarullo, am I pronouncing your name,
right, Tarullo?
Mr. Tarullo. Tarullo, correct.
Chairman Shelby. Thank you.
STATEMENT OF DANIEL K. TARULLO
PROFESSOR, GEORGETOWN UNIVERSITY LAW CENTER
Mr. Tarullo. Thank you, Mr. Chairman, Senator Sarbanes. I
think you have heard already this morning that there is a lot
of common ground, the common ground being that there needs to
be better risk management in banks; there needs to be a better
capacity in our bank supervisors to understand the risks that
banks are assuming; and that Basel I is admittedly imperfect
and becoming increasingly so.
But those facts do not translate into a policy conclusion
that we should adopt Basel II. And it seems to me, looking at
this as a former policymaker, there are two kinds of questions
that we and you should be asking. First, are we ready to move
ahead to Basel II, given our current state of knowledge and the
current conditions in the world? Second, will we ever be in the
position where Basel II is, on balance, the right approach to
bank capital regulation?
I am still somewhat agnostic on the second question,
although skeptical for a lot of the reasons that you and my
colleagues on the panel have articulated. I am very clear on
the first question: We are not ready to move forward; the bank
supervisory agencies are not ready to move forward; there are
too many unanswered questions.
One thing that I think academics and policymakers should
have in common is an appreciation of the law of unintended
consequences. When you take a regulatory or other action, some
things you anticipate, many things you do not. Some of those
unanticipated things can be very harmful indeed--witness our
history with the savings and loan crisis and the Latin American
debt crisis, both of which had some of their roots in prior
regulatory actions which, in retrospect, were good for the
economy but where the consequences had not been anticipated and
guarded against.
So, again, thinking as a former policymaker, how do I
understand what the agencies are proposing right now? Well,
they are saying first, we want to make the biggest change in
bank regulation in about 15 years. Second, we have tried to
figure out what the effect will be on bank capital. We are not
really sure. In fact, every time we do one of these tests, the
results surprise us. But we have now concluded the only way we
can really figure it out is to go ahead and implement the new
process, the new rules, even though we really admit we do not
understand what will happen.
Third, to the extent we do understand, there seems to be a
momentum toward capital reduction. And the agencies have not,
in my view, and I do not think they did this morning either,
articulated their theory of where capital needs to be and why.
This is not a science. There can be and should be
sophisticated quantitative aids to capital regulation. But
ultimately, regulatory capital levels are a judgment about the
amount of safety and soundness, the amount of bank stability
versus the constraints on banks moving money out the door to
more productive uses. That is a policy judgment. The market
needs to make its judgments as to how much risk it wants to
bear in lending to banks as well. So I think these things ought
not to be confused.
Finally, as you look at all this as a policymaker, you have
to say, okay, what are the costs and benefits we are going to
get immediately? Why do we have to move so quickly? I think Mr.
Dugan is absolutely right: Over the medium to long-term, he
needs to get his arms around what is going on in banks. But if
nothing happens tomorrow, we are not faced with a crisis. This
is not a circumstance in which people come to you as
legislators or to them as regulators and say ``look, we know we
do not understand everything that is going to happen, but we
have to move, because the bad things that are happening to us
are just too great.''
So, Mr. Chairman, Senator Sarbanes, I appended to my
testimony a list of just the core questions that I, as a former
policy coordinator, would have put to agencies proposing to
move forward and that I urge you as Members of the relevant
oversight Committee to put to the regulators. They may be able
to answer them. I genuinely mean that when I say they may be
able to answer them. But I think what has happened is there has
been this kind, as Senator Sarbanes was suggesting, a momentum
to finish. There is a kind of exhaustion. The regulators think
we have been doing this for 6 years; we are already looking at
another six. Let us say we are putting it in place right now,
get it off of the desk, and move on to the next set of
issues.''
That is where the role of a policy coordinator or a
Congress oversight committee comes in--to say we understand you
are tired of this; we understand it is putting a lot of strain;
we understand it is very difficult, but we still need to
understand what the consequences will be before we move
forward.
Thank you.
Chairman Shelby. Thank you. That is very articulate. I
believe that Senator Sarbanes and I are working together, and
we will on this issue. We have a lot of questions that we did
not have time to ask the regulators earlier this morning, and
some of your suggestions will be very helpful.
Ms. Wyatt, thank you for appearing here with us.
STATEMENT OF KATHERINE G. WYATT
HEAD, FINANCIAL SERVICES RESEARCH,
NEW YORK STATE BANKING DEPARTMENT
Ms. Wyatt. Good morning, Chairman Shelby, Senator Sarbanes,
and Members of the Committee. My name is Katherine Wyatt. I am
a Ph.D. mathematician and Head of the Financial Services
Research Unit at the New York State Banking Department. I have
followed the development of Basel II for the Banking Department
since 2000. I have studied the possible effects of the simpler
approaches under Basel II and worked with my Federal
counterparts in analyzing banks' implementation programs for
Basel II.
I appreciate the opportunity to testify today, because I am
concerned that adhering to the current timetable for
implementation that the agencies have proposed could lead to
far-reaching changes in how we measure capital without
sufficient understanding of the possible consequences.
The Federal agencies plan to issue the Basel II NPR in
early 2006, then to finish the rulemaking process for Basel II,
conduct a year of parallel run of current requirements and
Basel II, and have Basel II in effect all by January 1, 2009.
They also aim to have the domestic capital modifications for
non-Basel II banks or amended Basel I or Basel I-A in effect on
January 1, 2009, also.
I am afraid the pushing ahead to complete the rulemaking
process for two complex proposals in less than 3 years will not
allow time for essential review of either. Now, there are these
3 years of floors that they have talked about. However, banks
that adopt Basel II on January 1, 2009, will have already made
sizeable investments in systems, in data collection programs,
and I think it will be very difficult to make material changes
to the proposal after they become effective, and they are ready
to start.
I believe there are two large gaps in our understanding of
the impact of Basel II that must be addressed before we move to
a final rule for either Basel II or Basel I-A. First, we do not
know what the actual level of capital will be under Basel II
for any given bank. You have heard already this morning about
the results from QIS-4. They are crude. No one is willing to
say that this is what the level of capital will be or will not
be.
Also, Basel II has changed over time. It was 2\1/2\ years
ago that the ANPR was released. There has been nothing public
since then. Now, the timetable seems to involve going ahead
without another impact study of the fully specified proposal
after the NPR. Even more importantly, since capital
requirements under Basel II are based on the outcomes of
mathematical models, we need time to develop rigorous technical
guidelines for parameter estimation and tests of data
sufficiency to ensure that required levels of capital are
adequate.
Now, the second large gap in knowledge comes from the fact
that we have not addressed the changes that may be brought by
Basel II and Basel I-A across the banking system. Basel II's
impact has been studied primarily on large, complex banking
institutions. We have not fully studied the competitive effect
of Basel II on the close to 9,000 non-Basel II institutions in
the country.
It is essential that we study the effects of both Basel II
and the amended Basel I or Basel I-A that was released just
last month side-by-side before we go ahead.
I would like to speak first about banks' Basel II capital
calculations, if I could. Under Basel II, capital requirements
for credit exposures are based on the outcomes of a particular
mathematical model of default specified by supervisors. This
supervisory model is applied to complicated portfolios with a
host of adjustments, specifications, exclusions, and exceptions
that grew out of attempts to reconcile the model results with
existing international bank regulations. As I said, we will not
even know what the final U.S. version of these specifications
will be until early next year.
Basel II banks provide their own estimates of probability
of default, loss given default, exposure at default and
maturity as inputs into the Basel II formulas. These parameter
estimates depend on the data and other models used by the bank.
Now, a key premise of Basel II is that banks will have enough
reliable data to produce rigorous results from the model. I
think many will agree with me that this is often not the case.
The schedule for implementation must allow enough time for
supervisors to work with bank models, to understand different
parameter estimation techniques, and to gauge the sufficiency
of bank data. Otherwise, there is a real danger that the
estimation techniques that most large banks choose will become
the de facto best practices.
Unfortunately, Basel II could be gamed by choosing the
modeling techniques and data sets that will produce the lowest
capital requirements. We have already seen what the QIS-4
results showed. The strong possibility exists also that as the
distance between risk-based capital requirements and current
leverage under prompt corrective action capital requirements
grows, there will be increased pressure on bank regulators to
drop the leverage ratio requirements.
The bank supervisors I have talked with are very worried at
the prospect of dropping PCA requirements. They remember other
times when banks' predictions about the future did not come
true. They also point to their experience that well-capitalized
banks are profitable banks, can enjoy lower costs of funding,
and can more easily weather economic downturns.
Second, we just do not know enough about the impact of
Basel II more broadly on the U.S. banking system. The
Conference of State Banking Supervisors described some widely
shared concerns about the impact of Basel II in a letter to the
Federal agencies in September. CSBS wrote: As proposed, Basel
II creates significant differences between capital requirements
of banks that adopt Basel II and those that do not. The current
approach reduces the capital large institutions hold for
mortgages and small business loans among other assets. In a
very practical sense, the reduced capital requirements would
provide a pricing advantage for the larger institutions.
In a competitive economy, eventually, market forces will
likely drive these assets from smaller banks toward the Basel
II adopting banks, requiring nonadopting banks, the vast
majority of which are small community banks, to move to higher
risk areas of banking.
I am afraid that the available analysis does not adequately
answer these concerns. The Federal Reserve has posted on its
website several white papers covering some of the competitive
issues raised by the original bifurcated regime proposed by the
agencies. However, these papers are based on Basel II circa
2003, and both the Basel Committee and the Federal agencies
have made changes since then. The new, amended Basel I
proposal, of course, is not considered at all in this research.
The white paper's suggests that the impact on non-Basel II
banks may be minimal, but these papers are not definitive, and
other authors have disagreed with their findings.
We need to have a much better understanding of the
consequences of Basel II before it is implemented. We should
take the time now, both Federal and State banking regulators,
to fully test the impact of Basel II and the amended Basel I
proposals. In this way, we can work to safeguard the soundness
and profitability of the banking system, and we can ensure that
U.S. borrowers will continue to have the access to capital that
a strong U.S. banking system affords them.
Thank you very much.
Chairman Shelby. Thank you, Ms. Wyatt.
Dr. Kaufman, in your testimony, you indicate, ``there is
substantial empirical evidence of a negative relationship
between leverage ratios and bank insolvency but no such
evidence between risk-based capital ratios and insolvency.''
Could you discuss in a little more detail, help us understand
why is this the case?
Mr. Kaufman. Because insolvency basically deals with when
the value of the assets falls below the value of liabilities.
It has nothing to do with whether the value of risk-based
assets falls below the value of liabilities. And so, you have a
factor there that does not seem to fit in logically into the
framework. Take a bank that holds nothing but very low credit
risk assets, it would have a very low capital requirement. But
if anything happens, if the models make a mistake, then, with
the low capital, you burn right through the capital. Capital is
a buffer that is to burn through, and the more capital you
have, the longer it takes to burn through.
Chairman Shelby. No sponge to absorb it.
Mr. Kaufman. Right.
Chairman Shelby. Have you done any work or thinking about
the likely, if there is going to be macroeconomic impact on the
U.S. economy of the adoption of Basel I and Basel I-A?
Mr. Kaufman. I have not focused on Basel I-A. As has been
said here, the small banks in particular hold so much more
capital than is required to be held, I do not think it will
make much of a difference.
Chairman Shelby. Mr. Tarullo, do you have any thoughts
there?
Mr. Tarullo. Senator, there have been quite a few academic
studies trying to project whether Basel II will exacerbate what
is called the procylicality of banking regulation, meaning that
when things are bad, the banks withdraw more money from the
system, because they have to ramp up their capital
requirements. And I would just say that at this point, whether
Basel II will increase procyclicality is another of the
unanswered questions. The academic studies done to date have
differed, and some of it differs based on, as Dr. Wyatt was
suggesting, which assumption you are ultimately going to stick
into the model.
Chairman Shelby. Just to the panel, do you believe the
adoption of Basel II and Basel I-A leave banks with sufficient
capital or would, Mr. Isaac?
Mr. Isaac. Well, that depends on what they come out with in
the end and whether they maintain the leverage ratio. It seems
to me that the whole exercise they are going through, a lot of
people expect that capital requirements are going to go down,
and that was the premise of the----
Chairman Shelby. Something Senator Sarbanes raised earlier
and rightly so.
Mr. Isaac. And I agree. We see people testifying and making
speeches that we have to do this to keep our banks competitive
with foreign banks that have much lower capital ratios. So, I
believe the premise that everybody is operating on is that
capital ratios are going to go down if Basel II goes forward.
Small banks say that is not fair to them, so the regulators
publish Basel I-A, to lower small bank capital ratios as well.
In other words, we fix the inequity by lowering capital ratios
across the board.
Chairman Shelby. Could we be playing with fire, though?
Mr. Isaac. I think it is a prescription for disaster, and I
am strongly opposed to lowering capital standards in this
country.
Chairman Shelby. Professor Kaufman.
Mr. Kaufman. I think it all hinges, as has been said a
number of times, whether you lower the numerical definitions of
adequately capitalized banks under the leverage ratio. If you
do not do anything there, then, it really does not matter what
comes out of the risk-based Basel II analysis.
The other factor you want to be sure about is that you make
it quite clear to the banks and to the country as a whole that
FDICIA makes it very difficult to protect uninsured depositors
in bank failures--FDICIA coming out of this Committee a number
of years ago--and, therefore, that the market better hold
enough capital regardless of what the regulatory requirements
are.
Chairman Shelby. Mr. Tarullo.
Mr. Tarullo. Mr. Chairman, I would echo two things Mr.
Isaac said: First, we cannot know right now, because these are
inchoate proposals. Second, there does seem to be a lot of
momentum toward lower capital, both in the larger banks, in the
smaller banks that fear the competitive consequences, and I
noted in Europe, where the adoption of Basel II in the European
Parliament was touted by European Union officials as lowering
capital requirements for European banks.
Chairman Shelby. Ms. Wyatt.
Ms. Wyatt. Chairman Shelby, first, I would just point out
that when I have looked at capital ratios, it has struck me
that already, for many very large institutions, the leverage
ratio is the constraining ratio. We are already in the
situation where banks are much closer to the leverage ratio
than they are to the risk-based capital one.
I think we just have to know more about the actual
mechanics of the Basel II programs that banks will have before
we can say, and until we do have some assurance about what is
going to come out of these Basel II systems, it would be much
better to stay with the current capital requirements.
Chairman Shelby. Competition in the banking community: We
have large banks, we have a lot of small banks. As I mentioned
earlier, and you well know, all of you, our economy's job
creation machine is small and medium-sized business. Mr. Isaac,
could this have an effect on competition, the banking industry,
and then get into our economy?
Mr. Isaac. I think that if the large bank capital standard
is significantly reduced, and we do not do something about the
smaller banks, that is going to have a big impact on small
banks around the country, because they are not going to be able
to be on the same footing competitively, and it is going to
hurt them.
And they tend to reach out and take more risk when they are
getting hurt competitively, which creates more problems in the
system. I do not think the answer is to lower the capital for
the big banks and then respond by lowering the capital for the
small banks. I think the answer is keep the minimum standards
where they are and use Basel II as a management tool and as a
regulatory tool. I actually favor developing these models,
because I think it will improve supervision and management of
banks, but I do not see the tie between that----
Chairman Shelby. Lowering the capital, though, should be
left out of it.
Mr. Isaac. Exactly.
Chairman Shelby. Professor Kaufman.
Mr. Kaufman. I do not see that as being a very important
issue. As I pointed out, the small banks in particular hold so
much more capital than their regulatory requirement now. If you
go back to the pre-FDIC days and the pre-Fed days, even, when
we did not have capital regulation, you see exactly the same
picture as we see now: That the small banks held considerably
higher capital than the larger banks. They competed with each
other. They both survived. It is just the nature of the beast
that when you are smaller, you concentrate your risks more, and
you are more likely to hold higher capital. So, I do not think
that the adoption of Basel II will have significant competitive
implications.
Chairman Shelby. Mr. Tarullo, do you have any comment?
Mr. Tarullo. No, I do not.
Ms. Wyatt. I think we do not know.
Chairman Shelby. A lot of things we do not know about this.
Ms. Wyatt. And I am sorry to keep repeating that, but the
problem is it is not just large banks and small banks. It is
not just large banks against small community banks. There are a
whole bunch of banks in between. These large regional banks,
clearly, it seems to me, are competing with the large banks.
And their capital ratios seem to be moving closer to the
capital ratios that the large banks keep.
Also, I think there are questions about the kinds of
lending that it will be more profitable for Basel II banks to
do.
And if they have already invested in a lot of systems, if
they have sophisticated data collection programs, it will be
easier for them, perhaps, to do a certain kind of retail
lending that can be done where the loans can be pooled or even
small business loans where you could look at a small business
loan in a retail sense rather than as a small corporate loan. I
think there will probably be changes in the ways banks do
lending because of this, and I think that is one of the things
we have to look at.
Mr. Isaac. Mr. Chairman, could I just add one little
thought to that?
Chairman Shelby. Yes, sir.
Mr. Isaac. I think the answer to what is going to happen
competitively is pretty clear from the fact that something like
the 10 largest banks are required to implement Basel II if it
goes forward. A significant number of banks below the top 10
are moving to implement Basel II at great expense and not
because they want to opt in to Basel II, but because they fear
they will not be able to compete against the top 10 banks if
they do not implement Basel II.
We do have a system in which everybody is competing against
everybody right now, unlike the system we had in the 1970's
where everybody carved up their territories. So, I think we
have to be very careful about the competitive effects of this.
We are going to have a bunch of banks opting into Basel II
because they want to take advantage of lower capital, and I do
not think they should have lower capital.
Chairman Shelby. Senator Sarbanes.
Senator Sarbanes. Thanks very much, Mr. Chairman.
This has been an extremely helpful panel. We appreciate
your appearing today, and also, we appreciate the careful
thought that obviously went into the prepared statements. Mr.
Chairman, we should send these prepared statements to each of
the regulators and urge them to review them very carefully. I
suggested that to them when they were here.
Chairman Shelby. I think that is a good idea. I think it
would be probably something that the two of us could work
together with our staff and propound a number of serious
questions to all of our regulators that we did not have time to
do today.
Senator Sarbanes. Right.
Chairman Shelby. We could work on that together.
Senator Sarbanes. Fine.
I know we are going to have a vote shortly. I want to lay
the basis for my question, and then, I will put it to you. This
QIS-4 study which they did, released earlier this year, showed
that for the 26 participating U.S. banks, the aggregate capital
level would drop 15.5 percent. The median capital reduction was
26 percent. In other words, half of the participating banks
would have capital reductions of 26 percent or more, and one of
the banks had a capital reduction of 50 percent. But we
continue to move down this path.
I mean, you get this kind of study result that just seems
to me should bring you up short about what is going on, but the
beat goes on. So now, the banking regulators have put out this
revised plan for implementation of Basel II framework. By its
very terms, they are going to put on a minimum 3-year
transition period, during which the agencies would apply limits
on the amount by which each institution's risk-based capital
could decline with the application of Basel II, so it is going
to go 95 percent, 90 percent, 85 percent over a 3-year
transition period.
And then, the institution's primary Federal supervisor
would assess the institution's readiness to operate under the
Basel II base capital rules. They would make a decision on the
termination of the floors, in other words, the 85 percent
floor, on an institution by institution basis. So what they are
putting out there envisions going to 85 percent of the current
risk-based capital levels. I mean, it is built right into the
proposal.
Now, I am trying to search for where the motive force or
the dynamic element is coming from to keep this thing going
when you get these kinds of results, when we have very able
testimony here, Mr. Chairman, Mr. Seidman, Mr. Isaac are very
experienced; Mr. Kaufman, a leading academic; Dan Tarullo, we
know, and of course, Ms. Wyatt, Diane Taylor is your
superintendent.
Ms. Wyatt. Yes, she is.
Senator Sarbanes. And you all run a good department there
in New York, the New York State Banking Department, with a lot
of very able and competent people, and you are one example of
that.
And so, why is all of this being ignored? Why do we
relentlessly keep moving forward down this path when people are
raising a lot of red flags about what the implications of this
are? Does anyone on the panel have a theory, a hypothesis, as
to what is the driving force behind all of this? I see they all
want to answer. Very good.
Mr. Isaac. I will just speculate a little bit, because I do
talk to regulators a lot, and I think I know what is on their
minds, at least in many cases. I believe this whole effort is
well-intentioned. We have banks that are huge and increasingly
complex, and they have all sorts of risks that we are not sure
that they understand. And so, there is a desire--I think
Comptroller Dugan expressed it--to get our arms around these
banks. We need to understand them better, these great big,
complex institutions.
So, I think that is the motivation. I think somebody made a
very strategic mistake 10 years ago or whenever they started
talking about this in the hallowed halls of Basel, when they
decided to tie the Basel II effort to capital and to induce the
big banks to do it by saying, ``You got a shot at lower capital
if we come up with a system.''
I think it would have been much better if the regulators
just simply said, ``We do not know enough about you, you do not
know enough about you, and we are going to insist that you
develop these models; no inducements here. Just do these
models. We know it is going to cost you money, but that is the
price of being an insured, regulated bank.''
And then, we can all see what we are dealing with over a
period of 10 or 15 years; we will perfect these models, and
maybe then, somehow, that will affect regulation and capital
requirements and the like. But we should not start there. We
should keep the system in place that has brought us the best
and strongest banking system in the world and mandate that
Basel II be implemented for purposes of management information
and regulatory information so we can figure these banks out
better.
I think they are very well-intentioned. I just think they
got off on the wrong foot, and it is hard to back off of it.
Mr. Kaufman. I would agree with what Mr. Isaac just said.
In my prepared statement, I differentiated between public
policy tools and best practices, and Basel II basically is a
best practice as is Basel I. Given that Europe does not have
the prompt corrective action system that we have in this
country, they are trying to substitute, without really having
thought it through, Basel II for prompt corrective action. That
is not going to work.
What we also have going is the Europeans pushing the
Americans at this point. We also have natural bureaucratic
momentum that is set up. You either stop it, which is difficult
to do, or you just keep revising it and keep going. I gave a
talk some time ago in which I suggested if the Basel group
really wanted to meet in a city named B, maybe they should meet
in Baghdad or Bogota. That might stop the process.
[Laughter.]
Chairman Shelby. That is a good answer.
Mr. Tarullo. I think there is a lot to what Mr. Isaac said,
and I would just add the following: I observed frequently when
again I was in the position of watching agencies come forward
with proposals that there is an early period where different
ideas can get some traction. At some point, though, the
bureaucratics of the agency become set on one idea. And at that
point, every problem that comes along is a problem to be dealt
with, to answer it; let us fix this, let us deal with this
constituency, but we are going down this track, and it becomes
very difficult for someone to get perspective on it.
And I think that is exacerbated for bank regulatory
matters, because as you saw this morning, you have four
agencies plus the Federal Reserve Bank of New York, which, of
course, originally was in the lead on this entire issue. All of
them have to work together. There is not continuity of
leadership over the course of the exercise. And so, it does
acquire a momentum of its own. And I think Mr. Isaac has put
his finger on maybe what gave the push to that momentum in the
first instance.
Senator Sarbanes. Is there a view from the State level
about this?
Ms. Wyatt. Yes; I have been very fortunate that I have been
able to be an observer at many meetings, Federal interagency
meetings on Basel II, and we are very involved in Basel II at
the State level. We have many banks that are interested in
Basel II, the larger regional banks. And I think that a lot of
very useful things have come out of the Basel II process.
I mean, it has been going on for many years. I think we
have learned a lot more about banks, and their portfolios have
started to address some of the complexities that are here. The
problem just is, though, in my opinion, we are not quite ready
to go to implementation. I think we need to have a fully
specified NPR, something with all the details filled in, and
step back, look at it, look at its effect, look at Basel I-A
NPR when it is out. Just take a moment to make sure, more than
a moment, but to be sure before we go to the next stage.
Senator Sarbanes. Is there not a very important dimension
that a whole group of banks--it will just work its way right
down the pyramid--will say to bank regulators, well, you
allowed this Basel II for these 26 or whatever it is large
banks. Now, they are holding less capital. But I am still
having to hold the old levels of capital.
That is unfair. I am at a competitive disadvantage. You
have allowed the most complex institutions to drop their
capital levels. Why should we not be allowed to do the same? If
you do not allow us to do the same or something comparable, we
are going to be at a significant competitive disadvantage, and
will that argument not continue to be asserted right down the
size chain?
And that is not really being analyzed here. I mean, this is
all the focus, what does it do to the 26. But what is the
spillover or the carryover effect of this on the rest of the
institutions? Is that a substantial problem?
Mr. Isaac. I think it is a very clear problem. I agree with
you completely: It is going to ripple through the system
because banks below the top 10 are going to opt into Basel II.
I want to make another point here because I have a special
concern, having run the FDIC during the banking crisis of the
1980's and handled the biggest bank failure in our history,
Continental Illinois. Continental was the eighth-largest bank
with assets of $40 billion. It stressed the agencies--the Fed
and the FDIC--and it stressed the banking system.
I do not know what we will do if one or two of these really
big ones that we have today gets in trouble, and I think it is
not a time to be fooling around with lowering capital ratios.
It just is not. We need to understand a lot more, which is why
they should keep on moving on Basel II as a management tool and
as a regulatory tool, but we should not, under any
circumstances, be lowering capital ratios in these banks in my
opinion.
Senator Sarbanes. Well, Gene Ludwig, who had been a former
Comptroller, made the observation just recently that, we were
not in a down cycle. He thinks we may have a down cycle in 24
to 36 months and says, you cannot run these models testing it
unless somehow you work into the model the down part of the
cycle. So, I mean, just to underscore the point are making.
Mr. Isaac. I agree.
Senator Sarbanes. Mr. Chairman, I see the vote has started.
This has been a very helpful panel.
Chairman Shelby. Senator Sarbanes, you have been here a
long time. You have chaired this Committee, and we both were
here when Chairman Isaac was going through the crisis. We went
through the crisis, and we do not want to go through another
one.
Mr. Isaac. I do not either.
Chairman Shelby. We appreciate all of your testimony, your
input, and we are going to look closely at some of the
questions Mr. Tarullo and others have suggested, because
Senator Sarbanes has a lot of questions to be asked.
Thank you so much. The hearing is adjourned.
[Whereupon, at 11:41 a.m., the hearing was adjourned.]
[Prepared statement and response to written questions
supplied for the record follow:]
STATEMENT OF JOHN C. DUGAN
Comptroller of the Currency
U.S. Office of the Comptroller of the Currency
November 10, 2005
Introduction
Chairman Shelby, Senator Sarbanes, and Members of the Committee, I
appreciate this opportunity to discuss plans of the U.S. banking
agencies to update and enhance our regulatory capital program in two
fundamental ways: First, through the implementation of the
``International Convergence of Capital Measurement and Capital
Standards: A Revised Framework,'' generally known as the Basel II
framework; and second, through revisions to our existing domestic risk-
based capital framework for banks not adopting Basel II, generally
known as Basel I-A.
The primary impetus for the agencies' work to revise existing risk-
based capital rules is to enhance the long-term safety and soundness of
our banking system. While the U.S. banking industry continues to
operate profitably, supervisors must ensure that regulatory capital
rules remain relevant and appropriately address existing and emerging
safety and soundness challenges. For our largest banks, the fundamental
thrust of our efforts has been to develop a more risk sensitive
regulatory capital system better suited to the complex operations and
activities of these institutions. For banks not adopting Basel II, our
primary goal is to increase the risk sensitivity of our risk-based
capital rules without unduly increasing regulatory burden. Work in
these areas is again moving forward as the result of agreement by the
agencies, announced in a joint statement on September 30.
The joint statement included a revised timeline for U.S.
implementation of Basel II and a series of prudential safeguards to
ensure that capital levels similar to those that exist in our largest
banks today will be maintained over an extended transition period. The
statement also highlighted our expectation that the rules implementing
Basel II in the United States will be modified as necessary based on
experience with the new framework during that transition period, and
before the prudential safeguards expire.
The joint statement reflected a consensus by all the U.S. agencies
that implementation of the Basel II framework should move forward. Our
agreement to do so was based on several key premises:
<bullet> First and foremost, the Basel II framework offers necessary
and appropriate improvements to address recognized flaws in the
existing risk-based capital regime for our largest, most complex
banks. Basel II will promote significant advances in risk
management that will benefit supervisors and banks alike and that
will enhance the safety and soundness regime under which the
largest institutions operate.
<bullet> Second, to achieve its intended purpose, the framework will
have to be thoroughly tested and almost certainly adjusted. The
recent quantitative impact study (QIS-4) of estimated Basel II
results in large U.S. banks produced significant dispersion of
results across institutions and portfolio types and suggested a
material reduction in aggregate minimum required capital. Apart
from the notice and comment process, however, additional agency
study of the Basel II framework itself will do little to resolve
those concerns. Indeed, without seeing live systems in operation--
and subjecting them to supervisory scrutiny--we will not be able to
gain the level of comfort we ultimately must have in order to rely
on Basel II for regulatory capital purposes.
<bullet> Third, it is our intention to proceed deliberately, gaining a
better understanding of the effects of Basel II on bank risk
management practices and capital levels. Upcoming Basel II
rulemakings, therefore, will include a meaningful transition period
during which we can observe and scrutinize Basel II systems while
strictly limiting, through a system of simple and conservative
capital floors, potential
reductions in capital requirements. Based on the experience we gain
through supervisory oversight in the transition period, we will
incorporate any necessary revisions to the U.S. Basel II-based
rules before the transition period ends.
Because we believe that regulations must be tailored to the size,
structure, complexity, and risk profile of banking institutions, we
expect mandatory application of Basel II to be limited to large complex
institutions. However, we need meaningful but simpler improvements in
our domestic risk-based capital rules for banks that will not be
subject to Basel II. Our Basel I-A initiative is separate from but
complementary to the Basel II rulemaking process, and it is important
that the public be able to compare, contrast, and comment on definitive
proposals for both Basel II and Basel I-A in similar timeframes. We
believe that overlapping comment periods for these two rulemakings is a
critical element of our on-going effort to assess the potential
competitive effects of these proposals on the U.S. financial services
industry.
On this basis, the banking agencies agreed that it is both prudent
and necessary to develop and issue a notice of proposed rulemaking
(NPR) for Basel II implementation and to solicit comments from the
public. In order to do this, however, prudential safeguards are an
absolute necessity, and we recognize that further changes will take
place through future rulemakings.
The Need for Basel II in the United States
The implementation of Basel II in the United States remains
controversial, requiring banks and supervisors to balance sometimes
conflicting objectives regarding complexity of minimum capital
requirements, regulatory burden, competitive equity, alignment of
regulatory capital with better measures of risk, and recognition of
marked improvements in risk management capabilities. A fair question,
and one we have asked ourselves at various stages of this process is,
``Given all the difficulties and uncertainties associated with Basel
II, why move forward with it at all?'' While other sections of my
testimony explain how we plan to go forward, I also understand the need
to address why.
The 1988 Basel Accord, also referred to as Basel I, established a
framework for risk-based capital adequacy standards that has now been
adopted by most banking authorities around the world. The U.S. agencies
have applied rules based on the 1988 Basel Accord to all U.S. insured
depository institutions. Although Basel I was instrumental in raising
capital levels across the industry in the United States and worldwide,
it became increasingly evident through the 1990's that there were
growing weaknesses in Basel I. In particular, the relatively simple
framework has become increasingly incompatible with the increased scope
and complexity of the banking activities of our largest banking
institutions. The crude risk-weighting mechanisms of Basel I bear
little resemblance to the complex risk profiles and risk management
strategies that larger banks are capable of pursuing. The
misspecification of risk under Basel I creates inappropriate incentives
and arbitrage opportunities that undermine supervisory objectives. And
dealing with outdated and mismatched regulatory requirements is costly
to banks.
In response to these issues, the Basel Committee commenced an
effort to move toward a more risk sensitive capital regime. As the OCC
has noted in earlier hearings, we firmly support the objectives of the
Basel Committee and believe that the advanced approaches of the Basel
II framework--the advanced internal ratings-based approach (IRB) for
credit risk and the advanced measurement approaches (AMA) for
operational risk--constitute a sound conceptual basis for the
development of a new regulatory capital regime for large
internationally active banks. In a system in which some individual
institutions hold well over $1 trillion in assets, the flaws of the
current, overly simplistic risk-based capital system cannot be seen as
merely superficial or inconvenient.
It is important to understand that the supervisory benefits of
Basel II are found not only in the increased risk sensitivity in
regulatory capital requirements, but also in the significantly improved
bank risk management systems required to generate them. As the front-
line supervisor for national banks, which hold nearly 70 percent of the
Nation's banking assets, the OCC stands to gain significantly from
implementation of those systems, not only from improved risk
sensitivity of regulatory capital ratios, but also from the wealth of
internal information and analyses that banks will provide us under
Basel II. Banks will be better informed about the risks they face, and
supervisors will have information about those risks from both an
individual institution and industry perspective. Large banks have
already made substantial investments in the development of Basel II
systems. Without further guidance and proposed rules, however, progress
toward Basel II standards will be severely limited.
While clearly secondary to U.S. safety and soundness concerns,
another important consideration is the need for internationally active
banks to have similar capital regimes in the jurisdictions in which
they operate. The benefits of global comparability in regulatory
capital are not limited to level playing field considerations. Moving
forward with Basel II also enhances internationally active banks'
ability to interact on a meaningful and consistent basis with various
supervisory authorities while improving how supervisors interact with
one another. Without a common framework, our ability to gain useful
information and cooperation from foreign supervisors would be severely
constrained. We are very much aware that differences in the
implementation details, including the timeline, can create significant
challenges for banks operating in multiple jurisdictions. While some of
these differences are
unavoidable, the OCC and the other U.S. banking agencies will continue
to work closely with foreign-based regulators to address these issues
as they arise. The implementation of Basel II will ultimately serve to
increase the dialogue and coordination among national supervisors and
to enhance the level of cross-border cooperation for our largest banks.
In short, the continued safety and soundness of our banking system
demands that we move away from the current simplistic system to one
that more closely aligns capital with risk. Put another way, doing
nothing to change capital requirements would over time threaten the
safety and soundness of the banking system, especially with regard to
our largest banks that engage in increasingly complex transactions and
operations and hold increasingly complex assets. In these largest
banks, more closely aligning regulatory capital and risk management
systems with actual risk is a conceptually sound and prudent way to
move forward. That is the fundamental purpose of Basel II, and while
the framework may require significant changes over time, it is moving
in the direction required by safety and soundness concerns.
That is the essential reason why I believe we should support the
Basel II approach.
QIS-4 Results and Analysis
In previous Congressional testimony, in Basel Committee
deliberations, and in discussions with the industry and other
supervisors, the OCC has repeatedly emphasized that reforms to our
regulatory and supervisory structure must be adopted in a prudent,
reflective manner, consistent with safety and soundness and the
continued competitive strength of the U.S. banking system. In
furtherance of those standards, the U.S. agencies conducted an
extensive quantitative impact study, QIS-4, in late 2004 and early
2005.
It is well-known that QIS-4 helped us identify significant issues
about Basel II implementation that have not been fully resolved. Even
subsequent to additional agency analysis, the QIS-4 submissions
evidenced both a material reduction in the aggregate minimum required
capital for the QIS-4 participant population and a significant
dispersion of results across institutions and portfolio types. One
measure produced by QIS-4 is the estimated change in ``effective
minimum required capital,'' which represents the change in capital
components, excluding reserves, required to meet the 8 percent minimum
total risk-based ratio. This measure is independent of the level of
capital actually held by institutions and of their currently measured
capital ratios. Aggregating over the QIS-4 participants, the decrease
in effective minimum required capital compared to existing standards
was 15 percent, with a median decrease of 26 percent. As noted above,
the additional QIS-4 analyses also confirmed that dispersion in
results--with respect to individual parameter estimates, portfolios,
and institutions--was much wider than we anticipated or than we can
readily explain. In particular, the agencies' additional analysis
revealed a wide dispersion of results between institutions with respect
to individual credit exposures and selected portfolios, even when
controlling for differences in risk.
The agencies are in the process of preparing a more detailed
summary of results of our follow-up analyses of QIS-4 for public
release and are now conducting meetings with participants to discuss
observations about their particular submissions. There are, however,
some broad observations I can make today about the apparent underlying
causes of the significant reductions and wide dispersions in capital
requirements in QIS-4:
<bullet> The single most important conclusion from our analysis is that
differences in results between banks and within portfolios
evidenced in QIS-4 submissions did not correspond directly to
identifiable differences in risk. Banks' current estimates of key
parameters in the IRB approach--probabilities of default, loss
given default, and exposure at default--fall well-short of the
level of reliability that will be necessary to allow supervisors to
accept those estimates for risk-based capital purposes.
<bullet> Closely related is the observation that institutions are still
at widely varying stages of development of the systems and
processes necessary to implement the Basel II framework. This
finding is not intended to be a criticism of bank implementation
efforts; banks have dedicated significant staff and budget
resources to Basel II. Rather, these development efforts have been
hindered by the absence of definitive rules or final guidance in
the United States. Consequently, the full impact of Basel II
implementation remains to be seen, as do potential ramifications
for the U.S. banking system.
<bullet> Basel II results appear to be materially influenced by the
prevailing economic cycle, which suggests significant fluctuations
in capital requirements under the framework over the course of
economic cycles.
In short, the QIS-4 results and the inevitable questions they raise
are sources of concern for the banking agencies. The process for
implementing Basel II as established in the September 30 joint
statement is designed to provide the OCC and other agencies a complete
understanding of the framework's implications for the banking system.
We have concluded that some of the weaknesses identified in QIS-4 are
attributable to the fact that no ``live'' Basel II systems have been
built--in large part because we have not yet fully specified all the
requirements for such a system. We also believe that certain of the
concerns identified in QIS-4 will only be fully understood and resolved
as the Basel II framework is implemented through a final rule, final
supervisory guidance, and rigorous examiner scrutiny.
The Need to See Systems in Operation
QIS-4 was a voluntary, ``best efforts'' undertaking by participant
banks. The actual implementation of Basel II systems will be preceded
by stringent qualification assessments and, assuming qualification,
will be subject to regular on-site review by examination staff and
other subject matter experts. We expect to see less dispersion in
results for similar risks as banks more fully develop IRB and AMA
compliant methodologies, supported by enhanced data systems and subject
to rigorous
ongoing supervisory oversight and disclosure requirements. We remain
convinced that supervisors and the industry will both eventually reap
significant rewards--in the form of better risk management and better
information about risk--when Basel II systems are built and operating.
It became apparent as we analyzed QIS-4 results that we have
reached a point where more study of the conceptual underpinnings of
Basel II will yield little additional practical benefit. Rather, the
questions that we as supervisors still have about Basel II--and there
are several that are extremely important--can only be answered by
continuing to move toward implementation. Given the obstacles that have
not yet been cleared, though, I firmly believe that the only
responsible way to do that is in a carefully controlled manner, with
strong safeguards, during a significant transition period to see the
systems in actual practice.
We see only one pragmatic solution to resolve this inherent
stalemate between our insistence on understanding the effects, and
allowing for and encouraging the development of systems that will allow
us to gain that understanding. That is to proceed with the next steps
of Basel II implementation, but with a series of prudential safeguards
in place until we can observe approved Basel II systems in actual
operation and subject them to supervisory scrutiny. Only then will it
be possible to judge whether Basel II is operating as intended and to
make adjustments as necessary to ensure that it does.
Transition Provisions
The revised implementation plan announced by the agencies on
September 30 includes several key elements that allow for the progress
we believe is necessary, over time, for risk management and supervisory
purposes, while strictly limiting reductions in risk-based capital
requirements that might otherwise result from systems that have not
been proven.
The first element is a one-year delay in initial implementation,
relative to the timeline specified by the Basel II framework. As a
result, the ``parallel run,'' which is the prequalification period
during which a bank operates IRB and AMA systems but does not derive
its regulatory capital requirements from them, will be in 2008. The
parallel run period, which will last at least four quarters but could
be longer for individual institutions, will provide the basis for the
OCC's initial qualification determination for national banks to use
Basel II for regulatory risk-based capital purposes. Following initial
qualification, a minimum 3-year transition period would apply during
which reductions in each bank's risk-based capital would be limited.
These limits would be implemented through floors on risk-based capital
that will be simpler in design and more conservative in effect than
those set forth in Basel II. For banks that plan to implement the Basel
II framework at the earliest allowable date in the United States, we
expect to propose the following timetable and transitional
arrangements:
------------------------------------------------------------------------
Year Transitional Arrangement
------------------------------------------------------------------------
2008 Parallel Run
------------------------------------------------------------------------
2009 95 percent floor
------------------------------------------------------------------------
2010 90 percent floor
------------------------------------------------------------------------
2011 85 percent floor
------------------------------------------------------------------------
The OCC will assess national banks' readiness to operate under
Basel II-based capital rules consistent with the schedule above and
will make decisions on a bank-by-bank basis about termination of the
floors after 2011.
We also intend to retain the prompt corrective action (PCA) and
leverage capital requirements in the proposed domestic implementation
of Basel II. During the several years in which those provisions have
complemented our basic risk-based capital rules, U.S institutions have
thrived while building and maintaining strong capital levels--both
risk-based and leverage. This capital cushion has proved effective not
only in absorbing losses, but also in allowing banks to take prudent
risks to innovate and grow.
I have mentioned that the floors we intend to apply during the
transition period will be simpler in design and more conservative in
effect than those set forth in Basel II. I expect PCA requirements to
play a significant role in the floor requirements. For example, in
order to be ``well-capitalized'' for PCA purposes, a Basel II bank in
2009 (subject to a 95 percent floor) would be required to have a total
risk-based capital ratio of at least 10 percent, calculated under non-
Basel II rules but with a 5 percent reduction in risk-weighted assets.
The bank would also have to meet the 10 percent total risk-based
capital threshold on the basis of its Basel II results, with similar
dual calculations applying to the 6 percent well-capitalized threshold
for the Tier 1 risk-based capital ratio. PCA thresholds for the
leverage ratio would of course also remain in place as they are today.
While we intend to be true to the timelines above, we also expect
to make further revisions to U.S. Basel II-based rules if necessary
during the transition period (that is, before the system-wide floors
terminate in 2011), on the basis of observing and scrutinizing actual
systems in operation during that period. That will allow us to evaluate
the effectiveness of the Basel II-based rules on the basis of real
implementation and to make appropriate changes or corrections while the
prudential transition safeguards are still in effect. Of course, any
future revisions will also be subject to the full notice and comment
process, and we expect to look to that process where necessary to help
resolve difficult issues.
The revised timeline detailed in the September 30 joint statement
also makes possible greater coordination between the Basel II process
and the on-going effort to revamp risk-based capital rules governing
banks not adopting Basel II. The agencies expect that proposed rules
for the U.S. implementation of Basel II will be available in the first
quarter of 2006. As discussed in more detail below, the banking
agencies are also jointly seeking comments on a number of possible
revisions to our existing risk-based capital rules for banks not
adopting Basel II. After consideration of public comments on that
proposal, the agencies expect to move forward with a notice of proposed
rulemaking on Basel I-A in 2006. As I have made clear previously, it is
imperative that there be substantial overlapping comment periods on the
Basel II and Basel I-A proposed rules. This will permit regulators and
industry participants to directly compare the two proposals and assess
competitive effects and other issues in the development of comments.
Basel I-A
On October 20, the agencies jointly published in the Federal
Register an advance notice of proposed rulemaking (ANPR) seeking
comments on suggested broad revisions to our existing domestic risk-
based capital rules, which are based on the 1988 Basel Accord. I
believe the ANPR is a good first step in the direction of improving the
risk-based capital rules that apply to U.S. banks without the enormous
expense and massive complexity of the Basel II framework.
The modifications we are considering would:
<bullet> Increase the number of risk-weight categories to which credit
exposures may be assigned;
<bullet> Expand the use of external credit ratings as an indicator of
credit risk for externally rated exposures;
<bullet> Expand the range of collateral and guarantors that may qualify
an exposure for a lower risk weight;
<bullet> Use loan-to-value ratios, credit assessments, and other broad
measures of credit risk for assigning risk weights to residential
mortgages;
<bullet> Modify the credit conversion factor for various commitments,
including those with an original maturity of under 1 year;
<bullet> Require that certain loans 90 days or more past due or in a
nonaccrual status be assigned to a higher risk weight category;
<bullet> Modify the risk-based capital requirements for certain
commercial real estate exposures;
<bullet> Increase the risk sensitivity of capital requirements for
other types of retail, multifamily, small business, and commercial
exposures; and
<bullet> Assess a risk-based capital charge to reflect the risks in
securitizations backed by revolving retail exposures with early
amortization provisions.
Our primary goal in this rulemaking effort is to increase the risk
sensitivity of our domestic risk-based capital rules without unduly
increasing regulatory burden. This is no small challenge, and we cannot
easily accomplish that goal without substantial input from the public.
In crafting the current proposal, the agencies drew from discussions
with the banking industry, Congress, and our experiences in supervising
the current risk-based capital regime. It is important to acknowledge
that much important work on this proposal lies ahead. While some of the
modifications the ANPR presents are well-defined, there are some areas
that are not specified in great detail at this time. We are looking to
commenters to provide additional views and information on current risk
management practices to help refine these areas. So, I am eager to hear
from the industry and other interested parties, and I hope this public
comment process will begin a fruitful dialogue that will lead to more
definitive proposals for a more risk sensitive regime.
We recognize that a number of banks and industry groups are
concerned that banks operating under Basel II might gain a competitive
edge over banks not governed by the Basel II framework. That issue will
remain in the forefront as we more fully develop any proposals that
might stem from the Basel I-A ANPR as well as proposals for Basel II
implementation. It is almost a certainty that the level of risk
sensitivity we hope to achieve under Basel II is not possible in a
simpler risk-based capital regime. However, we need to be very mindful
of competitive equity issues, and we will endeavor to reduce gaps
between the two frameworks as much as possible given our overarching
priority to ensure that both frameworks move in the direction of
greater risk sensitivity. That will require, among other things, an
assessment of the quantitative effects of the Basel I-A proposals as
they become more fully developed. It is also critical for regulators
and interested parties to be able to review and compare definitive
proposals for Basel II and for other domestic capital revisions within
the same general timeframes.
Conclusion
The overarching challenge we face is to improve on the simplistic
Basel I risk-based capital regime. That regime is a poor arbiter of
risks being taken by banks, insufficient to the task of monitoring risk
in large, complex financial institutions, and long overtaken by events
in the marketplace. It is also a source of inefficiency in the
financial system. What we have learned through the development of Basel
II is that for institutions that have the scale and financial capacity
to do so, we can and should establish high standards of risk management
that can be used to improve the alignment of regulatory capital with
risk. Our Basel I-A efforts embody our belief that we can and should do
better in defining capital requirements for the vast majority of
national banks without massive complexity or enormous expense.
We are committed to improving risk sensitivity of the risk-based
capital rules for all institutions, but doing so in a way that is
tailored to the size, structure, complexity, and risk profile of the
institution, and that ensures safety and soundness. For the complex
operations of our largest globally active national banks, we believe
the Basel II framework holds great promise, and we remain committed to
the next steps of implementing it in the United States. For the vast
majority of national banks that will not use Basel II, we believe that
the Basel I-A proposal introduces enhancement in the risk sensitivity
of regulatory capital without unduly increasing regulatory burden.
We have undertaken this task with full awareness of the challenges
ahead. The OCC would not be pursuing these proposals if we did not
believe they would take the industry and us in the direction of not
only better risk-based capital calculations, but also better risk
management, and even more fundamentally, a stronger and safer banking
system.
----------
PREPARED STATEMENT OF SUSAN SCHMIDT BIES
Governor, Board of Governors of the Federal Reserve System
November 10, 2005
Chairman Shelby, Senator Sarbanes, and Members of the Committee on
Banking, Housing, and Urban Affairs, I thank you for the opportunity to
join my colleagues from the other Federal banking agencies to discuss
the current status of Basel II in this country, as well as the status
of proposed amendments to our existing Basel I-based capital rules.
Introduction
The Federal Reserve considers the maintenance of strong and stable
financial markets as an integral part of our responsibility and
critically related to safety and soundness of the participants in those
markets. Financial stability contributes to sustained economic growth
by providing an environment in which financial institutions,
businesses, and households can conduct their business with more
certainty about future outcomes. Part of maintaining a strong financial
system is ensuring that banking organizations operate in a safe and
sound manner with adequate capital cushions that appropriately support
the risks they take.
As many of you are aware, there have been two major developments
within the past 6 weeks regarding U.S. regulatory capital requirements
that apply to banking institutions. First, on September 30, the U.S.
banking agencies announced their revised plan for the implementation of
the Basel II framework in the United States. Second, the agencies
published for comment an advance notice of proposed rulemaking (ANPR)
pertaining to amendments to the existing Basel I-based capital rules
(the amended Basel I). Taken together, these proposals on Basel II and
the amended Basel I represent substantial revisions to the regulatory
risk-based capital rules applied to U.S. banking institutions, from the
very largest to the smallest. From the Federal Reserve's perspective,
these two initiatives, when implemented successfully, should produce a
much-improved regulatory capital regime in the United States that
enhances safety and soundness. The Federal Reserve considers the
ongoing discussion between the Congress and the U.S. banking agencies--
and, of course, with the banking industry and members of the public--to
be critical to the success of both sets of proposals.
Reasons for Pursuing Basel II
We have all witnessed the substantial changes in the U.S. banking
industry over the past decade, including growth in size and geographic
scope, expansion of activities, development of new instruments and
services, and greater use of technology. As a result, we have seen the
rise of very large entities with large geographic reach operating in
many lines of business and engaging in complex and sophisticated
transactions. The largest institutions have moved away from the
traditional banking strategy of holding assets on the balance sheet to
strategies that emphasize redistribution of assets and actively
managing risks. These dramatic changes to the risk profiles of many
banking organizations have only accelerated with the continued
evolution of many, often complex, financial tools, such as
securtitizations and credit derivatives.
Additionally, risk-management techniques employed by many banking
organizations continue to change, improve, and adapt to the ever-
changing financial landscape. For instance, operational risk was not
part of our risk-management thinking 10 years ago, but tools to
identify, measure, and manage it are now becoming prevalent. Also, the
lines between the banking book and the trading book have blurred
significantly and organizations continue to move resources and products
to optimize earnings and manage risks. And finally, global competition
has intensified significantly, as the ability of customers to choose
from a variety of local and international banking firms, as well as
nonbank competitors, has increased.
While the current Basel I-based rules have served us well for
nearly two decades, they are simply not appropriate for identifying and
measuring the risks of our largest, most complex banking organizations.
Basel I, even when periodically amended, must be straightforward enough
for even the smallest banking organizations to implement with relative
ease. Thus, the categories of risk used to determine capital are very
broad and are intended to capture the ``average'' risk levels across
the banking system for that generic exposure.
Large financial institutions tend to manage risk in more proactive
ways, and are able to take advantage of new innovations in financial
instruments to hedge, sell, or take on risk exposures to support their
business strategies and profitability targets. As a result, they are
able to remove balance sheet exposures for risks where they feel
regulatory capital is set too high, and thereby reduce minimum
regulatory capital. Smaller organizations generally do not have the
risk-management systems or scale of transactions to make these
practices economically viable.
While the balance sheet focus of Basel I is appropriate for most
banking organizations, the largest organizations have significant
exposures off the books, and these risk exposures need to be considered
explicitly in determining minimum regulatory capital for these
sophisticated organizations. Large organizations are increasingly
gravitating toward fee-based revenue streams. This is due to
securitizations of loan portfolios that retain the responsibility of
servicing the loans, buying and selling financial instruments for
customers, and growth in business lines where fees are generated by
transactions and account processing. These activities have little
exposure shown on the balance sheet at a moment in time, but failure to
operate complex systems and negotiate complex financial deals in a
sound manner can lead to large loss exposures given the volume of
activity that runs through the line of business. They also use
sophisticated models to manage credit, market, and interest rate risks.
Poor data integrity, model reliability or lack of sufficient controls,
can create losses when management action relies on the faulty results
of decision models.
Finally, the complexity of these organizations makes it more
difficult for executive management to view risk in a comprehensive way,
both in terms of aggregating similar and correlated risks, but also
identifying potential conflicts of interest between the growth of a
line of business and the reputation, legal and compliance risks of the
firm as a whole, In recent years, large financial institutions have
reported losses from breaks in these operating controls that in some
cases have exceeded those in credit or market risk.
The Basel II framework should improve supervisors' ability to
understand and monitor the risk taking and capital adequacy of large
complex institutions, thereby allowing regulators to address emerging
problems more proactively. It should also enhance the ability of market
participants, through public disclosures, to evaluate the risk
positions at those institutions by providing much better risk measures.
The advanced approaches under Basel II, which include the advanced
internal ratings-based approach (or A-IRB) for credit risk and the
advanced measurement approaches (or AMA) for operational risk, offer
particularly good improvements in terms of risk sensitivity, since they
incorporate advanced risk-management processes already used today by
best-practice institutions.
Indeed, the expected improvements in risk measurement and risk
management form the core of our reasons for proposing Basel II in the
United States. Its advanced approaches create a rational link between
regulatory capital and risk management. Under these approaches,
institutions would be required to adopt a set of quantitative risk-
measurement and sophisticated risk-management procedures and processes.
For instance, Basel II establishes standards for data collection and
the systematic use of the information collected. These standards are
consistent with broader supervisory expectations that high-quality risk
management at large complex organizations depends upon credible data.
Enhancements to technological infrastructure--combined with detailed
data--will, over time, allow firms to better track exposures and manage
risk. The emphasis in Basel II on improved data standards should not be
interpreted solely as a requirement to determine regulatory capital
standards, but rather as a foundation for more advanced risk-management
practices that would strengthen the value of the banking franchise. But
while the new framework would, in our view, provide useful incentives
for institutions to accelerate the improvement of risk management, we
believe in most areas of risk management institutions would continue to
have the choice among which methods they employ.
Thus, from a safety and soundness regulatory perspective, for these
large, complex financial organizations, regulators and market
participants need the information provided by the advanced framework of
Basel II.
Recent Developments with Basel II
The Federal Reserve considers the agencies' September 30
announcement relating to Basel II a good outcome and an example of
successful interagency cooperation. As you may recall, in April of this
year, the agencies announced jointly their reaction to initial results
of a fourth quantitative impact study pertaining to Basel II, known as
QIS-4. As the April statement indicated, we were concerned about
results from QIS-4 that showed a wider dispersion and a larger overall
drop in minimum regulatory capital requirements for the QIS-4
population of institutions than the agencies had initially expected.
The initial QIS-4 results prompted the agencies to delay issuance of a
notice of proposed rulemaking (NPR) for Basel II in order to conduct
further analysis of those results and their potential impact. The
agencies' reaction to the initial QIS-4 results, deciding to take
additional time to understand more fully the information provided by
the QIS-4 institutions, is an indicator of how seriously we are taking
Basel II implementation.
During the summer, the U.S. agencies conducted additional analysis
of the information reported in QIS-4. That analysis is for the most
part complete. Based on the new knowledge gained from the additional
QIS-4 analysis, the U.S. agencies collectively decided to move ahead
with an NPR but adjust the plan for U.S. implementation of Basel II.
Adjustments to the plan include extending the timeline for
implementation and augmenting the transitional floors, which should
provide bankers and regulators with more experience with Basel II
before it is fully implemented in the United States. In addition, the
agencies stated specifically in our joint press release that after
completing a final rule for Basel II, we intend to revisit that rule
prior to the termination of the transitional floors. That is, we expect
to perform additional in-depth analyses of the Basel II minimum capital
calculations produced by institutions during the parallel run and
transitional floor periods before we move to full implementation
without floors. This is consistent with the overall process we have
laid out for implementing Basel II. We want to ensure that the minimum
regulatory capital levels for each institution and in the aggregate for
the group of Basel II banks provide an adequate capital cushion
consistent with safety and soundness.
Probably the most important thing we learned from the QIS-4
analysis is that progress is being made toward developing a risk-
sensitive capital system. In terms of the specifics of the analysis, we
learned that the drop in QIS-4 capital was largely due to the favorable
point in the business cycle when the data were collected. While the
previous QIS-3 exercise was conducted with data from 2002, a higher
credit loss year, QIS-4 reflected asset portfolio, risk management
information and models during one of the best periods of credit quality
in recent years. We learned that the dispersion was largely due to
varying risk parameters used by the institutions, which was permissible
in the QIS-4 exercise, but also due to portfolio differences. That is,
banks have different approaches to risk-management processes, and their
models and databases reflect those differences.
We also learned that some of the data submitted by individual
institutions was not complete; in some cases banks did not have
estimates of loss in stress periods--or used estimates that we thought
were not very sophisticated--which caused minimum regulatory capital to
be underestimated. Based on the results of QIS-4, the Federal Reserve
recognizes that all institutions have additional work to do. In our
view, the findings did not point to insurmountable problems, but
instead identified areas for future supervisory focus. In that way, the
analysis was critical in providing comfort to enable us to move
forward.
It is also helpful to remember that the QIS-4 exercise was
conducted on a best-efforts basis. It was just one step in a
progression of events leading to adoption of the Basel II framework. We
certainly expect that as we move closer to implementation, supervisory
oversight of the Basel II implementation methodologies by our
examination teams would increase. Indeed, during the qualification
process we expect to have several additional opportunities to evaluate
institutions' risk-management processes, models, and estimates--and
provide feedback to the institutions on their progress. So while the
QIS-4 results clearly provided a much better sense than before of the
progress in implementing Basel II and offered additional insights about
the link between risks and capital, QIS-4 should not be considered a
complete forecast of Basel II's ultimate effects. It was a point-in-
time look at how the U.S. implementation was progressing.
Institutions participating in QIS-4 put a lot of time and effort
into assisting with the QIS-4 analysis. For that reason, we owe it to
the institutions to provide feedback prior to engaging in a detailed
public discussion of the findings. Those feedback sessions, a full
interagency effort involving an interagency agreed-upon presentation of
the results, are now underway and we expect them to be largely
completed by the end of this month. The agencies plan to release a
public document describing our findings shortly after these sessions
are completed, we hope by the end of the year.
Proposed Next Steps in the Basel II Process
I would now like to describe some possible next steps in the Basel
II process. To be clear, these thoughts represent our best estimates at
this time and could change, given the extensive opportunity for public
comment and additional interagency discussions to come. But I thought
it would at least be helpful to offer the Federal Reserve's
perspective.
First, we support the idea of finishing an NPR on Basel II and
related supervisory guidance as soon as possible, which right now looks
to be in the first quarter of 2006. We believe that the best way to
further augment our understanding of the impact of Basel II is to issue
the NPR and hear reaction from the Congress, the industry, and the
public. In addition, we are interested in issuing the NPR and related
supervisory guidance as soon as possible so that bankers can have a
better idea of supervisory expectations relating to Basel II. The NPR
will help bankers identify the areas where they need to strengthen
their risk-measurement processes as they continue to prepare for
adoption of Basel II.
After the end of the NPR comment period, the agencies plan to
review the comments and decide more specifically on how to move
forward. The agencies would then develop a strategy for issuing a final
rule on Basel II, of course taking into account comments received. Once
the final rule is issued, those institutions moving to Basel II would
complete preparations to move to a parallel run, a period in which
minimum regulatory capital measures under both Basel II and Basel I
will be calculated. Under the current timeline, the parallel run would
start in January 2008.
The parallel run period, which is intended to last for four
continuous quarters, should provide us with additional key information
about the expected results for Basel II on a bank-by-bank basis, as
well as the level of bank preparedness to operate under Basel II. Once
an institution conducts a successful parallel run, the relevant primary
Federal supervisor would then confirm the bank's readiness and give
permission for the institution to move to the first initial phase of
adoption, into the initial floor period. It is only after an
institution has operated to the primary supervisor's satisfaction in
the parallel run and each of the 3 years of floors that it would be
allowed to have its minimum regulatory capital requirements determined
by Basel II with no floors.
During U.S. implementation of Basel II, if at any stage in the
process we see something that concerns the banking agencies, we will
reassess and propose amendments to relevant parts of the framework. The
agencies have already decided to embed in the planned timeline the
possibility for a later revision to the initial Basel II rule (before
the floors are removed), since it is expected that new information
provided in the parallel run and floor years might point to a need for
adjustments to that initial rule. This is entirely consistent with the
path we have taken in the past regarding Basel I, to which there have
been more than twenty-five revisions since 1989. The Federal Reserve
considers all of the planned safeguards and checks and balances to be
sufficient for Basel II to be implemented in the United States
effectively, and with no negative impact on safety and soundness or the
functioning of banking markets.
Proposed Amendments to Basel I
As I noted, the Basel II proposal is not the only minimum
regulatory capital proposal being contemplated by the U.S. banking
agencies. We have issued an ANPR for amendments to Basel I that is
another important initiative in our efforts to update regulatory
capital rules. The regulatory capital rules to be amended by the ANPR
would apply to thousands of banking institutions in the United States,
while the Basel II proposal would likely only apply to 10 to 20 at
inception. The agencies are focusing considerable attention on the
potential interplay between the proposed Basel II rules and the
proposed Basel I amendments in order to ensure that the goals for each
are achieved.
The Federal Reserve's statement pertaining to the release of the
ANPR highlighted that the revisions are intended to align risk-based
capital requirements more closely with the risk inherent in various
exposures. The ANPR relates, in part, to some long-standing issues in
our current capital rules that have been identified (such as requiring
capital for short-term commitments). We also noted that the amended
Basel I is intended to mitigate certain competitive inequalities that
may arise from the implementation of Basel II rules (such as lowering
the risk weight for some residential mortgage exposures). In
considering these possible revisions, the U.S. agencies are seeking to
enhance the evaluation of bank portfolios and their inherent risks
without undue complexity or regulatory burden. In issuing the ANPR, an
advance notice, the agencies are emphasizing that views are still being
developed and additional comment from the banking industry and other
interested parties would be both beneficial and welcome before we move
forward. We are intentionally leaving a number of areas open in order
to solicit a broad range of comments before we narrow down the range of
possibilities.
The U.S. banking agencies have identified over the past several
years a number of issues that need to be addressed within our current
Basel I rules. The development of Basel II-based rules also creates the
need for the U.S. agencies to amend the current rules in order to
address issues relating to competitive impact. While we view that
impact as limited, we want to ensure that institutions not moving to
Basel II have equal opportunities to pursue business initiatives and
are not placed at a competitive disadvantage or otherwise adversely
affected. That is why we are being very careful to analyze the
potential results of these two efforts in tandem, and asking for the
Congress, the industry, and others to provide comments on the potential
effects of both initiatives.
We believe that the revisions to Basel I-based rules should benefit
most institutions by better reflecting current risk exposures in
regulatory capital requirements at little additional burden. Naturally,
regulatory capital requirements are usually not the binding constraint
for banking organizations. Nearly all institutions hold capital in
excess of the minimum required regulatory ratios, in many cases several
percentage points above, to satisfy rating agencies, debtholders and
shareholders, and counterparties in the market. By the same token,
pricing in the banking industry is not driven by regulatory capital,
but rather, as most would intuitively assume, by supply and demand and
business decisions made by bankers. But we think regulatory capital can
act as a useful gauge of risk-taking, even though it would not be the
deciding factor in business decisions.
With respect to the proposals for amended Basel I, as well as Basel
II, the Federal Reserve fully supports retention of the existing prompt
corrective action (PCA) regime, which the Congress put in place more
than a decade ago, as well as existing leverage requirements. In
addition to the safeguards planned for initiatives being discussed
today, we at the Federal Reserve take comfort that the PCA and leverage
requirements will continue to provide a level of protection for
depositors, consumers, and the financial system as a whole. These
regulations help to ensure a minimum level of capital at individual
institutions and in the aggregate that we consider to be absolutely
vital to the health of our banking system and the economy more broadly.
Importance of the Rulemaking Process
At the Federal Reserve--indeed, I think I can say among all the
U.S. banking agencies--we understand and respect the rulemaking process
and the legal requirements for implementing regulatory revisions. This,
of course, includes comment periods for each of our regulatory capital
proposals and transparency in our overall process. We encourage a
healthy debate about the agencies' proposed initiatives--including the
recently revised timeline for Basel II. We look forward to continuing
to engage the industry, the Congress, fellow supervisors, and others in
a discussion about what effects the Basel II framework and the Basel I
revisions might have on our banking system. The proposals are intended
to provide the right incentives for bankers, but if the proposals do
not achieve this goal, we want to know why. In the past, when we have
been provided with well-documented and convincing reasons for making a
change to the Basel II framework or the U.S. implementation process, we
have heeded those arguments. We expect to have the same posture
regarding comments on the proposed Basel I amendments. We continue to
recognize that vigorous discussion and debate produce a much better
product. And we expect to remain vigilant about the potentially
unintended and undesired consequences, particularly those that might
affect a certain class of banks.
Additionally, I would like to emphasize that from my perspective
the U.S. agencies continue to work well with one another on these
regulatory capital proposals in a general environment of cooperation
and good will. While the U.S. agencies naturally disagree on certain
policy matters and implementation issues from time to time, we at the
Federal Reserve are pleased with the outcomes to date and recognize
that all 4 agencies are making considerable contributions to the
overall effort.
Dialogue with the Industry
The extension of the U.S. timeline for Basel II, along with the
ongoing proposals for amended Basel I, obviously present some
challenges for U.S. institutions. We will continue our efforts to
ensure that we hear about these challenges and do our best to assist
institutions in meeting them. First of all, bankers must keep track of
the latest proposals and understand what they could mean for their own
institutions. For those institutions looking to prepare for adoption of
Basel II, making the manifold upgrades in risk-measurement and -
management systems--not the least of which is developing credible
databases--is even more difficult, especially since complete and final
supervisory expectations have yet to be released. But we certainly hope
that institutions do not lose momentum based on the revised timeline
for Basel II; indeed, that timeline reflects our assessment of the work
that still lies ahead.
While institutions might be challenged to move forward in certain
areas until the Basel II NPR and its associated supervisory guidance is
issued, we still believe that they can make strides in other areas. For
one, the agencies all along have emphasized the importance of
institution-specific implementation plans, which include gap analyses,
clearly defined milestones, and remediation plans. In other words, we
think that institutions could now continue development of the corporate
governance surrounding each institution's efforts in Basel II
implementation and focus on their individual implementation processes.
In addition, supervisors have begun to discuss individual QIS-4 results
with each participant; these discussions include specific feedback
about the institution's results and some general peer comparisons.
Additionally, we do recognize that the recent update to U.S.
implementation plans could generate some challenges for U.S.
institutions as they try to implement Basel II worldwide, as well as
for foreign banks operating in the United States. Overall, we think
these challenges are manageable and we can facilitate solutions to them
during the implementation process. While not downplaying potential
challenges, the U.S. agencies, in deciding to adjust implementation
plans, thought it was important to ensure that implementation in the
United States be conducted in a prudential manner and without
generating competitive inequalities in our banking markets. As before
the September 30 announcement, we continue to work with institutions
and foreign supervisors to minimize the difficulties in cross-border
implementation. Our support includes extensive discussion with other
countries in the Basel Accord Implementation Group, as well as more
informal, bilateral discussions with institutions and foreign
supervisors. Our view is that these cross-border issues do not
necessarily represent fundamentally new problems; while requiring some
work, these challenges are manageable. It is also useful to point out
that all Basel member countries have their own rollout timelines and
national discretion issues, not just the United States--which is
entirely appropriate. In order to assist institutions in resolving
their cross-border challenges, we are eager to hear specifics from
institutions so that we can develop targeted solutions.
Conclusion
Mr. Chairman, in my remarks today I have described the Federal
Reserve's views on suggested changes to the current regulatory risk-
based capital regime, namely the proposals for Basel II and amended
Basel I. I have outlined the need for change, the work completed to
date, and some of the lessons learned. In our view, recent exercises
such as QIS-4 have served as useful indicators of the progress being
made and the direction needed for these initiatives on regulatory
capital requirements. QIS-4 was part of an extended series of
activities to ensure that the suggested regulatory capital revisions
are implemented in an appropriate and prudent manner. From the Federal
Reserve's perspective, we should continue to move forward with the
activities I described, while seeking comment and listening to feedback
at every stage.
Our support for Basel II stems from the belief that it would
provide a much better measure of minimum regulatory capital at the
largest, most complex institutions, aligning capital with risks to
which these institutions are exposed. We also believe that Basel II
would bring about substantial improvements in risk management to those
institutions. At the same time, amending Basel I for the vast majority
of banking institutions in the United States could improve the
reflection of risks in Basel I-based rules without much additional
burden. Taken together, these initiatives should ensure adequate
minimum capital cushions, allow fair competition, maintain safety and
soundness, and enhance financial stability.
I am pleased to answer your questions.
----------
PREPARED STATEMENT OF DONALD E. POWELL
Chairman, Federal Deposit Insurance Corporation
November 10, 2005
Chairman Shelby, Senator Sarbanes, and Members of the Committee, it
is a pleasure to appear before you today on behalf of the Federal
Deposit Insurance Corporation to discuss current developments regarding
Basel II.
As you know, Basel II is an international effort to create
standards for capital requirements that would allow banking
institutions to use internal estimates of credit and operational risk
to determine their minimum risk-based regulatory capital requirement.
Basel II is intended to be a framework that is more risk-sensitive and
one that promotes a more disciplined approach to risk management at our
largest banks. Basel II has been developed to respond to concerns that
the regulatory arbitrage opportunities available under Basel I threaten
the adequacy of the regulatory capital buffer needed to ensure
financial system stability. It is important to remember that an
overarching objective of Basel II is to reinforce capital adequacy
standards by better aligning minimum capital requirements with risk and
thereby prevent an erosion of the aggregate level of capital in
individual banks and the banking system.
The FDIC supports these broad goals and is actively engaged in the
regulatory process to develop a new capital framework for the United
States. As the U.S. banking and thrift agencies move forward to
implement Basel II, we must ensure that the new capital framework does
not produce unintended consequences, such as significant reductions in
overall capital levels, the creation of substantial new competitive
inequities between certain categories of insured depository
institutions, or an expansion of the Federal banking safety net by
blurring the regulatory lines between banks and holding companies.
About 6 weeks ago, the U.S. agencies announced a plan for moving
forward with the implementation of Basel II in the United States. I
participated in and support that plan because Basel II has the
potential to represent positive change in capital regulation for our
largest banks. Basel II clearly requires a more sophisticated approach
to risk measurement by the adopting banks. At the same time, however,
the most recent quantitative impact study, QIS-4, showed both a very
large reduction in capital requirements for many banks, and large
differences in capital requirements for what appeared to be identical
risks. All the agencies agreed that the results of the impact study
were unacceptable and that more work remains to be done to address
these concerns.
QIS-4 was a comprehensive effort drawing upon data submitted by 26
of the largest U.S. banking organizations designed to provide the
agencies with an improved understanding of how Basel II affects minimum
required capital at the industry, institution, and portfolio level. A
comprehensive review of the QIS-4 results, conducted over the spring
and summer of this year, raised many questions and concerns. The
agencies' preliminary review of QIS-4 data indicates that, relative to
Basel I, minimum risk-based capital requirements under Basel II will be
reduced for most of the banking organizations in the study--
substantially in many cases--to levels that the FDIC does not consider
commensurate with the risks to which these institutions are exposed.
Further, the results indicate a wide dispersion of results at both the
banking organization and portfolio or business line level, including
material differences in capital requirements for identical, or
virtually identical, credit exposures.
These QIS-4 results pose a dilemma for the agencies. QIS-4 suggests
that the present framework will produce unacceptable capitalization
outcomes. Yet, committing to specific changes to the framework at this
time, without the benefit of further experience and industry systems
development, would be premature. That is why, on September 30, 2005,
the agencies announced that we will move forward with a Basel II Notice
of Proposed Rulemaking (Basel II NPR) that includes additional time for
bank systems development with added prudential safeguards. Those
safeguards include more conservative risk-based capital floors, and a
clear signal that changes to the framework will be made based on
further experience. Ultimately, changes to the Basel II framework are
likely to be required to avoid unjustified and imprudent reductions in
overall capital levels and to reduce the potential for wide variations
in capital requirements for similar types of exposures. While
improvements in risk management practices and risk profiles may justify
lower capital under Basel II, the FDIC believes that a correctly
calibrated Basel II standard will produce overall minimum risk-based
regulatory capital requirements that exceed the capital necessary to
maintain a rating of ``adequately capitalized'' under current prompt
corrective action (PCA) regulations that were mandated by the Federal
Deposit Insurance Corporation Improvement Act of 1991 (FDICIA).
My testimony will focus on how Basel II can be adopted and
eventually implemented in the United States given the concerns raised
by QIS-4. I will focus on the importance of the relationship between
the Basel II standards and the PCA regulations, specifically our
existing U.S. leverage requirements. My testimony will argue that the
QIS-4 results reinforce the need to revisit Basel II calibrations
before risk-based capital floors expire and to maintain the current
leverage ratio standards. Leverage requirements are needed for several
reasons including:
<bullet> Risks such as interest-rate risk for loans held to maturity,
liquidity risk, and the potential for large accounting adjustments
are not addressed by Basel II.
<bullet> The Basel II models and its risk inputs have been, and will be
determined subjectively.
<bullet> No model can predict the 100-year flood for a bank's losses
with any confidence.
<bullet> Markets may allow large safety-net supported banks to operate
at the low levels of capital recommended by Basel II, but the
regulators have a special responsibility to protect that safety-
net.
Under the current formulation of Basel II, the leverage ratio
standard will be more important than ever in guarding against losses to
the insurance funds resulting from insufficient capital at the
individual bank and the industry level.
Explaining the QIS-4 Results--Concerns Continue
Following a preliminary analysis of the QIS-4 results completed in
April, the agencies sshould determine the reasons for the significant
declines in required capital levels and dispersion in reported results.
Comprehensive analysis was needed to determine whether QIS-4 anomalies
reflect actual bank differences in risk, limitations in the design and
implementation of the QIS-4 study, variations in the stages of bank
implementation efforts (particularly related to data availability), or
whether the QIS-4 results indicate the need for adjustments to the
Basel II framework.
The agencies are not yet in a position to publish a comprehensive
summary of our analysis of the QIS-4 results. The agencies have,
nevertheless, determined that the QIS-4 results were driven to varying
degrees by all of the aforementioned factors. As was envisioned in the
design of Basel II, QIS-4 results show that the amount of required
regulatory capital does vary with the risk characteristics of
individual exposures. In many cases, however, variation in reported
capital requirements had more to do with differences in banks' risk
measurement methodologies, or the degree of their adherence to Basel II
requirements as provided in the QIS-4 instructions, than with true
differences in risk. In the design of QIS-4, the regulators did not
intend to be prescriptive about how banks measure their own risk and
recognized that the Basel II framework itself allows for considerable
variation in capital requirements for identical exposures.
In addition to the observed capital variation that reflected
differences in banks' internal risk assessments, the FDIC's analysis
suggests that much of the observed reduction in capital requirements
under QIS-4 is built into Basel II's formulas. That is, the regulatory
capital formulas in Basel II are inherently calibrated to produce large
reductions in risk-based capital requirements.\1\ Better data, or
better compliance by Basel II banks with the standards required by the
framework, would not in this view mitigate the large reductions in
capital requirements suggested by QIS-4. If anything, QIS-4 may
understate the reductions in minimum capital requirements that would
ultimately be expected under an up-and-running Basel II.
---------------------------------------------------------------------------
\1\ See for example, Kupiec (2004), ``Capital Adequacy and Basel
II,'' FDIC Center for Financial Research Working paper No. 2004-02, and
Kupiec (2005), ``Unbiased Capital Allocation in an Asymptotic Single
Risk Factor (ASRF) Model of Credit Risk,'' FDIC CFR Working Paper No.
2005-04.
---------------------------------------------------------------------------
Exhibits 1-3 in Appendix A provide a sense for the drop and
dispersion of capital requirements suggested by QIS-4. At many
institutions, the QIS-4 results show significant reductions in risk-
based capital. The Exhibit 1 table indicates the total minimum capital
requirement of all participating banking organizations falls by an
aggregate of 15 percent. Capital requirements declined by more than 26
percent in more than half of the banking organizations in the study.
Basel II sets capital requirements for selected portfolio groupings
(for example, wholesale commercial and industrial, retail, real estate
development, etc.). At the portfolio level, the Basel II capital
requirements for most portfolio groupings decreased substantially
across participating banking organizations. Most organizations reported
double digit declines in capital requirements for most loan portfolios,
with only a few portfolio categories posting increases (notably credit
card exposures). For example, Exhibit 1 indicates that capital
requirements for wholesale loans declined by 25 percent on average at
the subject banking organizations. Capital requirements for so-called
high volatility commercial real estate loans fell by 33 percent, while
capital requirements for other commercial real estate loans fell 41
percent. Capital requirements for small business loans fell by 27
percent. Capital requirements for mortgage and home equity loans fell
by 61 and 74 percent, respectively.
In addition to problems resulting from the significant decline in
the level of regulatory capital, concern persists over QIS-4 results
showing an inconsistency in the capital results for similar risks
across institutions. The Exhibit 2 and 3 charts show the wide variation
in capital requirements around the averages reported in Exhibit 1.
Further, in a sample of large corporate credits that had identical
lending relationships with many of the QIS-4 institutions, individual
banking organizations reported changes in minimum capital requirements
that varied widely. Using the QIS-4 results of a single reporting
institution to set a benchmark of comparison, QIS-4 participants
reported minimum capital requirements for these identical credits that
ranged from 30 percent less to 200 percent greater than the benchmark
bank's calculation. For representative mortgage products, banking
organizations reported risk weights that ranged from 5 percent to 80
percent on identical exposures. These results suggest that QIS-4
differences in minimum capital requirements are not entirely explained
by true risk differences in bank products. Rather, a substantial source
of the variation in Basel II capital requirements can be explained by
differences in the risk inputs that individual organizations assign to
identical exposures.
Overall, the QIS-4 study confirms that the regulatory capital
requirements set by the Basel II framework are very sensitive to
individual banks' subjective assessments of risk. Achieving consistency
in Basel II currently hinges on the hope that industry best practices
and better data will lead to reduced dispersion in the capital
treatment of similar loan portfolios across banks. At present, however,
the QIS-4 results show that there is little commonality in the
approaches the various banks are using to estimate their risk inputs.
While this inconsistency may, in part, be corrected with refinements to
internal systems and through improved regulatory guidance, differences
are also inherent in the proposed framework and suggest the need for
adjustments and safeguards going forward.
Notice of Proposed Rulemaking Will Set Forth Additional Prudential
Standards
The U.S. agencies intend to publish a Notice of Proposed Rulemaking
on the U.S. implementation of Basel II during the second quarter of
2006 (Basel II NPR). The Basel II NPR will propose an advanced internal
ratings-based approach that includes additional prudential safeguards
designed to address the QIS-4 results. These safeguards include:
<bullet> A limit on the amount by which an institution's risk-based
capital may decline as a result of Basel II. These floors will be
retained for a minimum 3-year transition period and established at
95 percent the first year, 90 percent the second year, and 85
percent the third year.
<bullet> The release of floors only upon approval of an institution's
primary Federal regulator.
<bullet> Continuing evaluation of revisions to the framework given
actual experiences over the transition period.
<bullet> The retention of existing PCA standards, including the
existing leverage ratio standards.
The FDIC continues to emphasize the importance of maintaining a
minimum capital standard embodied in a leverage ratio. The Basel II
standard is not intended to provide capital for all material risks. For
example, the interest rate risk associated with most loans that banks
hold to maturity, liquidity risk, and business risks such as the
potential for large accounting adjustments, are not factored into the
Basel II framework. Moreover, the framework relies on individual bank
risk exposure estimates that are, by their nature, prone to inaccuracy.
Further, these estimates are input into a regulatory model that is only
a simplified expression of the actual risks retained by large complex
banking institutions. These model risks, that are inevitable when banks
are required to estimate their own risk inputs for a simplified
regulatory capital model, may lead to inadequate regulatory capital
requirements under the Basel II framework.
Retaining the existing leverage ratio, a simple and effective
standard, is an important pillar of the safety and soundness regime.
The importance of the leverage ratio is highlighted by recent analysis
conducted by the FDIC that draws upon the QIS-4 results. This analysis
shows that under the current Basel II framework, the leverage ratio
will serve a more important role than ever in ensuring that adequate
levels of capital are maintained throughout the system.
Basel II and PCA: An Impending Conflict of Expectations
The FDIC has analyzed how the Basel II standard would compare to
U.S. capital standards currently applicable to insured institutions. We
found that as a set of quantitative capital standards, Basel II appears
to lower the bar considerably compared to current U.S. leverage and
risk-based capital standards embodied in the agencies' PCA regulations.
The QIS-4 exercise was conducted at the consolidated bank holding
company level. QIS-4 does not quantify the minimum regulatory capital
levels that may prevail under Basel II at the individual banks that
participated in the study. Moreover, the capital requirements reported
in Charts 1-3 in Appendix A are for total capital, which includes
elements such as loan loss reserves, subordinated debt, and certain
intangible assets that do not provide the same level of protection to
the insurance funds as does core, or tier 1, capital. To better
quantify the issues that are most directly relevant to the FDIC as
insurer, we therefore estimated the tier 1 capital requirements that
would apply at the 74 insured banks that are subsidiaries of the 26
QIS-4 reporting organizations. Details on this estimation methodology
are provided in Appendix B.
Analysis of the QIS-4 data completed by the FDIC shows that Basel
II produces minimum regulatory capital requirements that are
unacceptably low under the existing PCA standards implemented pursuant
to FDICIA. Using QIS-4 data, our analysis reveals that--should the
leverage ratio be removed under Basel II--the majority of QIS-4
institutions would be less than adequately capitalized (that is, under-
capitalized, significantly under-capitalized, or critically under-
capitalized) if they held only the level of capital generated by the
Basel II formulas.
As shown in the table on the next page, if the Basel II standards
are the only constraint on the banks' minimum levels of capital, the
majority of 26 banking companies participating in the QIS-4 study could
fall to levels currently considered less than adequately capitalized
under the PCA standards; that is, the minimum regulatory capital of
these institutions would fall below the 4 percent leverage ratio.
<GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT>
In other words, under the Basel II framework as currently
fashioned, the leverage ratio will become the effective, binding
minimum capital standard for most large U.S. banking companies. While
we are aware that minimum regulatory capital requirements can constrain
bank equity returns, we are not aware of any public policy studies or
other claims that the current level of regulatory capital requirements
is a barrier for the provision of additional banking services that are
beneficial for the public. In the FDIC's view, Basel II should be
calibrated in a manner that ensures that, for most banks in most
circumstances, the overall minimum risk-based capital requirements
(credit, operational, and market risk) exceed the minimum leverage
capital requirements that are currently set in FDICIA and its
implementing regulations.
In terms of the capital impact of an up-and-running Basel II, if
the present framework remains unchanged, the FDIC's analysis suggests
that the future will bring even greater declines in capital
requirements than are suggested by QIS-4. As described in Appendix B,
the risk inputs of banks for QIS-4 purposes appear on average very
conservative, more so than a strict reading of the framework would
require. Moreover, the QIS-4 declines in required capital are achieved
without fully factoring in capital reductions that can be achieved
using credit risk hedges and third party guarantees under a fully
implemented Basel II standard. These additional factors could generate
significant reductions in capital requirements beyond those that were
identified in the QIS-4 results. The FDIC does not believe that there
is adequate support for the agencies to conclude that the capital
reductions that likely will result from the current Basel II framework
are commensurate with the reductions in the investment risk exposures
of banks that will be engendered by improvements in risk management
occurring under Basel II. Indeed, unless it can be demonstrated that
Basel II will substantially reduce banks' credit risk exposure
profiles, the increase in allowed leverage could easily lead to higher
system-wide risks.
Even if there were no leverage requirement and no PCA regulations,
the FDIC would find the capital requirements coming out of QIS-4 to be
too low for many reporting institutions. Banks operating with the
benefit of a Federal safety net have operated at such capital levels
for a time, but ultimately at a great cost to that
safety net. In part because Basel II can be expected to generate such
low capital requirements, the leverage ratio will play a more important
role than ever in protecting the insurance funds.
In the view of the FDIC, the leverage ratio is an effective,
straightforward, tangible measure of solvency that is a useful
complement to the risk-sensitive, subjective approach of Basel II. The
FDIC is pleased that the agencies are in agreement that retention of
the leverage ratio as a prudential safeguard is a critical component of
a safe and sound regulatory capital framework. The FDIC supports moving
forward with Basel II, but only if U.S. capital regulation retains a
leverage-based component.
Expectations for Insured Banks under Basel II
The Federal safety net in the United States extends explicitly to
insured banks, not their holding companies. The absolute accountability
of insured institutions for their own governance, and for maintaining
an adequate level of capital, is of fundamental importance in
controlling the potential cost of that safety net. That is why a
critical element for the success of Basel II as a safety-and-soundness
initiative is maintaining appropriate expectations for insured banks.
In concrete terms, insured banks that adopt Basel II will need to
calculate and report a capital requirement that is appropriate for
their own risk exposures. Capital reductions derived from
diversification of exposures held in separate legal
entities may prove to be only hypothetical should one of the entities
become undercapitalized on a stand-alone basis. This does not mean that
holding companies will need to maintain separate and duplicative Basel
II infrastructures at every insured subsidiary. Indeed, to the extent
that regulators expect the accurate measurement of risk at the holding
company level, that would seem to require compatible systems at all
subsidiary legal entities. In terms of managing and controlling the
government's deposit insurance exposure, however, effective risk
control requires that capital calculations be geared to the unique
risks and exposures of each insured subsidiary.
Transparent Information--Ongoing Analysis Required
The FDIC is committed to transparency, and it is our belief and
expectation that the banks and their primary Federal regulators will
collaborate and share information in a manner that allows each agency
to address its concerns with regard to the new capital framework. As
outlined above, the QIS-4 study indicates that modifications of the
current Basel II framework are likely to be necessary to ensure that
regulatory standards require adequate bank capital and equal capital is
required for equal risk. In order to reach a prudent judgment regarding
the safety and soundness implications of any such proposed changes and
to ensure a level playing field within the United States, the FDIC and
the other banking regulatory agencies must obtain adequate information
regarding all participating banks' internal credit risk modeling
systems and resulting minimum capital requirements. From the FDIC's
perspective of assessing risks to the insurance funds, collaboration
must include access to information about the critical assumptions,
models and data used to implement capital requirements based on banks'
own estimates of risk.
Competitive Equity--Basel I-A Advance Notice of Proposed Rulemaking
Throughout the Basel II process, the FDIC has expressed concerns
about the potential detrimental effects that the new framework could
have on competition within the U.S. banking sector. Indeed, the QIS-4
results suggest that the competitive ramifications could be profound.
Absent modifications to the current and proposed risk-based capital
frameworks, the FDIC believes that the non-Basel II banking sector
could be placed at a competitive disadvantage to larger banks subject
to the Basel II framework. To address these concerns, the agencies have
issued an Advanced Notice of Proposed Rulemaking to begin the process
of developing an alternative for non-Basel II adopters (Basel I-A
ANPR).
The Basel II banks already enjoy a pricing advantage over their
smaller competitors due to their asset size, underwriting volume, and
related economies of scale. However, this pricing advantage could be
magnified by the reduced risk-based capital requirements of Basel II.
The higher capitalized non-Basel II banks may become more attractive
acquisition targets for Basel II adopters. Further, the results of the
QIS-4 exercise show that the advantage of the Basel II framework could
be the greatest in those areas where credit risks historically have
been the lowest.
Under the Basel II framework, capital requirements for residential
mortgages, home equity loans, and similar exposures drop significantly.
For example, risk-based capital requirements for single-family
residential mortgage exposures fall from 4 percent under current Basel
I standards to 1.5 percent under Basel II (based on the average risk-
weight reported in QIS-4). Moreover, Basel II, as seen in the QIS-4
results, greatly expands the disparity in minimum required risk-based
capital between lower risk and higher risk credits. For example, prime
mortgages will receive a much lower capital charge than subprime
mortgages under Basel II. In contrast, prime mortgages and subprime
mortgages are generally assigned to the same risk weight category under
existing risk-based capital rules. It is reasonable to assume that
there will be a similar disparity between capital requirements for
prime and subprime credit card exposures. As a result, without
mitigating changes to the competing frameworks, non-Basel II banks
could be placed at a severe competitive disadvantage to Basel II banks
in prime-grade markets while possibly gaining a competitive advantage
over Basel II banks in subprime markets. The end result of such
disparate capital treatments could be a migration of high risk credits
away from Basel II banks and toward non-Basel II banks. We must monitor
this potential change very carefully from a safety and soundness
perspective as well as monitor changes in the exposure of the insurance
funds.
In order to advance a full dialogue of the competitive concerns
associated with changes to the capital framework, the agencies issued
the Basel I-A ANPR that outlines potential changes to risk-based
capital regulations for all U.S. banks. The agencies are soliciting
comments on how to achieve greater risk sensitivity for capital in a
way that does not create undue burden for insured institutions and is
consistent with safety and soundness objectives.
The FDIC is aware that competitive equity concerns are not the same
for all banks. Some community banks choose to maintain large amounts of
risk-based capital--not because they operate in a risky manner, but
rather because they have lower risk appetites or tolerances. Therefore,
we are requesting comments in the Basel I-A ANPR concerning the
possibility of allowing these types of institutions to opt out of
proposed changes.
In addition to addressing potential competitive inequities and
recognizing industry advances in credit risk measurement and mitigation
techniques, the Basel I-A ANPR will also propose ways to modernize the
risk-based capital rules for all U.S. banks. Key components of the ANPR
ask for comment on:
<bullet> Increasing the number of risk-weight categories for bank
credit exposures.
<bullet> Expanding the use of external credit ratings as an indicator
of credit risk for externally rated exposures.
<bullet> Expanding the capital reductions available from the use of
collateral and guarantors.
<bullet> Adopting loan-to-value ratios and credit score measures to
assign risk weights to residential mortgages.
We believe that most, if not all, of these proposals can be applied
using information that is readily available to banks. However, we have
asked for comment on whether the trade-off of a more risk-sensitive
capital framework is justified by any possible burden generated by its
implementation.
Finally, we are asking for comment on any concerns not addressed by
the agencies in the ANPR. The FDIC is confident that by listening to
the needs and concerns of the banking community and other commenters, a
revised capital framework can be put in place for non-Basel II banks
that will mitigate many of the competitive equity concerns.
Conclusion
Going forward, the FDIC plans to issue the Basel II NPR, and
coordinate its issuance with a Basel I-A Notice of Proposed Rulemaking
(this will follow the ANPR) in a manner that will allow for some
overlap in the comment period for the two notices of proposed
rulemakings. This process will allow the two proposals to be compared
side-by-side so that the public can fairly determine the possible
competitive implications of the overall package of proposed changes to
U.S. capital regulation.
We are working diligently to ensure that, as originally envisioned,
the new regulatory capital framework articulated by Basel I-A and Basel
II enhances the safety and soundness of the U.S. banking system. The
U.S. agencies must continue to work closely together, share
information, reach conclusions on important changes to the proposed
framework, and reassess the impact of any such changes. The FDIC is
working with the other agencies to develop a framework that achieves
this broad objective and preserves a set of straightforward minimum
capital requirements to complement the more risk-sensitive, but also
more subjective, approaches of Basel II. We also want to maintain
competitive equity and achieve results under Basel II that are less
extreme and more consistently applicable across banks.
The FDIC, like the other banking agencies, will proceed with the
implementation of Basel II in an appropriately deliberative manner and
with full consideration of the comments of all interested persons.
<GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT>
PREPARES STATEMENT OF JOHN M. REICH
Director, Office of Thrift Supervision
November 10, 2005
Introduction
Good morning, Chairman Shelby, Ranking Member Sarbanes, and Members
of the Committee. Thank you for the opportunity to discuss the views of
the Office of Thrift Supervision on the development of the Basel II
capital framework in the United States for our largest U.S. financial
institutions and the parallel modernization of Basel I for our other
institutions.
The development of Basel II has been underway, internationally, for
a number of years. In the United States, the four Federal Banking
Agencies (FBA's) commenced the formal rulemaking process with the
issuance of an Advanced Notice of Proposed Rulemaking (ANPR) in 2003.
The parallel modernization of domestic risk-based capital requirements
stemming from Basel I (our current risk-based capital standards), aimed
at institutions that will not adopt Basel II, has been discussed among
the regulators for some time. It formally began, however, only a few
weeks ago with the issuance of an ANPR.
The goal of Basel II is to produce a risk-based capital system that
promotes effective risk management, maintains capital adequacy, and
increases the transparency of risks undertaken by our largest,
internationally active institutions. The goals of the FBAs are to
ensure that Basel II, as implemented in the United States, and the
parallel modernization of our current risk-based capital standards,
will enhance the risk management, capital adequacy, and risk
transparency of all our financial institutions, while maintaining the
safest and soundest banking system in the world. OTS will endorse the
implementation of systemic changes to our capital rules only if they
advance these goals.
Although we are more than 2 years from the start of a proposed 4-
year phase-in of Basel II, there are significant hurdles to overcome
before we can represent to you, Mr. Chairman, that it is ready to be
implemented. This is underscored by a recent quantitative impact study,
QIS-4, conducted by the FBA's. QIS-4 indicated that further,
significant revisions are needed before we can implement Basel II in
the United States. We also need to resolve difficult policy issues in
the modernization of our current risk-based capital standards. We
remain committed to meeting the challenges raised in the development
and implementation of both capital systems.
Today, I will address some of these issues and provide an update on
the approach to risk-based capital contemplated by Basel II, as well as
issues that our U.S. institutions are expected to face under a revised
Basel I-based framework.
Given that the banking and thrift industries are profitable and
well-capitalized, with few troubles in recent years, you may reasonably
ask what is broken and why do we need to fix it? The answers start with
the growing exposure of large and internationally active institutions
to steadily increasing risks that are not captured very well--if at
all--under our current risk-based capital framework.
Much of the public debate about Basel II has been about changes in
minimum regulatory capital; however, the new framework goes beyond
that. It focuses first and foremost on enterprise-level risk
management, and encourages ongoing improvements in risk assessment
capabilities. Basel II also provides for governance changes, induding
board of director accountability for risk management. Basel II may
allow for reduced minimum risk-based capital requirements for certain
institutions, where a reduction is justified and commensurate with real
and verifiable risk exposure. That would only be available, however,
where an institution can demonstrate--to the satisfaction of its
primary regulator--that its risk management capabilities and resultant
risk reduction merit such a change in capital requirements.
There is also a need to modernize Basel I risk-based capital
requirements in order to minimize competitive effects of adopting Basel
II and to better align capital requirements with the wide range of risk
profiles of domestic financial institutions.
Overview and Background of the Basel Process
Basel I
Basel I, issued 1988 by the Basel Committee on Banking Supervision
(BCBS), provided a set of capital principles designed to strengthen
capital levels at large, internationally active banking organizations,
and to foster international consistency and coordination.\1\ Although
Basel I applied only to the largest, internationally active banks in G-
10 countries (countries outside the G-10 were encouraged to adopt it
for their banks operating internationally), the themes of Basel I were
intended to apply to all banking organizations worldwide, of any size
and activity.
---------------------------------------------------------------------------
\1\ The BCBS identified two fundamental objectives at the heart of
its work on regulatory convergence under Basel I. As the BCBS 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.''
---------------------------------------------------------------------------
While OTS did not participate in developing Basel I, we applied it
to the institutions we regulate, along with the other FBA's. Throughout
implementation of Basel I, the FBA's developed risk-based capital
standards consistent with its underlying principles, but with
modifications intended to enhance risk sensitivity and conform to the
unique needs of the U.S. banking system.
When Basel I was issued, the BCBS recognized that it was only a
start, and that more refinement would take place over time. In today's
sophisticated financial marketplace, Basel I is a relatively simplistic
framework. For example, it makes no distinction between a well-
underwritten commercial credit to a strong borrower and a relatively
weak commercial credit to another borrower. Both are assigned the same
(100 percent) risk weight. Similarly, residential mortgages, which can
vary widely in quality, are assigned either a 100 percent risk weight
or, if prudently underwritten, a 50 percent risk weight (most 1-4
family residential mortgages). Currently, even the lowest credit risk
residential mortgages are subject to a 50 percent risk weight; whereas
the highest credit risk residential mortgages are subject to no more
than a 100 percent risk weight.
As stated by the BCBS, advances in risk management practices,
technology, and banking markets have made the 1988 Accord's simple
approach to measuring capital less meaningful for many institutions.
Likewise, improvements in internal processes, more advanced risk
measurement techniques, and more sophisticated risk management
practices have dramatically improved the monitoring and management of
risk exposures and activities. In short, the static rules of the 1988
Accord have not kept pace, and in fact, were not designed to keep pace
with advances in risk management.
Basel II
As financial instruments, systems, and products became more
complex, the BCBS began designing a new regulatory capital framework.
This framework, Basel II, incorporates advances in risk measurement and
management practices, and attempts to assess capital charges more
precisely in relation to risk, and in particular, credit risk and
operational risk. The international agreement articulating these
principles was issued in June 2004.
Basel II calls for institutions to measure and maintain internal
data about different loan types for credit and operational risk. These
requirements help to promote improved risk management systems. The
FBA's also expect institutions to continue to develop and improve their
internal economic capital models to more accurately measure their own
unique enterprise risk. While Basel I focused on measuring risk
exposure on an asset-by-asset basis, placing assets into simple,
broadly defined risk buckets, Basel II focuses on enterprise-wide risk
management, and encourages institutions continually to evaluate and
assess their risk exposure.
There are numerous reasons for our U.S. banking system to move
forward to a more sophisticated, risk-based framework for evaluating
capital adequacy in institutions implementing Basel II. At the same
time, it is important to identify ways to improve our Basel I-based
system for the thousands of institutions that will not adopt Basel II.
We believe that these objectives are not mutually exclusive, but rather
mutually dependent in order to prevent potential competitive inequities
between Basel II adopters and non-adopters.
Implementation of Basel II in the United States
While OTS supports the concepts, principles, and stated goals of
Basel II, implementation in the United States will occur only when the
FBA's are confident that it can be achieved in a manner that
affirmatively strengthens and does not undermine our financial system.
This requires us to maintain the safety and soundness of Basel II
adopters and implement a modernized Basel I-based system that treats
all U.S. institutions fairly and consistently regardless of the risk-
based capital regime that they follow.
The FBA's recently revised the proposed Basel II timeframes to
allow more time consistent with these principles. In addition to
delaying the start to 2008 and adding an additional phase-in year, the
FBA's provided for greater supervisory control over individual
institutions at each step of Basel II implementation,\2\ along with
progressively less binding capital floors until fully implemented in
2012.\3\
---------------------------------------------------------------------------
\2\ The new timeframes provide for a nonbinding ``parallel run'' of
the Basel II framework starting in 2008; however, institutions may
enter the parallel run only with the permission of their primary
regulator. This will require an institution to demonstrate to its
regulator that it has accurate and reliable systems in place for
enterprise-wide risk management. During the parallel run phase,
institutions seeking to implement the Basel II framework would also be
required to comply with existing Basel I requirements. Throughout
implementation, institutions would be subject to close supervisory
scrutiny and a strict leverage ratio requirement.
\3\ The phase-in schedule provides that, in the first year (as
early as 2009), an institution's capital reduction is subject to a
floor of 95 percent of the level calculated for risk-weighted assets
under Basel I. Reductions in risk-weighted assets would be limited to a
90 percent floor in the second year of implementation (as early as
2010), and an 85 percent floor in the third year (as early as 2011).
Supervisory approval is required in each successive year to go to the
next floor. During implementation, an institution's primary regulator
will closely monitor its systems for gathering and maintaining data,
calculating the Basel II capital requirement, and ensuring the overall
integrity, and safety and soundness, of the application of the Basel II
framework.
---------------------------------------------------------------------------
While the FBA's have agreed to proceed with these safeguards in
place, the Basel II implementation process remains dynamic. The FBA's
are currently working on a Notice of Proposed Rulemaking (NPR) as a
precursor to issuance of a rule implementing the Basel II framework in
the United States.
In conjunction with issuance of an NPR, the FBA's plan to issue
comprehensive proposed guidance consolidating previously issued
guidance on retail, corporate, and operational risk and including
issues not previously addressed, such as securitization, credit risk
mitigation, equity exposures, and various wholesale transactions.
Industry reaction and comment on the consolidated guidance will be very
important since it will be the first iteration of U.S. regulatory
policy on some of these subjects. In addition, this will be the first
opportunity for the industry to assess the adequacy of the guidance
based on the standards enumerated in the NPR. The FBA's plan to make
additional adjustments to the guidance after receiving industry
comments and to ensure consistency with a final rule.
The FBA's are currently working toward issuance of a final rule in
mid-2007, which is a critical timing issue for U.S. financial
institutions to have sufficient lead-time to prepare for a 2008
parallel run. We recognize that the 2007 final rule may not even be the
last word on Basel II. Rather, we anticipate that further rulemakings
may be necessary to refine the Basel II framework for use in the United
States based on our experiences during the parallel run and subsequent
implementation stages.\4\
---------------------------------------------------------------------------
\4\ It is also important to note that OTS, like the OCC, is subject
to Executive Order 12866, which requires executive agencies to
determine whether a proposed rule is a ``significant regulatory
action.'' OTS has determined that the Basel II NPR will be a
significant regulatory action based on the potential effects of the
rule. Thus, OTS is required to prepare a regulatory impact analysis of
the NPR, including an analysis of the need for regulatory action, the
costs and benefits of the NPR and alternative approaches, and the
potential impact on competition among financial services providers.
Pursuant to the Executive Order, the NPR and accompanying regulatory
impact analysis will be submitted to the Office of Management and
Budget for review prior to publication of the NPR.
---------------------------------------------------------------------------
Modernization of Our Current Risk-Based Capital Standards
On October 19, 2005, the FBA's issued an ANPR announcing the start
of the rewrite of Basel I-based domestic capital standards. OTS was an
early advocate of revising and modernizing the existingstandards. We
strongly support amending our current risk-based capital standards
simultaneously, or in close proximity to, Basel II. Our view is that
these revisions should encompass meaningful reforms, while avoiding
imposing costly analytical processes on smaller banks and thrifts.
Modifying the existing rules with more accurate riskweights allocated
to a wider range of asset buckets will significantly improve the
current framework. Applying commonly used risk criteria for identifying
different levels of risk will further enhance the existing framework.
This would provide a more granular, risk-sensitive system of
determining appropriate levels of capital, by asset type, and within
asset type.
Modernization of Basel I -based capital standards will also
mitigate potential competitive inequities that may arise with the
implementation of Basel II.
In considering revisions to our current capital rules, five general
principles have guided the FBAs. These are:
<bullet> Promoting safe and sound banking practices and maintaining a
prudent level of regulatory capital;
<bullet> Maintaining a healthy balance between risk sensitivity and
operational feasibility;
<bullet> Avoiding undue regulatory burden;
<bullet> Creating appropriate incentives for banking organizations; and
<bullet> Mitigating material distortions in the amount of regulatory
risk-based capital requirements for large and small institutions.
The recently issued ANPR for modernizing our domestic risk-based
capital standards focuses on a number of potential modifications. These
include increasing the number of risk weight categories for credit
exposures; using loan-to-value (LTV) ratios, credit assessments, and
other broad measures for assigning risk-weights to residential
mortgages--a particularly important issue for OTS; modifying the risk-
based capital requirements for certain commercial real estate
exposures; and increasing the risk sensitivity of capital requirements
for other types of retail, multifamily, small business, and commercial
exposures.
Two additional issues particularly important to OTS are expanding
the number of risk weight categories to which credit exposures may be
assigned; \5\ and continuing to consider private mortgage insurance
(PMI) in the risk-weighting of residential mortgages.
---------------------------------------------------------------------------
\5\ Current categories are 0, 20, 50, 100, and 200 percent, and
possible new and additional categories for consideration are 10, 35,
75, 150, and 350 percent.
---------------------------------------------------------------------------
As we consider modernizing our current risk-based capital
requirements to increase its risk sensitivity and minimize potential
competitive inequities, we realize that some banking organizations may
prefer to operate under the existing Basel I framework, unchanged, to
determine their minimum risk-based capital requirements. The ANPR
anticipates this option and expressly invites comment. We expect
additional comment about flexibility, and balancing safety and
soundness with regulatory burden concerns.
The QIS-4 Survey
There have been a series of structured and coordinated information
gathering exercises conducted internationally, referred to as
Quantitative Impact Studies (QIS). The most recent data collection,
QIS-4, was an important milestone in U.S. development of the Basel II
framework. The FBA's initiated QIS-4 to gauge the potential impact of
Basel II in the United States. Before discussing the results of QIS-4,
it is important to note its inherent limitations.
In October 2004, the FBA's released the QIS-4 materials, and 26
institutions responded to the study. The initial results of QIS-4
revealed a material drop in required risk-based capital compared to
Basel I requirements, and significant dispersion among the respondents
with regard to the Basel II minimum risk-based capital requirement. As
a result, this past April the FBA's delayed the planned 2005 issuance
of the Basel II NPR in order to provide time to conduct additional
analysis on the QIS-4 data.
Several important factors likely influenced the overall quality of
the QIS-4 data. These include unsettled and/or incomplete guidance on
Basel II from the FBA's, as well as the fact that many institutions are
still developing the data and systems required to fully implement Basel
II. The QIS-4 process was instructive, however, on the state of
readiness of--and need for additional preparation by--both the
regulators and the industry.
Notwithstanding these factors, several aspects of the data bear
further scrutiny. Chief among these was a material difference in
required risk-based capital levels among respondents that was not fully
explained by differences in their enterprise risk profiles. We are
particularly concerned with analyses indicating materially disparate
capital charges for credit exposures that generally pose comparable
levels of credit risk.
Another unsettling aspect of QIS-4 was a material drop in overall
risk-based capital. In fact, every category of credit risk showed
declines in capital requirements except for credit cards. This is
clearly an issue that requires further study to ensure that the Basel
II capital standards are adequate.
With respect to mortgage lending, the QIS-4 results demonstrated
large capital reductions for mortgage and home equity lending. While
this is an especially important result because of commercial banks'
existing concentration in mortgage-related assets,\6\ some of the
decline in capital for prime mortgage lending was expected because QIS-
4 did not address interest rate risk--typically, one of the most
significant risks for prime mortgage lending. In fact, the drop in
capital for prime mortgages highlights the importance of realigning
risk-based capital requirements to more closely correspond to actual
credit risks, while also taking into account interest rate risk-a
critical element in evaluating appropriate capital levels.\7\
---------------------------------------------------------------------------
\6\ Since 1995, commercial banks have increased their holdings of
residential-related mortgages 174 percent in real dollars, from $991
billion to $2.72 trillion. As a percentage of assets, commercial bank
holdings of residential-related assets have increased 40 percent, from
23.0 percent of assets in 1995 to 32.3 percent of assets today. By
contrast, thrifts have increased their holdings of residential-related
mortgages in real dollars by 62 percent, but as a percentage of assets
thrift holdings are actually 4 percent lower than in 1995, from 75.6
percent of assets in 1995 down to 72.5 percent of assets today.
\7\ By their very nature, conservatively managed mortgage lenders
typically have substantially lower credit risk exposure than lenders
concentrating in other retail lending activities. A major risk for
mortgage lenders, interest rate risk, is also greatly reduced by the
presence of sound and prudent interest rate risk management practices,
including access to the secondary mortgage market.
---------------------------------------------------------------------------
In recognition of the substantial interest rate risk associated
with many forms of mortgage lending, OTS has developed a rigorous
interest rate risk model. It is our experience, working with our
interest rate risk model for well over a decade, that savings
associations have modified their interest rate risk-taking behavior
based on information and tools provided by our model. How interest rate
risk is ultimately treated under Basel II is an important issue for OTS
and the thrift industry, as well as for banks that focus on mortgage
lending activities.
A disturbing result from QIS-4 was the sizable reduction in
required capital for horne equity lines of credit. Since the end of
2000, home equity lines of credit on institution balance sheets have
grown by an extraordinary 354 percent, to $534 billion. This is due, in
large part, to the low interest rate environment of the last several
years for mortgages and mortgage-related products. We are concerned
that the study results may reflect only our recent experience, and not
accurately portray risks present in a full economic cycle. This remains
an area deserving of additional attention as we move forward with our
rulemaking and the development of guidance on Basel II.
The results of QIS-4 suggest that Basel II remains a work in
progress in the United States, both for the PBA's and institutions that
intend to implement it. The PBA's are committed to creating a
regulatory framework that resolves the concerns raised by QIS-4. There
is a significant amount of work remaining to create a regulatory
structure that ensures risk sensitive results, promotes fair
competition, and ensures the continued safety and soundness of the U.S.
financial system.
Public Policy Concerns with Basel II and the Modernization of Current
Risk-Based Capital Standards
Timing
Implementing a more risk sensitive capital framework in the United
States is an important objective, but we must be vigilant not to harm
our existing banking system. Longstanding capital adequacy standards
combined with a well-established and highly respected supervisory
structure have delivered a banking system that is healthy and robust.
OTS supports Basel II, provided its implementation enhances our
existing banking system. It is important that we review this objective
at each step of the way toward Basel II implementation.
As we take the steps necessary to move to a more advanced and risk
sensitive capital framework, it is important to exercise caution and
allow for sufficient time to consider how best to proceed in
implementing Basel II in the United States. The safeguards we recently
added, including a parallel run year, followed by 3 years of capital
floors, and ongoing regulatory and supervisory review, will help us
proceed prudently toward Basel II implementation.
Competitive Considerations
Implementing more risk-sensitive capital requirements (without
undue burden) is as important for small community banking organizations
as it is for large, internationally active institutions. Achieving
greater risk sensitivity for one part of the banking system and not the
whole will create competitive distortions. While global capital
standards are important, we must avoid potential negative effects on
U.S.-based institutions not operating internationally.
It is important to maintain comparable, although not necessarily
identical, risk-based capital standards for all U.S. institutions with
respect to lending activities that have the same risk characteristics.
Although our largest institutions should receive capital treatment
commensurate with their ability to reduce risk via diversification and
technology, community banking organizations should not be competitively
disadvantaged in the process. Competitiveness issues raised by Basel II
necessitate an across-the-board examination of capital standards for
all our institutions. This provides an opportunity to reexamine our
current risk-based capital standards, and to take any appropriate steps
to reduce potential competitive inequities for community banking
organizations.
OTS is pleased that the modernization of Basel I-based capital
standards, an initiative we championed, has evolved into a commitment
by all the FBA's. The goal of this initiative is to achieve greater
risk-sensitivity without undue complexity. We believe this can be
accomplished by, among other things, increasing the available asset
``risk-buckets'' first enunciated under Basel I, and by applying
commonly understood criteria for assessing the relative risk within and
among various loan types.
As previously described, while Basel II includes minimum risk-based
capital requirements for credit and operational risk, it does not
include specific capital requirements for interest rate risk. OTS
believes that this significant risk, especially important in mortgage
products, should be addressed by the FBA's consistently. If the Basel
II construct for interest rate risk is maintained, it will be important
to prepare comprehensive interagency guidance on how we expect this
risk to be measured and managed by U.S. institutions implementing Basel
II.
Leverage Requirements, Prompt Corrective Action, and other Safeguards
An issue garnering significant attention under Basel II is the
interrelationship--and tension--between a risk-insensitive leverage
ratio and risk-based capital requirements. On the one hand, the
increased risk sensitivity offered by Basel II is intended to align
risk-based capital requirements more closely with a banking
organization's own internal capital allocation. A leverage requirement
is fundamentally different, however, in that it constrains the extent
to which an institution can leverage its equity capital base. In
effect, a risk-insensitive leverage requirement--a backstop protecting
the Federal deposit insurance funds--potentially operates as a
disincentive for an institution to invest in the least risky assets,
while not constraining its investment in high-risk assets.
Prompt corrective action (PCA), which includes the leverage ratio
constraint, was instituted in the late 1980's in response to the need
for more aggressive and timely supervisory intervention in the face of
stressed and declining capital levels. PCA provides a graduated capital
structure for identifying categories of capital adequacy based on both
leverage ratio and risk-based capital.\8\
---------------------------------------------------------------------------
\8\ The FBA's currently define a ``well-capitalized'' institution
as having Tier 1 capital of 5 percent, ``adequately capitalized'' is
set at 4 percent, ``under-capitalized'' at less than 4 percent,
``significantly undercapitalized'' at less than 3 percent, and
``critically under capitalized'' at less than 2 percent Tier 1 capital.
---------------------------------------------------------------------------
PCA and its leverage ratio are based on institution-wide levels,
rather than individual asset risks as prescribed by Basel II.
Notwithstanding Basel II's focus on risk sensitivity, an institution
with a concentration of low risk assets will be constrained by the
leverage ratio; as a result, its capital will not be risk sensitive.
Conversely, the leverage ratio may impose no restraint on a relatively
high-risk institution, yet that institution would be constrained,
presumably, by an effective risk-sensitive standard. Thus, a risk-
insensitive leverage ratio works against a financial institution's
investment in low-risk assets.
Today's expanding universe of off-balance-sheet activity also goes
untouched by existing leverage requirements. Thus, a regulatory capital
system with a leverage ratio not sensitive to risk operates as the
principal binding capital constraint on financial institutions, rather
than a backstop measure. Such a system may perversely motivate low
credit risk lenders to pursue riskier lending.
Along with other prudential safeguards, leverage is an important
capital buffer. OTS remains firmly and unequivocally committed to
maintaining an appropriate leverage ratio that is sufficiently rigorous
and also flexible enough to address the unique operating
characteristics of all types of lenders.
Conclusion
OTS supports the goals and objectives of Basel II, and we are
committed to implementing a more risk-sensitive capital framework for
all our regulated institutions. While it is important that the United
States continue to move forward on Basel II, we should proceed in a
cautious, well-studied, and deliberate manner. The revised timeframe
for Basel II is consistent with this goal; however, it is critical that
all interested parties--including the industry, Congress and the
regulators--continue an active, open, and thorough dialogue regarding
Basel II. We will continue to work together with the Members of this
Committee, the other FBA's, and with our international colleagues on
these issues.
We stand ready to make any necessary adjustments to ensure that
domestic implementation of Basel II, in conjunction with the
modernization of Basel I-based capital standards, appropriately enhance
capital adequacy and risk management in the United States. Most
importantly, we will continue to seek assurance, Mr. Chairman, that
these efforts not yield unintended consequences to our U.S.
institutions.
Thank you, Mr. Chairman and Ranking Member Sarbanes, for holding
this important hearing, and for the continued interest and hard work of
you and your colleagues and staffs on these important issues. We will
be happy to provide any additional information that you may require
regarding the ongoing Basel II and Basel I rewrite processes.
----------
PREPARED STATEMENT OF L. WILLIAM SEIDMAN
Former Chairman, Federal Deposit Insurance Corporation
Former Chairman, Resolution Trust Corporation
November 10, 2005
Mr. Chairman and Members, thank you for inviting me to this
important hearing. I have not been before this distinguished Committee
for a long time, so please excuse me if I am a little rusty.
Basel II has been billed as an important safety and soundness
initiative that is needed to correct the deficiencies of Basel I. Basel
II will force banks to hold capital, it is said, for the hidden and
undercapitalized risks Basel I allowed them to take.
I was a part of the delegation at Basel I, which had as its
principal objective creating common capital minimums for banks around
the world. It has achieved that objective. Now Basel II wishes to
improve that minimum standard through use of economic models to
evaluate risk.
This is a good sentiment--however, unfortunately, as the agencies
announced in April 2005, the results of a quantitative impact study
(QIS-4) of proposed Basel II showed material reductions in minimum
required capital for the population of U.S. institutions that submitted
their capital estimates for the study. The study showed a 15.5 percent
decline in the average bank's minimum required capital and more than
half of the participating banks posted declines in excess of 26
percent. Subsequent studies produced by the Federal Deposit Insurance
Corporation suggest that minimum risk-based capital standards at many
Basel II eligible banks could fall far below the leverage standards set
by prompt corrective action (PCA) for well-capitalized banks. The
original intent of Basel II was to more closely align minimum
regulatory capital with actual risk, not to materially reduce overall
capital levels within the banking system.
The QIS-4 study also showed a significant amount of dispersion in
minimum capital requirements across institutions and across different
portfolio types. Some dispersion in capital results may be expected
because the Basel II framework allows for a significant degree of
subjectivity in developing inputs to the capital risk formulas.
However, the substantial variance shown in QIS-4 results for categories
that should be fairly similar among different institutions--retail
lending for example--raise questions as to the prudence of adopting a
capital regime that results in such a wide range of perceptions of risk
across banking institutions.
It is far from certain that the risk measurement systems in banks
have become so precise that they can operate safely at reduced capital
margins mandated by Basel II and calculated by banks in the QIS-4
study. Compared to Basel I, the Basel II capital standard may better
align capital with the risks taken by banks, but the new standard is
far from foolproof. The Basel II standard omits some significant risks
faced by banks. Interest rate risk was a key part of the early stages
of banking and the thrift crises of the 1980's, yet the Basel II
standards do not cover interest rate risk. The Basel II capital
standard is also based on estimates from the banks' own data. Make no
mistake, these are estimates and are subject to errors. Moreover, Basel
II uses these imperfect estimates in a capital formula that, while
complicated and derived from financial theory, is almost certainly an
oversimplification of the actual risks taken by banks. So can we be
reasonably sure that the Basel II standards would provide sufficient
capital for banks? Not in my judgment.
We learned a number of lessons during the bank and thrift crisis
years of the late 1980's and early 1990's. One of these lessons is that
even sophisticated models can prove to be wrong when faced with
unanticipated volatility and changing conditions that invalidate bank
model assumptions. Models calibrated to fit performance in good times
often will perform badly when market conditions deteriorate. Model
risk, an unavoidable by-product of sophisticated risk measurement
practices, underscores the importance of retaining capital safeguards,
such as the leverage ratio and prompt corrective action minimum equity
capital standards.
While the details of the agencies' QIS-4 analysis have not yet been
made public, there are some publicly disclosed Basel II minimum capital
level results that can be compared to existing regulatory benchmarks,
and these comparisons should raise concerns. My specific concerns focus
on minimum capital requirements for residential mortgages. You are
aware no doubt that some prominent officials in the banking regulatory
community have been vocal about the need to raise minimum regulatory
capital requirements for the housing GSE's, Fannie Mae and Freddie Mac,
which currently are about 2.5 percent for mortgages held on the GSE
balance sheets. There seems to be a growing consensus in these halls
that the safety and soundness of the financial system may be enhanced
by increasing the minimum capital position that GSE's are required to
maintain. At the same time that bank regulators have weighed in on GSE
regulation, a recent study by the Federal Reserve Board \1\ stated that
the Basel II minimum capital requirement for GSE conforming mortgages
held in banks would be below 50 basis points. I fail to understand how
a bank regulator may reconcile these two positions. Common sense
dictates that if 2.5 percent is too low for mortgages held in the
GSE's, then it must also be too low for mortgages held in banks.
---------------------------------------------------------------------------
\1\ ``An Analysis of the Potential Competitive Impacts of Basel II
Capital Standards on U.S. Mortgage Rates and Mortgage Securitization,''
by Diana Hancock, Andreas Lehnert, Wayne Passmore, and Shane M.
Sherlund (HLPS).
---------------------------------------------------------------------------
I am concerned by reports that Basel II may result in significantly
lower risk-based capital requirements for many of the largest U.S.
banks. The Basel II process requires banks to calculate expected
defaults and losses from these defaults. Regulatory guidance
notwithstanding, it seems possible that Basel II banks' may use their
experience over the past few very good years to make these assessments
and this may lead a regulatory capital minimum that is too low in the
case of bad times. My experience taught me that the minimal equity
ratios prior to the banking crisis, then around 4 percent in large
banks, were grossly inadequate for the problems that followed. At least
two of our largest banks would have failed if there had been the
slightest reduction in capital minimums. I do not know where Basel II
capital levels will actually be, but I fear that using the
extraordinarily benign recent period to calculate future risk will
result in banks that are systematically undercapitalized when troubles
arise.
To their credit the regulators are saying they are not comfortable
with these QIS-4 results and would not allow banks to operate at these
levels of capital. The regulators indicate they will keep the leverage
ratio and get around to fixing the Basel II formulas later.
Maybe they will. Meanwhile, we are going to see a lot of screaming
from U.S. banks because they are putting systems in place--at great
expense--that will be, for most of them, irrelevant to determining
their overall capital. U.S. financial institutions believe they are
already constrained by our leverage ratio, unlike their foreign
counterparts. The regulators do not believe low capital is a
competitive advantage, quite the opposite. Our banks are not only well
capitalized, but they also earn record profits year after year. But
those banks who do measure performance by their current return on
equity probably are going to believe we have given the rest of the
world's banks a huge advantage.
That is why I fear Basel II is putting in place a dynamic that
cannot be controlled, and will ultimately lead to significant reduction
in the capital of our banking system, and significantly increase the
cost of the Federal safety net. I do not doubt the good intentions of
the regulators today who say they will keep this process under control.
However, they cannot control the actions of future agency heads. In
addition, those agency heads are going to be under tremendous pressure
because of what the regulators are doing today.
----------
PREPARED STATEMENT OF WILLIAM M. ISAAC
Former Chairman, Federal Deposit Insurance Corporation
and Chairman, The Secura Group
November 10, 2005
Mr. Chairman and Members of the Committee, it is a distinct honor
and privilege to appear before you today to discuss the implications of
the proposed Basel II capital accord currently under consideration by
U.S. regulators and those in other major countries.
I will explain my rationale below, but let me say up front that I
have grave reservations about the proposed Basel II capital regime. I
believe it carries the potential to do enormous harm to the U.S.
banking system, which is the strongest, most profitable, and most
innovative banking system in the world.
No regulator, legislator, or banker who lived through the banking
crisis of the 1980's in this country can ever forget the lessons of
that period--one of which is that capital very much matters. U.S.
regulators and bankers have spent the last quarter century building the
best banking system in the world. We should not risk everything we have
worked so hard to create by relying on theoretical risk models to
determine the appropriate level of capital for our banks.
International regulators developed risk-based capital rules (called
Basel I) in the late 1980's to apply to all banks. They offered two
primary justifications for Basel I: (a) capital rules should apply
uniformly to banks throughout the world in order to level the playing
field, and (b) capital requirements should correspond to the level of
risk in individual banks. One could hardly quarrel with these
objectives, although it was far from clear that Basel I was the most
appropriate way to achieve them.
Despite its stated objectives, Basel I has not come close to
fostering parity in worldwide capital standards. At year-end 2004, the
20 largest banking companies outside the United States had a median
capital-to-assets ratio of 3 percent compared to the 6 percent median
ratio (excluding goodwill) of the 20 largest U.S. banking companies.
The capital level of these foreign banks was below where the
capital of large U.S. banks was in the late 1970's, before U.S.
regulators began a major push to increase large bank capital levels.
Only 2 of the 20 largest foreign banks met minimum U.S. standards at
year-end 2004.
Despite their overwhelming use of leverage, the median return on
equity of these large foreign banks was comparable to the median return
of major U.S. banks. In short, large U.S. banks have much stronger
balance sheets and enjoy much higher operating margins than their
foreign counterparts.
Most career regulators in the United States were skeptical about
Basel I from the beginning. They were reluctant to place too much faith
in rigid formulas and doubted the claims that Basel I would bring about
international parity.
As a result of these concerns, U.S. regulators overlay on Basel I a
minimum capital to assets standard (that is, leverage ratio) to guard
against potential flaws in Basel I and to prevent capital from falling
too sharply. Moreover, U.S. regulators retained the ability to override
the Basel I standards and demand more capital whenever warranted. I
have no doubt that the comparative strength of the U.S. banking system
today is due in no small part to the fact that U.S. regulators did not
succumb to the pressure to lower their regulatory standards to
international norms.
Almost from the day Basel I was adopted, the Basel Committee at the
Bank for International Settlements began agitating for a more
``sophisticated'' version of the regime, which has come to be known as
Basel II. The rationale behind Basel II is that the largest banks in
the world are too complex for the relatively simple capital standards
of Basel I. Champions of Basel II believe the largest and most
sophisticated banks have the ability to construct models to assign
capital to their various risk exposures and that these models will be
superior to any rigid tests the regulators might apply.
Basel II is very controversial, as one might imagine. My major
concerns about Basel II are that:
<bullet> Basel II is based on inadequate and unreliable data. No large
bank has detailed information on losses going back as much as 10
years. It is virtually impossible to build reliable models with
such a paucity of information, particularly when the decade that
the available information covers is the most prosperous in banking
history. Moreover, most of the large banks barely resemble what
they looked liked 10 or 15 years ago. They are amalgamations of
countless mergers of disparate cultures, businesses, and
information systems. How one could build a reliable model based on
the performance of a business that did not exist 15 years ago is
difficult to fathom. Adding to the confusion, Basel II expects
banks to build models for ``operational risks.'' There are two
categories of operating risks: (a) those you can predict, price,
and insure against, and (b) those that are not predictable. Losses
of the first type have never been a systemic problem in banking,
and losses in the second category cannot be modeled.
<bullet> Basel II will be used to reduce large bank capital ratios and
either place smaller banks at a competitive disadvantage or force
regulators to lower smaller bank capital ratios. Neither option is
acceptable public policy. The regulators conducted a Qualitative
Impact Study (QIS-4) on Basel II earlier this year, which produced
some very disturbing results. QIS-4 indicated that Basel II would
allow capital levels to drop by at least 26 percent at half of the
big banks, some falling by as much as 50 percent. The regulators
responded to this, in part, by rushing out with proposed revisions
to Basel I, which will presumably allow small-bank capital to
decline. Alice in Wonderland would love it, but no self-respecting
bank supervisor should. I do not believe a compelling case has been
or can be made for reducing capital in the banking system.
<bullet> Basel II is so complex it cannot be adequately understood by
senior bank managements, boards of directors, regulators, or the
public. I have been in the banking world all of my adult life and
served at the FDIC for nearly 8 years. Moreover, I have served on
the boards of three financial institutions. I think I know
something about what it takes to run a bank right and why banks
fail. The good ones have very good and experienced managements,
strong and independent boards of directors, and a healthy respect
for what can go wrong. They diversify their risks, put in place
strong control systems, maintain solid balance sheets, and are
always asking themselves what will happen if their assumptions are
flawed. Running a successful bank (and successfully regulating
banks) is an art that uses modeling and other forms of science as
tools. Basel II employs exceedingly complex models (constructed
largely by economists and mathematicians who have never made, much
less collected, a loan), which very few people in any bank or any
regulatory agency will understand. Making matters worse, Basel II
will likely foster complacency and a false sense of security, as
some bank managers and boards of directors and even some analysts
place unwarranted reliance on the models. If we allow Basel II to
elevate questionable science well above the art of management,
someday somewhere we almost certainly will pay a big price. We need
look no further than the debacle at Long Term Capital Management in
1998 to see what mischief can be caused by ``brilliant''
mathematicians and Nobel Prize winning economists who are given
free rein to make huge bets based on their models. The Federal
Reserve found it necessary to pressure 14 banks and brokerage firms
to invest nearly $4 billion in new equity to prevent LTCM's
collapse and avert a possible panic in the worldwide financial
markets.
All models, of necessity, look backward--that is, they use
historical data to predict future events. They are accurate, most of
the time, if properly constructed and manipulated. But they can result
in spectacular failures when they are poorly constructed, use
inadequate data, or rely on false assumptions.
Large banks pronounced with great certitude in the 1970's and early
1980's that loans to less developed countries were riskless because
sovereign nations could not afford to default and thereby lose their
access to international credit markets. Had Basel II existed then, I
suspect the risk weighting assigned to sovereign debt would have been
close to zero.
I participated in serious and urgent interagency planning in 1984-
1985 on how to handle the potential simultaneous collapse of the
largest U.S. banks in the event of widespread defaults on sovereign
debt. Later in the 1980's Citibank, alone, charged off some $3 billion
of LDC loans in one fell swoop. So much for theories and models.
In important ways, Basel II is deja vu. The regulators developed a
three-tier system for capital adequacy in the 1970's. Community banks
were required to maintain capital equal to at least 8 percent of
assets. Regional banks were allowed to go as low as 6 percent on the
theory they were better managed and more diversified, and their size
made them less vulnerable to catastrophic fraud losses. Money center
banks had no minimum standard--they were OK so long as they did not get
out of line with their peers.
Believe it or not, the theory advanced by both banks and top
regulators was that capital was not particularly relevant in banks as
large and sophisticated as the money center banks. The typical money
center bank had capital in the range of 3 percent during this period,
with a couple falling below 3 percent.
The FDIC, during my tenure, regarded it as fundamentally unfair to
require smaller banks to maintain more capital than larger banks. Of
even greater concern was the increasingly heavy reliance of larger
banks on volatile money market sources of funding. We were concerned
that at the first sign of trouble this money would flee, rendering a
bank helpless. We believed that a strong balance sheet was of even
greater importance in banks with volatile funding.
The debate among banks and regulators was heated, not unlike today.
The theories went out the window in 1984 when Continental Illinois, the
eighth largest bank in the country, with one of the lowest capital
ratios, lost nearly half of its funding virtually overnight.
It was all the FDIC and Federal Reserve could do to stem the
outflow and keep the infection from spreading to banks throughout the
world. Together, these two agencies advanced in the range of $20
billion to Continental Illinois, and the FDIC took the unprecedented
step of guaranteeing that no creditor of Continental Illinois would
suffer any losses or even any delay in receiving its money.
Capital was no longer a theoretical exercise. The regulators
quickly agreed that no bank, no matter how seemingly strong and well-
run, would be allowed to maintain less than 5 percent tangible equity
to assets. Banks would be required to maintain capital above that level
based on a case-by-case evaluation of risk through the bank supervisory
process.
Basel I was developed a few years later to quantify capital
measures to a greater extent. While I have never been a fan of Basel I,
at least U.S. regulators maintained a minimum capital to assets
standard and Basel I did not discriminate based on the size of the
bank.
Some argue that a U.S.-imposed capital floor places U.S. banks at a
disadvantage vis-a-vis foreign banks. This argument was advanced in the
1970's and 1980's, particularly with respect to the Japanese banks,
which were growing with reckless abandon and comprised nearly all of
the 10 largest banks in the world. We hear little about the Japanese
banks today, except that they are a drag on the Japanese economy. They
operated with little capital, had very low profits, and pursued growth
for growth's sake. This business model cannot be sustained
indefinitely.
I have not found a single professional bank supervisor who is
enthusiastic about Basel II. For that matter, I have not found a single
bank CEO who is enthusiastic about Basel II. Indeed, I have spoken with
large bank CEO's who complain that Basel II is inferior to their own
procedures, which give great weight to seasoned bankers using their
judgment. They will implement Basel II if required to do so by their
regulators, but they intend to continue to rely on their own models and
procedures, which have withstood the test of time through various
business cycles.
Capital regulation, in my judgment, should be simple and easily
understood. It is foolhardy to adopt a capital regime that will be
virtually impossible for senior managements, boards of directors,
regulators, and market participants to understand.
There is nothing wrong with the current capital regime in the
United States that cannot be fixed with some relatively simple
modifications to Basel I. Moreover, it is critically important that the
leverage ratio and other capital tests be maintained at their current
levels.
This is not to say that the Basel II exercise has been for naught.
Basel II-type models no doubt provide useful information to the
managements of banks and their regulators. Banks should be encouraged
to develop and improve such models. But the last thing U.S. regulators
should do is engage in a ``competition in laxity'' with supervisors in
other countries by lowering U.S. capital standards to international
norms.
I thank you again for providing this public forum on one of the
most profound public policy issues currently confronting our Nation's
banking system.
----------
PREPARED STATEMENT OF GEORGE G. KAUFMAN
Co-Chair, U.S. Shadow Financial Regulatory Committee
and John F. Smith Professor of Finance and Economics
Loyola University Chicago
November 10, 2005
Mr. Chairman, it is a pleasure to testify before this Committee on
the public policy implications for the health and safety of the banking
system and the U.S. macroeconomy of the proposed Basel II capital
standards. My bottom line is that Basel II represents only minor
improvement over Basel I as a public policy tool for enhancing
financial stability in the United States and has the potential for
weakening the more comprehensive structure that is currently in place
in the United States.
The Basel proposals will apply to banks (depository institutions)
in both the United States and many other countries in order to achieve
greater harmonization in capital standards among countries. If adopted
in the United States, the proposal will be incorporated within our
system of structured early intervention and resolution (SEIR), which
includes both prompt corrective action (PCA) and a legal closure rule
at positive capital at which point a bank is placed in receivership.
But most other countries do not have such a system effectively in place
and Basel needs to be evaluated on its own merits. For my remarks, I
will focus primarily on the United States, but periodically refer to
other countries.
To evaluate Basel II objectively, it needs to be compared to an
alternative structure for enhancing bank stability. Such an appropriate
alternative is the system of SEIR, which was introduced in the United
States in FDICIA in 1991 in response to its banking crisis of the
1980's. Unfortunately, much of the discussion of Basel II in the United
States has neglected to incorporate the existence of this structure.
Basel and SIER/PCA have different histories. Each should be
evaluated on the basis of what it was initially intended to do and not
necessarily as a substitute for the other. Unfortunately, the different
underlying histories often appear to be forgotten and each tends to be
evaluated on a basis for which it was not primarily designed.
Currently, in the United States, this confusion may be setting up a
battle with possible serious adverse consequences for long-term
financial stability. For the rest of the world, the proposed reliance
on Basel may be taking attention away from developing more effective
means of enhancing financial stability.
Basel (the Basel Committee on Banking Supervision, which meets at
the facilities of the Bank for International Settlements in Basel,
Switzerland) developed in the 1970's from a need to facilitate the
sharing of information among bank regulators and supervisors in
different countries on internationally active banks operating in their
countries, which were expanding rapidly (Herring and Litan, 1995). In
large measure, this need was initially motivated by the large
international costs related to the collapse of the medium-sized
Herstatt Bank in Germany, that operated heavily in the foreign exchange
market and lost. The bank was legally closed by the
German authorities at the end of the business day in Germany after it
received payments from its foreign counterparties, many of whose
business day closed later, for foreign exchange transactions, but
before it paid these counterparties at the end of their business day.
Ironically, the international repercussions reflected primarily a
regulatory and not a market failure. The timing of the closure
effectively shifted losses from German depositors and the German
deposit insurance agency to banks outside Germany. Basel's objectives
were expanded in the 1980's to developing international capital
standards to promote both safety among large internationally active
banks in light of large losses from LDC lending and competitive
equality across countries with respect to capital ratios. The latter
was aimed particularly at Japanese banks, which were expanding their
foreign market share rapidly on perceived very low capital bases. This
was viewed as giving them a competitive advantage.
The capital standards constructed in Basel I (1988) effectively
resembled guidelines at the time for ``best practices'' in bank capital
management, in particular with respect to incorporating credit risk
exposures. Individual assets were weighted by one of four risk
classification weights (buckets) and summed. The resulting risk-
weighted assets were then divided into capital to obtain risk-based
capital ratios. The minimum suggested overall capital ratio for a bank
was set at 8 percent, which most banks were then able to satisfy. But
the scheme provided no provisions for enforcing this capital standard,
replenishing shortfalls, or resolving an insolvent institution at least
or even low cost. Thus, the usefulness of the structure for public
policy was limited, although it did increase the sensitivity and
knowledge of bank managers and bank regulators to measuring and
managing risk.
In contrast, in the United States at the same time, emphasis was
not on developing best practices schemes for banks but on developing
public policy measures to prevent a reoccurrence of the large-scale
failure of thrift institutions and commercial banks in the 1980's,
which imposed high cost on their insurance agencies and, for thrift
institutions, also on the taxpayers (Benston and Kaufman, 1994). The
structure was designed to turn troubled institutions around before
insolvency, primarily through
recapitalization or merger with healthier institutions, and, failing
that, as a last resort to legally close and resolve them at lowest cost
to the insurance agency and potentially taxpayers. This is to be
achieved through increasing both market discipline and regulatory/
supervisory discipline.
Market discipline was enhanced by increasing the number of de facto
at-risk claimants through severely curtailing the use of the misnamed
``too-big-to-fail'' (TBTF) policy--which, in reality, dealt not with
failure but with protecting de jure uninsured claimants (Kaufman,
2004b). TBTF was transformed into a harder to
invoke ``systemic risk exemption'' (SRE). Supervisory discipline was
enhanced by establishing SEIR with PCA, a legal closure rule at
positive capital, and least cost resolution with enforcement provisions
that affect both the regulators and the banks.
Capital is the primary, but not the only, measure that triggers
regulatory sanctions on troubled institutions, which are structured
both to resemble the sanctions typically imposed by the market on
floundering firms in nonregulated industries and to become
progressively harsher and more mandatory as the condition of the bank
deteriorates, culminating in legal closure and receivership. These
provisions make it more difficult for regulators to forebear and force
speedier actions to replenish capital if it declines below minimum
target levels. Capital is viewed as the owners' funds that they ``have
to play with''. The less capital, the more restrictive the game becomes
and as capital approaches zero, the legal closure rule is invoked, the
game is declared over, and least cost resolution is commenced.
The usefulness of capital to absorb losses in this framework is
related to the size of the bank. The usual measure of firm size is its
total assets. Capital divided by the bank's total assets is the so-
called leverage ratio. Under FDICIA, PCA specifies three capital
ratios--tier 1 (basically equity) leverage ratio, tier 1 Basel risk-
based ratio, and total capital Basel risk-based ratio--and five capital
tranches or zones ranging down from ``well capitalized'' to
``critically undercapitalized.'' (The major provisions are summarized
in Table 1.) The minimum capital levels necessary to be classified
``adequately capitalized'' are set at 4 percent, 4 percent, and 8
percent for the three capital measures, respectively.
The development of the Basel II was primarily motivated by a desire
to correct two perceived weaknesses in Basel I. In the process, the
number of pillars was expanded from one to three. The recommended
changes are intended to:
<bullet> Enhance the accuracy of risk-based capital (Pillar I), and
<bullet> Introduce means for enforcing minimum regulatory capital
ratios (new Pillars II and III).
But both improvements are relatively weak.
Risk-based capital determined by market weights and forces is
necessary for managing banks efficiently, but it is difficult for
regulators to replicate accurately. The revised capital requirement in
Pillar I attempts to do so for credit risk primarily for large banks.
In the United States, with the approval of their regulators, these
banks will be permitted to use their own internally generated credit
risk ratings (including probability of default and loss if default) to
compute their risk-based capital from a model provided by the
regulators. (I will not comment on the Basel proposal for smaller banks
or the recently modified proposal by U.S. regulators for most U.S.
banks. Many of these banks maintain such high capital ratios that they
will be effectively unaffected by these plans.) But, as has been
discussed in a number of
recent statements by the Shadow Financial Regulatory Committee, by
increasing complexity, Pillar I does not necessarily make the
regulation more accurate (Shadow 2000, 2002, and 2003). Increased
complexity is likely, however, to both increase compliance costs and
reduce understanding, particularly by the bank CEO, board of directors,
and possibly even the CFO and by bank supervisors. Simplicity often
trumps complexity in producing desirable outcomes through greater
understanding. Furthermore if the risk weights are incorrectly
selected, as is likely, the opportunity for gaming by the banks
increases. Bank management and supervisors may be outgunned by highly
technical Ph.D. model builders. The revision also introduces capital
charges on a bank's operational risk for the first time.
As noted, Basel II introduces two new pillars--Pillar II:
Supervisory Review and Pillar III: Market Discipline--to increase the
public policy usefulness of the structure (Kaufman, 2004a). However,
Pillar II contains few specifics. It is intended to supplement Pillar I
in determining appropriate capital for credit risk exposure and to
expand regulatory concern to interest rate risk. But it focuses
primarily on general principles and does not consider the wide
variation in supervisory competence across countries. Most importantly,
Pillar II contains neither mandatory PCA type measures to replenish
capital and turn troubled institutions around before insolvency nor a
legal closure rule and least-cost resolution provisions to guide the
supervisors' actions. Much talk and little required action.
Pillar III is not really about market discipline, but rather about
creating transparency and disclosure. Mandatory disclosure would not be
as necessary if there were more truly at-risk claimants, who would be
expected to demand more transparency and exert more market discipline.
Indeed, Pillar III would be far more
useful in enhancing market discipline if it focused on increasing de
facto at-risk claimants. This could be achieved by reducing the
likelihood of invoking TBTF/SRE through making it more difficult to do,
as in the United States since FDICIA, and by encouraging or requiring
banks to issue truly at-risk subordinated debt (Shadow Financial
Regulatory Committee, 2000). Nor does Pillar III introduce any cost-
benefit criteria to evaluate whether the benefits of each additional
item to be disclosed exceeds the costs of collecting and processing it.
As a result, Basel II provides only partial and flawed improvements
over Basel I as a tool for public policy to achieve the goal of
enhanced financial stability. Pillar I remains basically a ``best
practices'' guide for internal bank management and not a public policy
instrument.\1\ Indeed, while there is substantial empirical evidence of
a negative relationship between leverage ratios and bank insolvency,
there is no such evidence between risk-based capital ratios and bank
insolvency (Evanoff and Wall, 2001). In no other industry do analysts
compute risk-weighted assets or risk-based capital ratios for
individual firms. But they do compute and investors use leverage
ratios. Risk-weighted assets are an inferior scaler to total assets to
gauge how much capital is available to a bank before the value of its
assets declines below the value of its liabilities and it becomes
insolvent. In sum, Pillars II and III are vastly inferior to PCA/SEIR
with a strong legal closure rule at positive capital to minimize both
the number and cost of bank failures.\2\
---------------------------------------------------------------------------
\1\ Indeed, a recent article argues that the internal ratings model
underlying the regulation is outdated even as it is being proposed
(Thomas and Wang, 2005).
\2\ A similar conclusion appears to have been reached by Goodhart
(2004).
---------------------------------------------------------------------------
Unfortunately, Basel II may be on the verge of causing major
mischief in the United States that could weaken financial stability
over the longer-term for large banks intending to use the advanced
internal ratings approach. It appears that the 4 percent risk-based
tier I capital requirement ratio can be achieved under Basel II for
many of these banks with lower capital than currently is required both
under Basel I and is required to be classified as an ``adequately
capitalized'' bank according to the 4 percent tier I leverage ratio.\3\
Consequently, the leverage ratio is likely to become the binding
constraint for these banks and prevent a reduction in required
regulatory capital. The FDIC has recently concluded that ``U.S.
policymakers will be confronted with a choice between ignoring the
results of Basel II or substantially weakening the PCA requirements''
(FDIC, 2004). Although almost all U.S. banks currently maintain capital
ratios at well above the regulatory requirements--indeed, the FDIC
reported that bank equity capital ratios at midyear 2005 climbed to
their highest level since 1939, more than twice the ratio required to
be adequately capitalized--some banks appear to be lobbying U.S.
regulators to lower the numerical threshold capital leverage ratio to
qualify as adequately capitalized to below 4 percent, say to 3\1/2\
percent or lower. Congress in FDICIA delegated to the appropriate
Federal regulators the setting of the numerical thresholds for all
tranches but the minimum critically undercapitalized closure trigger of
2 percent equity capital. Some large banks are also arguing that the
current leverage ratio requirements put them at a disadvantage with
their competitors in the rest of the world, who are not subject to
these ratios.
---------------------------------------------------------------------------
\3\ This occurs because 4 percent times a bank's risk-weighted
assets yields a tier I capital number which, when divided by the bank's
total assets, may be less the 4 percent.
---------------------------------------------------------------------------
For U.S. regulators to cave in to such pressure in order to have
Basel II adopted would be a big and costly mistake both for the U.S.
macroeconomy and for the banks themselves. Considerable evidence
suggests that even a 4 percent equity leverage ratio is lower than that
maintained by almost all domestic nonbank competitors of banks, who are
not similarly regulated nor covered by a safety net (Kaufman, 1992 and
Kwast and Passmore, 1999). When industry leverage ratios decline below
6 percent bank failures increase, particularly when the economy is in a
recession. Indeed, on the whole, there is a negative relationship
between leverage ratios and defaults in all industries (Molina, 2005).
With respect to individual large banks, there is no evidence either
that equity capital increases the bank's overall cost of funds or that
there is an inverse relationship between bank capital ratios and bank
return on either assets or equity.
A time series analysis for U.S. banks shows a weak positive
relationship between bank capital and profitability (Berger, 1995). A
cursory cross-section analysis across the world's largest banks also
shows a positive relationship between capital and profitability since
the 1980's. Recently, United States, United Kingdom, Australian, and
Spanish banks have both high capital ratios and high profitability,
while German, Swiss, and Japanese banks have both low capital ratios
and low profitability, although some adjustment may need to be made for
the possibility of simultaneity in the direction of causation (Table
2). Nevertheless, this helps to explain why capital ratios actually
maintained by U.S. banks are considerably higher than the regulatory
requirements. They are signaling strength to both depositors and
borrowers. Among the largest 1,000 banks in the world in 2003, U.S.
banks accounted for only 15 percent of aggregate assets, but 22 percent
of aggregate tier 1 capital and fully 37 percent of aggregate pretax
profits. For Japan and the EU countries, capital accounted for a lower
percentage of aggregate capital for all 1,000 banks than did their
assets as a percent of aggregate assets and profitability even less.
For the remaining countries, capital was at least as high a percentage
of the aggregate as were assets and was about the same percentage as
profitability (Table 3).
The results of the recent quantitative impact studies for large
banks by the U.S. regulators raise concerns not only because they show,
on average, lower regulatory capital requirements than currently but
also because they show a large variance among individual banks. This
suggests that the individual bank models generating credit risk weights
may have flaws in construction and that they are very sensitive both to
the quality of data available for each bank and to the sample time
period over which they are empirically tested. The models may not yet
be ready for prime time!
In sum, adoption of Basel II for large banks in the Unites States
is likely to have little effect on the banks and the economy if the
current numerical threshold values for the leverage ratios for
adequately and well-capitalized banks in PCA are not reduced and if
market forces operate to maintain current capital ratios. If, however,
because the new Basel II risk-based requirements can be met, on
average, with lower leverage ratio numbers, the numerical definitions
for adequately and well-capitalized banks were reduced, there are
likely to be adverse longer-term consequences for both the banks
themselves and the economy as a whole. The integrity of SEIR and PCA
should not be compromised for the sake of harmonizing bank capital
standards across countries.
Adoption of Basel II by itself in other countries could undermine
their adoption of better public policy structures and thereby increase
both the likelihood and costs of financial instability. It is time,
therefore, for these countries to reconsider the benefit-cost tradeoff
of Basel versus a U.S.-type of structure resembling SEIR and PCA. But
the Basel process has not been totally negative. It has greatly
improved the measurement and management of risk by both bankers and
regulators and thus enhanced financial stability worldwide. Basel
should be maintained as an ongoing process to develop ever better bank
best practices schemes for internal management purposes, but it should
not be halted and put in place. It is the process, not the end result,
that will provide the major benefits.
<GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT>
PREPARED STATEMENT OF DANIEL K. TARULLO
Professor, Georgetown University Law Center
November 10, 2005
Mr. Chairman, Senator Sarbanes, I appreciate your invitation to
testify today. I am currently Professor of Law at Georgetown University
Law Center and the Nomura Visiting Professor of International Financial
Systems at Harvard Law School. I teach, among other things, Banking
Regulation and International Economic Law. At present I am at work on a
book on Basel II. As you know, I held several economic policy positions
in the Clinton Administration, ultimately as Assistant to the President
for International Economic Policy. I testify today purely in my
individual capacity as an academic, with no client interests or
representation.
Basel II would, if implemented, represent the greatest change in
the regulation of commercial banks since the early 1980's. The savings
& loan debacle that followed those changes is a cautionary tale of the
potential for unintended and unanticipated consequences from any major
regulatory change. Even policies that eventually prove worthwhile can
entail significant transitional problems. In the case of the advanced
internal-ratings-based (A-IRB) approach of Basel II, there are major
additional grounds for concern: The complexity of the rules, the opaque
manner in which they will be implemented, the absence of reliable
information on the impact those rules will have on capital levels, and
uncertainty on how the new regime will affect global financial
stability.
These and other difficulties have stretched the Basel II process
years beyond the original target date for completion. They have also
bedeviled efforts by bank regulators to implement into U.S. law the
final Basel II rules, as released in June 2004. In late September, the
agencies delayed once again their timetable for implementation. In
their notice of delay, the agencies expressed their intention to issue
a notice of proposed rulemaking early next year. My principal
recommendation is that the agencies not proceed with an implementation
regimen unless and until they are able to answer more convincingly the
basic questions surrounding the impact of Basel II--most importantly,
its effect upon minimum regulatory capital levels.
Because Congress has previously given the banking agencies broad
authority to regulate bank capital levels, no legislation is needed to
implement this international arrangement. But, as the regulators have
already been responsive to the concerns of banks throughout the Basel
II process, one would hope they will be at least as responsive to a
considered request for caution by Congressional committees of relevant
jurisdiction.
While delay will understandably upset those financial institutions
that have
prepared for Basel II in anticipation of substantial reductions in
their capital requirements, there is no compelling reason to brush
aside the important unanswered questions. Our banks today are sound and
they are profitable. There is no crisis requiring action in the face of
incomplete information. While it is certainly important to encourage
large banks to adopt the best available methods for risk management,
this is insufficient justification for a leap into the unknown on
capital requirements. As I explain in the next section, capital
requirements have become so central to prudential bank regulation that
proposals for significant change should be subjected to the full
scrutiny and debate that would accompany any major modification of
government policy.
In the remainder of my testimony I will first review the key role
played by capital requirements in prudential bank regulation, next
identify the goals that the Basel Committee has itself set for the
revised Accord, and then discuss my doubts that Basel II will realize
these goals. Finally, I will elaborate my reasons for counseling
caution and suggest some steps this Committee might take to assure that
Basel II does not compromise the safety and soundness of large banks,
both in the United States and abroad.
The Importance of Capital Regulation
Regulatory monitoring of bank capital levels has existed in the
United States since at least the early years of the twentieth century.
Although the sophistication of that monitoring evolved fairly steadily,
explicit minimum capital requirements were not imposed by U.S. bank
supervisors until the 1980's. In the intervening years, capital
requirements have become central to prudential regulation. As a result,
the complete overhaul of capital regulation contemplated in Basel II is
of greater significance to the safety and soundness of the U.S. banking
system than the Gramm-Leach-Bliley Act or, indeed, any other
legislation since the Federal Deposit Insurance Corporation Improvement
Act of 1991 (FDICIA).
Bank capital is generally thought to play three interconnected
roles. First, it provides a buffer against bank losses arising from bad
loans or any other cause. Capital reserves permit the bank to absorb
these losses without becoming insolvent. Second, capital mitigates
creditor losses if the bank nonetheless fails. Third, capital holdings
create a disincentive for banks to take excessive risks. If a bank has
low or zero capital, its owners and managers will be tempted to put the
funds available from deposits toward risky, but potentially high-
return, uses. They have literally nothing to lose. The higher the
bank's capital levels, the more they have to lose, and thus the more
prudent their lending policies should be.
These effects of capital can be found in any corporation, but the
government usually does not require minimum capital levels in
companies. We generally leave it to those who lend money to
corporations to find ways to protect themselves. Where banks are
concerned, however, there are powerful reasons to regulate capital
levels. The existence of federally insured deposits means that the
government is, in effect, one of the biggest creditors of most banks.
Since the premiums charged for deposit insurance are not in any strong
sense calibrated for the riskiness of the bank, capital requirements
serve as a substitute to protect the FDIC, and ultimately American
taxpayers.
In addition, there is a widespread perception in financial markets
that many large banks are considered ``too big to fail'' by bank
supervisors. Bank failures can in some instances lead to much greater
harm to the economy as a whole than the bankruptcy of even the largest
nonfinancial corporations. Most important, of course, is the
possibility of a systemic effect, in which the failure of one large
bank produces dislocations that reverberate through the financial
system. In the worst case, the result could be a freezing up of credit
availability and thus a shock to the economy. Creditors of banks
thought to be too big to fail may not insist on high enough levels of
bank capital because they believe that the Federal Reserve Board will
bail out a very large bank before it becomes insolvent. Capital
requirements help compensate for this form of moral hazard effect.
There is nothing particularly new in the role that regulatory
capital can play. What is relatively new is the reliance placed on
capital regulation to assure bank safety and soundness. For half a
century following Depression era reforms, banks were quite
circumscribed by legal restrictions and technological limitations. The
activities they could engage in, the interest rates they could pay
depositors, and the places where they could do business were all
severely constrained. Banks were, in an important sense, protected from
themselves, or at least from any inclination they might have to take
business or financial risks. They were also the beneficiaries of a more
or less guaranteed market for their services as financial
intermediaries between savers and users of capital.
Changes in the financial environment, such as the expansion of
capital markets and the rapid growth of money market funds, provided
vigorous competition to banks at both ends of their business. What
followed is well-known. Geographic, interest rate, and activities
restrictions were all relaxed or eliminated. Banks undertook new
activities and many took more risks in their traditional lines of
business. There ensued some fairly dramatic problems, notably the
weakening of large money center banks during the Latin American debt
crisis of the early 1980's and the S&L crisis later in the decade. If
the safety and soundness of banks were to be ensured, a new approach
was obviously needed. Congress decided upon capital regulation in the
wake of the Latin American defaults and, in FDICA, built on that
approach by mandating the important practice of prompt corrective
action. Regulators in many European countries had already implemented
various forms of risk-based capital requirements.
Today, we rely on capital regulation as the most important single
element of bank regulation. We use capital levels to determine when
supervisory intervention is required. We use capital levels to decide
when banks are strong enough to engage in the nonbanking activities
permitted by Gramm-Leach-Bliley. And, of course, we rely on capital
levels to provide a buffer against loss and insolvency from the complex
and varied activities of a large, modern bank.
This very brief survey of the role and history of capital
regulation is, I hope, sufficient to show the importance of knowing
with some certainty the impact of the proposed regulatory changes
before actually adopting them. It is never possible to predict all the
effects that a new regulatory scheme will have but, the more important
that scheme is to ensuring a social good as important as bank safety
and soundness, the more we should know before it is prudent to proceed.
The Basel II Process
The first Basel Accord on capital adequacy had two aims: First, to
increase the stability of the international financial system by
ensuring that internationally active banks had sufficient capital
cushions. This arose directly from the travails of U.S. and other banks
arising from the Latin American sovereign debt crisis. Congress had
insisted that, in exchange for the partial bail-out these banks were to
receive through IMF assistance to the defaulting sovereigns, the banks
be required to increase their capital levels. Regulators in a number of
other countries shared a concern that the capital levels of many large
banks were too low. Second, the Accord was intended to promote greater
competitive equality among internationally active banks from different
countries. American and British banks in particular believed that the
more generous safety nets for banks provided by the Governments of
Japan and France gave banks in those countries access to private
capital at a lower cost. As noted earlier, bank creditors who believe
that the borrowing bank will be saved from insolvency by its government
will demand a lower risk premium in their lending to that bank. With a
lower cost of funds, the bank will in turn be able to lend profitably
at lower rates than its competitors from countries with less generous
safety nets.
To what degree were these aims realized? It is difficult to isolate
the effects of Basel I, as implemented around the world, from economic
conditions, market developments, and other regulatory factors.
Regulators in the Basel Committee countries were already converging
around the view that risk-based capital requirements should be an
important element of bank regulation. Thus, Basel I essentially
ratified and harmonized a regulatory trend; it did not initiate a major
regulatory change. Whatever the reasons, it is indisputable that
capital levels rose following adoption of the Accord. In the
intervening years, large international banks have remained remarkably
sound, despite two international financial crises and rapid change in
financial services industries. The major exception, of course, has been
the Japanese banking system, but the origins of its problems
considerably predated implementation of Basel I.
There is little reason to believe that the Accord has, to any
significant extent, created the proverbial level playing field among
internationally active banks. We do not have the benefit of
comprehensive empirical work on this question. Still, on the basis of
some limited work on the question, several scholars have concluded
that, in all likelihood, national differences in tax, accounting, and
other regulatory measures outweigh any leveling achieved by harmonized
minimum capital standards. This is not to say that such standards have
no effect on the competitive position of banks--only that it is
probably modest compared to other factors.
The shortcomings of Basel I have been well-rehearsed throughout the
Basel II process, and I need not repeat them in detail today. Suffice
it to say that the possibilities for regulatory arbitrage prompted the
Basel Committee to undertake a substantial revision of the original
Accord in the late 1990's. Its first effort, released in 1999, was
roundly criticized by large banks and others as sharing many of the
same flaws as Basel I. Subsequently the Committee completely reoriented
its work, producing the Advanced Internal Ratings Based (A-IRB)
approach to capital regulation. In the course of this shift, the
Committee adopted an additional aim for Basel II--to bring regulatory
capital requirements more in line with the quantified credit risk
assessment and management practices of many large banks. U.S. and other
supervisory agencies on the Committee had concluded that the complexity
of banking transactions and the speed with which the creditworthiness
of counterparties may change had rendered Basel I anachronistic.
Supervisors commented that the Basel I capital requirements were based
on such crude assessments that they really did not provide a useful
picture of the risk profile of a large, complex bank.
It is important to note that, in reorienting the revision of the
Accord toward basing regulatory capital requirements upon banks'
internal credit modeling, the Basel Committee repeatedly emphasized
that its goal was neither to raise nor to lower the aggregate level of
regulatory capital in the banking system.
The regulatory agencies that constitute the Basel Committee--
including U.S. supervisors--have thus identified their purposes for
internationally harmonized minimum capital requirements: (1) stability
in the international banking system, to be achieved through minimum
capital levels that reflect (2) a closer alignment of regulatory
capital with the ``economic'' capital determined by banks themselves on
the basis of sophisticated risk management systems to be optimal for
their situations, which (3) does not result in significant changes in
aggregate levels of regulatory capital, while (4) promoting competitive
equality among internationally active banks. My misgivings about the
process arise in substantial part because, at this juncture at least,
Basel II does not stack up well against these stated aims.
Unanswered Questions Concerning Basel II
Earlier in my testimony I referred to the savings & loan debacle of
the 1980's. Let me emphasize that I do so not because I predict that
implementation of Basel II would lead to a comparable calamity, but
because that episode is an object lesson in the potential for
unintended and unanticipated consequences resulting from major
regulatory change. Even the most thorough analysis cannot eliminate
completely the possibility for such consequences. But, in the absence
of pressing
circumstances that require a response to deal with an immediate threat,
major regulatory change should generally proceed only after analysis
has yielded a reasonable degree of confidence as to the likely effects
of that change. In the case of Basel II, I believe that we are not at
that point of reasonable confidence. To the contrary, some of what we
do know gives additional grounds for concern.
Uncertainty as to Impact on Capital. The matter can be stated
simply: As we sit here today, no one knows what the impact of the Basel
II formulas will be on the regulatory capital requirements applicable
to the large banks that will adopt the A-IRB approach. Each time U.S.
banking supervisors have performed a so-called Quantitative Impact
Study (QIS) to attempt an assessment of these effects, the results have
surprised them. The latest of these exercises, QIS-4, suggested that
minimum capital levels would fall by at least a quarter, and in some
cases much more, at half the bank holding companies participating in
the study. This clearly was not the result the Federal Reserve Board
expected.
Let us be clear. I am speaking here not about unintended
consequences of regulatory change, such as shifts in market behavior of
regulated entities. I am speaking of the intended and direct
consequences of the new rules upon the amount of capital the banks must
hold. The inability to determine what minimum capital levels will
actually be required under the A-IRB approach is a major reason why, on
September 30, the banking agencies again delayed the implementation
schedule for Basel II in the United States. Yet, even as they announced
the delay, the agencies indicated their intent to move forward with
implementing regulations early next year. In an apparent effort to
reassure those of us concerned that capital levels will decline
significantly under Basel II, they also indicated that they will limit
the amount by which minimum capital can decline in the first 3 years
that Basel II is in effect. Finally, the banking agencies state that a
bank's primary Federal supervisor will decide whether to terminate the
capital floors at the end of the transition period.
At present, then, the position of the banking agencies is
essentially as follows: We do not really understand what the impact of
the Basel II formulas will be on minimum capital levels, but we think
the only way we will find out is to implement the new rules. We will
not let regulatory capital fall too quickly in the first few years.
After that, we will take a look at what happens and, if we see the
need, instruct banks to hold higher capital levels than the Basel II
formulas would require.
This plan is at best premature. It might be defensible if the
overall impact of the regulatory change were broadly understood, with
some second-order effects unclear based on the test runs. Here, though,
the banking agencies cannot say what the most basic impact of the Basel
II rules will be. The agencies may respond that they can raise increase
minimum capital levels down the line if capital appears to be dropping
too much. In the present context, this argument is not persuasive. The
very determination of the Fed, in particular, to proceed with
implementation in the face of all doubts and uncertainties leads one to
question whether the agencies would admit errors and effectively
supersede Basel II in a few years. Moreover, one must expect that the
large banking organizations would vigorously oppose any move to, as I
am sure they would put it, increase their capital requirements above
what their credit risk models indicate is necessary.
The presumption, I believe, should run in the other direction.
Implementation should not proceed until the impact and implications of
Basel II are much better understood, and have been much more thoroughly
vetted. It is hardly unreasonable for Congress and the public to expect
that there be at least one impact study that bears out the expectations
and predictions of the banking agencies before they proceed with
implementation.
Questions About Credit Risk Models. The inability to specify the
effects of Basel II formulas on regulatory capital is of even greater
concern in light of the anticipated reliance on credit risk models as
the basis for determining minimum capital levels. The concern arises
both from the technical state of credit risk models and from the
supervisory challenge in overseeing the use of those models.
Credit risk modeling is a relatively new undertaking, at least in
its comprehensive form. Any model is, of course, only as good as its
inputs. If the credit risk parameters supplied by banks are unreliable,
even a well-constructed model will give a misleading picture of actual
risk. One difficulty is the potential for intentional distortion of
model input. Even assuming good faith on the part of the banks, the
relative dearth of useful historical data is cause for concern. There
is generally less than a decade's worth of historical data available
from which to generate the values incorporated into the model.
Additionally, it is considerably more difficult to backtest credit risk
models than market risk models. While the prices of traded securities
change daily, defaults are relatively unusual events and tend to occur
in clusters because of adverse macroeconomic conditions. Because there
has been no serious recession during the last decade, there has been
little opportunity to stress test the models.
The banking agencies have acknowledged that even the credit risk
models used at the largest banks do not yet possess the
``sophistication and robustness'' that would be necessary to rely upon
them for regulatory purposes. Basel II takes account of the relatively
undeveloped state of the art of credit risk modeling by imposing its
own formulas into which only four bank-generated variables are fed
(probability of default, the bank's exposure at default, expected loss
if default occurs, and the maturity of the asset). Yet even oversight
of this process presents a new kind of supervisory challenge. The
complexity of, and differences among, bank models will require a highly
specialized expertise within the banking agencies in order to oversee
compliance of A-IRB banks with their capital requirements. The banking
agencies have assembled teams of experts to supervise the
qualification, implementation, and operation of models in the A-IRB
banks. To my knowledge, though, the agencies have not provided detailed
information on the numbers of experts they have employed or projections
as to how many bank models the teams will be able to examine before
they are stretched too thin.
Questions about the ability of our banking agencies to supervise
the use of internal credit models for regulatory capital calculations
naturally raise the question of who can monitor the supervisors. The
difficulties raised by the complexity of the bank models are compounded
by the fact that much of the information contained in the models will
be proprietary to the bank. Congress, academics, and other interested
observers will thus not be in a position to assess how good a job the
banking agencies are doing. Nor will creditors of banks be able to make
their own informed assessment of the bank's risk and capital position,
thereby limiting a source of market discipline that might contribute to
bank safety and soundness. Finally, the opaque nature of the
supervisory process for internal credit risk models means it is not
clear whether and how U.S. banking agencies will be able to determine
if their foreign counterparts are effectively supervising their own A-
IRB banks. Basel II would hardly be contributing to even the limited
equalization of competitive conditions that can be effected by capital
requirements if those requirements are not being rigorously enforced in
some countries.
There are no easy solutions to these difficulties. It is perhaps
understandable that, after years of work, the regulators have grown
impatient with the seemingly endless technical challenges and want to
get on with implementation, making further needed changes as they go
along. However, in my judgment, there are still too many questions
outstanding for the agencies to proceed.
A Downward Spiral for Capital Levels? The problem, of course, is
not just that the agencies cannot say what the effects of Basel II will
be. It is also that such indications as we have from the imperfect
impact studies suggest that capital levels could decline significantly.
Recall that the Basel II process began with assurances from the
supervisors that aggregate capital levels would not decline
significantly. It appears that this assurance has been abandoned by the
supervisors. Instead, about the only thing we can be sure of is that
the A-IRB approach would produce significant declines in regulatory
capital.
Indeed, the Basel II process seems to have acquired a disturbing
capital-reducing momentum. Large banks appear to regard Basel II as an
agreement between the banks and the regulators, whereby the banks will
make the investments necessary to qualify for the A-IRB approach and
the regulators will reduce required capital levels. I note with some
concern that, when the European Parliament was considering new capital
regulations based on Basel II, at least one European Commission
official was reported to have touted the benefits of reduced capital
requirements for European banks of between =80 billion and =120 billion
(approximately $94-$141 billion at current exchange rates).
After seeing the risk weights that will be applied to residential
mortgage and small business lending under Basel II, the 9,000 U.S.
banks that will not be applying the A-IRB rules became concerned that
they will be disadvantaged in competing with the A-IRB banks in those
lending markets. Their complaints have prodded the agencies to proceed
with plans for a so-called Basel I-A for all but the largest twenty or
so banks that will adopt A-IRB. Existing capital rules based on Basel I
will be modified to provide more ``risk sensitivity''--a euphemism for
reduced capital requirements--for these categories of loans. The
banking agencies' advanced notice of proposed rulemaking provides only
possible means to this end, rather than a concrete proposal.
Ironically, some of the ideas advanced by the agencies are contained in
the ``standardized approach'' of Basel II--a revision of the original
Basel I approach that U.S. regulators had previously rejected as
insufficiently risk-sensitive to be worth implementing. While we do not
know exactly where this process will eventually take the banking
agencies, it is clear that reductions in minimum capital requirements
will be the result.
The Basel I process began out of concern that capital levels at
many internationally active banks were too low. In the years following
adoption of Basel I, capital levels did generally increase. The Basel
II process began out of concern that capital holdings may not be
sufficiently related to actual risks incurred, particularly in large
complex banking organizations, but with the stipulation that aggregate
capital levels would not change significantly. What we seem to be
getting is a kind of downward spiral in capital levels--large banks
expect reduced regulatory capital requirements in exchange for more
complicated risk assessment systems; changes are made in the Basel II
formulas that appear to reduce capital requirements; banks that will
not be adopting the A-IRB approach seek, and are granted, lower capital
requirements in order not to be at a competitive disadvantage with the
large banks; meanwhile, the large banks see Europe moving ahead with
Basel II implementation under the promise of lower capital levels and
do not want to be left behind. Federal Reserve Board officials even
suggested last spring that simple leverage ratios would eventually be
eliminated, thereby lowering capital requirements even further for
banks whose risk-weighted requirements will drop substantially under
Basel II. While the Fed backed off this position following an outcry
from Members of Congress, the floating of this proposal showed the
momentum that has been generated for reducing capital levels.
This momentum is all the more disturbing because the regulators
have not given us any explanation of why they believe capital levels
should be lower than at present, if indeed they believe so. Capital
requirements reflect a judgment as to the optimum trade-off between
making more bank resources available for investment in productive
activities and the costs that will be borne by the public fisc and the
economy if banks fail (or are propped up by the government so as not to
create problems in other parts of the financial system). This trade-off
is a policy judgment; it cannot be reduced to a formula. In some
respects, the judgment is based on the intuition of those with
knowledge and experience of banks and banking systems. The bank
regulatory agencies have not explained their theory of where capital
levels should be, and why. Instead they have focused mostly on
technical issues. They leave those of us observing the process worrying
that the overriding goal of the Basel II process has become simply
getting the new rules in place.
Some Practical Steps Forward
While I share the skepticism of many academics and former
policymakers that the A-IRB methodology of Basel II is the best
approach to capital regulation, I cannot say today that I have
concluded it should be abandoned. However, there are simply too many
important unanswered questions--theoretical, policy, and practical--to
make proceeding with implementation a prudent course of action. My core
recommendation to you is that Members of Congress urge the banking
agencies not to proceed with implementation until at least the more
important such questions have been satisfactorily answered:
<bullet> Do the agencies believe that, in general, minimum capital
levels are too high and, if so, what is the basis for that belief ?
<bullet> Have the agencies been able to predict with reasonable
precision the levels of capital that the A-IRB approach will
require of large banks, and then to confirm their prediction
through studies that are well-conceived and executed?
<bullet> Have the agencies done scenario planning to anticipate the
effects, unintended and otherwise, of Basel II upon the financial
system as a whole? For example, are significant portions of certain
kinds of assets likely to migrate out of banks into the unregulated
sector? If so, might new risks of financial disruption be created?
<bullet> What do the agencies regard as the effective capacity of the
specialized examiner teams that will be overseeing the use of bank
models under A-IRB? How will the agencies determine whether other
countries are successfully monitoring bank capital levels under the
A-IRB approach?
<bullet> Have the banking agencies consulted with institutional
investors, ratings agencies, independent analysts, and other market
actors concerning the amount of
information about their credit modeling that A-IRB banks will be
required to disclose?
<bullet> Do the banking agencies continue active study of alternative
approaches to capital regulation, such as the varieties of
proposals based on market discipline or on so-called precommitment?
I add this question because, although I do not believe that any of
those approaches is at present appropriate as the sole basis for
capital regulation, I cannot but wonder whether devoting some of
the time and resources spent on developing the A-IRB approach might
have revealed workable variants of other ideas. It seems to me that
we should continue to explore these and other alternatives since,
as many experienced observers believe, the A-IRB approach may not
ultimately prove feasible.
It may well be that the agencies already have good answers to some
of these questions, in which case they should be able to satisfy some
of these concerns fairly quickly. With respect to the core issue of
what capital levels will result from the Basel II rules, I believe that
more work almost certainly needs to be done.
I have one other suggestion for you to urge upon the banking
agencies if and when Basel II is implemented. The heart of the A-IRB
approach involves using banks' internal estimates of the probability of
default and a few other variables as the basis for risk-weighting
minimum capital requirements. As mentioned earlier, there is not much
of a track record on how accurate the predictions will be, especially
in a time of economic turbulence. It will thus be important to
rigorously and thoroughly evaluate the performance of the models on an
ongoing basis. While I expect that the banking agencies will themselves
pay heed to this subject, it would be useful for all concerned that
there be an independent examination of this critical issue. Banking
agency staff with credit model expertise will presumably be fully
occupied in supervision, and will not have time for a complete
assessment of past performance. In any case, it will be reassuring for
Congress and the public to have an independent evaluation to complement
the views of the banking agencies. Just as Congress has urged, and even
mandated, evaluations of the efficacy of certain government spending
programs in accomplishing their stated aim, Congress should encourage
the banking agencies to contract with expert outsiders to conduct
periodic evaluations of the bank models and of the supervisory
requirements for the A-IRB banks.
Conclusion
Banking agencies in the United States and other Basel Committee
countries have invested an enormous amount of work in the Basel II
process. There is little question but that, as a result of this effort,
we understand far better the challenges of risk management and
prudential regulation in large banks operating in the financial
services environment of the 21st century. It is clear, to me at least,
that the Basel I standards cannot indefinitely remain a viable method
of capital regulation. But that does not mean they should be abandoned
before we have enough information to know that their replacement will
be a net improvement. The mere fact that so much has been invested by
the banking agencies in the A-IRB approach is not itself a
justification for moving forward. Indeed, the accomplished economists
who work in those agencies should know better than most of us the old
economic axiom that the sunk costs you spent yesterday should not
affect your assessment of the best way to spend additional resources
going forward.
There needs to be more information developed by the agencies and
more public debate on the central questions concerning Basel II before
we make such dramatic changes in the regulation of our largest banks.
It is noteworthy that the leadership of all five of the key
institutions has turned over, or is in the process of doing so, since
the Basel II process got underway in earnest (I include the Federal
Reserve Bank of New York as a key actor, in addition to the quartet of
the Federal Reserve Board, the Office of the Comptroller, the Federal
Deposit Insurance Corporation, and the Office of Thrift Supervision.) I
am hopeful that the new leadership is well-positioned to build on the
work that has been done, while heeding the concerns of this Committee
and others, to formulate sound and effective capital policies for all
our banks.
Thank you for your attention. I would be pleased to answer any
questions you might have.
----------
PREPARED STATEMENT OF KATHERINE G. WYATT
Head, Financial Services Research Unit,
New York State Banking Department
November 10, 2005
Good morning, Chairman Shelby, Senator Sarbanes, and Members of the
Committee. My name is Katherine Wyatt. I head the Financial Services
Research Unit at the New York State Banking Department. I have followed
the development of Basel II for the Department since 2000, studied the
possible effects of the simpler approaches under Basel II, and worked
with my Federal counterparts in analyzing banks' implementation
programs for Basel II. I appreciate the opportunity to testify today,
and ask that a copy of my written statement be included in the record.
I am speaking to you today because I am very concerned that
adhering to the proposed timetable for implementation of Basel II in
the United States will lead to far-reaching changes in the how we
measure capital without sufficient understanding of their possible
consequences.
Although the New Basel Capital Accord (Basel II) has been discussed
in the United States since 1999 (at least) there has been insufficient
study of its impact on the U.S. banking system, in part because the
proposal itself has changed over time, and in part because the Federal
banking agencies decided to restrict its implementation to the most
complex approach and to impose it on only the largest internationally
active banking organizations. For example, the best study we have, the
fourth Quantitative Impact Study, involved about 20 large
internationally active banks and used a version of the Basel II
proposal that did not have important details that have been promised
for the Basel II NPR, due in Q1 2006.
The banking system in the United States comprises about 9,000
different institutions and is rich in different-sized banks with
different business models. Because we have only considered Basel II in
terms of large complex internationally active institutions, we do not
know what the competitive impact of Basel II will be. We do not know
what the effect of large banks calculating capital requirements one
way, while smaller banks in the same market calculate requirements
another way will mean. We do not know what the effect on smaller banks
that are public companies and are concerned with return on equity will
be if their competitors are allowed to hold less capital. We must carry
out a comprehensive impact study--across the entire banking system--
before regulations are adopted that could have far-reaching effects for
banks and their borrowers.
The Federal agencies plan to finish the rulemaking process for
Basel II, conduct a year of parallel run of current requirements and
Basel II, and to have Basel II in effect, all by January 1, 2009. They
also aim to have the Domestic Capital Modifications for non-Basel II
banks (Amended Basel I) in effect on January 1, 2009. I am afraid that
pushing ahead to complete the rulemaking process for two complex
proposals in less than 3 years will not allow time for essential review
of either.
The Basel II Notice of Proposed Rulemaking will have been in
development for 2\1/2\ years when it is released; we need sufficient
time to study this complex proposal. It will be very important to look
at Basel II and the Amended Basel I
proposals side-by-side, and to study their impact on the U.S. banking
system. I am concerned that supervisors will not have enough time to
consider the possible consequences of these sweeping changes before
they become effective--and before revision is much more difficult.
It is true that the agencies have included 3 years of ``floors'' in
their implementation schedule. There will be graduated limits on
capital requirements, and banks will be able to move fully to Basel II
only at the end of these ``floor'' years. However, I believe it will be
very difficult to make fundamental changes in Basel II after the
January 2009 effective date. Banks that adopt Basel II in 2009 will
already have made substantial investments in systems and data
collection processes, and will surely object strongly to making changes
after the effective date.
The Agency projection also assumes that all the necessary
documentation for supervision of banks following these revised
regulations--guidance, reporting requirements, and examination
procedures--will be developed in this very short period of time.
I believe that there are two large gaps in our understanding of the
impact of Basel II that must be addressed before we move to a Final
Rule for either Basel II or Amended Basel I.
First, we do not know what the actual level of capital will be
under Basel II for any given bank, or across all Basel II banks.
Preliminary QIS-4 results showed a broad range of required capital
amounts, even for similar portfolios. Also, Basel II has changed over
time; the agencies' timetable seems to involve going ahead with
implementation without an impact study of the fully specified proposal.
Even more importantly, since capital requirements under Basel II are
based on outcomes of mathematical models, we need time to develop
rigorous technical guidelines for parameter estimation and tests of
data sufficiency, to ensure that required levels of capital are
adequate.
The second large gap in knowledge comes from the fact that we have
not addressed the changes that may be brought by Basel II (and Amended
Basel I) across the entire banking system. Basel II's impact in the
United States has been studied primarily on large complex banking
institutions. However, we have not fully studied the competitive effect
of Basel II on the close to 9,000 non-Basel II banks in the United
States. The Amended Basel I proposal for non-Basel II banks was
released only last month, and it is essential that the effects of these
two proposals be studied side-by-side.
Use of Models in Calculating Capital Requirements
I would like to speak first about banks' Basel II capital
calculations. Under Basel II, capital requirements for credit exposures
are based on the outcomes of a particular mathematical model of default
specified by supervisors. This supervisory model is applied to
complicated portfolios, with a host of adjustments, specifications,
exclusions, and exceptions that grew out of attempts to reconcile the
model results with existing bank portfolios and existing international
bank regulation. (The final U.S. version of these adjustments and
specifications should be released early next year.) Basel II banks
provide their own estimates of probability of default, loss given
default, exposure at default, and maturity as inputs to the Basel II
formulas; these parameter estimates depend on the data and other models
used by the bank.
A key premise of implementation of Basel II is that banks will have
enough reliable data to produce rigorous results from the model. I
think many would agree with me that this is often not the case. The
variation in required capital estimates for similar exposures found in
QIS-4 is quite possibly due to problems of insufficiency of data. We
also need to make sure that the modeled capital requirements are
adequate when times are bad--the history of the last several years in
the United States is of good times.
In contrast to the treatment of credit risk, Basel II allows banks
to choose their own model to calculate capital requirements for
operational risk. Here, even more variation in results is possible,
particularly since there are even fewer data for operational loss
events than for credit losses. The Basel II ANPR advises banks to use
``expert opinion'' scenarios to fill out data points in their modeling,
thus providing even more opportunities for selection in modeling
techniques.
Unfortunately, Basel II could be gamed by choosing the modeling
techniques and data sets that will produce the lowest capital
requirements. As well as a very broad range of required capital,
preliminary QIS-4 results showed decreases as great as 74 percent for
some bank portfolios. The strong possibility exists, also, that as the
distance between risk-based capital requirements and current leverage
ratio under prompt corrective action (PCA) capital requirements grows,
there will be increased pressure on bank regulators to drop the
leverage ratio requirements. For many large banks, satisfying the well-
capitalized leverage ratio is already the constraining capital
requirement, rather than meeting risk-based capital ratios.
The bank supervisors I have talked with are very worried at the
prospect of dropping PCA requirements--they remember other times when
banks' predictions about the future did not come true. They also point
to their experience that well-capitalized banks are profitable banks,
can enjoy lower costs of funding, and more easily weather economic
downturns.
I am concerned that without direction from supervisors, capital
requirements could differ widely according to the parameter estimation
methods used by banks, and depending on banks' own data sources. It is
essential that the schedule for implementation allows enough time for
supervisors to work with bank models, to understand different parameter
estimation techniques, and to gauge sufficiency of data. Supervisors
will then be able to develop necessary technical guidelines for the
estimation process and to set the restrictions and constraints
necessary to ensure adequate required capital. If this time is not
allowed, there is a real danger that the estimation techniques ``most
large banks choose'' will become the de facto ``best practices.''
Allowing this supervisory review period will also ensure that the
necessary examination procedures and guidance will be developed, both
for Basel II and Amended Basel I. We need time to understand and assess
these proposals, and once they are accepted, we need time to develop
examination materials, to provide necessary training for examiners,
particularly in the supervision of Basel II banks, and to provide
support for bankers.
It can take several years for a bank to develop the systems
necessary for adoption of Basel II. Some of the largest banks have
already begun this process, in order to be able to adopt Basel II at
its effective date. I am concerned that maintaining the current
timetable will intensify pressure to keep the Basel II proposal
``open'' enough so that banks that have begun implementation projects
will not have to make radical changes. This could make it very
difficult to institute material changes in the future, when banks have
committed even more sizeable resources to their Basel II systems. We do
not know enough now about the consequences of Basel II to go ahead, in
an attempt to justify the expenditures a few banks have already made.
Impact of Changes in Capital Requirements across U.S. Banking System
Both bankers and supervisors are concerned about the impact of
Basel II on the U.S. banking system. As pointed out in a letter sent by
the Conference of State Bank Supervisors to the Federal Agencies in
September of this year,
Implementing the risk-based capital requirements depicted in
the recent studies could have profound competitive implications
and significantly harm the banking industry in general and non-
Basel II banks in particular. As proposed, Basel II creates
significant differences between capital requirements of banks
that adopt Basel II and those that do not. The current approach
reduces the capital large institutions hold for mortgages and
small business loans, among other assets. In a very practical
sense, the reduced capital requirements would provide a pricing
advantage for the larger institutions. It will be difficult for
smaller banks to compete for mortgages and small business loans
and certainly difficult for these institutions to hold such
assets in their portfolio. In a competitive economy, eventually
market forces will likely drive these assets from smaller banks
toward the Basel II adopting banks, requiring nonadopting
banks, the vast majority of which are small community banks, to
move to higher-risk areas of banking.
In addition, with substantially lower capital requirements,
larger institutions could acquire community and mid-tier banks
without much cost involved by immediately lowering the acquired
bank's required capital to a level that is allowed by Basel II
banks. The lower capital requirements and the magic of the
current Basel II mathematics promote the incentive for
consolidation within the banking industry.
CSBS strongly urged the Federal banking agencies ``to conduct
further analysis of potential capital changes that would ensue from
adopting the current Basel II proposal.''
I am afraid that the publicly available analysis does not
adequately answer these concerns. The Federal Reserve has posted on its
website several ``White Papers,'' covering some Basel II portfolios and
some of the competitive issues raised by the original ``bifurcated''
regime proposed by the agencies. However, these papers are based on
Basel II circa 2003, and both the Basel Committee and the Federal
agencies have made changes to their proposals since then. The new
Amended Basel I proposal, of course, is not considered at all in this
research. The ``White Papers'' suggest that the impact on non-Basel II
banks may be minimal, but these papers are not definitive, and other
authors have disagreed with their findings.
We need to have a much better understanding of the consequences of
Basel II before it is implemented. We should take the time now--both
Federal and State banking regulators--to fully test the impact of Basel
II and Amended Basel I proposals. In this way, we can work to safeguard
the soundness and profitability of the banking system and ensure that
U.S. borrowers will continue to have the access to capital that a
strong U.S. banking system affords them.
I hope that these remarks are helpful to the Committee and would be
pleased to answer any questions that you have.
RESPONSE TO WRITTEN QUESTIONS OF SENATOR SHELBY
FROM SUSAN SCHMIDT BIES
Q.1.a. Basel II would impose new demands on bank regulators. It
would require bank regulators to not only approve banks'
internal assessments of their risks, but also to make judgments
on whether banks are holding enough capital in light of all the
risks banks face. This means bank regulators are going to need
personnel who are well trained in risk analysis.
How many additional personnel does your agency expect it
will need to hire in order to implement Basel II? What type of
expertise will your agency need to obtain? Do you foresee any
problems in recruiting?
A.1.a. The Federal Reserve has been preparing for the
implementation of Basel II for some time. Preparation includes
planning for reallocations of existing staff, hiring of new
staff, and planned training enhancements, all started as far
back as 2001. It is difficult to quantify exactly the number of
personnel associated with Basel II as our planning is tightly
integrated with broader continuing initiatives to expand and
augment System expertise in areas such as risk management and
quantitative analysis. That is, we are not hiring supervisory
staff only for Basel II implementation purposes. In fact, even
without the movement to an enhanced risk-based capital
framework, supervisors have, for some time now, been addressing
the implications for Federal Reserve staffing and specialized
training of quantitative risk management models and techniques
as their use at financial institutions has grown around credit,
operational and other risks.
With regard to expertise, staffs at Reserve Banks are
working with staff of the Board of Governors to support Basel
II and risk-related policy requirements. Quantitative risk
units have been established within the Federal Reserve System
to serve the System's need for specialized expertise in this
critical area. Specific expertise includes staff with risk
management, quantitative analysis, and modeling skills, and
technically trained industry experts with market and
operational risk, corporate credit, and retail credit risk
experience.
In an effort to appropriately prepare Federal Reserve
System staff for the implementation of the Basel II framework,
a number of training programs are already in place to ensure
that supervisors will be ready to meet the challenges posed by
U.S. implementation of the framework at the local level. Also
important are coordination efforts across Federal Reserve
districts to maximize the efficient deployment of experts and
cooperative interagency training efforts to share expertise
among banking supervisors.
Recruiting in a competitive environment will continue to be
a challenge. This would have been the case even without Basel
II as use of quantitative models at banking organizations
continually expands. The Federal Reserve is prepared to use all
tools in place to ensure that qualified staff with the
requisite skills are recruited and retained and that there is
sufficient investment in training for existing staff, as
further elaborated in responses 1b and 1c.
Q.1.b. Will your agencies be able to successfully compete
against banks in the recruitment of personnel? Do you have any
concerns that banks will simply be able to hire the best
Ph.D.'s in risk modeling and outgun their regulators?
A.1.b. The need to ensure successful recruiting and retention
of specialized staff--not only for overall supervision but also
for Basel II as well--has consistently been a focus for the
Federal Reserve. The quality and effectiveness of our
supervisory efforts depend most importantly on the quality and
knowledge of our staff. Over the years, we have been successful
in recruiting and retaining the staff necessary to carry out
our responsibilities in an effective manner, despite growing
demand and competitive salary pressures for our well-regarded
and highly skilled examiners and staff. Basel II, of course,
has been a key focus for the last 24-36 months. As the private
sector has come to recognize the value of these skills, the
demand for specialists with risk-management skills has
increased the total compensation packages that our staff can
earn in the private sector. The ``value proposition'' of
working at the Nation's central bank in a policy and public
service role, in addition to factors such as competitive
benefits packages and work/life balance options are partial
offsets to external market pressures. Recruiting for
specialized expertise, especially quantitative modeling and
analysis, is necessarily expensive, and pushes the limits of
existing pay structures. Successful recruits will typically
command the upper end of the pay scale in some cases. The
research staffs at the Federal Reserve Banks and the Board
provide a significant pool of internal expertise in areas such
as risk modeling and quantitative analysis. As in times past,
we will use research staff with quantitative expertise to
assist risk specialists and examiners on the supervision staff,
as needed.
Q.1.c. How much in additional annual costs do you expect to
incur in implementing Basel II? Other than personnel costs,
what other costs will your agency incur in implementing Basel
II? Are there any large one-time costs?
A.1.c. Over a 5-10-year timeframe, the Federal Reserve will
make considerable investments related to Basel II, including
the cost of course development and training hours for
examiners. There will also be some costs associated with
collection and analysis of new data related to the Basel II
implementation.
The total training requirements will depend, in part, on
how many banks opt-in to Basel II. As noted in our response to
1a, these efforts have been in progress for several years and
are integrated tightly with broader initiatives to ensure our
supervisors are well-trained to meet the complexities and rapid
change in supervised organizations. In addition, both Reserve
Banks and Board supervision functions are continually
reprioritizing resources to address our most pressing concerns
including the effective supervision of the most complex
organizations. As a result, we have reallocated staff
development funds to augment our training programs around risk-
management techniques including those related to an enhanced
risk-based capital framework. Two courses have been developed
for examiners and quantification experts at the Reserve Banks
related to credit risk measurement and quantification for
corporate and retail activities. A third course, developed
jointly with the Office of the Comptroller of the Currency, the
Federal Deposit Insurance Corporation, and the Office of Thrift
Supervision, focuses on operational risk and provides a
description of the range of advance measurement approach (AMA)
practices, including specific examples of how banks might
implement their AMA processes. An agreement with the BIS
Financial Stability Institute for online learning provides a
wide variety of tutorials on subjects related to capital
adequacy and will require a modest outlay of funds over the
next 2 years. We do not currently anticipate any large one-time
expenses related to Basel II. However, we are finding it
necessary to use enhanced compensation and benefits to attract
and retain staff with critical skills in quantitative risk-
management techniques and other important supervisory skills
areas.
Q.2.a. In his testimony, Dr. Tarullo posed several questions
that he believes need to be answered by U.S. banking agencies
before implementation of Basel II begins. Please provide
answers to his questions with respect to your agency, which are
as follows:
``Do the agencies believe that, in general, minimum capital
levels are too high and, if so, what is the basis for that
belief ?"
A.2.a. For most banking organizations, minimum regulatory
capital levels today are adequate. However, for our largest,
most complex banking organizations, Basel I-based capital rules
have become increasingly inadequate and do not provide
supervisors with a conceptually sound framework for assessing
overall capital adequacy in relation to risks, including which
institutions are outliers, and how capital adequacy may evolve
over time. That said, we do know that, for some individual
banking organizations minimum regulatory capital levels
required to be held against some of the assets they hold, and
the inherent risks in those assets and operations, may be too
high. Similarly, there are also some banking organizations
whose assets and operations are riskier than average and thus,
for particular assets, their minimum capital requirements may
be too low. The current framework does not distinguish between
those organizations that are taking on greater risk versus
those that are not. That is one of the limitations of Basel I
and why a more risk-sensitive approach is needed for our
largest, most complex banking organizations.
One should remember that simplifying assumptions are built
into minimum regulatory capital requirements, such as Basel I,
so that they can apply to a wide range of institutions.
Accordingly, they cannot get to a level of granularity that
will provide a tight linkage between risks that are being taken
and the amount of capital needed for every single banking
organization. For the largest, most complex banks, Basel II
attempts to provide more granularity and risk sensitivity, but
it still only looks explicitly at three types of risk in Pillar
l--credit, market, and operational. As such, there are still
risks that are not explicitly accounted for in the Basel II
risk-based minimum capital numbers, such as liquidity,
concentration, interest rate, and legal-compliance risks. That
is why Basel II has a component of capital evaluation in Pillar
2.
Minimum regulatory capital requirements simply cannot
explicitly reflect all the risks that every banking institution
undertakes. To do so would, indeed, be excessively burdensome
and not cost effective. Minimum regulatory capital measures
provide floors below which capital should not fall. The
agencies have instead relied significantly on the supervisory
process to supplement minimum regulatory capital requirements
and ensure that a banking organization holds an amount of total
capital that is adequate in relation to all the risks faced by
the organization. Pillar 2 of Basel II reinforces the
importance of a bank's internal and supervisory assessments of
its overall capital adequacy. The agencies also have relied,
and will continue to rely, on a minimum leverage ratio
requirement to supplement the risk-based capital requirements
for both Basel I and Basel II.
Successful internal capital adequacy processes (economic
capital) at the largest banking organizations emphasize the
need to identify and measure all relevant risks. Most
institutions maintain a ``cushion'' above internally generated
capital measures for various reasons. The actual total capital
a bank holds above the regulatory minimum reflects bank
management's assessment of risks the institution faces, and the
related amount of capital the institution needs to cover all of
its unexpected losses. Sometimes investors, counterparties, and
customers demand a stronger credit rating at the bank than that
implied by the minimum regulatory capital requirements. Many
banks also want to remain prepared to promptly pursue mergers
and new business expansion opportunities as they arise, which
requires maintaining capital above the regulatory minimum.
Banks may also wish to hold additional capital in anticipation
of reduced ability to raise it in future periods. Such
justifications for holding capital above the regulatory minimum
are entirely appropriate, indeed necessary for determining the
appropriate level of total, rather than minimum regulatory,
capital for an individual institution. In the supervisory
process, we check to see that all minimum regulatory capital
requirements are met, but we also analyze a banking
organization's total capital and its rationale and plans for
maintaining that capital in order to assess whether its capital
is adequate given its risk profile.
Q.2.b. ``Have the agencies been able to predict with reasonable
precision the levels of capital that the A-IRB approach will
require of large banks, and then to confirm their prediction
through studies that are well-conceived and executed?''
A.2.b. The Basel II implementation plan for the United States
has incorporated from the outset an ongoing process for
evaluating both the quantitative impact of Basel II on banks'
minimum risk-based capital requirements and the quality of
banks' systems for calculating Basel II risk parameters. One
element of this process has been several quantitative impact
studies (QIS) conducted over the past several years. The most
recent QIS exercise, QIS-4, provided valuable insights into
banks' minimum regulatory capital relative to their risk
profiles, the varying methodologies used to measure risk, and
where banks' risk-measurement and -management capabilities are
stronger or weaker relative to their peers.
Consistent with the agencies' expectations, QIS-4 provided
somewhat limited insights into what would be the minimum
required capital under Basel II when fully implemented. Like
its predecessors, QIS-4 was conducted on a ``best-efforts''
basis without close supervisory oversight, and it reflects only
a point-in-time look at banks' portfolios and risk-measurement
and -management systems (in this case, as of the second or
third quarter of 2004). An important finding was that banks'
systems for estimating Basel II risk parameters were only
partially developed at the time of QIS-4 and would not meet
supervisors' expectations for reliability. As a result, the
QIS-4 results may not represent accurately the minimum capital
charges that would be generated by Basel II when banks' systems
are fully compliant with their risk measurement and management
standards. Indeed, QIS-4 showed that while banks have been
making steady progress in improving the quality of these
systems, no institution would have been qualified to implement
Basel II at the time of QIS-4.
Uncertainty surrounding the precise quantitative impact of
Basel II has been incorporated into the agencies'
implementation plan for the United States as outlined in our
September 30, 2005, press release. The plan recognizes
explicitly that the agencies cannot be certain about the
quantitative impact of Basel II until banks have fully
implemented the systems needed to support the framework. Yet,
banks cannot build up these systems until the agencies have
released a notice of proposed rulemaking (NPR) and accompanying
examination guidance setting forth our qualification standards.
There may also be additional changes in the final rule that
banks will need to take into account. More precise estimates of
the quantitative impact of Basel II should become available
during the ``parallel run'' period (beginning January 2008) and
the transition years (from 2009 to 2011). The agencies will
periodically review and analyze data collected from the Basel
II banks during these periods, as we continue to evaluate the
quantitative impact of Basel II. The agencies will use this
information to determine whether any modifications to the Basel
II framework, particularly to Pillar 1, are needed, for
example, to preclude any unwarranted decline in banks' minimum
risk-based capital requirements. To ensure adequate time to
make such adjustments, over the 3 years following the parallel
run, Basel II would be phased-in by constraining each bank's
consolidated risk-based capital charge to be at least 95
percent, 90 percent, and 85 percent of Basel 1 levels in 2009,
2010, and 2011, respectively. In addition to these transitional
safeguards, the current prompt corrective action (PCA)
framework and minimum leverage ratio requirement will remain in
place. Taken together, we believe these measures will ensure
that banks maintain prudent capital levels throughout the
transition period and beyond.
Q.2.c. ``Have the agencies done scenario planning to anticipate
the effects, unintended and otherwise, of Basel II upon the
financial system as a whole? For example, are the significant
portions of certain kinds of assets likely to migrate out of
the banks into the unregulated sector? If so, might new risks
of financial disruption be created?''
A.2.c. It is important to note that one of the motivating
factors for pursuing capital reform is to reduce existing
financial market distortions that are the result of banks'
efforts to arbitrage Basel
I-based rules, such as by securitizing high-quality assets and
retaining lower-quality, yet higher-yielding, assets. That
said, the Federal Reserve takes the potential effects produced
by Basel II very seriously. Throughout the Basel II process, we
have remained vigilant about the impact of our decisions. This
includes conducting analyses, such as our series of white
papers about the effects of Basel II on certain aspects of the
U.S. banking system. We have also remained very attentive
throughout the Basel II process to comments from the industry,
the Congress, and other interested parties about the potential
effects of Basel II on the financial system. When information
from any of these sources or evidence from risk models in use
identified modifications that should be made, we have pushed
for adjustments to strengthen the proposed capital requirements
of Basel II.
To the extent that Basel II better reflects risk-taking by
banks, minimum regulatory capital should be more consistent
with the view toward risk and capital that rating agencies and
investors have been using both for banks and their competitors.
In general, it is our view that there will not be disruptions
to the financial system as a result of Basel II, nor should
there be a sizeable migration of assets outside the banking
system to non-bank entities. But, as with any change in capital
rules, bankers will evaluate options and optimize their use of
capital, so some adjustments are likely. For example, the
introduction of PCA in the early 1990's encouraged even the
largest banks to hold more additional excess capital so they
would qualify to be deemed ``well-capitalized''.
In most cases, our analysis suggests that it is usually not
minimum regulatory capital levels that significantly affect
where assets are held, but rather the capital allocations that
banks (and others) make internally within their organization,
so-called economic capital. Moreover, most banks, and
especially the smaller ones, hold capital far in excess of
regulatory minimums for various reasons (as outlined in
question 2a, above). Thus, in normal circumstances, changes in
a bank's own minimum regulatory capital requirements for
particular assets generally would not have a sizeable effect on
the level of actual capital the bank chooses to hold and would
not necessarily have sizeable affects on internal capital
allocations. In short, we do not expect Basel II to
significantly change the market realities among banks, or
between banks and non-bank financial entities.
Naturally, if we see evidence that distortions might arise
from Basel II, we will seek to review and analyze the specific
situation and make adjustments to the U.S. capital proposals
accordingly. As you know, the agencies have embedded in their
plans for United States implementation of Basel II a number of
safeguards, given some of the uncertainty that still remains
about Basel II's ultimate effects. These include the parallel
run period, 3 years of transitional floors, and additional
rulemakings. And, as noted elsewhere, we will perform periodic
reviews and analysis of the quantitative impact of Basel II on
individual banks and in the aggregate.
Q.2.d. ``What do the agencies regard as the effective capacity
of the specialized examiner teams that will be overseeing the
use of bank models under A-IRB? How will the agencies determine
whether other countries are successfully monitoring bank
capital levels under the A-IRB approach?''
A.2.d. As noted in question 1 above, the Federal Reserve has
been preparing for some time to prepare for policy and
implementation requirements associated with Basel II. We
already have considerable expertise and experience related to
evaluation of quantitative risk models in the supervision of
our largest and most complex banking organizations. As you
know, U.S. supervisors are still developing their Basel II
proposals, associated supervisory guidance, and proposals for
Basel II data collection. These taken together will provide a
lot of detail to both the examination staff and the banking
organizations using Basel II, and will support consistency
across U.S. supervisors and clarify our standards to other
countries.
To address concerns about inconsistent application of the
Basel II framework across countries, the Basel Committee
several years ago established the Accord Implementation Group
(AIG), made up of senior supervisors from each Basel member
country. The AlG gathers regularly to share best practices and
develop ways to foster consistent application of Basel II
across national jurisdictions. As discussions about
implementation become more detailed and more focused on
specific banking groups, they move out of the AIG and evolve
into bilateral supervisory relationships dealing with practical
details. The AIG has also published a substantial amount of
information so that other supervisors and bankers can benefit
fTom some of the AIG's discussions and have a better sense of
the expectations of AIG members. No doubt there will be some
differences in application of Basel II across banking systems
with different institutional and supervisory structures. All
AIG members, and certainly the Federal Reserve, would remain
alert to this issue and work to minimize differences in
application. However, it is helpful to remember that issues
relating to cross-border supervision already exist today, and
U.S. supervisors have extensive experience in addressing such
issues. With Basel II, as is currently the case, host
supervisors will still examine the legal entities in their
country. Accordingly, U.S. bank subsidiaries of foreign banks
would be operating under United States rules and foreign bank
subsidiaries of U.S. banks would be operating under foreign
host-country rules. As part of our home-host communication, we
will keep foreign supervisors informed about how operations
outside their jurisdiction affect the entities they supervise,
and do this with a minimum of burden on the consolidated
organization. We expect the reverse to be true as well. And, of
course, we continue to maintain an ongoing dialogue with
institutions for which we are the home or the host supervisor.
The sooner that institutions raise specific issues of concern
with us, the quicker we will be able to work on solutions and
the smoother the transition is going to be.
Q.2.e. ``Have the banking agencies consulted with institutional
investors, rating agencies, independent analysts, and other
market actors concerning the amount of information about their
credit modeling that A-IRB banks will be required to disclose?
[If so, what were the views of these market actors as to the
adequacy of the contemplated disclosure requirements?]''
A.2.e. The Pillar 3 disclosures were issued for public comment,
both formally and informally, on several occasions. In
addition, representatives of the agencies consulted with market
participants in developing the disclosures. In general, market
participants, such as debt and equity analysts, support
enhanced public disclosures of risk information and capital
adequacy by banking organizations. The version of Pillar 3
included in the June 2004 text from the Basel Committee
balances the desire for additional disclosures by market
participants and the reporting burden associated with such
disclosures. This version leverages off of the information
institutions will need in order to implement Basel II and is
substantially streamlined from earlier Pillar 3 proposals.
Q.2.f. ``Do the banking agencies continue active study of
alternative approaches to capital regulation, such as the
varieties of proposals based on market discipline or on so-
called precommitment?''
A.2.f. The agencies, on an ongoing basis, consider ways to
improve the existing risk-based and leverage capital regimes.
Over the past 15 years or so, the agencies have adopted over 28
modifications to the existing risk-based and leverage capital
regime. In addition to this ongoing incremental maintenance and
refinement, the Basel II process is one example of developing
an entirely new approach to capital regulation--part of this
process has included considering alternative approaches and
reaching consensus on the approaches determined to be most
appropriate. Further, the Federal Reserve has ongoing dialogue
with other agencies and standard setters such as the SEC and
FASB to consider ways to enhance or refine our regulatory
approaches.
With regard to market discipline, the Federal Reserve Board
and Treasury submitted a report to Congress on ``The
Feasibility and Desirability of Mandatory Subordinated Debt''
in December 2000. The link to the report is: http://
www.Federalreserve.gov/boarddocs/rptcongress/debt/
subord_debt_2000. pdf.
In that report, we noted that the Board will continue, and
explore opportunities to enhance, its use of data from
subordinated debt, equity, and other markets to evaluate the
current and expected future condition of large depository
organizations. With respect to subordinated debt, the Board
will continue, as part of the supervisory process, to monitor
both yields and issuance patterns of individual institutions.
The Federal Reserve System continues to study market
discipline and the use of market data and has an ongoing
program of regularly reviewing market data, including
subordinated debt spreads, credit default swap premiums, and
stock price data, as part of our off-site surveillance program.
With regard to ``precommitment''-based proposals, the
agencies did evaluate proposals incorporating this concept
prior to the adoption of the Market Risk Amendment. However,
the agencies decided not to incorporate such proposals given
various shortcomings, including that supervisors would be
required to impose higher capital requirements on firms after
the fact, rather than ensuring that there is adequate capital
on an ex-ante basis.
Q.3. Although Basel I-A aims to reduce any competitive
advantage Basel II may confer on large banks, have you
conducted any analysis to determine whether Basel I-A itself
creates any competitive issues within the subset of banks that
are intended to be covered by it? If not, do you intend to do
so?
A.3. In the Basel I ANPR, the agencies articulate five broad
principles to guide the developing revisions. These principles
are to (1) promote a safe and sound banking system, (2)
maintain a balance between risk-sensitivity and operational
feasibility, (3) avoid undue regulatory burden, (4) create
appropriate incentives for banking organizations, and (5)
mitigate material distortions in the amount of minimum risk-
based capital requirements for large and small institutions. As
you know, the Basel I changes are still in the developmental
stage and as work progresses, competitive equity issues will
continue to be a primary area of focus--not only between the
Basel I and Basel II frameworks, but also within each of those
frameworks. Staffs have done some preliminary analysis using
existing Call Report data to understand the potential impacts
of various risk-weight changes and combinations of risk-weight
changes. As the proposed Basel I modifications continue to
evolve, additional analyses will be performed to ensure that
the agencies understand the possible effects of each
modification that the agencies consider proposing.
Q.4. What consideration has been given to the fact that there
are huge differences between the 7,000 or so banks that will
participate in the Basel I-A framework?
A.4. The agencies' existing risk-based capital rules generally
apply to banks, thrifts, and bank holding companies of all
types and sizes. As we continue developing modifications to the
current rules, with the guiding principles set forth in the
response to question 3 in mind, we will endeavor to develop a
package of revisions that enhances risk sensitivity without
being unnecessarily burdensome. The current regime is a broad-
brush approach that results in a modestly risk-sensitive
framework that can be used by a variety of institutions. We
expect that a modified Basel I-based set of rules also will be
applicable to a wide range of institutions with different risk
profiles and management and measurement approaches. As part of
the Basel I ANPR issued last year, the agencies specifically
sought comment on whether the proposed modifications should be
applied to all non-Basel II banks or whether some or all non-
Basel II banks should be given a choice of applying the
existing risk-based capital rules or the risk-based capital
rules that emerge from the process to amend Basel I.
Q.5. In Dr. Kaufman's testimony, he stated that there is
``substantial empirical evidence of a negative relationship
between leverage ratios and bank solvency,'' but ``no such
evidence between risk-based capital ratios and bank solvency.''
Do you agree with Dr. Kaufman? If so, does this conclusion
undermine the Basel II framework?
A.5. The Federal Reserve does not believe this is an accurate
application of extant research as it relates to Basel II.
Research employing the Basel I capital ratios is not at all
relevant for evaluating fundamentally different risk-based
capital ratios that are designed to be significantly more risk-
sensitive than Basel I, such as those under Basel II when fully
implemented. In this regard, the Basel II framework should
provide better insights into banks' overall capital adequacy on
an aggregate and individual basis, and how this evolves over
time, thus reinforcing the goal of the PCA regime to intervene
as early as possible at problem institutions. Accurate data for
assessing the efficacy of the Basel II capital ratios, however,
will not be available until banks have fully implemented the
systems needed to support the Basel II framework and have
access to more detailed information about supervisory
expectations for various aspects of Basel II that will be
included in the NPR and associated supervisory guidance.
Fortunately, minimum capital ratios are only one
supervisory tool the banking agencies use to monitor and assess
the overall condition of U.S. banking organizations. We have
tailored risk-focused examination practices that will continue
to be applied to ensure we have a solid and accurate
understanding of each institution subject to the agencies'
jurisdiction. In both Basel I and Basel II, supervisors are
able to intervene to require stronger capital above the
minimums when internal controls, weak earnings performance, or
loss exposures are a concern.
Q.6. Have you undertaken any analysis of the initial start-up
costs and annual compliance costs that Basel II will impose on
banks?
A.6. Most of the information we have to date is anecdotal. On-
site supervisory examination teams have ongoing dialogue with
institutions that are expected to be moving to the Basel II
advanced approaches to get a sense of the cost implications.
That information seems to change periodically. Further, some
institutions have reported to us that the delay in issuing the
NPR and modifications to the framework will increase their
costs.
In addition, through the QIS-4 supervisory questionnaire
process, some institutions answered questions about compliance
costs related specifically to ongoing compliance as well as
start-up and recurring costs but the responses that were
received were not comprehensive or easily comparable across
institutions and therefore were not included in the QIS-4
results.
Finally, much of the infrastructure development needed for
Basel II is consistent with supervisory expectations for
advanced risk measurement and management in general, so some
element of cost, while consistent with, is not necessarily
exclusive to Basel II. As a continuing part of the Basel II
process, the agencies will seek to obtain more complete and
reliable information.
Q.7. To what degree will the significant costs necessary to
comply with Basel II act as a barrier to entry that prevent
banks from growing and becoming a bank large enough to qualify
to use Basel II? In effect, will Basel II cement the market
positions of the largest banks? If not, why?
A.7. When evaluating the costs to U.S. banks of adopting Basel
II, it is important to note that all large or complex banks
need to have in place appropriate systems for measuring and
managing their risks. Banks that are not large and do not have
complex products or portfolios do not need complex measures of
risk-taking either for sound risk management purposes or for
minimum regulatory capital requirements. Much of the cost of
adopting Basel II is associated with developing and
implementing such systems for internal management purposes, and
so we would expect such costs to be incurred by a large bank
even in the absence of Basel II. Indeed, the risk-management
approaches underlying Basel II, including economic capital
precepts, reflect the general direction that large, well-
managed banks have been moving to since the mid-1990's.
The issue is the same for a non-core (that is, opt-in)
bank. As an existing supervisory issue, regardless of whether
there is a Basel II, we would expect a large and growing bank,
especially one nearing the size criteria that is envisioned in
the Basel II context, to have risk measurement and management
systems appropriate for their size and risks, including
internal economic capital methodologies for managing their
businesses. What Basel II does is ask them to make sure that,
as they build more robust risk management systems, those
systems can also generate various risk parameters to permit
calculation of risk-based capital using the supervisory
formulas. So as a bank continues to grow toward the Basel II
criteria, it should be both improving its risk management and
preparing to have its systems generate required Basel II
criteria. Improving risk management systems as a bank grows in
size or complexity should be considered as a cost of doing
business.
There may also be concerns that Basel II will reduce the
capital requirements for some banks and not others. That is one
of the reasons why the banking agencies have proposed to amend
Basel I so that competitive effects can be minimized. As noted
above, the Federal Reserve recognized that competitive effects
were possible and undertook several studies designed to assess
the possibility that adoption of Basel II would provide
adopters with a substantial competitive advantage over non-
adopters. While these studies point to a few areas in which
some competitive effects may occur, their overall findings
strongly suggest that competitive effects of Basel II are
likely to be modest. One study examined whether a reduction in
regulatory capital requirements resulting from adoption of
Basel II would prompt large adopting banks to acquire smaller,
non-adopting banking organizations. A second study addressed
whether changes in regulatory capital requirements brought
about by Basel II would provide large adopting banks with a
competitive advantage in small business lending; the third and
fourth studies addressed similar questions as they apply to
mortgage lending and credit card lending, respectively. A fifth
study addressed the likely effect of explicit operational risk
charges on processing banks that adopt Basel II. And even
though anticipated effects are likely to be modest, because
these studies, public comments, and other factors pointed to
some unintended competitive consequences from the
implementation of Basel II, the U.S. banking agencies are
developing a revised Basel I that is partially aimed at
addressing such concerns.
Q.8. Please explain the type of information that banks will be
required to disclose under Pillar 3 with respect to how banks
calculate their capital requirements? Will banks be required to
disclose proprietary capital requirement models?
A.8. Consistent with the Pillar 3 disclosure requirements in
the Basel Committee 2004 mid-year text, the agencies will be
proposing, at about the same time as the NPR, to collect
information, both qualitative and quantitative, related to
scope of application (that is, what entities within an
organization are using the Basel II-based capital rules), an
institution's capital structure, capital requirements as
calculated under the rules, as well as disclosures related to
specific risk areas including: Certain credit risk-related
exposures, credit risk mitigation, securitization, market risk,
operational risk, equities and interest rate risk. Under the
Pillar 3 requirements, banks will not be required to disclose
propriety information related to their capital models.
Q.9. If Basel II is implemented, do you favor at any point in
the future reducing the leverage ratio for well-capitalized and
adequately capitalized banks or reducing the statutory capital
thresholds for prompt corrective actions?
A.9. The Federal Reserve is not in favor of any attempt to
reduce or eliminate the leverage ratio, nor to reduce or
eliminate statutory thresholds for PCA. These existing
requirements would provide additional protection against
possibly unfavorable outcomes in Basel II implementation. From
a minimum regulatory capital perspective, not all risks have
been explicitly incorporated into the Basel II-based framework
in terms of a specific capital charge. To be clear, we believe
that the leverage ratio and PCA rules should remain in place
for banks operating under both Basel I and Basel II rules.
The Federal Reserve also would note that all of the banking
agencies would need to concur before bank leverage requirements
and PCA thresholds could be eliminated. Moreover, since
Congress has mandated some aspects of these requirements for
banks, changing them would require an act of Congress.
Q.10. In his testimony, Mr. Isaac points out that compared to
European banks, ``large U.S. banks are much better capitalized
and far more profitable in terms of their operating margins.''
Do you agree or disagree with Mr. Isaac's observation, and what
is your rationale for your position?
A.10. The strength and sustained earnings power of the U.S.
banking industry over the past several years have indeed been
remarkable. U.S. banks have regularly delivered record profits
and higher shareholders' equity. In this setting, most
comparisons of U.S. banking organizations with many foreign
banks will favor the U.S. institutions. The chart below reports
aggregate capital measures for the seven U.S. bank holding
companies and 28 private-sector European banks with assets
greater than $200 billion, based on public disclosures and
shown on a merger-adjusted basis. Crude equity-to-assets ratios
(shown as bars) were significantly higher for the U.S. banking
organizations. Gauged by the total risk-based capital ratio,
however, capital levels at the largest American and European
banks were roughly comparable over the last decade. During the
10 years ending in 2004, U.S. banks reported a total risk-based
capital ratio of 11.63 percent on average, only 29 basis points
above that for European banks. Although in most of these years
the average ratio for U.S. banks was higher, in others (1997 to
1999) the average ratio for European banks exceeded that for
U.S. banks. Both sets of results contrast sharply with data for
1994, which show sharply higher capital ratios for U.S. banks
that reflected buttressing at many U.S. banks against still-
significant asset quality problems and a buildup of capital
ratios at European banks during the early years of the Basel I
regime.
<GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT>
Although there are many differences between the balance
sheets and risk profiles of U.S. and European banking
organizations, the difference that seems to be most important
to this comparison of capital ratios is that the European banks
tend to have relatively higher concentrations in government
bonds and other assets that receive low risk weights under the
current Basel standards. The risk-based capital ratios take
account of this difference in risk profiles--albeit crudely--
while the equity-to-assets ratios do not.
All 35 of the large institutions in the data above, and
indeed nearly all banking organization in these countries, hold
capital well in excess of their current regulatory minimum
requirements and are judged to be adequately capitalized--or
better--by their regulators. As such, the differences in
capital levels are best understood as the result of choices
made by the organizations' management rather than a competitive
advantage or disadvantage. Minimum capital requirements are
just one of many factors bank managements consider in
developing their capital policies. Banks may choose the size of
their capital ``cushions'' with many objectives in mind,
including positioning themselves as desirable counterparties,
managing funding costs, optimizing their risk profile, and
taking advantage of potential growth opportunities. Market
discipline plays an especially significant role for these 35
institutions because all are active in global financial markets
and are thus subject to the scrutiny of their counterparties
around the world. Supervisors review the capital cushions and
their rationale as part of the larger assessment of capital
adequacy, but the size of the cushions is based on managements'
own assessments and judgments.
Turning to profitability, the chart below displays
aggregate return on average assets (ROA, a reasonable proxy for
overall operating margins) for the same group of large U.S. and
European banking organizations, showing that on this basis U.S.
banks indeed have generated consistently higher profit margins.
Market participants and supervisors generally prefer to measure
profitability as return on equity, that is, as the ultimate
return to shareholders. On this basis, a differential in
profitability is still evident although less pronounced and not
present in every year.
Since 1994, return on equity averaged roughly 16 percent
for U.S. banks versus 12 percent for European banks.
<GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT>
Q.11. Do you believe that U.S. banks are operating at a
significant disadvantage vis-a-vis foreign banks due to the
higher capital requirements imposed on U.S. banks? If you do,
please explain in detail and quantify the competitive effects
on U.S. banks, including giving your assessment of how low U.S.
capital requirements would need to go in order to level the
playing field with large foreign banks.
A.11. Minimum regulatory capital requirements imposed on
internationally active U.S. and foreign banks have been
substantively the same since the implementation of the 1988
Basel Accord. The 1988 Accord was intended to bring global
commonality to the
assessment of capital adequacy and to promote stronger capital
positions in banks globally. There exist some differences in
implementation across local jurisdictions that arise in large
part from differences in banking system structures, accounting
practices, and capital instruments. Nonetheless, the risk-based
capital ratios provide reasonably comparable indicators of
capital adequacy.
Such assessments and comparisons between U.S. and local
jurisdiction requirements are performed routinely by the
Federal Reserve as part of its responsibility to supervise
State-chartered branches and agencies of foreign banks
operating in the United States. Foreign banks intending to do
business in the United States are expected to meet the same
general standards, experience, and reputation as required for
domestic institutions.
As noted in the response to question 10, large banking
organizations in the United States and Europe--and indeed
nearly all banks in these countries--have operated at capital
levels well in excess of these regulatory minimum standards for
many years. The size of these capital cushions is chosen by
these institutions' management after considering a number of
factors, including the
demands and expectations of the banks' counterparties and
customers, and investors, the needs of the institutions'
particular business strategies, the institutions' funding mix
and costs, the overall risk posture of the institutions, and
flexibility to accommodate future business growth or
acquisitions. Supervisors review the capital cushions and their
rationale as part of the larger assessment of capital adequacy,
but the size of the cushions is based on managements' own
assessments and judgments.
Despite the convergence of capital standards and guidelines
for meeting capital equivalency for foreign banks operating in
the United States, there remains the perception that higher
capital requirements are imposed on U.S. banks, placing
domestic banks at a competitive disadvantage vis-a-vis foreign
banks. Such perception may be based on the additional leverage
requirement for U.S. banks based on the ratio of Tier 1 capital
to total assets. These additional requirements support the
safety, soundness and resilience of the U.S. banking system.
Actual capital adequacy ratios remain well above all regulatory
minimum levels, including Tier 1 leverage requirements and
standards for being well-capitalized under PCA, at essentially
all U.S. banks and all of the largest European institutions. As
such, there is no indication that capital adequacy requirements
have significant effects on the competitive balance between
U.S. and foreign banks.
Q.12. If capital standards are reduced for the largest U.S.
banks, do you believe that will place U.S. banks that do not
implement Basel II at a competitive disadvantage vis-a-vis
banks that do implement Basel II? If so, how do you suggest
that competitive inequity be remedied?
A.12. The Basel I ANPR presented options to make minimum
regulatory capital more reflective of risk. The regulators are
proposing to recalibrate capital by portfolio type to reflect
higher or lower risk exposures based on current information. A
better alignment of capital to risk in Basel I would mitigate
the potential differences between that framework and the more
risk sensitive Basel II framework. As noted earlier, one of the
objectives of the amended Basel I ANPR was to begin the process
of making enhancements to the existing minimum risk-based
capital rules for all banking institutions operating in the
Unites States, in part, to mitigate competitive inequities that
may arise between organizations using the Basel I-based rules
and those using the Basel II-based rules. Overall, we believe
that potential competitive inequities arising from the two sets
of rules will be limited. As we go forward, we want to ensure
that institutions not moving to Basel II have fair
opportunities to pursue business initiatives and are not
adversely affected. We will continue to work on these
initiatives in tandem and will be seeking input from the
industry, Congress, and others along the way, particularly
related to the potential effects of both initiatives.
Q.13. Have you considered requiring the largest U.S. banks to
implement Basel II-type models as part of their risk management
programs without making Basel II a capital-regulation device?
If you have considered and rejected such an approach, please
explain why you rejected it.
A.13. We have considered whether we can just continue to
encourage the improvement in risk modeling at banks and stop
there, that is, not tie risk models to capital. While
improvements in the methodology of risk models and the
transparency of better risk modeling in business decisionmaking
are very useful, it is our view that one cannot stop there.
Banks that have similar models of risk can have very different
inherent levels of risk. That is, each institution has to
decide what level of risk it is willing to accept to run its
business. Some organizations are willing to take on much
greater levels of risk than others. Some lines of business and
some products are inherently riskier than others, while some
organizations may be more adverse to accept risk than others.
For safety and soundness reasons, bank supervisors must be
sure that a bank with greater exposure to riskier lines of
business, products, and customers holds more capital than a
bank that is more risk adverse and designs its business plan to
minimize risk-taking. That is, just looking at risk models and
not tying capital to the measured risk exposures does not
provide the backstop that supervisors need to ensure that each
institution has the appropriate capital in place before the
unexpected loss occurs. Capital should be based on risk
exposures, and the evolving risk-modeling methodologies provide
improved tools to better determine the appropriate level of
capital.
That is why the Basel II framework contains minimum
regulatory capital requirements in Pillar 1, to tie risks to
levels of capital. In Pillar 1, the framework seeks to achieve
a balance between risk sensitivity and prudence by building on
certain firm-specific inputs, but deriving capital outputs from
a regulatory prescribed capital model. Pillar 1 is, of course,
complemented by Pillar 2 (supervisory review) and Pillar 3
(market discipline). Together, the three pillars should provide
banks with the appropriate incentives to move toward leading
risk-measurement practices and provide supervisors and the
marketplace with a conceptually sound framework for assessing
how capital and risk evolve over time, which
institutions are outliers, and, ultimately, whether the banking
system is becoming more or less sound from a capital adequacy
perspective.
Q.14. Have you compared the capital requirements for the
participating banks that came out of your Quantitative Impact
Study with the judgments of your professional bank examination
staff on the capital needs of those same banks? How closely
correlated were the results?
A.14. The QIS-4 exercise involved not just Basel II experts
from within the U.S. banking agencies, but also members of the
on-site teams responsible for day-to-day supervision of the
entities participating. On-site supervision staff received and
reviewed the information submitted in QIS-4 by their respective
institutions. But, they did not review the measurement systems
and methodologies used by the banks to construct the
information. The incorporation of on-site teams into the QIS-4
process is consistent with our overall objective of
``normalizing'' Basel II work into regular supervision for
relevant institutions as we move closer to implementation. A
major portion of the QIS-4 exercise was a qualitative
questionnaire filled out by participating institutions, in
which they described what stood behind the numbers they
produced (the data sources, methodologies, etc.). On-site teams
were asked to compare their knowledge and understanding of the
institution with what was contained in the questionnaires.
In general, on-site supervision staff agreed with the
conclusions of Basel II experts about the QIS-4 results namely,
that the drop in QIS-4 required minimum risk-based capital was
largely due to the favorable point in the business cycle, and
the dispersion among institutions was largely due to the
varying risk parameters and methodologies they used.
Importantly, we also learned that some of the data submitted by
individual institutions was not complete; in some cases banks
did not have estimates of loss in stress periods or used ad hoc
estimates, which might have caused minimum regulatory capital
to be underestimated. As we have said before, we would not be
comfortable qualifying any institution for Basel II based on
QIS-4 results, since it was just a point-in-time look at how
banks are progressing and was conducted on a ``best-efforts''
basis, rather than in ``full compliance'' with the Basel II
rules, without the benefit of additional supervisory oversight.
In the future as banks move to the parallel run and prepare to
qualify for Basel II certification, the supervisory oversight
process will play a significant role in determining whether the
measurement and risk management systems used by the
participating banks are consistent with the Basel II
requirements. There is obviously more work to be done. The
recognition of the amount of work still required is reflected
in the U.S. agencies' plan for a one-year delay before Basel II
is implemented and a one-year extension of transitional floors
once it is.
Q.15. Basel II represents an attempt to apply the same formulas
for measuring commercial loan credit, consumer loans, and
commercial real estate loan risk to banks in the United States
and to banks in other developed countries. What evidence do you
have that the formulas will assure that U.S. banks will hold
sufficient capital to see them through a variety of economic
scenarios, such as those that have occurred in the United
States over the last 25 to 50 years? Are the markets and
economic environment similar enough that the same formulas can
be applied safely to large banks from the United States,
Canada, the United Kingdom, France, Germany, Japan, etc.?
A.15. The Basel II framework contains formulas for calculating
risk weights based on bank inputs. In developing Basel II, it
was necessary to create ``average'' risk-weight functions to
apply across countries that plan to implement it. While these
formulas do not necessarily fully capture each institution's
risk profile perfectly, they are the best approximation at this
time for producing similar capital requirements given similar
risks across a range of institutions. One of the key features
of the Basel II framework is that many specifics with respect
to implementation are left to national discretion. As a result,
certain aspects of the framework, which could include some
parts of the formulas, will be tailored to national
jurisdictions. Moreover, to the extent that there are
differences in the risk-taking of banking institutions across
the major developed countries, the Basel II framework was
designed to take these differences into account through the
resulting differences in bank-provided risk parameters,
including estimates of probability of default, loss given
default, and exposure at default.
Given that the Basel II framework is risk sensitive,
minimum regulatory capital requirements should be expected to
fluctuate to a certain extent as economic conditions change.
This aspect of the framework is necessary and desirable,
sending signals to both banks and supervisors about changing
credit conditions and potentially aiding the evaluation of the
relationship between actual and minimum capital levels at any
point in the economic cycle. That said, the Federal Reserve is
working diligently to ensure that U.S. banks will hold
sufficient capital to see them through a variety of economic
scenarios. Indeed, the framework has been constructed with
safeguards in Pillars 1, 2, and 3 to help ensure that minimum
capital requirements remain prudent at all points in the
economic cycle. As we gain additional experience from banks'
implementation of the revised framework, we will continue to
assess the adequacy of the framework, including the behavior of
minimum capital requirements over the economic cycle. We also
will have other supervisory tools to use in conjunction with
Basel II, things like the
leverage ratio and PCA.
As part of the U.S. rulemaking process, the U.S. agencies
are seeking comment on the entire framework, including its
formulas, to determine its appropriateness for U.S.
institutions. The U.S. agencies have already demonstrated their
willingness to deviate from other countries' implementation
strategies by extending the start date for Basel II in the
United States and calling for an additional year of
transitional floors. And, should it become necessary, the
Federal Reserve would be prepared to consult with our
colleagues at home and in other Basel member countries to make
changes to the framework itself, in keeping with the objective
of ensuring the safety and soundness of the U.S. banking
system.
Q.16. The agencies' current transitional plan for implementing
Basel II contemplates allowing capital levels to fall to an 85
percent ``floor''. How can the agencies know, even before the
``parallel run'' contemplated for 2008, that a drop of 15
percent in the capital of banks would be consistent with safety
and soundness considerations?
A.16. It is important to remember that the floors described in
the Basel II proposals apply to individual institutions, not
the banking system as a whole. The floors require a financial
institution to hold capital at the higher of the minimum
regulatory capital determined by its risk models or the
transition floor. It is only when an individual institution can
demonstrate to supervisors that it has a lower risk profile
than Basel I implies, that supervisors will then consider
allowing that institution's minimum regulatory capital to
decline to the floor levels proposed under Basel II. And after
the 3 years of floors, supervisors will continue to monitor
institutions' risk profiles to see if the declines in minimum
regulatory capital are still warranted. Of course, for some
institutions, minimum regulatory capital may increase under
Basel II--reflecting the greater risk sensitivity of the new
framework.
Additionally, the implementation of Basel II will progress
in stages, and only after supervisors achieve comfort at the
end of each stage will we move on to the next. The agencies'
response to QIS-4 is a good example of how we took extra time
before moving forward, and then did so with a revised plan. We
expect to use the same type of prudence for the period of
transitional floors, and by that time we will have much more
information than today about Basel II and its effects. Even
before moving to the transitional floors, banks will have to
demonstrate their ability to produce credible Basel II minimum
capital measures; that is, they will be able to move to the
first year of floors only if the primary supervisors are
satisfied with what they see during the parallel run. The first
year of floors is set at 95 percent (implying at most a 5
percent decline for any given institution). If the primary
supervisors are not comfortable with a bank moving to the
second-year floor (at 90 percent), they might retain the 95
percent floor for that institution. Similarly, primary
supervisors will have to give approval for an institution to
move from the 90 percent to the 85 percent floor. The primary
supervisors will also have to approve a banking institution
moving from the 85 percent floor to full implementation without
a floor.
Q.17. Basel II's capital formulas are based on highly complex
models that are to be implemented on a bank specific basis.
While Basel I has been criticized for its oversimplification,
the sheer complexity of Basel II has raised some very serious
concerns. In his written testimony, Mr. Isaac states, ``Basel
II is so complex it cannot be adequately understood by senior
bank managements, boards of directors, regulators, or the
public.'' Professor Kaufman states, ``Increased complexity is
likely, however, to both increase compliance costs and reduce
understanding, particularly by the bank CEO, board of
directors, and possibly even the CFO and by bank supervisors.''
Do you share Mr. Isaac's and Professor Kaufman's concerns? If
you share their concerns, what recommendations do you have for
the regulators as to how to fix this problem?
A.17. Basel II is designed to reflect the risk-management
practices of large, internationally active financial
institutions. These institutions have become more complex in
the sophistication of their services and business practices, as
well as in their organizational structures. As a result,
effective risk management has been evolving to support these
innovative financial structures. Indeed, Basel II may not look
that complex to many of the risk managers at institutions
today. Most banking organizations involved in complex financial
instruments should already possess an understanding of advanced
risk concepts and should have implemented effective risk-
measurement and -management practices. Indeed, the largest,
most complex institutions in the United States are already
employing
models as complex as, or more complex than, those in Basel II,
underscoring the need to move to a capital adequacy framework
that is more closely aligned with how banks measure and mange
risk internally.
As prudent supervisors, all of the U.S. banking agencies
require any organization employing sophisticated financial
practices or using complex financial instruments to have
adequate risk management to address its risk positions, as well
as governance and control structure commensurate with those
activities. This includes having knowledgeable staff to set
risk limits effectively, and clearly communicate risk positions
and risk-management strategies to executive management and
boards of directors.
The advanced approaches of Basel II are complex and have
many moving parts, which is one of the reasons that in the
United States we are proposing to require only the largest,
most sophisticated financial institutions to adopt Basel II.
For these organizations, the incremental cost of adopting Basel
II advanced approaches, while admittedly significant, should be
relatively modest compared with the significant risk-management
investments they have already made.
It is our view that via increased disclosures as part of
Pillar 3, Basel II will actually augment the general
understanding of banks' capital adequacy in relation to
underlying risks by providing increased information on banks'
risks and how they are measured and managed.
Q.18. According to FDIC data, the equity to capital ratio of
insured U.S. financial institutions reached a low of slightly
above 5 percent in the late 1980's, during the S&L crisis, and
then rose steadily through the 1990's reaching the level of
over 10 percent, at which it stands today. Profits--as measured
by return on assets--which had fallen dramatically during the
period of low capital, also began a steady rise in the 1990's,
rising to well over 1 percent for the last decade, reaching
record levels. When looked at together, it becomes clear that
large profits and strong levels of capital can coexist. Do you
agree that strong levels of banks capital can and do coexist
with strong bank profitability?
A.18. Profitability has a direct relationship to capital
levels. As banks' profits recovered in the early 1990's, the
amount of earnings retained after paying out dividends also
rose. Retained earnings contribute significantly to increased
levels of equity and total capital.
Profitability and healthy capital levels are both important
to a safe and sound banking system. Earnings and capital
adequacy are two components of the FFIEC's CAMELS supervisory
rating system. Strong and consistent profits provide
competitive returns to shareholders and allow for internal
generation of capital while healthy capital levels allow banks
to weather potential adverse conditions, protecting the deposit
insurance funds and aiding funding and liquidity by making the
bank more attractive to market participants as a counterparty.
Moreover, strong capital means that bank owners have a
significant stake in ensuring that the institution is managed
in a prudent manner. Consistently strong earnings and capital
also suggest that institutions have prudently balanced their
appetite for risk-taking with return, an objective that the
Federal Reserve believes will be further advanced by the Basel
II framework.
The chart below depicts aggregate capital ratios and
profitability (measured as return on assets, or ROA) for
insured commercial banks in the United States since 1984. The
lines in the chart show three capital measures, including a
simple equity-to-assets ratio and, since the implementation of
the 1988 Basel Accord (which took effect in 1992), the more
risk-sensitive total risk-based capital ratio and the Tier 1
leverage ratio.
<GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT>
In the late 1980's, the simple equity-to-assets ratio was
just a little over 6 percent. The implementation of the Basel
Accord and the PCA framework in the United States were part of
a broader supervisory emphasis on strengthening capital
positions. The positive
results of this emphasis are borne out by steady increases in
equity-to-assets and leverage ratios over time, and aggregate
risk-based capital ratios that are well in excess of minimum
standards. Implicit in the aggregate data shown here is that
nearly all insured commercial banks have consistently
maintained regulatory capital ratios that exceed minimum
standards for being well-capitalized under the PCA framework.
Profitability also improved in the 1990's for many reasons,
including recovery from the downturn in commercial real estate
in the 1980's, innovation and the growth of new business lines,
expansion of bank activities in the capital markets arena, and
improved operating efficiency.
On balance, strength in both profitability and capital
levels is important to the soundness and resiliency of a
banking system. During the most recent economic downturn,
capital and reserve levels provided a cushion for banks to
absorb losses and weather the credit cycle. Banks that are
strongly capitalized may enjoy greater flexibility and a
broader range of choices in how to handle adverse credit or
operational conditions. That ultimately may have a beneficial
impact on their long-term profitability.
Q.19. Under the Administrative Procedure Act, agencies are
required to consider public comment prior to issuing new final
regulations. How can the agencies fully comply with this
requirement? Doesn't the fact that the agencies reached an
international agreement on the new capital framework
effectively limit their ability to fully consider public
comment and modify proposed regulations in light of that
comment?
A.19. The Federal Reserve has been, and intends to continue,
complying fully with the letter and spirit of the
Administrative Procedure Act (APA) in connection with the
pending U.S. rulemaking to implement Basel II. Although not
required by the AP A, the US. banking agencies issued an ANPR
on the U.S. implementation of Basel II in August 2003 to ensure
that any proposals by the agencies to implement Basel II in the
United States would reflect comments from banking
organizations, Members of Congress, and other interested
parties.\1\ Comments received on the ANPR influenced
significantly the June 2004 Basel II document agreed to by the
U.S. banking agencies and other G--10 bank supervisory
authorities and will influence in important ways the
forthcoming NPR on the United States implementation of Basel
II. We note that the Basel II framework has changed
considerably since the first proposal by the Basel Committee in
the late 1990's, in material part due to the valuable comments
of U.S. financial institutions and other interested parties in
the United States.
---------------------------------------------------------------------------
\1\ 68 Federal Register 45900 (August 4, 2003).
---------------------------------------------------------------------------
The June 2004 Basel II document is not a binding agreement
that has the force of law in the United States. In addition,
while it lays out important principles, it leaves to national
discretion much of the specific details regarding
implementation. The U.S. agencies expects that the forthcoming
NPR would be generally consistent with the June 2004 Basel II
document but likely will vary in some of its details from the
June 2004 Basel II document to accommodate some special
features of U.S. banking organizations and the U.S. financial
markets. In addition, if comments received on the NPR suggest
that changes should be made to the June 2004 Basel II document
or to any U.S. final rule that implements Basel II, the Federal
Reserve will work with the appropriate parties to make those
changes.
Although not required by the APA, the Federal Reserve and
the other U.S. banking agencies also have sought public comment
on a draft of the supervisory guidance that would accompany any
U.S. rule that implements Basel II.\2\ The agencies have
revised this draft guidance to reflect comments from U.S.
banking organizations and intend to seek a second round of
comments on the
guidance soon after issuance of the NPR. The agencies have
endeavored, and will continue to endeavor, to be as transparent
as possible about all aspects of Basel II implementation in the
United States, including the important risk-measurement and -
management infrastructure requirements of the Basel II
framework.
---------------------------------------------------------------------------
\2\ 68 Federal Register 45949 (August 4, 2003: 69 Federal Register
62748 (October 27, 2004).
---------------------------------------------------------------------------
Finally, it is useful to remember that Basel I has been
amended 28 times in the United States since it was introduced,
often in order to keep up with advancements in the industry;
the recent changes relating to certain securitization
transactions are a good example. Today's U.S. Basel I rules
differ from Basel I rules in other countries, appropriately
reflecting the characteristics of national banking systems. We
expect to tailor Basel II rules to the U.S. environment in a
similar fashion.
Q.20. The Basel II framework ignores a significant risk:
Interest rate risk. Many banking assets, such as mortgage
servicing rights, are extremely sensitive to changes in
interest rates, and the value of assets such as these may
fluctuate dramatically when interest rates change. How do you
propose to take interest rate risk into account under the Basel
II framework?
A.20. The Basel II framework addresses interest rate risk
substantively as part of the Pillar 2 process. Specifically to
support the Pillar 2 approach, the Basel Committee issued in
July 2004 a paper, ``Principles for the Management and
Supervision ofInterest Rate Risk.'' This paper combines a
principles-based approach along with a quantitative benchmark.
The first 13 principles address the need for effective interest
rate risk measurement, monitoring and control functions within
the interest rate risk management process. Principles 14 and 15
address the supervisory treatment of interest rate risk in the
banking book.
The paper provides guidance to help supervisors assess
whether internal measurement systems are adequate. If
supervisors determine that a bank has insufficient capital to
support its interest rate risk, they must require either a
reduction in the risk or an increase in the capital held to
support it, or a combination of both. Supervisors are cautioned
to be particularly attentive to the capital sufficiency of
``outlier banks''--those whose interest rate risk in the
banking book leads to an economic value decline of more than 20
percent of the sum of Tier 1 and Tier 2 capital following a
standardized interest rate shock or its equivalent. Individual
supervisors may also decide to apply additional capital charges
to their banking system in general.
In practice, the approaches developed in this paper,
supporting Pillar 2, are consistent with the existing
supervisory practices of the Federal Reserve and other Federal
banking and thrift regulators. Additionally, if the agencies
begin to see evidence that the existing treatment of interest-
rate risk in Basel II is inappropriate, inaccurate, or
inadequate, they can propose changes.
Q.21. Due to the tremendous devastation caused by Hurricanes
Katrina and Wilma this fall, it is estimated that over 100,000
homes have been lost. Would the Basel II framework have
provided institutions with enough capital buffer to handle a
calamity of this magnitude?
A.21. Over the past several years, as part of broader risk-
management improvements, banks have been upgrading business
continuity and contingency planning for natural disasters or
other large-scale events. Preparations for Y2K, lessons learned
from September 11, and experience with recent natural disasters
have aided in these efforts. We believe that Basel II would
provide even more benefits to institutions in this particular
area because of its expected risk-management enhancements.
These enhancements would likely apply to managing both
operational risks and credit risks. For example, institutions
should plan for the possibility that their business operations
could be disrupted by a natural disaster in a certain location.
Similarly, as they estimate possible losses during stress
conditions, institutions should take into account their credit
risk concentrations and the possibility that obligors in an
entire geographic area could be adversely affected and suffer
economic and financial difficulties. In fact, as part of their
work related to Pillar 2, institutions should conduct analyses
of the buffer they hold above minimum regulatory capital levels
to ensure that actual capital held reflects all risks to which
that individual institution is exposed. In general, Basel II
should provide institutions with a better sense of the capital
needed to support risks associated with natural disasters or
other large-scale stress events than they currently have, as
well as an improved ability to manage those risks.
Q.22. Last week, on November 18, 2005, the Financial Times
published an editorial by Harold Benink and Jon Danielsson
titled, ``There is a Chance to Correct the Defects of Basel
II.'' Harold Benink is Professor of Finance at RSM Erasmus
University in the Netherlands. Jon Danielsson is Reader in
Finance at the London School of Economics. They concluded:
``Instead of requiring bank capital to be risk sensitive,
banking regulation should simply require the use of high
quality risk models in banks without using their output to
determine capital. Minimum capital is better calculated as a
simple fraction of bank activity in broad categories.'' What is
your response to this suggestion?
A.22. See response to question 13. Regarding the suggestion
that minimum regulatory capital is better calculated as a
simple fraction of bank activity in broad categories, this is
essentially what the current Basel I capital framework does.
However, the Federal Reserve believes that this simple
framework has become increasingly inadequate for large,
internationally active banks offering ever-more complex and
sophisticated products and services. Today's capital framework,
left unchanged, could eventually undermine the safety and
soundness of the financial system. It encourages banks to hold
on to high-risk assets and shed low-risk assets and does not
provide incentives for better risk-management techniques and
processes.
Basel II builds on the risk-management approaches of well-
managed banks and creates strong incentives for banks to move
toward leading risk-measurement and -management practices. It
establishes a coherent relationship between how supervisors
assess regulatory capital and how they supervise the banks.
Thus, it will
enable examiners to better evaluate whether banks are holding
prudent capital levels, given their risk profiles, and to
better understand differences across institutions. For safety
and soundness purposes, prudential supervision requires that
sound bank management requires the ability to model and measure
risk, and that minimum regulatory capital reflect the amount of
risk exposure that the individual institution chooses to
accept.
Q.23. In his testimony, Mr. Isaac states, ``I have not found a
single professional bank supervisor who is enthusiastic about
Basel II.'' We have also heard of a similar lack of enthusiasm
on the part of the supervisors who will actually be responsible
for supervising Basel II. What is your response? Separately,
what gives you confidence that regulators will be up to the
task of understanding and policing Basel II?
A.23. In our view, most U.S. bank supervisors are enthusiastic
about risk-management improvements expected through the
application of Basel II and fully support the concepts behind
the framework. In fact, supervisors already work with bankers
at large,
complex institutions to advance their ability to employ more
sophisticated and quantitative risk measures and corresponding
risk-management practices. While this is the first time that we
are using bank inputs as determinants of minimum regulatory
capital requirements for credit and operational risk exposures,
we do have the experience of the Market Risk Amendment, which
has been implemented successfully in terms of promoting better
risk management and measurement of market risk exposures. That
said, bank supervisors, as a group, are skeptical and cautious-
indeed, they are trained to be so. So it not surprising that
some bank supervisors would have some reservations about
immediately adopting a new capital framework whose full effects
are not yet known. Furthermore, U.S. bank supervisors will want
to ensure that the risk-measurement and -management
improvements expected through the application of Basel II
actually prove fruitful before allowing institutions to use
internal inputs for regulatory capital.
Accordingly, the U.S. agencies are taking a careful and
deliberate approach to adopting Basel II in this country, and
will continue to review and analyze its potential impact with
each successive step in the process. The extensive comment
period, the parallel run, the transitional floors, the
opportunity for additional review and analysis, and additional
rulemakings all indicate that we will not move to Basel II
unless we have solid evidence that it is meeting its objectives
and promoting safety and soundness of the U.S. banking system.
During each of these stages, supervisors will have full
access to institutional information, including data sources and
methodologies, in order to render a judgment about banks'
preparedness. The input from these supervisors will continue to
be a critical component of the judgments about elements of the
Basel II proposal as well as individual bank's readiness to
move to Basel II.
We should also note that U.S. bank supervisors have for
many years been involved with analyzing and examining
sophisticated risk-measurement and -management practices at
large, complex organizations. At many levels, our examination
staff has experience with practices similar to or even more
sophisticated than those described in Basel II. Our knowledge
and experience gives us some confidence in both the banks'
abilities to achieve the risk-measurement and -management
enhancements expected in Basel II, as well as confidence in our
own ability to supervise banks under Basel II. But as always,
we, as good supervisors, will still need to see actual results
before we can be completely comfortable.
RESPONSE TO WRITTEN QUESTIONS OF SENATOR SHELBY
FROM DONALD E. POWELL
Q.1.a. Basel II would impose new demands on bank regulators. It
would require bank regulators to not only approve banks'
internal assessments of their risks, but also to make judgments
on whether banks are holding enough capital in light of all the
risks banks face. This means bank regulators are going to need
personnel who are well-trained in risk analysis.
How many additional personnel does your agency expect it
will need to hire in order to implement Basel II? What type of
expertise will your agency need to obtain? Do you foresee any
problems in recruiting?
A.1.a. The FDIC's staffing needs relative to Basel II derive
from two roles: (i) its role as primary Federal banking
supervisor of
insured depository institutions that opt-in to Basel II, or
that otherwise adopt Basel II by virtue of being bank
subsidiaries of organizations that are required to adopt Basel
II; and (ii) its role as deposit insurer of other banks
adopting Basel II.
It is anticipated that most of the staffing needs relative
to Basel II derive from the first role. There are roughly 20
FDIC-supervised insured banks that have either expressed an
interest in opting in to Basel II or are subsidiaries of
mandatory banking organizations. Basel II implementation for
these banks will involve a hybrid supervisory approach whereby
traditional examiner judgment and risk analysis skills will be
supplemented with targeted quantitative review of risk
management models.
Our approach to meeting our Basel II staffing needs is to
reallocate and train existing examiners that already possess
sound risk analysis and judgmental capabilities, and target
other specialists to address non-traditional examination
functions. This has been supplemented with a limited amount of
external hiring of individuals with academic or practitioner
backgrounds in quantitative risk measurement. Appropriate
staffing levels, as well as the efficiency and effectiveness of
internal processes, will continue to be monitored on an on-
going basis. We anticipate additional staffing requirements as
the implementation process goes forward.
Q.1.b. Will your agencies be able to successfully compete
against banks in the recruitment of personnel? Do you have any
concerns that banks will simply be able to hire the best
Ph.D.'s in risk modeling and outgun their regulators?
A.1.b. In part because of their ability to offer more lucrative
compensation packages, large banks will indisputably enjoy
certain
advantages in recruiting qualified and motivated risk modeling
professionals both from the private sector and from regulatory
agencies. This is by no means a new issue created by Basel II,
but it may become more pronounced in the coming years as banks
build out required Basel II systems.
Two countervailing considerations must be kept in mind,
however. First, the agencies have traditionally been able to
attract and retain highly qualified individuals, in both
traditional examination disciplines and more quantitative
disciplines, for whom public service is a satisfying career
path. Second, it must be emphasized that the evaluation of
proprietary bank models will be a defined and narrow subset of
overall Basel II implementation activities. Traditional
examination functions of assessing overall risk, adequacy of
banks systems, internal controls, and management oversight will
continue to play a lead role in the overall supervisory
process.
Q.1.c. How much in additional annual costs do you expect to
incur in implementing Basel II? Other than personnel costs,
what other costs will your agency incur in implementing Basel
II? Are there any large one-time costs?
A.1.c. It is difficult to isolate implementation costs related
to Basel II, in part because Basel II represents only one
aspect of the overall increasing complexity associated with the
supervision and insurance assessment of large insured
depository institutions.
Over the next few years, the FDIC's costs will be primarily
associated with staff time devoted to the rulemaking processes,
guidance development and training, industry outreach, and on-
site
examination processes. The cost associated with recruiting and
retaining qualified staff will also likely increase in the
coming years.
Q.2.a In his testimony, Dr. Tarullo posed several questions
that he believes need to be answered by U.S. banking agencies
before implementation of Basel II begins. Please provide
answers to his questions with respect to your agency, which are
as follows: ``Do the agencies believe that, in general, minimum
capital levels are too high and, if so, what is the basis for
that belief ?
A.2.a. The FDIC does not believe that minimum capital levels
are too high. There is little evidence that the historically
high capital levels in the United States have limited business
opportunities for banks or the availability of credit. Indeed,
the industry as a whole has had record earnings over the past
few years.
Market analysts also do not believe that lowering capital
standards would benefit the industry. For example, Standard &
Poor's recently (November 30, 2005) released a report on ``Bank
Industry Risk Analysis: United States.'' In that report, the
analysts observed that ``A final uncertainty for the capital
strength of the sector is the adoption of the Basel II capital
accords. While Standard & Poor's wholeheartedly endorses
efforts to link capital levels and risk, we do not believe that
the industry as a whole is overcapitalized. For this reason,
material reductions in capital would be viewed as a negative
for the industry's strength.''
Q.2.b. Have the agencies been able to predict with reasonable
precision the levels of capital that the A-IRB approach will
require of large banks, and then to confirm their prediction
through studies that are well-conceived and executed?
A.2.b. No. There are diverging views about the levels of
capital that the A-IRB approach will ultimately require of
large banks. There have been a number of major interagency
studies of this issue in which potential adopters of Basel II
provided their own
estimates of specified risk parameters to lengthy and detailed
spreadsheets designed to calculate the resulting A-IRB capital
requirements. The results of such studies remain subject to
considerable uncertainty, as discussed below.
The agencies have stated an expectation that capital
requirements under the advanced approaches of Basel II would
not, on
average, be substantially less than current risk-based capital
requirements, although capital requirements for individual
banks might show more variation based on their risk profiles.
The two most recent Quantitative Impact Studies, QIS-3
conducted in 2002-2003, and QIS-4 conducted in 2004-2005,
reported, based on banks own estimates, average reductions in
risk-based capital requirements of 6 percent and 15.5 percent
respectively for the participating U.S. banks.
A number of points need to be considered in evaluating
these estimates. First, the data on which they are based is
imperfect, based as they are on bank systems and models that
are in various stages of development, and made in many cases
without a clear understanding of what the Basel II systems
requirements and supervisory expectations ultimately will be.
Some observers have pointed in particular to a lack of
compliance by QIS-4 reporters with requirements to incorporate
stressed conditions and economic losses into estimates of loss
given default, tending, at least from this consideration alone,
to understate capital requirements.
Second, QIS-3 was conducted closer to a period
characterized by recession conditions in the corporate sector
while QIS-4 was conducted during more benign economic
conditions. From this difference alone one would expect a
greater reduction in capital
requirements to be reported in QIS-4 than was reported in QIS-
3. This, in fact, was the case.
Third, most QIS-4 banks reported their exposures as if they
received no capital benefit from guarantees, collateral, or
hedging, because their information systems were not yet
configured to allow them to do so. Were the systems in place to
allow the banks to benefit from the credit risk mitigation
already in place, the reductions in reported capital
requirements would have been greater, at least from this factor
alone.
Fourth, the FDIC has published analysis that suggests that
both QIS-3 and QIS-4 fundamentally overstated the capital
requirements likely to be produced by Basel II, once it is
actually up and running. This analysis has argued that
participating banks in both studies used risk inputs that were,
on average, more conservative than what the Basel II framework,
taken literally, would appear to require. Details of this
analysis are available in an article in the FDIC's ``FYI''
series published in December 2003, and in appendix B of
Chairman Powell's November 2005 testimony before the Senate
Committee on Banking, Housing, and Urban Affairs.
Finally, it should be noted that the magnitude of reduction
in capital requirements depends significantly on both the
capital concept being used and the measurement concept being
used. The QIS-4 aggregate results described above refer to the
average reduction--5.5 percent--in total capital requirements.
The median reduction, in contrast, was 26 percent. The
reductions in tier 1 capital requirements (tier 1 capital
excludes loan loss reserves, most subordinated debt, and other
elements not fully available to absorb losses on a going-
concern basis) were greater. The average QIS-4 reduction in
tier 1 capital requirements was 22 percent and the median
reduction in tier 1 capital requirements was 31 percent.
Q.2.c. Have the agencies done scenario planning to anticipate
the effects, unintended and otherwise, of Basel II upon the
financial system as a whole? For example, are significant
portions of certain kinds of assets likely to migrate out of
banks into the unregulated sector? If so, might new risks of
financial disruption be created?
A.2.c. There is little research on the effect of regulatory
capital standards on competition within the broader financial
industry. It is difficult, if not impossible, to disentangle
the effects of capital standards from other differences between
segments of the industry. For example, over the past 5 or 10
years, securities affiliates of banks have gained market share
at the expense of more traditional securities firms. The
reasons for this shift are not clear but this trend at any rate
does not suggest that these large banking organizations have
suffered from a regulatory capital disadvantage to date. Going
forward under Basel II, it is not clear whether there would be
any important classes of assets that would be forced to be
capitalized at higher levels within a bank than would be the
case if those assets were financed outside the Federal safety
net.
Q.2.d. What do the agencies regard as the effective capacity of
the specialized examiner teams that will be overseeing the use
of bank models under A-IRB? How will the agencies determine
whether other countries are successfully monitoring bank
capital levels under the A-IRB approach?
A.2.d. It should be noted that ``overseeing the use of bank
models under A-IRB'' will be but one aspect of Basel II
reviews, and this process is not new to regulators (for
example, market risk reviews). The FDIC has established
examination programs and processes designed to assess all
material risks within an institution. These existing processes
will be tailored to incorporate Basel II qualification and
ongoing validation through the implementation of appropriate
guidance, procedures, and training, as well as recruitment or
reallocation of specialized resources.
With regard to international processes, the FDIC and other
U.S. regulatory agencies have fostered sound supervisory
working relationships with many jurisdictions. The FDIC is
actively involved in international working groups focused on
Basel II implementation issues, such as the Accord
Implementation Group (AIG), to ensure that high-level
principles and protocols are established and that key issues
are raised in an effort to reach convergence. Additionally,
many jurisdictions, including the United States, have
established ``supervisory working groups'' as a forum to vet
key implementation issues at the staff level.
Q.2.e. Have the banking agencies consulted with institutional
investors, ratings agencies, independent analysts, and other
market actors concerning the amount of information about their
credit modeling that A-IRB banks will be required to disclose?
[If so, what were the views of these market factors as to the
adequacy of the contemplated disclosure requirements?]
A.2.e. The agencies have consulted with investors, analysts,
and ratings agencies regarding the disclosures for Basel II A-
IRB banks through the Basel II ANPR and the Basel consultative
papers which included solicitations for comment on the
disclosure
requirements. In addition, we encourage the industry to provide
comment on market disclosures following the publication of the
Notice of Proposed Rulemaking.
The agencies have previously sought input on the disclosure
requirements through meetings with banks, ratings agencies, and
other groups in connection with our participation on the Basel
Committee's Transparency Group. The investor and rating agency
communities have been generally supportive of the agencies'
proposed disclosures. In addition, on June 4, 2002, the FDIC
sponsored the ``Enhancing Financial Transparency'' symposium,
which provided a forum for leading experts from the private and
public sectors, including Wall Street experts, to discuss
issues pertaining to market transparency and disclosure.
Further, on July 31, 2002, the FDIC cosponsored the ``Rise of
Risk Management: Basel and Beyond'' symposium with Credit
Suisse First Boston that also included a discussion of
disclosure policy as it relates to Basel II.
Q.2.f. Do the banking agencies continue active study of
alternative approaches to capital regulation, such as the
varieties of proposals based on market discipline or on so-
called precommitment?
A.2.f. In the Basel II Advance Notice of Proposed Rulemaking
(ANPR) published in August 2003, the agencies indicated that if
further analysis or comment suggested there would be
significant unintended competitive ramifications as a result of
the adoption of a bifurcated regulatory capital framework, the
agencies would consider changes to both the general domestic
capital rules as well as changes to the Basel II framework. The
further analysis and comments received since that time have not
caused the agencies to
consider changes to the Basel II framework. Consequently, the
agencies' efforts continue to be focused on the domestic
rulemaking process for Basel II and the modification of current
risk-based capital rules. Market discipline is included as one
of the key aspects in the Basel II framework, the ``third
pillar'' that requires public disclosures of Basel II
information for each institution.
Q.3. Although Basel I-A aims to reduce any competitive
advantage Basel II may confer on large banks, have you
conducted any analysis to determine whether Basel I-A itself
creates any competitive issues within the subset of banks that
are intended to be covered by it? If not, do you intend to do
so?
A.3. ``Basel I-A'' refers to the agencies' publication of an
advance notice of proposed rulemaking (ANPR) that seeks
industry comment on a wide variety of issues including
competitive inequities and regulatory burden.
The ANPR proposes modifications to the existing risk-based
capital rules where quantitative factors used to measure the
risk associated with a given product or exposure can be readily
articulated. However, there are certain areas where risk-
measurement factors are not well-defined or universally
applied, such as with unrated commercial loans and certain
retail loans. The agencies have requested comments in those
areas. As a result, the agencies will wait until the comments
received on the ANPR have been thoroughly analyzed before more
fully developed risk-based capital proposals can be considered.
Until the ANPR comments have been analyzed and more definitive
proposals developed, it is probably premature to speculate
about competitive effects within the subset of banks operating
under Basel I-A.
Notwithstanding, the FDIC is concerned with competitive
issues among various Basel I-A institutions as well as between
Basel I-A and Basel II institutions. To seek more information,
the FDIC placed the following request in the Financial
Institution Letter that transmitted the ANPR to FDIC-regulated
institutions:
The FDIC recognizes that the proposals under consideration
might not be suitable to the entire universe of institutions
that will most likely not adopt the Basel II approaches.
Institutions vary considerably in size and capital levels. Some
institutions may be more inclined to remain on the existing
risk-based capital framework rather than adopt a more risk-
sensitive framework.
The FDIC encourages all commenters to carefully consider
the implications of the proposals included in the ANPR. In
addition to comments on the specific proposals set forth in the
ANPR, the FDIC welcomes any alternatives or suggestions that
would facilitate the development of fuller and more
comprehensive proposals applicable to a range of activities and
exposures.
The FDIC believes that the industry will provide
significant feedback should the proposals included in the Basel
I-A ANPR suggest that competitive equity issues may arise
between Basel I-A institutions.
Q.4. What consideration has been given to the fact that there
are huge differences between the 7,000 or so banks that will
participate in the Basel I-A framework?
A.4. The FDIC believes that the Basel I-A framework should be
designed to reduce the competitive inequities associated with
Basel II while ensuring that capital remains adequate for the
risks inherent in these institutions. However, since the
beginning of the Basel I-A process, the FDIC has realized that
these stated goals are difficult given the large disparity in
asset size and operating complexity associated with the 8,000+
institutions that would apply the Basel I-A framework. The FDIC
wants to ensure than any additional burden that could be
generated by these proposals is commensurate with the benefit
derived.
Therefore, in drafting the Basel I-A ANPR, the FDIC
proposed that a great deal of flexibility be included in the
proposals. For example, the ANPR sought comment on whether
certain banks, especially community banks operating with
capital ratios well in excess of their minimums, should be
given the flexibility to opt out of the Basel I-A framework in
whole or in part.
Q.5. In Dr. Kaufman's testimony, he stated that there is
``substantial empirical evidence of a negative relationship
between leverage ratios and bank insolvency,'' but ``no such
evidence between risk-based capital ratios and bank
insolvency.'' Do you agree with Dr. Kaufman? If so, does this
conclusion undermine the Basel II framework?
A.5. Dr. Kaufman is correct that there is ``substantial
empirical evidence'' that banks with low leverage ratios are
more likely to fail. There also is some academic evidence that
banks with low risk-based capital ratios are more likely to
fail, and some research at the FDIC has verified this result.
However, it should be noted that most bank failures occurred
before risk-based capital standards were implemented.
Q.6. Have you undertaken any analysis of the initial start-up
costs and annual compliance costs that Basel II will impose on
banks?
A.6. FDIC staff conducts routine outreach with regulated
institutions that might be subject to Basel II. Cost estimates
associated with designing and implementing systems and
recruiting, training, or reallocating personnel are across the
spectrum and are by no means firm at this juncture.
Q.7. To what degree will the significant costs necessary to
comply with Basel II act as a barrier to entry that prevent
banks from growing and becoming a bank large enough to qualify
to use Basel II? In effect, will Basel II cement the market
positions of the largest banks? If no, why?
A.7. The FDIC believes that the costs of implementing a
qualified Basel II capital system would act as a significant
barrier to adoption of the new framework for many institutions.
In order to use the advanced internal risk-based approaches of
Basel II, expensive data management systems and complex
modeling must be employed; the cost of such systems would be
too great for smaller and medium-sized banks to afford. The
recently proposed modifications to the domestic risk-based
capital rules attempt to partially mitigate this effect and
even the playing field between Basel II and non-Basel II banks
in the United States by establishing a more risk sensitive
framework based on risk measures more suited to the operations
of community banks.
Q.8. Please explain the type of information that banks will be
required to disclose under Pillar 3 with respect to how banks
calculate their capital requirements? Will banks be required to
disclose proprietary capital requirement models?
A.8. The risks to which a bank is exposed and the techniques
that it uses to identify, measure, monitor, and control those
risks are important factors that market participants consider
in their assessment of the institution. In addition to
reporting their capital ratios and minimum capital
requirements, the agencies expect banks to disclose information
that includes the weighted average estimated probability of
default, loss given default, exposure at default, and expected
loss for each category of wholesale and retail exposures. The
agencies also will request information on credit risk mitigants
and off-balance-sheet exposures.
The Pillar 3 disclosure requirements as summarized above
provide for an appropriate balance between the need for
meaningful disclosure and the protection of proprietary and
confidential information. Accordingly, the disclosure
requirements have been structured in a manner that should limit
the extent to which proprietary and confidential information is
revealed. The agencies have solicited comments on the
disclosure requirements though the Basel II ANPR process as
well as through the issuance of Basel consultative papers.
There is a need to strike a balance between obtaining
sufficient information disclosure and protecting proprietary
information and the agencies will carefully consider comments
received on these issues. In addition, the agencies have
previously sought input on the disclosure requirements through
meetings with banks, ratings agencies, and other groups in
connection with our participation on the Basel Committee's
Transparency Group.
Q.9. If Basel II is implemented, do you favor at any point in
the future reducing the leverage ratio for well-capitalized and
adequately capitalized banks or reducing the statutory capital
thresholds for prompt corrective actions?
A.9. No. The Basel II minimum capital measure fails to provide
for any protection against interest rate risk, liquidity risk,
and other material risks present in our banking system. Credit
risk, market risk, and operational risk are modeled under Basel
II, but such modeling is subject to a great deal of
uncertainty. Given the existence of deposit insurance and other
Federal bank safety net supports, clear constraints are needed
on the proportion of a bank's
activities that can be financed with debt.
The FDIC believes that the leverage ratio and the risk-
based capital ratios are complementary measures that work well
together, each compensating for the other's shortcomings. The
leverage ratio sets the minimum amount of qualifying equity at
stake in the financial institution. The risk-based capital
ratio operates to establish limits on the amount of credit risk
that a financial institution can undertake. Even in a situation
where the risk undertaken by an institution is very low, moral
hazard can still occur if the amount that the equity holders
have at risk is insufficient. Hence from the perspective of an
insurer, the leverage ratio plays a predominant role in
constraining moral hazard.
Q.10. In his testimony, Mr. Isaac points out that compared to
European banks, ``large U.S. banks are much better capitalized
and far more profitable in terms of their operating margins.''
Do you agree or disagree with Mr. Isaac's observation, and what
is your rationale for your position?
A.10. The FDIC agrees with Mr. Isaac's observation. Data
presented by The Banker magazine in their annual review of the
``Top 1000 World Banks'' as well as data that are presented by
Fitch Rating's international bank database, ``Bankscope,''
generally indicate that large U.S. banks are better capitalized
and have higher operating margins than many of their European
counterparts.
Q.11. Do you believe that U.S. banks are operating at a
significant competitive disadvantage vis-a-vis foreign banks
due to the higher capital requirement imposed on U.S. banks? If
you do, please explain in detail and quantify the competitive
effects on U.S. banks, including giving your assessment of how
low U.S. capital requirements would need to go in order to
level the playing field with large foreign banks.
A.11. No. Given the many differences in the global financial
system, such as tax, accounting, and legal frameworks, it is
difficult to isolate the effects of regulatory capital
requirements on international bank competition. This difficulty
is compounded by the fact that the definition of which items
constitute capital differs in each country to a greater or
lesser degree. It is clear, however, as the question suggests,
that U.S. banks face higher capital requirements than foreign
banks face because of U.S. Prompt Corrective Action regulations
(and because of a series of U.S. regulatory tightenings of the
Basel I risk-based capital standards not emulated overseas).
Yet, as indicated in the answers to other questions, this does
not appear to have compromised U.S. banks' ability to prosper
and compete effectively.
Q.12. If capital standards are reduced for the largest U.S.
banks, do you believe that will place U.S. banks that do not
implement Basel II at a competitive disadvantage vis-a-vis
banks that do implement Basel II? If so, how would you suggest
that competitive inequity be remedied?
A.12. Because the United States historically has maintained
regulatory capital standards that have been, for the most part,
the same for all banks, there is no historical evidence on
which to base judgments about the competitive effects of
applying vastly different regulatory capital regimes to
different banks.
Our judgment is that Basel II could place community banks
and thrifts at a competitive disadvantage vis-a-vis Basel II
banks in the United States since the A-IRB approach available
under Basel II would likely yield lower capital charges on many
types of products offered, such as residential mortgage loans,
retail loans, commercial loans, and commercial real-estate
loans as reported in QIS-4. Without a change in the risk-based
capital requirements imposed upon non-Basel II adopters, Basel
II would appear to open up sizable gaps between the capital
requirements of small and large banks for many activities. The
potential exists that investors in small banks would receive
suboptimal returns on these capital-disadvantaged assets
compared to the returns that owners of large banks could earn
on the same assets, resulting in a disequilibrium that would be
corrected over time by the movement of those assets toward the
large banks.
The ANPR was published in October to make capital
requirements for the remaining 8,000 banks in the industry more
risk sensitive. If this Basel I-A initiative ultimately proves
to offer only modest capital reductions to most banks, and if
Basel II proves to be as substantial a reduction in risk-based
capital requirements as the QIS-4 results suggest, bank
regulators will be faced with difficult decisions with regard
to regulatory capital. The FDIC has
expressed a preference in Congressional testimonies (May and
November 2005) that these issues ultimately be resolved by
finding ways to achieve less extreme results under Basel II,
including recalibration of underlying formulas if results
similar to QIS-4 persist during the parallel run and transition
period.
Q.13. Have you considered requiring the largest U.S. banks to
implement Basel II-type models as part of their risk management
programs without making Basel II a capital-regulation device?
If you have considered and rejected such an approach, please
explain why you rejected it.
A.13. A number of large U.S. institutions currently employ
internal economic capital allocation models to manage risk and
make business line decisions, and have for some time. The
regulators believe that banks' use of such models--the
information they generate and the processes and internal
controls around their validation--are of value to supervisors.
Regulators' use of these approaches could take two
fundamentally different directions. One is to base an explicit
system of minimum capital requirements on these approaches. The
other would be to retain the simpler existing capital
requirements while making the validation of risk-measurement
information generated by models a supervisory or ``Pillar 2''
requirement. Both approaches are conceptually defensible and,
we believe, either can be made to work.
Q.14. Have you compared the capital requirements for the
participating banks that came out of your Quantitative Impact
Study with the judgments of your professional bank examination
staff on the capital needs of those same banks? How closely
correlated were the results?
A.14. FDIC examination staff was very much involved in
designing, planning, and implementing QIS-4 and in analyzing
the results. The FDIC remains concerned with the capital levels
and the dispersion of results in QIS-4 data. Minimum tier 1
capital requirements reported by most of the 26 organizations
were not at levels the FDIC would regard as consistent with
safe and sound banking were banks actually to operate at such
levels under an up-and-running system of capital regulation.
QIS-4 results also reflected substantial dispersion of capital
requirements for what appeared to be identical risks. Our
expectation is that meaningful work remains to be done during
the parallel run and transition years to address these issues.
Q.15. Basel II represents an attempt to apply the same formulas
for measuring commercial loan credit, consumer loans, and
commercial real estate loan risk to banks in the United States
and to banks in other developed countries. What evidence do you
have that the formulas will assure that U.S. banks will hold
sufficient capital to see them through a variety of economic
scenarios, such as those that have occurred in the United
States over the last 25 or 50 years? Are the markets and
economic environments similar enough that the same formulas can
be applied safely to large banks from the United States,
Canada, the United Kingdom, France, Germany, Japan, etc.?
A.15. If we were faced with the prospect of sole reliance on
Basel II as it now exists, without benefit of the leverage
ratio, we would not have confidence that the formulas would
require enough capital to see banks through a variety of
scenarios. The preservation of the leverage ratio and the
additional work that will be done during the Basel II
transition years will be crucial, in our judgment, to ensuring
the success of this framework.
With regard to cross-country comparisons, it should be
noted that Basel II is not a treaty, and each country will
implement the framework in its own way that makes sense given
the various legal, tax, and accounting frameworks, and other
differences. Also, while it may be technically correct that
``the same formula'' determines capital requirements across
countries, its implementation leaves much up to the discretion
of supervisors. It is a premise of the Basel II initiative that
the supervisors will ensure adequate capital in each instance.
Q.16. The agencies' current transitional plan for implementing
Basel II contemplates allowing capital levels to fall to an 85
percent ``floor.'' How can the agencies know, even before the
``parallel run'' contemplated for 2008, that a drop of 15
percent in the capital of banks would be consistent with safety
and soundness considerations?
A.16. The agencies' have stated the expectations, as described
above, that Basel II might result in a modest overall aggregate
reduction in capital requirements, with the impact on capital
requirements at individual banks depending on risk profiles. We
are not aware of any expectation that risk-based capital
requirements will fall 15 percent or more in aggregate, and do
not believe such a decline would be acceptable. The agencies
have not ruled out a 15 percent reduction in risk-based capital
requirements for individual banks if warranted by their risk
profiles.
In this regard, it should be noted that the agencies have
stated that the declines in capital requirements reported in
QIS-4 are unacceptable, and if similar results persist
throughout the transitional implementation period, further
measures would be taken to remedy such declines.
Q.17. Basel II's capital formulas are based on highly complex
models that are to be implemented on a bank specific basis.
While Basel I has been criticized for its over simplification,
the sheer complexity of Basel II has raised some very serious
concerns. In his written testimony, Mr. Isaac states, ``Basel
II is so complex it cannot be adequately understood by senior
bank managements, boards of directors, regulatory, or the
public.'' Professor Kaufman states, ``Increased complexity is
likely, however, to both increase compliance cost and reduce
understanding, particularly by the bank CEO, board of
directors, and possibly even the CFO and by bank supervisors.''
Do you share Mr. Isaac's and Professor Kaufman's concerns? If
you share their concerns, what recommendations do you have for
the regulators as to how to fix this problem?
A.17. The FDIC shares these concerns. Given its ambition to
measure all forms of credit risk, operational risk, and market
risk, the Basel II framework is, inevitably, complex. It is
likely that there will be few individuals who will read and
understand, in its entirety, the U.S. Notice of Proposed
Rulemaking implementing Basel II. As discussed below, some of
the adverse effects of this complexity may be mitigated through
appropriate written supervisory guidance, through the retention
of the leverage ratio requirements of U.S. Prompt Corrective
Action regulations, and through a transparent approach to
implementation of the framework.
First, supervisory guidance will be important in helping
banks cut through some of this complexity--and supervisory
guidance will follow the publication of any NPR.
Second, the leverage-ratio capital requirement that is part
of the agencies' Prompt Corrective Action regulations will
ensure that the need for a clear-cut minimum of regulatory
capital is not lost in the complexity of Basel II. Basel II is
a formula-driven regime that produces a capital requirement
that some may portray as having the mantle of science. There is
a danger that the complexity of this framework may obscure the
reality that representations of its scientific certainty are
incorrect.
Finally, a transparent approach to the implementation of
Basel II may help mitigate the adverse effects of complexity.
That means public dissemination of the results of risk-based
capital calculations and summary views of the inputs to those
calculations. It means sharing of supporting confidential
detail behind the capital calculations among the agencies to
promote consistency and best practices. It may ultimately,
perhaps, involve some way for regulators to share with the
industry some type of information about approaches that are
being taken under the framework to the measurement of risk.
Q.18. According to FDIC data, the equity to capital ratio of
insured U.S. financial institutions reached a low of slightly
above 5 percent in the late 1980's, during the S&L crisis, and
then rose steadily through the 1990's reaching the level of
over 10 percent, at which it stands today. Profits--as measured
by return on assets--which had fallen dramatically during the
period of low capital, also began a steady rise in the 1990's,
rising to well over 1 percent for the last decade, reaching
record levels. When looked at together, it becomes clear that
large profits and strong levels of capital can coexist. Do you
agree that strong levels of bank capital can and do coexist
with strong bank profitability?
A.18. As indicated in the answers to questions 2, 10, and 11,
the FDIC agrees that strong levels of bank capital can and do
coexist with strong bank profitability.
Q.19. Under the Administrative Procedure Act, agencies are
required to consider public comment prior to issuing new final
regulations. How can the agencies fully comply with this
requirement? Doesn't the fact that the agencies reached an
international agreement on the new capital framework
effectively limit their ability to fully consider public
comment and modify proposed regulations in light of that
comment?
A.19. The Basel Committee is not a standard setting body but a
committee of technical experts that makes recommendations from
time to time about best practices. Individual countries
consider these recommendations and may elect to act on them, or
not, as appropriate. The U.S. agencies have made clear
repeatedly at the Basel Committee the seriousness with which
they take the notice and comment process in the United States.
The FDIC is committed both to fully consider all comments
and to take action to address those comments as appropriate.
Q.20. The Basel II framework ignores a significant risk:
Interest rate risk. Many banking assets, such as mortgage
servicing rights, are extremely sensitive to changes in
interest rates, and the value of assets such as these may
fluctuate dramatically when interest rates change. How do you
propose to take interest rate risk into account under the Basel
II framework?
A.20. The FDIC, the Federal Reserve, and the OCC historically
have not promulgated explicit capital regulations to cover the
interest rate risk of banks' exposures held outside of trading
accounts. This tradition continues with Basel II.
The Basel II framework establishes a ``three-pillar''
approach to bank capital regulation. Pillar 1 sets the
standards for computing regulatory capital requirements,
consisting of credit, market, and operational risk. Pillar 2 is
a supervisory review process that examines factors not
considered under Pillar I, such as board oversight, internal
controls, and assessment of risk to ensure capital adequacy. As
noted earlier, Pillar 3 encourages market discipline through a
public disclosure process.
Interest rate risk is not captured in the minimum
requirement calculation set forth in Pillar 1. Rather,
supervisors plan to incorporate interest rate risk, and other
risks such as liquidity and
concentration risks, within the ``assessment of risk for
capital adequacy'' process under Pillar 2. The FDIC is aware of
the importance of interest rate risk when assessing the
appropriate level of capital to an institution and directs its
examiners to consider this risk in relation to capital
adequacy. In fact, the ``S'' component of the CAMELS rating,
which measures sensitivity to market risk, specifically takes
into consideration a bank's interest rate risk.
Q.21. Due to the tremendous devastation caused by Hurricanes
Katrina and Wilma this fall, it is estimated that over 100,000
homes have been lost. Would the Basel II framework have
provided institutions with enough capital buffer to handle a
calamity of this magnitude?
A.21. The current well-capitalized position of U.S. banks was a
strength of the local banks impacted by the hurricane. It is
our intention that Basel II not result in a significant decline
in capital that would place U.S. banks in a position to cope
less effectively with unforeseen circumstances.
With respect to mortgages, it must be noted that the QIS-4
reported extraordinarily large reductions in regulatory capital
requirements. Without some type of clarification or regulatory
refinement during the transition years, our concern is that
Basel II will prove to undercapitalize substantially the risk
inherent in mortgage portfolios.
Q.22. Last week, on November 18, 2005 The Financial Times
published an editorial by Harold Benink and Jon Danielsson
titled, ``There is a chance to correct the defects of Basel
II.'' Harold Benink is Professor of Finance at RSM Erasmus
University in the Netherlands. Jon Danielsson is Reader in
Finance at the London School of Economics. They concluded:
``Instead of requiring bank capital to be risk sensitive,
banking regulation should simply require the use of high
quality risk models in banks without using their output to
determine capital. Minimum capital is better calculated as a
simple fraction of bank activity in broad categories.'' What is
your response to this suggestion?
A.22. As indicated in the answer to question 13 above, relying
on simpler measures of capital adequacy for regulatory purposes
while using the supervisory process to promote the rigorous use
of banks' internal capital models would be an alternative, and
defensible, way for the agencies' supervisory programs to
benefit from banks' use of internal models.
As we noted in response to question 9, the FDIC believes
that both the leverage ratio (``the simple fraction of bank
activity in broad categories'' referenced above) and the risk-
based capital ratios have important roles to play in ensuring
safe and sound banking.
Q.23. In his testimony, Mr. Isaac states, ``I have not found a
single professional hank supervisor who is enthusiastic about
Basel II.'' We have also heard of a similar lack of enthusiasm
on the part of the supervisors who will actually be responsible
for supervising Basel II. What is your response? Separately,
what gives you confidence that regulators will be up to the
task of understanding and policing Basel II?
A.23. As indicated in responses to previous questions, the FDIC
has serious concerns about the impact on capital under Basel II
as evidenced in the QIS-4 results. The FDIC, along with the
other regulators, believes these results were unacceptable and
must be addressed if Basel II implementation is to proceed.
Supervisors at the FDIC are committing significant
resources to policy development and implementation strategies
for Basel II. This includes collaboration with key FDIC
research staff. The FDIC staff involved in this effort bring a
level of engagement and professionalism to this task that gives
the agency confidence that it will be up to the task of
implementing the new framework while acknowledging the
challenges that will be posed.
RESPONSE TO WRITTEN QUESTIONS OF SENATOR SHELBY
FROM JOHN M. REICH
Q.1.a. Basel II would impose new demands on bank regulators. It
would require bank regulators to not only approve banks'
internal assessments of their risks, but also to make judgments
on whether banks are holding enough capital in light of all the
risks banks face. This means bank regulators are going to need
personnel who are well-trained in risk analysis.
How many additional personnel does your agency expect it
will need to hire in order to implement Basel II? What type of
expertise will your agency need to obtain? Do you foresee any
problems in recruiting?
A.1.a. OTS anticipates that it will need more examiners,
quantitative experts, and other supervisory personnel with
expertise in the support areas of finance, economics,
accounting, and law. Although OTS cannot predict how the Basel
II proposal may change as a result of the notice and rulemaking
process, under the current draft of the Basel II NPR, OTS would
be the primary Federal
supervisor of one core banking organization, several that have
indicated a desire to opt-in, and several subsidiaries of Basel
II organizations, holding companies, or broker-dealers. While
hiring has
already begun, OTS is prepared to expand its staff of trained
professionals throughout the period leading to the full
implementation anticipated in 2012.
OTS believes it can meet the challenge with a combination
of training programs and external staff recruitments. While the
recruitment environment is and will continue to be competitive
for such professionals, OTS has already demonstrated recruiting
success, at both entry and senior levels, and we are confident
we will continue to do so. OTS has begun and will continue to
enhance both internal and external training for many of our
current staff at all levels, in order to ``retrofit'' our
current resources to the work of ensuring safety and soundness
under capital rules as they are revised.
Q.1.b. Will your agencies be able to successfully compete
against banks in the recruitment of personnel? Do you have any
concerns that banks will simply be able to hire the best
Ph.D.'s in risk modeling and outgun their regulators?
A.1.b. As stated in the answer to Question 1.a., OTS believes
it will successfully meet its hiring needs in a competitive
environment, and has already recruited talented professionals
at many levels. OTS offers a stimulating work environment, and
a competitive compensation package that includes excellent
benefits and quality-of-life considerations when compared to
the private sector. This should help OTS attract and retain
additional staff in the necessary disciplines.
Q.1.c. How much in additional annual costs do you expect to
incur in implementing Basel II? Other than personnel costs,
what other costs will your agency incur in implementing Basel
II? Are there any large one-time costs?
A.1.c. OTS estimates that it has incurred costs of nearly $4
million from fiscal year 2002 through fiscal year 2005 related
to the development and implementation of the new Basel II
Accord and proposed NPR. For fiscal year 2005 alone, the cost
approached $2
million, or between 1 percent and 2 percent of the OTS budget.
While some part of the start-up costs may be characterized as a
one-time investment, the annual expenditure, thus far, has
increased each year. OTS anticipates that the annual cost
should decline as the revised framework is established and the
agency begins to reap economies of scale in building expertise
around that framework.
In general, costs fall in four areas:
<bullet> Training. Development of internal training, as well as
attendance at external training sessions.
<bullet> Travel. Both prequalification and qualification
meetings will require travel, as will national and
international regulatory meetings, on-site examinations,
and other supervisory activities.
<bullet> Data systems. Development of data systems to handle
new reporting requirements.
<bullet> Administrative. Rulemaking, guidance development, and
industry outreach.
Q.2.a. In his testimony, Dr. Tarullo posed several questions
that he believes need to be answered by U.S. banking agencies
before implementation of Basel II begins. Please provide
answers to his questions with respect to your agency, which are
as follows:
``Do the agencies believe that, in general, minimum capital
levels are too high and if, so, what is the basis for this
belief ?
A.2.a. OTS does not believe that minimum capital levels are too
high. However, OTS believes that the current capital rules,
which involve a relatively simple risk-bucketing system, are
limited as tools for measuring risks of sophisticated and
complex financial activities.
Q.2.b. Have the agencies been able to predict with reasonable
precision the levels of capital that the A-IRB approach will
require of large banks, and then to confirm their prediction
though studies that are well-conceived and executed?
A.2.b. QIS-4 was conducted on a best efforts basis. There were
data quality issues that reflected institutions' various stages
of development of their models and systems. In addition, the
rulemaking and implementation processes were also in the early
states such that the QIS-4 was done without institutions having
a clear understanding of what the Basel II systems requirements
and supervisory expectations would be.
OTS participated in QIS-4, which facilitated policy
analysis regarding appropriate capital levels, and provided
information for peer group capital comparisons between the core
institutions. The QIS-4 results helped the agencies to better
understand the impact of a more risk-sensitive approach for
regulatory capital standards. This effort enabled the agencies
to better gauge appropriate capital levels at banking
organizations, but there are diverging views on the levels of
capital that the A-IRB approach will ultimately require.
Moreover, all aspects of the Basel II NPR will be subject to
change as a part of the notice and comment rulemaking process,
which may also impact the capital levels that are ultimately
required. The agencies anticipate more studies in the
intervening 6 years before a Basel II final rule is expected to
be fully implemented in the United States.
Q.2.c. Have the agencies done scenario planning to anticipate
the effects, unintended and otherwise, of Basel II upon the
financial system as a whole? For example, are significant
portions of certain kinds of assets likely to migrate out of
banks into the unregulated sector? If so, might new risks of
financial disruption be created?
A.2.c. In the latest draft of the Basel II NPR, implementation
would not occur until after a parallel run, which is projected
to be in 2008. The parallel run would take place the first year
after the final rule's effective date. After that time period,
there would be 3 years of floor capital levels imposed--95
percent of the Basel I-based capital requirement the first
year, 90 percent in the second year, and 85 percent in the last
year. Of course, the parallel run, floor periods, and floor
capital levels may ultimately change as a result of public
comment on the Basel II NPR. However, it would be under these
periods of time that scenario planning would take place.
Until full implementation, it is too attenuated to
anticipate the potential consequences of Basel II
implementation, intended or unintended. As stated previously,
however, the agencies are committed to ongoing refinement
throughout the process, including after implementation, as
necessary, and would make changes if such scenario planning
demonstrates changes are warranted.
Q.2.d. What do the agencies regard as the effective capacity of
the specialized examiner teams that will be overseeing the use
of bank models under A-IRB? How will the agencies determine
whether other countries are successfully monitoring bank
capital levels under the A-IRB approach?
A.2.d. OTS has experience in overseeing the use of models and
validation processes. In anticipation of the proposed new
capital framework, OTS is already developing specialized
examination teams with the training necessary to oversee the
use of A-IRB. The Basel II Accord is an international agreement
with a multilateral commitment that each country will supervise
the measurement and management of capital within its own
financial institutions. International consultative processes
have been ongoing throughout the development of Basel II to aid
that goal. Theses will, no doubt, continue.
Q.2.e. Have the banking agencies consulted with institutional
investors, rating agencies, independent analysts, and other
market actors concerning the amount of information about their
credit modeling that A-IRB banks will be required to disclose?
[If so, what were the views of these market actors as to the
adequacy of the contemplated disclosure requirements?]
A.2.e. The Pillar 3 disclosures have been issued for public
comment, both formally and informally, on a number of
occasions, and will be subject to further public comment
following the issuance of the Basel II NPR. In addition,
representatives of the Federal banking agencies consulted with
market participants in developing the disclosures. In general,
market participants, such as debt and equity analysts, have
supported enhanced public disclosures of risk information and
capital adequacy by banking organizations. The version of
Pillar 3 included in the June 2004 text from the BIS balances
the desire for additional disclosures by market participants
and the reporting burden associated with such disclosures. This
version leverages off of the information institutions will need
in order to implement Basel II, and is substantially
streamlined from earlier Pillar 3 proposals.
Q.2.f. Do the banking agencies continue active study of
alternative approaches to capital regulation, such as the
varieties of proposals based on market discipline or on so-
called precommitment?''
A.2.f. The agencies have studied the other approaches and have
determined that the direction we are moving in, and that will
be enunciated more fully in the upcoming Basel II NPR and
guidance, is the best approach available. However, we also are
committed to flexibility and will consider all new ideas,
including new approaches and refinements suggested by public
commenters on the Basel II NPR. We understand that none of us
can predict what may evolve over the next half-decade for
incorporation into the Basel II rules in the United States. And
we have said consistently, this will be an iterative process,
even after we have ``finalized'' the Basel II rules. We are
prepared to introduce, for comment, any changes that serve
safety and soundness.
Q.3. Although Basel IA aims to reduce any competitive advantage
Basel II may confer on large banks, have you conducted any
analysis to determine whether Basel I-A itself creates any
competitive issues within the subset of banks that are intended
to be covered by it? If not, do you intend to do so?
A.3. When the agencies issued the Basel I-A ANPR, we recognized
the competitive issues between large banking organizations,
small community institutions, and regional banking
organizations. We specifically invited and anticipate comments
on competitive issues among and between all kinds of charters
and different institutions, by size and specialty. Since the
Basel I-A ANPR does not propose specific capital requirements,
it is probably premature to do an impact analysis at this time,
but we will analyze the capital impact of the approaches in the
Basel I-A NPR after these approaches are formulated. OTS senior
economists and examiners will also continue their analysis of
the models-based approaches in the Basel II NPR, and will
analyze the approaches adopted in the Basel II Final Rule, with
a careful eye on competitive concerns.
Q.4. What consideration has been given to the fact that there
are huge differences between 7,000 or so banks that will
participate in the Basel I-A framework?
A.4. OTS recognizes that a broad-based capital rule will not
address the idiosyncrasies of all 7,000 community banks and
thrifts. However, the Basel I-A rulemaking will likely benefit
financial institutions that have sound risk management systems
in place, and we expect it will also benefit smaller community
institutions that are less sophisticated. One of the primary
goals of the Basel I-A rulemaking is to better align capital
commensurate with risk. The Basel I-A ANPR suggested changes
attempt to meet this goal without undue burden to community
banks and thrifts. These improvements would be available to all
and are designed to benefit all by better aligning risk with
capital. However, in further recognition of the difference
among institutions, the Basel I-A ANPR specifically asked for
comment on allowing institutions the flexibility to remain
under the simpler and well-established Basel I-based rules,
which in most instances would result in those institutions
holding greater risk-based capital. OTS will carefully evaluate
the comments received when we develop the Basel I-A NPR.
Q.5. In Dr. Kaufman's testimony, he stated that there is
``substantial empirical evidence of a negative relationship
between leverage ratios and bank insolvency,'' but ``no such
evidence between risked-based capital ratios and bank
insolvency.'' Do you agree with Dr. Kaufman? If so, does this
conclusion undermine the Basel II framework?
A.5. We partially agree with Dr. Kaufman's analysis that the
current risk-based capital system is insufficiently risk-
sensitive. While eliminating a leverage ratio is not a viable
option at this time, this observation supports our efforts in
implementing a more risk sensitive regime, rather than
undermining it.
Q.6. Have you undertaken any analysis of the initial start-up
costs and annual compliance costs that Basel II will impose on
banks?
A.6. As part of the regulatory impact study required for the
Basel II NPR, OTS continues to analyze such costs. Thus far,
only anecdotal information using a variety of metrics is
available regarding start-up costs. Compliance costs will be
assessed in future estimates.
Q.7. To what degree will the significant costs necessary to
comply with Basel II act as a barrier to entry that prevent
banks from growing and becoming a bank large enough to qualify
to use Basel II? In effect, will Basel II cement the market
positions of the largest banks? If not, why?
A.7. OTS does not believe that cost will create a significant
barrier to future growth for thrifts. Because the Basel II NPR
is designed to reflect sound risk management practices for the
largest and most complex financial institutions, these
institutions have already incurred costs to develop
sophisticated risk management practices and systems. However,
we expect significant additional cost at some institutions to
modify existing information systems, policies, and procedures
to meet the requirements that may be contained in the Basel II
final rule, as well as to train or hire staff. As other thrifts
grow and engage in more complex activities, OTS expects these
institutions to also incur costs to develop more risk sensitive
management practices and systems. We expect there would be
marginal additional expense associated with compliance with the
Basel II final rule beyond what was spent to develop such
systems.
From an international perspective, the largest foreign
banks already have systems and practices in place, and are
competing with U.S. financial institutions, large and small, in
every corner of the financial industry, including the wholesale
and retail business. OTS believes that our largest thrifts will
find it necessary to improve their risk management practices
for competitive reasons, regardless of a Basel II final rule.
Q.8. Please explain the type of information that banks will be
required to disclose under Pillar 3 with respect to how banks
calculate their capital requirements? Will banks be required to
disclose proprietary capital requirement models?
A.8. Consistent with Pillar 3 requirements discussed in the
Mid-Year Text, the agencies intend to propose to collect, both
qualitatively and quantitatively, the following disclosures:
Scope of
Application; Capital Structure; Capital Adequacy; Credit Risk,
Credit Risk Mitigation, Securitization Approaches; Market Risk,
Operational Risk, Equities, and Interest Rate Risk. The
proposed disclosure requirements would be structured to limit
the extent of public disclosure of confidential and proprietary
information. In addition, the agencies will request comments
from institutions and other interested persons on the
appropriateness of the proposed disclosure requirements in
connection with the Basel II NPR.
Q.9. If Basel II is implemented, do you favor at any point in
the future reducing the leverage ratio for well-capitalized and
adequately capitalized banks or reducing the statutory capital
thresholds for prompt correction actions?
A.9. At this time, OTS opposes elimination of the leverage
ratio, and we urge great caution in any consideration to lower
it. We do, however, believe it is imperative that we engage in
a dialogue with the financial industry and other stakeholders
to rethink what a leverage ratio should be, as we make these
major changes in risk-based capital standards. We believe the
agencies should, as part of the eventual adoption of changes to
risk-based capital in the Basel II and Basel I-A rulemakings,
consider how best to synthesize a leverage ratio with these new
standards, especially as we become more comfortable with the
safety and soundness of the new rules in the years to come. We
oppose reducing the statutory capital thresholds for prompt
corrective action.
Q.10. In his testimony, Mr. Isaac points out that compared to
European banks, ``large U.S. banks are much better capitalized
and far more profitable in terms of their operating margins.''
Do you agree or disagree with Mr. Isaac's observation, and what
is your rationale for your position?
A.10. OTS agrees with the observation, and further adds that
the U.S. markets have experienced favorable credit conditions
over the last several years. Many foreign banks have been
attracted to the U.S. market for that reason. Only the United
Kingdom has neared the United States in terms of economic
growth among major competitors. The changes in the current
draft of the Basel II NPR would bring oversight and capital
measurement between Europe and the United States into better
alignment, which may level the playing field for U.S. banking
organizations with their foreign competitors here.
Q.11. Do you believe that U.S. banks are operating at a
significant competitive disadvantage vis-a-vis foreign banks
due to the higher capital requirements imposed on U.S. banks?
If you do, please explain in detail and quantify the
competitive effects on U.S. banks, including giving your
assessment of how low U.S. capital requirements would need to
go in order to level the playing field with large foreign
banks.
A.11. Merely lowering capital is not the key to
competitiveness. Competitiveness is enhanced by improvements to
management and measurement of risks within a capital system.
The system should adequately reflect risks so that capital
requirements themselves do not prevent U.S. banking
organizations from competing in various markets, here and
abroad.
Q.12. If capital standards are reduced for the largest U.S.
banks, do you believe that will place U.S. banks that do not
implement Basel II at a competitive disadvantage vis-a-vis
banks that implement Basel II? If so, how would you suggest
that competitive inequity be remedied?
A.12. In an effort to alleviate disadvantages that smaller
institutions may experience, the agencies are pursuing rule
changes in the Basel I-A rulemaking. Modernizing risk-based
capital is necessary, through the Basel I-A and the Basel II
rulemakings, for all U.S. banking organizations. Through the
notice and comment periods, the banking agencies will address
any competitive inequities.
Q.13. Have you considered requiring the largest U.S. banks to
implement Basel II-type models as part of their risk management
programs without making Basel II a capital-regulation device?
If you have considered and rejected such an approach, please
explain why you rejected it?
A.13. Although considered, OTS believes that risk management
programs and capital requirements are mutually reinforcing
concepts. Thus, allowing appropriate adjustments for capital
based on these enhanced programs is more likely to provide the
needed incentive to cause institutions to move toward more risk
sensitive management behaviors. OTS believes we can promote
both safe and sound risk management systems and adequate
capital levels, which will allow thrifts to successfully
compete in the inter-connected global markets.
Q.14. Have you compared the capital requirements for the
participating banks that came out of your Quantitative Impact
Study with the judgments of your professional bank examination
staff on the capital needs of those same banks? How closely
correlated were the results?
A.14. OTS examination staff was involved in the planning and
implementation of QIS-4, and agreed that, upon review of the
results, there was a fair amount of dispersion caused by
differences in data and methodologies across participating
institutions. OTS expects to engage examination staff in the
additional work that will be done during the transition years
to address these issues.
QIS-4 was a best efforts exercise. Nevertheless, the
results informed us about changes, enhancements,
clarifications, and significant adjustments the agencies need
to make to the framework to better align capital with risk as
we move forward.
QIS-4 focused on the Pillar 1 capital calculation, which
consists of capital allocated to credit risk and operational
risk. It did not include an assessment of overall capital
adequacy under Pillar 2. In order to make a supervisory
assessment regarding the adequacy of capital compared to the
comprehensive risk profile of an institution, it would be
necessary to include both Pillar 1 and Pillar 2 capital. Thus,
it would have been inappropriate to compare QIS-4
results to the total capital that a financial institution is
expected to retain. A judgmental assessment of capital
requirements calculated under the QIS-4 exercise can only be
performed based on a review of credit risk separate from other
risks associated with the lending operations. Subject to that
premise, we found that, for most portfolios, the calculation
for the credit risk component of the lending operation that is
in the current draft of the Basel II NPR was consistent with
the perceived risk based on supervisor and examination
experience.
Q.15. Basel II represents an attempt to apply the same formulas
for measuring commercial loan credit, consumer loans, and
commercial real estate loan risk to banks in the United States
and to banks in other developed countries. What evidence do you
have that the formulas will assure that U.S. banks will hold
sufficient capital to see them through a variety of economic
scenarios, such as those that have occurred in the United
States over the last 30 or 50 years? Are the markets and
economic environments similar enough that the same formulas can
be applied safely to large banks from the United States,
Canada, the United Kingdom, France, German, Japan, etc.?
A.15. The agencies expect that banking organizations will
manage their regulatory capital positions so that they remain
at least adequately capitalized during all phases of the
economic cycle, including economic downturns. The current draft
of the Basel II NPR would require a banking organization to
segment retail portfolios by similar risk characterizations,
and group commercial loans by internal ratings categories. The
past performance of the retail loan segment or commercial loan-
rating category would be determined through review and analysis
of internal or external data, and the banking organization
would produce an estimate of the economic loss expected if the
obligor were to default during economic downturn conditions.
The formulas would incorporate economic downturn conditions in
terms of the aggregate default rates for the exposure's
supervisory asset class in the exposure's supervisory
jurisdiction (United States or another developed country).
Accordingly, the proposal would require banking organizations
to use a loss given default (LGD) estimate that reflects
economic downturn conditions for purposes of calculating the
risk-based capital requirements.
Q.16. The agencies' current transitional plan for implementing
Basel II contemplates allowing capital levels to fall to an 85
percent ``floor.'' How can the agencies know, even before the
``parallel run'' contemplated for 2008, that a drop of 15
percent in the capital of banks would be consistent with safety
and soundness considerations?
A.16. Experience has taught us that existing Basel I risk-based
capital rules do not always afford a capital requirement
commensurate with the risk of the asset. In addition, by virtue
of the 100 percent risk-weight ceiling, Basel I-assigned risk
weights are not
always sufficient to address the credit risk of higher risk
assets. Under the current draft of the Basel II NPR, each core
or opt-in banking organization would have to adequately measure
and model risk. An institution that has a low risk portfolio,
for example a portfolio of traditional, amortizing home
mortgage loans, would appropriately be able to reduce its risk-
weighted assets, in some cases down to the floors. Another with
a higher risk profile would, in theory, see its risk-weighted
assets increase. As part of the implementation process, each
core or opt-in institution would have to receive the approval
of its regulator before moving forward from the parallel run,
and into each of the next three periods of potentially
reduced capital requirements. The floors would be assessed on
an institution-by-institution basis. The floors are intended to
ensure that any reduction in capital (if at all, and then, only
as approved by the primary regulator) can be no more than an
incremental percentage reduction. The agencies will seek public
comment on the proposed floor and floor amounts. All of these
provisions would be subject to change based on the analysis of
the comments. Moreover, if the regulators do not believe that a
banking organization's indicated reduction in capital is
warranted, they would not approve the move from one level to
the next. Within each level of these steps, the OTS's foremost
concern is safety and soundness.
Q.17. Basel II's capital formulas are based on highly complex
models that are to be implemented on a bank specific basis.
While Basel I has been criticized for its over-simplification,
the sheer complexity of Basel II has raised some very serious
concerns. In his written testimony, Mr. Isaac states, ``Basel
II is so complex it cannot be adequately understood by senior
bank managements, boards of directors, regulators, or the
public.'' Professor Kaufman states, ``Increased complexity is
likely, however, to both increase compliance costs and reduce
understanding, particularly by the bank CEO, board of
directors, and possibly even the CFO and by bank supervisors.''
Do you share Mr. Isaac and Professor Kaufman's concerns? If you
share their concerns, what recommendations do you have for the
regulators as to how to fix this problem?
A.17. While the new framework is complex, so is modern
enterprise-level risk management at the largest U.S. banking
organizations. The current draft of the Basel II NPR is based
on a value-at-risk approach to capital. The value-at-risk
model, and similar approaches, is already widely used at the
more sophisticated institutions. In fact, one of the most
important components of the proposal would be the ``use'' test
that requires institutions to build on their own internal risk
management models in meeting the proposed requirements.
The current draft of the Basel II NPR reflects the
complexity of many large institutions. Thus, it would supply
the agencies with a sufficient and sophisticated tool to
measure these complex businesses. Under Basel I-based rules,
the agencies had concern we were missing an accurate assessment
of complex activities. However, the new proposal is not
suitable for all institutions. The appropriate question then is
whether the senior management at the largest institutions would
be able to understand proposed requirements. Although there is
a danger that Basel II could reduce understanding if not
properly implemented, in general, we do not share Mr. Isaac's
and Dr. Kaufman's concern that the proposal is beyond the
comprehension of senior management at such institutions. In
addition to supporting guidance that would be issued
following the publication of the Basel II NPR, and which should
mitigate some of the complexities, OTS will ensure that
management and boards are properly addressing these
complexities. As regulators, we will enhance our own expertise
so we can provide the required oversight.
Q.18. According to FDIC data, the equity to capital ratio of
insured U.S. financial institutions reached a low of slightly
above 5 percent in the late 1980's, during the S&L crisis, and
then rose steadily through the 1990's reaching the level of
over 10 percent, at which it stands today. Profits--as measured
by return on assets--which had fallen dramatically during the
period of low capital, also began a steady rise in the 1990's,
rising together, it becomes clear that large profits and strong
levels of capital can coexist. Do you agree that strong levels
of bank capital can and do coexist with strong bank
profitability?
A.18. Strong levels of capital and profits can coexist. The
goal of the current draft of the Basel II NPR is not to reduce
capital, but to make it more risk-sensitive, so that we do not
inadvertently advantage or disadvantage any type of lending by
requiring too little or too much capital for the risks
involved.
Q.19. Under the Administrative Procedure Act, agencies are
required to consider public comment prior to issuing new final
regulations. How can the agencies fully comply with this
requirement? Doesn't the fact that the agencies reached an
international agreement on the new capital framework
effectively limit their ability to fully consider public
comment and modify proposed regulations in light of that
comment?
A.19. OTS will fully comply with the requirements of the
Administrative Procedure Act. The timelines for the Basel II
NPR and the Basel I-A NPR are being coordinated to permit
industry consideration of, and public comment on, the two
rulemakings along overlapping timeframes. OTS will consider all
comments, and take
action to address those comments as appropriate. The
international accord does not limit the agencies' ability to
make appropriate revisions in response to public comment. The
agencies would not be bound to adopt all aspects of the Basel
II Accord. Rather, the Basel II Accord provides a range of
options for determining the capital requirements for credit and
operational risk, allows national supervisors to select
approaches that are most appropriate for their
financial markets, and specifically recognizes the need for
national discretion in the implementation of various matters.
Q.20. The Basel II framework ignores a significant risk:
Interest rate risk. Many banking assets, such as mortgage
servicing rights, are extremely sensitive to changes in
interests rates, and the value of assets such as these may
fluctuate dramatically when interest rates change. How do you
propose to take interest rate risk into account under the Basel
II framework?
A.20. OTS agrees that an interest rate component is critical.
In fact, OTS has employed an interest rate risk model since
1992, and has comprehensive policies regarding appropriate
interest rate risk management. Interest-rate risk is addressed
under Pillar 2 of the Basel II Accord, and the agencies plan to
specifically invite comment on interest rate risk in the Basel
II NPR. Moreover, OTS is committed to developing interagency
guidance to ensure that interest rate risk is measured and
capitalized adequately, and that it is benchmarked for
consistent treatment by the supervisors for all banking
organizations, especially mortgage lenders. Interest rate risk
in the trading book would also be captured as part of the
market risk capital requirements of Pillar 1. While OTS does
not currently have a market risk rule, OTS intends to propose a
market risk rule in 2006.
Q.21. Due to the tremendous devastation caused by Hurricanes
Katrina and Wilma this fall, it is estimated that over 100,000
homes have been lost. Would the Basel II framework have
provided institutions with enough capital buffer to handle a
calamity of this magnitude?
A.21. The well-capitalized position of thrifts was a source of
strength for local community thrifts impacted by Katrina or
Wilma. Capital, however, cannot protect an institution against
every kind of conceivable catastrophic event, and it is not the
only tool to ensure an institution can endure such events.
Other tools include the ability to mitigate and manage risk by
ensuring strong underwriting, and by requiring flood, property,
and casualty insurance, as appropriate.
Q.22. Last week, on November 18, 2005, the Financial Times
published an editorial by Harold Benink and Jon Danielsson
titled, ``There is a chance to correct the defects of Basel
II.'' Harold Benink is Professor of Finance at RSM Erasmus
University in the Netherlands. Jon Danielsson is a Reader in
Finance at the London School of Economics. They concluded:
``Instead of requiring bank capital to be risk sensitive,
banking regulation should simply require the use of high
quality risk models in banks without using their output to
determine capital. Minimum capital is better calculated as a
simple fraction of bank activity in broad categories.'' What is
your response to this suggestion?
A.22. This suggestion ignores a financial institution's ability
to manage risks, and places a minimum capital standard on all
institutions regardless of their sophistication. It is this
type of conservatism in capital standards that could price
larger institutions out of global markets and severely limit
their ability to compete. The United States has a somewhat
unique regulatory environment in that it requires a simple
leverage ratio as a backstop. The leverage ratio requirement
provides for a simple fraction of financial activity as
Messer's Benink and Danielsson advocate.
The simplicity of broad categories, while attractive, can
mask significant differences in actual risk. For example, all
qualified residential mortgages receive a 50 percent risk
weight under the Basel I-based rules. However, actual risk can
vary among mortgages in that category. Because of this
structure, banking organizations face misaligned incentives.
Q.23. In his testimony, Mr. Isaac states, ``I have not found a
single professional bank supervisor who is enthusiastic about
Basel II.'' We have also heard of a similar lack of enthusiasm
on the part of the supervisors who will actually be responsible
for supervising Basel II. What is your response? Separately,
what gives you confidence that regulators will be up to the
task of understanding and policing Basel II?
A.23. OTS supports moving forward cautiously with the Basel II
NPR. We also recognize that anything new and complicated often
receives a mixed response from those familiar with systems that
are more routine and simple. However, as supervisors we must
adapt to change to keep our supervision relevant,
sophisticated, and effective. In order to succeed in an
increasingly complex financial world, banking organizations are
developing sophisticated
internal models and financial risk management approaches to
manage their businesses. They participate as intermediaries in
transactions that would not have been contemplated a decade
ago. As supervisors of those organizations, we must also remain
dynamic and enhance our knowledge of capital measurement and
management in a faster, more complex, financial world. If we
fail to do so and remain in a 1980's risk-based capital system,
we will miss those risks, and safety and soundness will suffer.
The current draft of the Basel II NPR promises complex and
new approaches to capital measurement and management at our
thrifts. As stated in response to earlier questions, we are
hiring and training staff in order to ``retrofit'' capital
supervision, risk analysis, and risk assessment to the
challenges posed to our institutions, and hence to us. In
conjunction with the careful and deliberative implementation
plan outlined by the agencies, OTS is confident it can meet
these challenges.