[Senate Hearing 114-115]
[From the U.S. Government Publishing Office]
S. Hrg. 114-115
MEASURING THE SYSTEMIC IMPORTANCE OF U.S. BANK HOLDING COMPANIES
=======================================================================
HEARING
BEFORE THE
COMMITTEE ON
BANKING,HOUSING,AND URBAN AFFAIRS
UNITED STATES SENATE
ONE HUNDRED FOURTEENTH CONGRESS
FIRST SESSION
ON
EXAMINING THE APPROPRIATE CRITERIA THAT THE FEDERAL RESERVE AND OTHER
REGULATORS COULD USE TO DETERMINE WHETHER AN INSTITUTION POSES A
SYSTEMIC RISK TO THE FINANCIAL SYSTEM
__________
JULY 23, 2015
__________
Printed for the use of the Committee on Banking, Housing, and Urban
Affairs
[GRAPHIC NOT AVAILABLE IN TIFF FORMAT]
Available at: http: //www.fdsys.gov /
____________
U.S. GOVERNMENT PUBLISHING OFFICE
97-533 PDF WASHINGTON : 2016
______________________________________________________________________________________
For sale by the Superintendent of Documents, U.S. Government Publishing Office,
http://bookstore.gpo.gov. For more information, contact the GPO Customer Contact Center,
U.S. Government Publishing Office. Phone 202-512-1800, or 866-512-1800 (toll-free).
E-mail, gpo@custhelp.com.
COMMITTEE ON BANKING, HOUSING, AND URBAN AFFAIRS
RICHARD C. SHELBY, Alabama, Chairman
MICHAEL CRAPO, Idaho SHERROD BROWN, Ohio
BOB CORKER, Tennessee JACK REED, Rhode Island
DAVID VITTER, Louisiana CHARLES E. SCHUMER, New York
PATRICK J. TOOMEY, Pennsylvania ROBERT MENENDEZ, New Jersey
MARK KIRK, Illinois JON TESTER, Montana
DEAN HELLER, Nevada MARK R. WARNER, Virginia
TIM SCOTT, South Carolina JEFF MERKLEY, Oregon
BEN SASSE, Nebraska ELIZABETH WARREN, Massachusetts
TOM COTTON, Arkansas HEIDI HEITKAMP, North Dakota
MIKE ROUNDS, South Dakota JOE DONNELLY, Indiana
JERRY MORAN, Kansas
William D. Duhnke III, Staff Director and Counsel
Mark Powden, Democratic Staff Director
Dana Wade, Deputy Staff Director and Senior Counsel
Jelena McWilliams, Chief Counsel
Laura Swanson, Democratic Deputy Staff Director
Graham Steele, Democratic Chief Counsel
Dawn Ratliff, Chief Clerk
Troy Cornell, Hearing Clerk
Shelvin Simmons, IT Director
Jim Crowell, Editor
(ii)
C O N T E N T S
----------
THURSDAY, JULY 23, 2015
Page
Opening statement of Chairman Shelby............................. 1
Opening statements, comments, or prepared statements of:
Senator Brown................................................ 2
WITNESSES
Robert DeYoung, Capital Federal Professor in Financial Markets
and Institutions, University of Kansas School of Business...... 3
Prepared statement........................................... 25
Responses to written questions of:
Chairman Shelby.......................................... 37
Deborah Lucas, Sloan Distinguished Professor of Finance, and
Director, MIT Center of Finance and Policy, Sloan School of
Management..................................................... 5
Prepared statement........................................... 26
Jonathan R. Macey, Sam Harris Professor of Corporate Law,
Corporate Finance, and Securities Law, Yale Law School......... 7
Prepared statement........................................... 29
Michael S. Barr, Roy F. and Jean Humphrey Proffitt Professor of
Law, University of Michigan Law School......................... 8
Prepared statement........................................... 31
(iii)
MEASURING THE SYSTEMIC IMPORTANCE OF U.S. BANK HOLDING COMPANIES
----------
THURSDAY, JULY 23, 2015
U.S. Senate,
Committee on Banking, Housing, and Urban Affairs,
Washington, DC.
The Committee met at 9:32 a.m., in room SD-538, Dirksen
Senate Office Building, Hon. Richard C. Shelby, Chairman of the
Committee, presiding.
OPENING STATEMENT OF CHAIRMAN RICHARD C. SHELBY
Chairman Shelby. The hearing will come to order.
Today, we will hear from experts on the best criteria and
methods to determine the systemic importance of U.S. banks.
For nonbanks, Dodd-Frank set up a process governed by a
council of Federal regulators to determine if an institution is
systemically important. As imperfect as this process is, there
is no such process for banks. Instead, Dodd-Frank deems a bank
systemically risky if it has $50 billion or more in total
assets. Moreover, once a bank reaches this arbitrary threshold,
it is automatically designated as systemically important. Under
this automatic framework, where there is no clear exit from the
designation, a bank has little incentive to reduce its level of
systemic risk.
Many experts have expressed concerns about the arbitrary
$50 billion threshold as an automatic cutoff for systemic risk.
Many of us share their concerns. In March, financial regulators
testified right here that there are currently banks above $50
billion that were regulated as if they were systemically risky,
even though they were not considered to be so. This regulatory
framework should not capture institutions whose failure would
not lead to systemic contagion. Doing so has a true cost to the
financial system.
First, it imposes a layer of regulation on financial
institutions that lend primarily to small businesses and local
or regional communities.
Second, it unnecessarily spreads too thin the important
resources of our financial regulators. This does not make the
financial system safer.
As I have said before, systemic risk is difficult to
measure, but 5 years after Dodd-Frank, the law that mandates
systemic risk regulation, we have better tools to assess it and
we should use them.
Last week at a hearing here in this Committee, Chairperson
Yellen of the Federal Reserve testified that she would support
giving some flexibility to the Federal Reserve to determine
which banks should be subject to enhanced standards based on
their set of multiple criteria. In fact, the Federal Reserve
uses a similar approach to determine the systemic importance of
banks in its regulation of bank capital.
Earlier this week, the Federal Reserve finalized a capital
surcharge rule for the Nation's largest and most systemically
risky banks. This rule incorporates a framework based on many
factors, including not only size, but also interconnectedness,
cross-jurisdictional activity, substitutability, and
complexity. According to the Fed, these five broad categories,
quote, ``are viewed as good proxies for and are correlated with
systemic importance.''
Today, I look forward to hearing the views of our panel of
witnesses on measures that can be used by regulators to
determine if a bank poses a systemic risk. Improving such
measures will allow our regulators to focus their resources on
the systemically important banks in order to protect American
taxpayers and the U.S. economy from the next financial crisis.
Senator Brown.
STATEMENT OF SENATOR SHERROD BROWN
Senator Brown. Thank you, Mr. Chairman, for holding this
hearing. It was almost a year ago I held a similar Subcommittee
hearing in which Professor DeYoung testified. Thank you again.
I thank him and all of our witnesses for being here today.
Tuesday, as we know, is the fifth anniversary of the day
that President Obama signed the Wall Street Reform Act into
law. Our country was emerging from a devastating economic
crisis, one caused in large part by financial institutions that
ran wild and regulators that did little or nothing about it.
Some Americans have recovered, but it is a slow process and
every household's story is different. Wall Street Reform
stabilized and strengthened our economy despite dire Republican
predictions.
The financial crisis was caused by poor mortgage
underwriting, lax capital standards, lax liquidity standards,
inadequate risk management, regulators that failed to challenge
the banks that they supervised. Congress through Dodd-Frank
crafted a reasonable response, directing agencies to institute
standards for capital and liquidity and risk management and
stress testing to lower the likelihood and the costs of large
bank holding company failures, called for heightened rules for
banks over $50 billion in total assets, 31 of the largest bank
holding companies. It urged regulators not to take a one-size-
fits-all approach, allowing for tailoring based upon a variety
of factors, so the $50 billion bank would not be treated the
same way as a $2 trillion bank.
On Tuesday, for example, as the Chairman said, the Fed
finalized a rule to increase capital standards. That rule
applied to the eight largest United States banks. Many of the
powers in Title I of Dodd-Frank were not new, but after
regulators failed to use their authority leading up to the
crisis, Congress wanted to ensure that regulators used their
authorities in ways that have teeth. The new rules were not
meant to cover only systemically important or too-big-to-fail
banks. In fact, these words are not even used in the law.
Enhanced prudential standards are intended to respond to the
last crisis, more importantly, though, to prevent the next one.
We all agree that a regional bank is not systemic in the
same way that a large money center bank is. The failure of one
regional bank, assuming it is following a traditional model,
will not threaten the entire system.
But, as we have heard in past hearings, the failure of a
single large institution can create systemic risk, but so can
multiple failures of similar small or midsize institutions, as
we saw in 2008. Systemic importance is also about the
importance of an institution to homeowners and small businesses
in the economic footprint where that bank operates. Congress
should only open up Dodd-Frank if it can identify real problems
affecting actual institutions, and it should be careful to do
so without undermining safety and soundness or consumer
protection.
That is why I am concerned by Title II of the bill that was
passed by this Congress along party lines in May and the
language that was included in an appropriations markup
yesterday. Secretary Lew said this proposal was, quote,
``designed to gut the heart of Dodd-Frank,'' unquote. So, if
the goal is to have something signed into law, we need to take
a more modest approach.
I would appreciate hearing today which specific prudential
standards are inappropriate for regional banks and why, and
whether the concerns being raised stem from implementing
regulations or from the law itself. We need to strike the right
balance. If the Fed should use its authority to tailor its
regulations to the institutions and activities that it thinks
present the most risk, but it should not become complacent and
take its eyes off of all possible sources of risk.
I thank the witnesses for joining us today.
Chairman Shelby. Thank you, Senator Brown.
Our witnesses today include, and we will start with
Professor DeYoung. He is the Capital Federal Distinguished
Professor of Finance at the University of Kansas.
Professor Deborah Lucas, the Sloan Distinguished Professor
of Finance and Director at the MIT Center for Finance and
Policy.
Professor Jonathan Macey, the Sam Harris Professor of
Corporate Law, Corporate Finance, and Securities Law at the
Yale Law School.
And the Honorable Michael Barr, who is no stranger to this
Committee, the Roy F. and Jean Humphrey Proffitt Professor of
Law at the University of Michigan Law School.
All of your written testimony will be made part of the
hearing record.
We will start with you on the left, Professor DeYoung. You
are recognized.
STATEMENT OF ROBERT DEYOUNG, CAPITAL FEDERAL PROFESSOR IN
FINANCIAL MARKETS AND INSTITUTIONS, UNIVERSITY OF KANSAS SCHOOL
OF BUSINESS
Mr. DeYoung. Thank you, Chair. You asked us to share our
perspective on which factors are important for determining the
systemic risk of bank holding companies and to provide explicit
examples of which rules and regulations or factors might be
inappropriate, and I will get to the latter during the
discussion. I will get to the former during my remarks here.
Bank size is, of course, the most immediate consideration.
Larger bank holding companies tend to have more volatile
earnings, tend to be less liquid, tend to be more
interconnected, and tend to be more difficult to value in a
resolution. But, by itself, as we all have discussed, the bank
size is neither a necessary nor a sufficient indicator of its
systemic risk. Drawing a bright line at $50 billion, or at $200
billion, or at any other place, will capture some nonsystemic
banks.
A good example is Washington Mutual, which held over $300
billion in assets at the time of its failure in 2008. The FDIC
was able to resolve WAMU without systemic consequences, without
Government financial support. So, resolvability, in addition to
assets, is another important factor in addition to bank size
for determining whether or not a banking company poses a
systemic threat.
The bill in question here would redraw the bright line at
$500 billion of assets, but it is not as bright a line as that
seems. It would also rely on the Federal Reserve and the
Financial Stability Oversight Council to evaluate the systemic
importance of banking companies below this asset size
threshold. This approach would automatically define the six
largest bank holding companies in the U.S. as systemically
important and, of course, not so coincidentally at all, on
Monday, the Federal Reserve announced systemic risk capital
surcharges on these same six firms.
For smaller firms, the Fed and the FSOC will be free to
consider multiple indicators of systemic risk other than asset
size--off-balance sheet positions, earnings volatility,
interconnectedness, cross-country exposures, and many others.
A good example, I think, of this type of multifactor
approach could be found in a recent policy brief from the
Office of Financial Research. Now, I am not in a position to
endorse the exact formulations within the OFR methodology, but
I do strongly endorse the general approach that it takes. It
uses predefined weights to translate each bank's size, business
activities, financial complexity, and interconnectedness into a
quantitative score that represents each bank's relative
systemic importance. This approach applies the same filters to
every banking company, so in a way, human discretion does not
play a role in determining the relative outcomes.
Now, the natural concern is that one or more banks that
pose systemic threats would be mistakenly left off this list,
and in order to err on the side of caution, we should maintain
a low asset size threshold. I understand this concern, but
given what we have learned, I believe it is somewhat
unwarranted. In any case, mistakenly putting nonsystemic banks
on the list imposes costs, as well, and we have to recognize
those costs. The size of a banking company is just one
potential indicator of systemic risk.
For example, consider four U.S. bank holding companies that
are each similar sized, between $300 and $400 billion: U.S.
Bank Corp., PNC, Bank of New York Mellon, and State Street. In
the Office of Financial Research's scoring exercise, two of
these banks, U.S. Bank Corp. and PNC, get relatively low
systemic risk scores because they practice traditional banking.
They hold a diversified portfolio of loans. Those loans are
fully funded by stable core deposits. They have very little
off-balance sheet exposures, and their clientele is almost
completely domestic.
The other two banks, in contrast, the Bank of New York
Mellon and State Street, get relatively high systemic risk
scores in this method because they hold very few loans, rely on
relatively unstable deposit funding, have large cross-country
exposures, and provide infrastructure and logistics that are
essential for the smooth operations of securities markets.
Of course, under such an approach, the Fed and the FSOC
would still have to determine where to draw the line. That is
where discretion happens. I strongly suspect that these
agencies will err on the side of caution when drawing this
line, and I think we can look at the example of MetLife, which
was designated as a SIFI, as a case study of this.
In closing, I want to reemphasize one of the factors I
talked about before, and that is the importance of
resolvability in determining a bank's systemic importance. If a
bank holding company can be resolved without causing
disruptions in financial markets or contagion to other banks,
either through regular bankruptcy or through orderly
liquidation authority, then that bank should not be considered
to be systemically important.
It is not the job of bank regulators to prevent
insolvencies at poorly run banking companies. I think we could
all agree that poorly run banking companies should exit the
market and stop wasting society's scarce resources. Our goal
should be a safe resolution for these banks, not additional
regulatory and supervisory safeguards that, by keeping poorly
run banks out of trouble, keeps them operating and keeps them
in business.
So, I will end there. Thanks for your time this morning. I
hope my remarks are useful. I look forward to answering your
questions.
Chairman Shelby. Thank you.
Professor Lucas.
STATEMENT OF DEBORAH LUCAS, SLOAN DISTINGUISHED PROFESSOR OF
FINANCE, AND DIRECTOR, MIT CENTER OF FINANCE AND POLICY, SLOAN
SCHOOL OF MANAGEMENT
Ms. Lucas. Thank you. Chairman Shelby, Ranking Member
Brown, distinguished Members of the Committee, thank you for
inviting me to speak with you today. I have been asked, too, to
comment on the appropriate criteria for determining whether a
bank holding company poses the systemic risk to the financial
system.
Banks deemed to be strategically important financial
institutions, or SIFIs, are subject to a higher level of
oversight and, often, higher capital requirements. Those
measures reduce the likelihood of distress and spill-overs to
the financial system, but also entail additional costs for the
banks. Ideally, banks would only be designated as SIFIs when
the financial stability benefits outweigh those costs.
Unfortunately, those cost-benefit tradeoffs are difficult
to quantify. Major systemic risk events are rare, but extremely
costly. History may be a poor guide to the future.
The good news is that, despite the challenges, the results
of recent analyses using new data suggest that the current
criteria used for SIFI designation could be improved upon in
several ways. I have two main conclusions.
The first is that the threshold for automatic SIFI
designation for bank holding companies could be raised
substantially from its current level of $50 billion of assets
without significantly increasing systemic risk. That conclusion
rests on the findings of several regulatory and academic
studies that use a variety of approaches to identify SIFIs. It
also reflects the common sense observation that the very
largest bank holding companies are enormously more complex and
interconnected than their midsized or even large peers.
What is striking about those analyses is that quite
different measurement approaches come to very similar
conclusions, with just eight of the largest U.S. bank holding
companies standing out for their likely systemic importance.
The smallest of those, State Street, has assets now of about
$280 billion, which is more than five times the current $50
billion threshold for SIFI designation.
Consistent with that emerging evidence, and as Senator
Shelby and Senator Brown noted, the Federal Reserve issued a
white paper last week that contemplates replacing the $50
billion asset size threshold with one of three alternatives
that effectively would increase the cutoff to at least $250
billion. The Fed's analysis also suggests the possibility of
setting a threshold based on the relative systemic risk score
rather than setting a dollar-size cutoff. Such an approach
would have the advantage of automatically adjusting over time
and certainly deserves further consideration.
My second conclusion is that it would be advisable for
regulators to use several criteria in addition to asset size to
more accurately identify SIFIs. There seems to be general
agreement that size alone is not the best proxy for an
institution's contribution to systemic risk, and financial
regulators in the U.S. and abroad have identified five broad
categories of factors to consider, including size,
interconnectedness, substitutability, complexity, and cross-
jurisdictional activity. Several of the analyses I referred to
earlier incorporate those criteria into the risk scores used to
identify the most systemically risky bank holding companies.
Nevertheless, incorporating multiple criteria involves
several significant challenges. The first is creating well-
defined metrics for each criterion. The second is designing a
weighting scheme that determines the relative importance of
each in an overall risk score. And making those choices,
considerations, include data availability, stability of
outcomes, avoiding excessive complexity, and preserving
transparency.
Choosing a weighting scheme is particularly difficult.
There is not a precise definition nor even complete agreement
about what makes a financial institution systemically risky,
and there is little evidence about the relative importance of
the different criteria or their predictive accuracy. It is,
nevertheless, promising that the various approaches now under
consideration point to a consistent set of bank holding
companies as SIFIs and that asset size is highly correlated
with all of the leading measures.
However, the metrics that regulators are beginning to adopt
are still new and evolving. Hence, I think it is advisable to
allow some latitude for revising the methodology as new data
becomes available and as market practices and perceived risks
change over time.
I only have a few seconds left, so I would like to use this
opportunity to just briefly discuss what I see as the most
serious deficiency in systemic risk oversight as it is
currently conducted, and that is the exemption of major
Government-run financial institutions from SIFI designation
and, hence, from any formal oversight by systemic risk
regulators.
Government financial institutions, particularly Fannie Mae,
Freddie Mac, FHA, and so forth, are collectively much larger
than the bank holding companies currently classified as SIFIs.
They satisfy most of the other criteria for SIFI designation,
such as high degrees of interconnectedness. So, these sorts of
considerations support the idea that Government financial
institutions are an important source of systemic risk and,
hence, also should fall under FSOC's mandate.
Thank you very much.
Chairman Shelby. Thank you.
Professor Macey.
STATEMENT OF JONATHAN R. MACEY, SAM HARRIS PROFESSOR OF
CORPORATE LAW, CORPORATE FINANCE, AND SECURITIES LAW, YALE LAW
SCHOOL
Mr. Macey. Thank you, Senator Shelby and Ranking Member
Brown and Members of the Committee, former law professor
colleagues. It is a professor to be here to talk about whether
it is appropriate to continue to assume that all banking
companies with more than $50 billion in assets are systemically
important and, therefore, subject to a heightened level of
prudential regulation.
Currently under consideration is a proposal that would move
the automatic threshold to $500 billion and then authorize the
Fed and the Financial Stability Oversight Council to evaluate
the systemic importance of banking companies below that $500
billion asset size threshold. This bill would reduce from 36 to
6 the number of financial institutions subject to the automatic
designation as systemically important, and I support this
approach for the following five reasons.
First, the bill would reduce some of the distortive effect
of the current regulatory regime, which provides incentives for
midsize banks to stop growing in order to avoid the SIFI
designation and provides incentives for institutions above the
threshold to grow until they approach the size of the so-called
big six in order to be able to amortize the additional cost of
regulation placed on such institutions designated as
systemically important.
Second, the bill would inject a greater degree of
intellectual rigor into the SIFI designation process. In
particular, regulators would not be able to focus solely on an
arbitrary measure and would have to look at the kinds of
objective factors that Professor Lucas was describing, and I
think it is useful to remember that when Dodd-Frank was
initially proposed, it was marketed as eliminating the
longstanding practice of treating some institutions as too big
to fail. But, if we look at what I regard as the flawed process
by which MetLife was designated as a SIFI, we do have a need to
impose better analytics and more intellectual rigor on the
designation process.
Third, I think that the bill would promote fairness by
reducing reliance on an arbitrary line of demarcation that
nobody has been able to support or defend, either analytically
or empirically.
Fourth, I think the bill would reduce some of the
pathologies in bank regulation that Dodd-Frank created. The
financial system is more concentrated, more interconnected, and
more opaque than it was before the financial crisis. Much of
this, I acknowledge, happened during the financial crisis, but
things are not getting any better. Massive concentration caused
by Bank of America acquiring Countrywide and Merrill Lynch,
JPMorgan acquiring Washington Mutual--although that was a good
deal, I agree with that--and also acquiring Bear Stearns, and
Wells Fargo's acquisition of Wachovia. Now, the six largest
financial institutions hold over 60 percent of all of the
assets in the financial system and hold a virtual 100 percent
market share of shadow banking activities.
For people like me who think that the administrative State
should be subject to the rule of law, Dodd-Frank poses
significant challenges. Never has so much rulemaking authority
and discretion been granted so broadly. As I have observed
before, Dodd-Frank, for all of its merits, it is not really
directed at people. It is an outline, a very long outline,
directed at bureaucrats and it instructs them to make still
more regulation and to create still more bureaucracies. And,
the efforts to designate mutual funds and other businesses that
really do not provide any systemic risk are really illustrative
of that.
And, fifth, the--for people like me who think the best way
to avoid having financial institutions that are too big to fail
is to reduce to zero the number of institutions that are too
big to fail, the proposed legislation provides positive
incentives for banks to be smaller and negative incentives on
banks to become larger. Like the Fed's new capital requirements
for the eight largest financial institutions, the proposed
statute imposes some cost on the very largest financial
institutions, which I support.
Regulators, left to their own devices, have incentives to
increase the list of systemically important financial
institutions. These incentives are unfortunate. Regulators
should be given incentives to reduce, not to expand, the list
of SIFIs, and if the concept of systemic risk is to have any
meaning, it must be the case that reducing systemic risk by
reducing the number of firms that pose such risk is an
important goal for any regulator.
Thank you.
Chairman Shelby. Thank you.
Professor Barr.
STATEMENT OF MICHAEL S. BARR, ROY F. AND JEAN HUMPHREY PROFFITT
PROFESSOR OF LAW, UNIVERSITY OF MICHIGAN LAW SCHOOL
Mr. Barr. Chairman Shelby, Ranking Member Brown,
distinguished Members of the Committee, it is my pleasure to
appear before you today, 5 years after enactment of the Dodd-
Frank Wall Street Reform and Consumer Protection Act.
That Act was passed in response to the worst financial
crisis since the Great Depression. In 2008, the United States
plunged into a severe financial crisis that shuttered American
businesses, that cost millions of households their jobs, their
homes, and their livelihoods. The crisis was rooted in years of
unconstrained excess and prolonged complacency in major
financial capitals around the world. The crisis demanded a
strong regulatory response.
I want to focus today on aspects of prudential oversight
established in the Dodd-Frank Act. Under the Act, the Fed is
directed to provide for a graduated system of regulation with
increased stringency depending on the risk that the firm poses
to financial stability. The Fed may tailor these prudential
standards for individual firms or categories of firms.
The enhanced prudential measures include risk-based capital
requirements and leverage limits, liquidity requirements, risk
management, resolution planning, credit exposure reporting,
concentration limits, and annual stress tests.
The Fed under the Act is not required to apply these more
stringent standards to bank holding companies with assets under
$50 billion. Annual firm-led stress tests, however, are
required for firms between $10 and $50 billion in size, and
publicly traded bank holding companies $10 billion and above
must establish risk committees.
None of these enhanced measures apply to about 95 percent
of banks, the category commonly described as community banks,
those under $10 billion in assets, more than 6,000 banks in
communities all across the country.
Graduated standards are already at work. Fed stress testing
applies to the largest firms in the country, the 31 firms with
assets of $50 billion and above. The largest and most complex
banks face more stringent standards. The Fed, for example,
imposes a supplementary leverage ratio, a countercyclical
capital buffer, and detailed liquidity coverage rules on only
14 firms with over $250 billion in assets. The eight largest
banks are subject to even tougher standards, including capital
surcharges, more stringent leverage ratios, and long-term debt
requirements. This graduated approach makes sense.
Some have argued that the size threshold for heightened
prudential standards should be substantially increased, while
others have argued that banks should not be subject to any
heightened standards unless they are specially designated as
systemic. Both approaches, in my judgment, are mistaken.
First, some have mistakenly said that the Act describes
firms with $50 billion in assets as systemic, but that is
simply not the case. There is no automatic designation under
the Act. Congress set the $50 billion threshold as a floor, to
establish a floor under which smaller firms would know that
they are not subject to the new rules. But, the rules were not
meant to apply only to the very few largest firms in the
country. They are not intended to apply only to systemically
important firms. They are designed, as I said, to work in a
graduated and tailored way.
Second, others have argued that bank holding companies
should have to be designated for heightened supervision by the
same process FSOC uses for nonbank firms. But, that runs
counter to the purposes of nonbank designation. Bank holding
companies should not be required to be designated in order to
be supervised. Bank holding companies are already supervised by
the Fed, and the Fed already has the authority to impose
heightened prudential standards on such firms on a graduated
basis as they increase in size and complexity.
The reason for the designation process for nonbank
financial institutions is that such institutions were not
subject to meaningful consolidated supervision by the Fed at
all. Firms such as Lehman Brothers and AIG could operate with
less oversight, more leverage, and riskier practices.
Recognizing that policing the boundaries of financial
regulation is critical to making the financial system safer,
the Dodd-Frank Act established a process for bringing such
nonbank financial institutions into the system of regulatory
oversight. It makes little sense to require designation of
firms that are already supervised by the Fed, and it will
dramatically slow down and disrupt the Fed's existing
oversight.
None of these changes would help community banks. There is
undoubtedly much that could be done to reduce the regulatory
burden on the smallest banks. For example, small community
banks would benefit from clear safe harbors and short plain
language version of rules that apply to them, longer exam
cycles, and streamlined reporting.
Today, the U.S. financial system is more resilient, but
there is still much more work to do together. Thank you.
Chairman Shelby. The Federal Reserve currently employs a
multifactor test to determine if a bank is globally
systemically important for the purposes of determining capital
requirements. I will direct this to you, Professor Lucas and
Professor Macey. In your opinions, what are the greatest
benefits of using criteria like this rather than solely the $50
billion asset threshold to regulate systemic risk?
Ms. Lucas. Well, the reason is, as some of the examples
cited, demonstrated that there are financial institutions that
are larger than $50 billion who, nevertheless, operate as very
traditional banks. There is nothing particular about their
activities that would suggest singling them out as being
systemically important. As people have noted, you can get
systemic importance when a lot of banks act in the same way,
but it does not make sense to apply special regulations to
banks that in most respects act like much smaller institutions.
Chairman Shelby. Professor Macey.
Mr. Macey. Yes. I mean, it seems kind of straightforward to
me, simple, really. If I am running a bank, because being
designated as systemically important is costly, if there were a
multifactor approach and not a bright line $50 billion
approach, then I could take steps to avoid being systemically
important without shrinking dramatically, and I think that is
the kind of incentives we, as a regulatory--as people thinking
about regulation--want to give to people running banks, that we
want them to engage in activities that do not impose systemic
risk, to, all else equal, decline or refrain from excessive
engagement in activities that are systemically risky, and this
sort of multifactor test is the only way to get there. Or, to
put it differently, having a bright line cutoff at $50 billion
eliminates that incentive.
Chairman Shelby. Thank you.
Professor DeYoung, how does resolvability relate to
systemic risk, and is it possible for a bank to be large in
terms of total assets but still be easy to resolve if they had
some challenges?
Mr. DeYoung. Yes. We often equate the two terms,
systemically important and too big to fail, correct. But, now,
I would not necessarily divide things up that way. I would say
that if a bank can be resolved, then I think what I stated was
that I do not think we should--we should not consider that bank
to be systemically important.
Now, what does it take for a bank to be resolved? A bank
needs to have assets that are easily valued, right. We need to
have buyers who can take a look at that bank, or FDIC
evaluation staff and take a look at that bank----
Chairman Shelby. Sure.
Mr. DeYoung. ----and figure out what the assets are worth,
play that off against the liabilities, do this to some high
degree of accuracy, not perfectly, but a high degree of
accuracy, and at that point, we have a value for the bank.
Once we have a value for the bank, two things could happen.
Another bank could purchase that failed bank, or we could have
a resolution process in which the bank's assets are sold off in
pieces, because, as I said, they are easy to value.
Now, if we have an organization that has much off-balance
sheet activity, a lot of counterparties, derivatives that are
traded over the counter which are not always easily valued, any
kind of assets or liabilities that are traded in thin markets,
these are the kind of banks that----
Chairman Shelby. That situation makes everything more
complex, does it not?
Mr. DeYoung. Yes, that is exactly right. And a key--I want
to come back just to the key--is that at that point, we cannot
value the bank, in which case makes resolution very difficult,
because we cannot find a buyer for the bank or we cannot find--
we do not know how big the hole is, right. We do not know what
the cost would be to resolving that bank.
Chairman Shelby. Thank you.
Mr. DeYoung. So, banks like that need to operate under
different rules.
Chairman Shelby. Professor Lucas, what are the risks of
grouping banks whose failure would not be contagious to the
system with banks who are systemically risky institutions?
Ms. Lucas. I do not think it is a risk to the system to
include those smaller banks, but I think the integrity of the
regulatory process should not draw into its net institutions
that do not need to be there. So, that is basically the
argument. And, actually, to take what Dr. Barr said and turn it
a little bit, those smaller banks are already heavily regulated
by the Federal Reserve, and so if I did not believe that there
was already a substantial amount of oversight, I might not be
arguing for lifting the limit, but because there is, it is not
clear that you need this additional layer of regulation.
Chairman Shelby. Professor DeYoung--my last question--you
said in your testimony that even large banks with total assets
of over $300 billion might have little systemic risk. Other
witnesses give some examples. Could you explain to the
Committee how a bank might be so large and yet exhibit little
systemic risk. It is because of what kind of banking they are
doing and the risk they take?
Mr. DeYoung. Yes. We speak about traditional banking, and
this is an excellent question. If you look at a bank that is
very large and then you take a look at its balance sheet and it
has got the loans, maybe the mortgage loans, maybe the business
loans, maybe the credit card loans, whatever, and they are
performing, the other side of the balance sheet shows how those
loans are funded. If these loans are funded with stable deposit
liabilities, which we tend to call core deposits, these are
deposits that will not run if there is some kind of financial
crisis, so we will not have a liquidity problem, OK, so that we
will not have a liquidity problem there.
On the other side of the balance sheet, these loans are
easy to value in whole or in part, depending on what kind of
loans they are. So, once again, I get back to my point that if
a bank can be valued, then it becomes resolvable and
nonsystemic.
Chairman Shelby. Thank you.
Senator Brown.
Senator Brown. Thank you, Mr. Chairman.
Professor Barr, thank you for pointing out, in spite of
what we hear from many in this town and at many different
hearings, that Dodd-Frank does not actually designate banks
systemically important, that it is just not part of Dodd-Frank
and to suggest that is not showing sufficient intellectual
rigor and insight and understanding or something worse than
that.
I want to ask Professor Barr a series of yes or no
questions, if I could, pretty simple questions.
Is it a good thing that large banks have more capital and
less leverage?
Mr. Barr. Yes.
Senator Brown. Is it a good thing for large banks to have
more liquidity than they did before the crisis?
Mr. Barr. Yes.
Senator Brown. Comptroller of the Currency Tom Curry has
made it his mission, in part, to install a more enhanced
prestige and stature with higher compensation for a risk
officer at medium-sized and large banking institutions. Is that
a good idea?
Mr. Barr. Yes.
Senator Brown. And, I assume that means large banks should
have a whole strong risk management structure to them?
Mr. Barr. Yes.
Senator Brown. Professor DeYoung mentioned resolvability.
This question, again, is for you, Professor Barr. Should large
banks be able to detail how they can fail safely?
Mr. Barr. Yes.
Senator Brown. Is it appropriate for large banks to conduct
regular stress tests?
Mr. Barr. Yes.
Senator Brown. Dodd-Frank contains all these provisions, so
it sounds like Dodd-Frank, in your mind, has made the financial
system stronger?
Mr. Barr. Yes.
Senator Brown. Thank you.
One more yes or no question, and I want to ask you a little
bit more detail to test your reasoning ability, which you have
not yet--you have showed in your testimony, but not yet in the
questions and answers.
Factors like capital structure and riskiness and complexity
and financial activity size, other risk-related factors, are
they appropriate criteria for use as a basis for crafting
prudential standards?
Mr. Barr. I think they are. I think the important thing is
we do not need them in the particular provision that has been
subject to controversy in this hearing thus far because, as I
said, bank holding companies are already subject to
supervision. You do not need them to bring them into the system
of supervision. They are useful tools to then decide, once
firms are supervised, what is the appropriate level of capital,
how stringent should the regulation be, what is the supervisory
expectation with respect to risk management of the firm, how do
you deal with resolvability. All those are really important
factors.
I agree that they are and should be graduated and that the
very largest firms that are the most complex firms, that are
firms that are the most interconnected, should have the highest
capital requirements, for sure. That is what is already in the
Dodd-Frank Act. It is what is already in Fed regulation. And, I
think, you know, one of the problems sometimes is that Dodd-
Frank is too big to read.
Senator Brown. You were looking forward to using that line
today.
Mr. Barr. I was.
Senator Brown. That was very well done.
[Laughter.]
Senator Brown. Let me explore what you just said about the
way that Dodd-Frank authorizes the Fed to tailor its standards.
A Fed official said in 2012, quote, ``Dodd-Frank was spot on in
requiring the Fed to make sure that we do not apply a one-size-
fits-all approach to every bank holding company above $50
billion.'' Discuss, if you would in the last couple minutes,
how well you think has the Fed tailored its rule sufficiently
and fairly and precisely enough for bank holding companies over
$50 billion in assets, or could they do more. Give me thoughts
and suggestions.
Mr. Barr. Well, I think, overall, I have been impressed
with the Fed's ability to tailor and provide a graduated
approach under the rule. As I said, for the eight largest bank
holding companies, quite stringent regulation, capital
requirements, liquidity rules, and stress testing. Slightly
less stringent but still quite tough rules over 250. And a more
graduated approach between 250 and 50. I think that is
appropriate. There may be some additional measures, simplifying
stress tests for firms between 50 and the 250 range that could
be done within the existing framework.
And then, I think, really, the area where I would like to
see the most work done is for small banks. I think that small
banks face regulatory burden that could be addressed both by
the Fed and the other regulators in a productive way under
current law, and it is the small banks that, I think, are
facing, really, the kind of burden we ought to be worried about
and they need clearer rules, more safe harbors, and a lighter
touch.
Senator Brown. Yesterday, the U.S. Senate in a vote has
declared that community banks are now one billion instead of
the 10 billion that I thought we mostly agreed on here. When
you say small banks, are you saying a billion or are you saying
ten billion? The ten billion is legislation that we have worked
on here, but the Senate yesterday spoke fairly resoundingly
that it is now one billion. Your thoughts?
Mr. Barr. Well, I think, generally speaking, there is some
variation, but people think of ten billion as the marking point
below which firms are thought of as community banks. And then
there is some gradation within that. I mean, a firm--a bank
that is a $900 million bank needs a lot lighter touch than a
firm that is close to a $10 billion bank. So, I think there
needs to be graduation, even within the community bank
standard, but ten billion and below is generally thought of as
in the category of community bank.
Senator Brown. Thank you.
Chairman Shelby. Senator Toomey.
Senator Toomey. Thank you, Mr. Chairman.
My friend from Ohio, the Ranking Member, began with a
celebration of the anniversary of Dodd-Frank, so I thought I
would just share my observations on this occasion, as well,
which is that after a crisis which was caused by the Federal
Government, monetary policy and lending regulations and
mandates that created a housing bubble, we discovered that, in
the crisis, that we did not have an adequate resolution
mechanism for the failure of a large complex institution. I
mean, that was pretty clear.
Rather than addressing that problem, which I think should
have been done through reforms of the bankruptcy code, we
created Dodd-Frank, and what we have to show for that now is
big banks are now essentially public utilities, completely
controlled by regulators who operate with enormous
subjectivity, stifling innovation, reducing liquidity in all
kinds of important markets.
Medium banks, medium-sized banks have been saddled with all
kinds of costs, which means they are--and regulations--which
means they are necessarily lending less than they otherwise
would be lending.
And, we have managed to completely eliminate--we have
completely destroyed the entire de novo banking industry of
America. While we used to routinely launch 100, 200 new
community banks every year all across America, the last 5 years
since Dodd-Frank, through this morning, we have had one de novo
community bank in America, which I think can only be attributed
to some combination of the outrageous monetary policy and the
unbelievable level of regulation. That is, I think, a pretty
disturbing outcome.
But, I want to get to the questions addressed at this
hearing specifically. I wonder if anybody on the panel could
name a single $50 billion bank in America--just name one--the
failure of which would result in a measurable impact on
American GDP. Is there one bank that comes to mind, a $50
billion bank?
Mr. Barr. I think, Senator, if I might say, if you get a
series of smaller banks that fail at the same time----
Senator Toomey. OK. OK.
Mr. Barr. ----they can have an impact in the economy----
Senator Toomey. So----
Mr. Barr. ----and that is true for the largest
institutions, too.
Senator Toomey. Got you. OK. So, nobody has named a single
bank. I think you are implicitly suggesting that probably the
answer to my question is there is not a single bank, but if
many banks all failed simultaneously.
So, now, let us ask a different question. What is the
chance--Professor Macey, maybe you could address this--do you
think there is any chance at all that the regulation of these
banks that do not individually pose any systemic risk creates a
risk correlation that might actually enhance the risk of a wave
of failures? What I am getting at is, certainly, the regulators
can identify some risks and they will surely force these banks
to go at great lengths to avoid those risks. Is there any risk
that regulators, being human, might not see a risk that is out
there, but will have driven all these individually unrisky
banks to a very similar profile and have actually increased the
risk that multiple failures could occur for some reason that
they are not anticipating? Is there any risk of that at all?
Mr. Macey. I think there is a huge risk. I think it is even
larger than the problem associated with the inevitable fact
that regulators are not perfect, that once a regulation is
promulgated, rational financial institutions will respond by
looking for the most profitable unregulated niches. This
reaction, in turn, creates a kind of lemmings problem which is
really the quintessential kind of essence of systemic risk,
because what is dangerous is if you have a whole bunch of banks
entering into the same line of business at the same time, if
that line of business, like the residential mortgage-backed
securities or CDOs, turns sour, then you have a--by definition,
then you have a systemic risk problem of major proportions.
Senator Toomey. So, I just want to underscore this point
that you are making, which is that when we add this additional
layer of regulation on institutions that are not individually
systemicly risky, we actually increase the risk that we will
have a widespread problem.
Mr. Macey. Unless we can invent a world in which regulated
entities do not respond to regulation in ways that are----
Senator Toomey. Which is, of course, inconceivable. All
right. Let me ask a specific----
Mr. Macey. Did not even have it in the Soviet Union.
Senator Toomey. Right. Let me ask a question of Professor
DeYoung. I am going to run out of time here. You mentioned PNC.
PNC is roughly $350 billion. If you look at their activity,
they look at lot like a community bank. They have almost no
international activity. They have a very small derivatives
portfolio. What they do is they take deposits and they make
loans to consumers and small- and medium-sized businesses, and
yet they are currently going to be subject to the liquidity
coverage rules that was meant by Basel to apply to much larger,
multinational, international, and complex institutions. Is it
not the case that the liquidity coverage ratio, when applied to
someone who does not pose this risk, necessarily means less
lending will occur?
Mr. DeYoung. Well, I share your observation that PNC is a
very large but very traditional bank, right, we say a very
traditional community--or maybe a very large community bank,
and they are not systemically important in the ways we think of
other banks of similar size. You mentioned liquidity. The
liquidity risk at banks like this is low because they do not
fund themselves--they are not funded with market--market
instruments that will fail to refinance when there is financial
market distress. They are funded with deposit customers who
have multiple reasons for staying with the bank.
So, on the issue of liquidity, this is one of those
potentially inappropriate regulatory answers. I share your
concern that liquidity coverage ratios and net stable funding
ratios, when applied to banks whose main business is lending,
will reduce their lending capacity. I think this is, obviously,
a true thing, either by reducing the amount of loans they can
hold or by increasing their cost of funding, one way or the
other.
I will also point out that there is no academic study yet--
I know there are some studies underway, one of which I have
just begun--that takes a look at the effect of liquidity
minimums on banks that are also constrained with capital
minimums. We have not imposed binding liquidity minimums on
banks in the past. We have always--supervisors have always and
bankers have always known this is important and they have
informally made sure liquidity was good. But, once you have
binding liquidity requirements, along with binding capital
requirements, now you have two constraints on a bank's balance
sheet, and, frankly, we do not know how that is going to play
out because we have not observed it before.
Senator Toomey. Thank you. Thank you, Mr. Chairman.
Chairman Shelby. Senator Warren.
Senator Warren. Thank you, Mr. Chairman, and thank you all
for being here today.
You know, there is a lot of talk about Section 165 of Dodd-
Frank, which requires the Fed to impose some tougher rules on
banks with more than $50 billion in assets. The main question
seems to be whether a bank with more than $50 billion in assets
poses more risk than a smaller bank. It is an interesting
theoretical question, but we are not engaged in a theoretical
exercise here. We are dealing with the very practical issue of
trying to keep the financial system from melting down, because
when it did in 2008, it cost this economy an estimated $14
trillion. That is a lot on the cost side.
In theory, the Fed could tailor its rules to fit each one
of the 7,000 banks in the country, but we do not live in
theory. We live in the world. We know that is impossible and
that Congress is going to have to give the Fed some basic
guidance on where they should direct their attention. We have
to draw some lines. The question is, how do we draw lines to
ensure the safety of the system?
So, I want to follow up on Senator Toomey's question. In
recent weeks, I have asked both Professor Simon Johnson of MIT
and Chair Yellen of the Fed whether or not the failure of two
or three banks of $50 billion in assets could pose a systemic
risk and both said yes. So, Professor Barr, you have been doing
yes/no questions. Do you agree----
[Laughter.]
Senator Warren. ----with Chair Yellen and with Simon
Johnson on this?
Mr. Barr. Yes, I do. I think that if there are multiple
institutions of that size that are failing at the same time, it
is usually an indication that there is broader weakness in the
financial system, and that is why it is important for the
Federal Reserve to be regulating for resiliency across the
financial system and not just at the very largest firms.
Senator Warren. OK. So, you talk about--it sounds to me
like we have consensus that two or three $50 billion banks
could pose a systemic risk. Let us as the auto-correlation
question that Senator Toomey asked. First, I want to ask it the
other way around. Did we see correlated risk before Dodd-Frank,
Professor Barr?
Mr. Barr. I think there was undoubtedly correlated risk in
the financial system leading up to the financial crisis of
2008. We had widespread use of mortgage assets, for example,
throughout the financial system as collateral for repo
transactions, securities financing transactions, and other
items, and underlining special purpose vehicles used in
derivatives transactions. So, there was a significant degree of
auto-correlation in the lead-up to the crisis. And, of course,
during the crisis, most asset classes became correlated and
that crushed the system.
Senator Warren. That is right. Indeed, if we had not had
correlation, we would not have had the collapse, would we. Is
the Fed aware of the problem of correlation?
Mr. Barr. I think they are quite aware of it.
Senator Warren. You think they are quite aware of the
problem and try to cope with it. This is part of what they look
for in their regulations, right?
Mr. Barr. Correct.
Senator Warren. OK. So, thank you. On the other hand, I
want to look at the other half of this. The $50 billion banks
generally pose less risk than a $1 trillion bank or a $2
trillion bank. It would make no sense for the Fed to require
the same rules and impose the same rules on a $50 billion bank
as it does on a $2 trillion bank.
So, Professor Barr, you helped write Dodd-Frank, so let me
ask. Does the Fed currently have all the legal authority it
needs to tailor the rules so that a $2 trillion bank is subject
to much tougher regulation than a $50 billion bank?
Mr. Barr. Yes, they have all the authority they need, and,
in fact, they have exercised that authority to impose massively
tougher rules on the largest institutions than on smaller ones.
Senator Warren. OK. And, then, one more practical question
on this. Let us say Congress raises the threshold to $250
billion or $500 billion, as has been suggested, but gives the
Fed discretion to impose tougher standards on banks below the
threshold. That is, you move the threshold and then say you can
impose tougher standards.
Professor Barr, do you think it is likely that the Fed
would actually use that discretion to apply tougher standards
to banks below the threshold?
Mr. Barr. I worry about whether they would, in fact, do
that. I mean, I think Congress decided in the Dodd-Frank Act in
a number of instances that the Federal Reserve had too much
discretion in the past, and in this instance and in a number of
other instances reined in Fed discretion, and I think that was
a wise choice.
Senator Warren. All right. Well, you worry about it, and I
have to remember what hangs in the balance is the entire
economy, not just of the United States, but the world. You
know, Congress chose a very practical approach in Section 165.
Any bank that hits $50 billion in assets, a bank that is one of
the 40 or so largest banks in this country, will generally be
subject to some tougher rules. But that bank can go to the Fed
and make the case for tailoring the rules to fit its specific
risks.
If the Fed is not doing a good job of using its existing
authority to tailor the rules appropriately, then Congress
should demand that the Fed do a better job. I am willing to
hold the Fed's feet to the fire to do what the statute says.
But simply raising the $50 billion threshold and cutting a
whole bunch of big banks loose is a dangerous overreaction, and
if it goes badly, it is the American people who will end up
paying.
Thank you, Mr. Chairman.
Senator Crapo. [Presiding.] Senator Cotton.
Senator Cotton. Thank you.
Professor Macey, I want to touch on a couple points in your
testimony in which you say that the Shelby bill would reduce
from 36 to 6 the number of financial institutions subject to
the automatic cutoff and that you support it for a few
different reasons. One of those reasons, and I quote from your
testimony, is for people like you who believe the
administrative State should be subject to the rule of law,
Dodd-Frank poses significant challenges. Never has so much
rulemaking authority and regulatory discretion been granted so
broadly. As I--you--previously argued, laws classically provide
people with rules. Dodd-Frank is not directed at people. It is
an outline directed at bureaucrats and it instructs them to
make still more regulations and to create more bureaucracies.
Could you please elaborate on this analysis.
Mr. Macey. Yes. Thank you for giving me the opportunity.
This hearing has been something of an epiphany for me because
people universally have been talking about what I call the
spoke regulation, which is regulation directed at a particular
firm, as being a good thing. The idea of having a one-size-
fits-all approach is bad. What was called tailored regulation
is good.
I understand that Dodd-Frank makes it very difficult to
have any other kind of regulatory approach, but generally
speaking, it is not really consistent with the rule of law or
what I think is kind of the American way, that you have
regulation that is directed at particular firms. The idea is,
you know, firms in the economy should be treated the same way.
You know, you look at the designation of General Electric
Capital Corporation as a SIFI. Putting aside the merits of
that, what then ensued was a bunch of corporate governance
rules for General Electric Capital that defined things like,
you know, independent director for that entity different than
the way an independent director would be defined at JPMorgan
Chase or some other firm.
You know, I think it is a healthy thing, I think it is
important to say to the extent that Dodd-Frank compels us to do
this, it is an unfortunate cost or consequence of the
regulation. So, I think that the ultimate goal that we need to
think about is to think about ways in which we can reduce the
number of institutions that are systemically important and
thereby subject to this kind of particular bespoke regulation
rather than embrace this idea of bespoke regulation as the new
normal. Thank you.
Senator Cotton. Professor Macey, when you describe bespoke
regulation, I have to say, I do not hear the term regulation,
or I do not hear anything like the rule of law, which is to
prescribe standards of conduct that will apply prospectively
with general application to all actors. When I hear you say
bespoke regulation, I hear arbitrary discretion in the hands of
regulators and bureaucrats.
Mr. Macey. Right. Well, I think that that is a tremendous
danger, that, you know, I am kind of with the Federalist 10
idea that enlightened statesmen will not always be at the
helm----
Senator Cotton. Shocking, I know.
Mr. Macey. ----and that seems to be true of bank regulatory
agencies as well as other places. So, I share your view
entirely.
Senator Cotton. Do you think our political and financial
elites over the last, say, eight or 10 years, have demonstrated
the ability to conduct such a bespoke regulation in an
effective manner in forums like the FSB and the FSOC, the IMF,
the Federal Reserve, and so forth?
Mr. Macey. You know, you kind of--I think one should hope
for the best, expect the worst. The reality is that the people
who are promulgating these regulatory reactions are moving back
and forth to the banking sector and they are not moving back
and forth randomly to financial institutions. They are moving
back and forth to the largest ones. And, I think that as the
CEO of JPMorgan Chase recently said to his shareholders, this
being all things to all people and the biggest possible firm is
good for us, and he is right.
Senator Cotton. Mm-hmm. Well, I do not mean to question
anyone's integrity or motives, just to say that in the
incredibly complex international financial markets, it is hard
for me to imagine any one person or any small group of people
have all of the wisdom and especially all of the knowledge
necessary to engage in such kind of one-off case-by-case
decisions in a prudent manner as opposed to laying out clear
criteria well in advance that is well known to all market
players.
Mr. Macey. I agree, and I do think that markets have a
certain element of wisdom that bureaucracies and individual
people cannot manage to reflect, and I think it would be nice
if regulation reflected that notion a little bit more, in my
opinion.
Senator Cotton. Thank you.
Senator Crapo. Senator Heitkamp.
Senator Heitkamp. Thank you, Mr. Chairman.
It has been interesting, because there has been a lot of
rewriting of history, I think, today, and a lot of concern for
the sense that this, almost for some of the panel members, that
what happened in 2008 did not happen, and it did not happen
because people made bad decisions, people who were acting in a
regulatory environment, but also had an obligation, in many
cases a fiduciary obligation, to actually be honest about what
their products were. And, so, I am a little perplexed by this,
although I tend to share an attitude that we have an obligation
to constantly look back on a regulatory scheme and say, is this
working? Is this right?
But, to not go down the rabbit hole too much, Professor
Macey, is it possible for Congress to legislate broadly, the
end result of which would be only one entity would fall within
a constitutional classification?
Mr. Macey. Sure. I think----
Senator Heitkamp. That is----
Mr. Macey. Oh, sure.
Senator Heitkamp. ----the only point I wanted to make, that
we best not get so embroiled in the consequences and look
instead at the regulation.
And, so, this hearing is about 165 and, obviously, a
critical component, and I am curious--and I am going to open
this up to anyone--when we look at the tailored application of
165, and I think the intent probably was that we cannot simply
just always put a monetary value and assume that we are going
to achieve the intended result. Congress does this very often,
maybe perhaps too often, turfing a lot of responsibility to the
regulators, and then sits in panels like this and complains
because the regulators have done what we gave them the
authority to do.
And, so, if--you know, we will let you play Fed for the day
and talk about the current exercise of 165 policy, I guess,
Professor Lucas, and say, where do you think the Fed is getting
it right and where are they getting it wrong, and if we were
going to not look at a broad sweeping change of 165 and the
categories of 165, where should we be looking that makes the
most amount of sense?
Ms. Lucas. OK. So, I am sympathetic to much of what you
said and I think that the reason that I came down where I did,
which was that it would be reasonable to raise the threshold,
is that there are some things that the Fed is doing where,
although if I did believe that it would significantly reduce
systemic risk, it would not bother me, but I believe that when
you do something like ask a very simple but fairly large bank
to undergo stress tests, that is a fairly significant
regulatory burden that will not result in any reduction at all
of systemic risk.
And, I think that just for general respect for the
regulatory system, you want to set up the rules so that you do
not annoy or impose costs on institutions----
Senator Heitkamp. With no benefits.
Ms. Lucas. ----where it is certainly not necessary. So, I
think it is the stress testing and just the heightened
examination.
So, again, I think that it comes down to the transparency
that is already there for those banks, or as Dr. DeYoung put
it, the resolvability. It is not clear to me that the Fed does
not already know everything it needs to know about those banks
to deal with the systemic risk, whether they fail individually
or collectively, because the--you know, if you think about what
these regulations are doing, it is actually extremely small.
So, we are talking about it like it makes a big difference,
but, in fact, it is a very small difference, because even the
ones that are subject to higher capital requirements, it is a
very small increment to their capital requirements. So, it is
not going to make much difference to the total amount of
failures and----
Senator Heitkamp. Mr. Barr, do you have input there, too?
Mr. Barr. I think that, overall, the graduated tailored
approach the Fed has taken makes a lot of sense, and it has
particularly been effective at the very largest institutions--
--
Senator Heitkamp. Could you give your response to Professor
Lucas's point about, you know, a lot of this is ``make work''
and it does not add to the quality of the regulation in terms
of preventing systemic risk.
Mr. Barr. You know, my experience with stress testing is
that it makes a big difference inside the firm in terms of
improvements in risk management, attention to appropriate
capital planning, organizational structure, and data integrity.
So, I think it actually makes quite a big difference to risk
management at the firm and I think it would be a mistake to not
apply that stress testing approach more broadly in the economy.
Senator Heitkamp. I am out of time, but I think what you
can see here is obviously a difference of opinion. There is not
anyone on this panel--I hope there is not anyone on this panel
who wants to impose burdens that do not have a public good,
instead are just ``make work,'' and that is the balance we are
at. We have litmus tests that set a target to provide
certainty. I am sympathetic to the argument that we are not
really--that is not always necessarily the right indicator of
what we need to do, but it does provide a bright line.
With that said, the response to that when we are looking at
systemic risk may be to give the regulators more authority,
which I have a sense here some of the folks would not be
particularly supportive of. It is objective versus the
subjective and it is a tough balance. But, I was not here when
Dodd-Frank was written, but I am certainly interested in
hearing how we can make it better and how we can streamline it,
especially for the small community banks.
So, thank you. It has been a really engaging panel.
Senator Crapo. Thank you.
I will take my turn at the questions now, and I want to
follow up on this same line of questioning, and to do so go
back to last week when we had Chair Yellen, Janet Yellen, here
in front of us. I reminded her that I asked this same question
to Governor Tarullo, I believe it was last year now. It has
been a while back. And, the question basically was, is there
some flexibility that we can have that would actually help to
reduce burdens on the regulatory system that we are imposing in
this context but still maintain the necessary prudential
standards and protection.
Governor Tarullo and Chair Yellen, in my opinion, gave the
same answer. I am going to quote what Chair Yellen said in the
hearing last week, where I asked her the question of whether
some kind of an adjustment of the $50 billion threshold would
be livable or appropriate, even. Her answer was yes, that she
would be open to a, what she called a modest increase in the
threshold, and she wanted to make it very clear that in her
concept, the banks that were below the threshold would still be
subject to a significant amount of regulatory authority.
I am going to use her own words here. She said, ``I guess
the reason I would be open to it is that, as he indicated,
Governor Tarullo, and as you just stated, we do have some
smaller institutions that under Section 165 are required to do,
for example, supervisory stress testing and resolution
planning, and for some of those institutions, it does look from
our experience like the costs exceed the benefits.''
As I hear that, when I hear that the costs of a rule or a
system exceed the benefits, I can extrapolate that into a lot
of things, but one of the things that it extrapolates into in
this context is that the consumers are going to be paying a
higher price for their services in this industry if we require
this.
She went on to talk--I am skipping down a little bit. She
said, ``At present, every firm over $50 billion has to do
things like supervisory stress testing, and I think that what
we have found is, in some cases, the burden associated with
that for many of those firms really exceeds the benefit to
systemic stability.''
Now, she--to be careful here, I want to make it clear that
she said that she thought the Fed ought to have the authority
to look carefully at the risk profiles of all the banks that
they are regulating, and for some of those banks that may fall
below whatever threshold Congress might set, there may be a
risk profile for that particular bank or a set of banks that
should have heightened scrutiny and perhaps even be required to
do stress testing, or whatever it may be, but that not every
single solitary bank under any standard, just an arbitrary
dollar number, should be subject to the same, what I will call,
rigid rule.
I would just like to have each of you comment on that. We
will start on the left with you, Mr. DeYoung. I have already
used up three of my 5 minutes, so if you guys could each be
relatively brief, I would appreciate it.
Mr. DeYoung. OK. I will attempt to be brief. I think banks
should do stress tests without being asked to do them. I think
a poorly run bank will not do a stress test and it will not be
forewarned or guarded and will not be able to prepare against
stress. So, I do not think stress tests are a bad thing or make
work or a waste of time. For many firms that are not
systemically important, though, there should be--I have stated
before, there should be no--the Federal Reserve, I think, would
have no interest in applying that to firms for which there is a
zero, a zero marginal benefit in terms of its systemic
importance.
In terms--I just want to mention an offer that was made in
the American Banker by Tom Hoenig a couple of weeks ago in an
op-ed, that for small banks that have traditional balance
sheets and do not have a lot of off-balance sheet activities
and do not have over-the-counter derivatives, Vice Chairman
Hoenig said we should roll back even the Basel III increments
on higher capital. So, I think there is an example there of
graduated supervision and applying these things appropriately.
Of course, Mr. Hoenig is not in a position to deliver on this
promise, of course, but I think----
Senator Crapo. Understood.
Mr. DeYoung. ----up and down the size of banks, there is
room for a graduated authority and regulation.
Senator Crapo. Thank you.
Professor Lucas.
Ms. Lucas. OK. I will be very brief. I basically agree with
you, but I do think it is important to really leave open the
possibility for the Fed to use discretion. Particularly, they
have to be able to do that quickly when events are unfolding
that might create systemic risk at a very short time scale.
But, with that proviso, I think it is quite safe to raise the
limits for the reasons you said.
Senator Crapo. Thank you.
Professor Macey.
Mr. Macey. Yes. I think one can divide all these
regulations up into basically two categories. One are
regulations that presume--that in order to be effective require
the regulator to be smarter than the bankers and to figure out
when the bankers are engaging in risky behavior that they are
kind of trying to hide.
And then the second category of regulation, which is the
category that I like, are regulations which incentivize the
regulated entities to do the right thing, that is to say, to
the extent that shareholders of financial institutions have to
internalize or bear the cost of a bank failure, then I would
believe those firms are going to do what Professor DeYoung was
talking about and have incentives to do these stress tests
themselves or take other steps to be meaningfully prudential.
And, we have seen--so that good regulations have that
characteristic.
And, we have seen--if we take, for example, risk-based
capital requirements or that this is a private sector invention
that was a terrific idea, but once it got internalized in a
regulation it became kind of ossified, I am not opposed to
them. I think they are better than nothing, but they never
really got up to the, I think, to the promise that the
technology initially promised. I think the same is exactly true
for so-called value at risk, VAR, models.
So, I think we just need to regulate with incentives rather
than regulate from a central planning point of view.
Senator Crapo. Thank you.
Mr. Barr.
Mr. Barr. Senator Crapo, I think we need to focus the
attention on the regulatory relief that the smallest banks
need. I think the banks under a billion, banks under ten
billion, often face regulatory burdens that are quite difficult
for them to handle with very small compliance staff. So, I
think if we can focus attention on the need to get longer exam
cycles for strong compliant institutions at that level, clear
safe harbors from rules where appropriate, much shorter plain
language versions or regulations so they do not have to hire an
army of consultants to comply with them, I think that is really
the area that ought to be the focus, and I think we are doing
OK on the larger institutions, I really do.
Senator Crapo. All right. My time has more than expired.
Did you have any more questions?
Senator Brown. [Shakes head side to side.]
Senator Crapo. All right. That concludes the questions. I
want to thank this panel. Chairman Shelby had to leave for
another committee which he chairs, and so he wants to also give
you his thanks for being an excellent panel. He told me when I
came in to relieve him that we had an outstanding panel of
experts here that we could well learn from. We appreciate you
bringing your expertise to us today. Thank you.
This hearing is adjourned.
[Whereupon, at 10:48 a.m., the hearing was adjourned.]
[Prepared statements and responses to written questions
supplied for the record follow:]
PREPARED STATEMENT OF ROBERT DEYOUNG
Capital Federal Professor in Financial Markets and Institutions,
University of Kansas School of Business
July 23, 2015
Thank you for the opportunity to address the Committee. The
Chairman has asked me to share my perspective on which factors are
important for determining the systemic risk of bank holding companies.
I am pleased to do so.
Bank size is the most immediate consideration. Larger banking
companies tend to have more volatile earnings, tend to be less liquid,
tend to be more interconnected, and tend to be more difficult to value.
The raw data shows that bank failure during and after the financial
crisis was clearly correlated with bank asset size.
But by itself, a bank's size is neither a necessary nor a
sufficient indicator of its systemic risk. Regulators currently treat
all banking companies with more than $50 billion of assets as
systemically important. But this single-factor, bright line approach
will is far too simple. A good example is Washington Mutual, which held
over $300 billion in assets at the time of its failure in 2008. The
FDIC was able to resolve WAMU without systemic consequences and without
Government financial support. So resolvability is another important
factor, in addition to bank size, for determining whether or not a
banking company poses a systemic threat.
The Shelby bill would redraw the bright line at $500 billion in
assets, and rely on the Federal Reserve and the Financial Stability
Oversight Council to evaluate the systemic importance of banking
companies below this asset size threshold. This approach would
automatically define the six largest bank holding companies in the U.S.
as systemically important--not coincidentally, on Monday of this week
the Federal Reserve announced systemic risk capital surcharges for
these same six firms. For smaller firms, the Fed and FSOC would be free
to consider multiple indicators of systemic risk other than asset size,
such as off-balance sheet positions, earnings volatility,
interconnectedness, and cross-country exposures. Both sets of banks
would be subject to enhanced regulatory and supervisory treatment.
A good example of this type of multifactor approach can be found in
a recent policy brief from the Office of Financial Research (OFR 15-01,
February 12, 2015). While I am not in a position to endorse the exact
formulations within the OFR method, I strongly endorse its general
approach. It uses predefined weights to translate each bank's size,
business activities, financial complexity, and interconnectedness into
a quantitative score that represents each bank's relative systemic
importance. This approach applies the same risk filters to every
banking company, so human ``discretion'' plays no role in determining
the relative outcomes. And while the calculations may appear
complicated, both the results and the reasoning are transparent.
The natural concern is that one or more banks that pose systemic
threats could be mistakenly left off the list, and to avoid this we
should err on the side of caution and maintain the $50 billion
threshold. I think this concern is unwarranted; and in any case,
mistakenly putting nonsystemic banks on the list imposes costs as well.
The size of a banking company is just one potential indicator of
systemic risk, it is an incomplete and sometimes misleading indicator.
For example, consider four U.S. bank holding companies, each with
assets in the neighborhood of $300 to $400 billion: U.S. Bancorp, PNC,
Bank of New York Mellon, and State Street. In the OFR's scoring
exercise, U.S. Bancorp and PNC get relatively low systemic risk scores
because they practice traditional banking: they hold diversified
portfolios of loans, fully funded by stable deposits, have very little
off-balance sheet exposures, and their clientele is almost completely
domestic. In contrast, Bank of New York Mellon and State Street get
relatively high systemic risk scores, because they hold very few loans,
rely on relatively unstable deposit funding, have large cross-border
exposures, and provide infrastructure and logistics that are essential
for the smooth operations of securities markets.
Of course, the Fed and FSOC would still need to determine where to
draw the line between SIFI and non-SIFI. I strongly suspect that these
agencies will err on the side of caution when drawing this line. The
designation of MetLife as a SIFI provides a case study.
In closing, I want to reemphasize the importance of resolvability
in determining a bank's systemic importance. If a bank holding company
can be resolved without causing disruptions in financial markets or
contagion to other banks--either through regular bankruptcy or via
orderly liquidation authority--then such a bank should not be
considered systemically important. It is not the job of bank regulators
to prevent insolvencies at poorly run banking companies. I think we can
all agree that poorly run banks should exit the market and stop wasting
society's scarce resources. Our goal should be safe resolutions for
these banks--not additional regulatory and supervisory safeguards that,
by keeping poorly run banks out of trouble, keeps them operating and in
business.
Thank you for your time this morning. I hope that my remarks have
been useful. I look forward to your questions.
______
PREPARED STATEMENT OF DEBORAH LUCAS
Sloan Distinguished Professor of Finance, and Director, MIT Center of
Finance and Policy, Sloan School of Management
July 23, 2015
Chairman Shelby, Ranking Member Brown, distinguished Members of the
Committee, thank you for inviting me speak with you about the
appropriate criteria for determining whether a financial institution
poses a systemic risk to the financial system. \1\
---------------------------------------------------------------------------
\1\ The views expressed are my own and do not represent those of
the MIT Center for Finance and Policy.
---------------------------------------------------------------------------
My main focus today is on that issue as it applies to bank holding
companies (BHCs). My basic conclusions are that: (1) the threshold for
automatic SIFI designation for BHCs could be raised substantially from
its current level of $50 billion in assets without measurably
increasing systemic risk; and (2) it would be advisable for regulators
to use several criteria in addition to asset size to more accurately
identify SIFIs. In fact, regulators have been exploring multifactor
approaches for SIFI designation, and those methods appear to be able to
more accurately identify the institutions most likely to cause
contagion than a crude size cutoff. However, best practices in this
area are still evolving. Any formulaic approach that regulators adopt
may need to be revised as new data become available and as market
practices change.
I also would like to use this opportunity to briefly discuss what I
see as the most serious deficiency in systemic risk oversight as it is
currently conducted. That is the exemption of major Government-run
financial institutions from SIFI designation, and hence from any formal
oversight by systemic risk regulators. Those Government institutions--
such as Fannie Mae, Freddie Mac, FHA, the Federal student loan
programs, and perhaps State and local pension funds--are collectively
much larger than the BHCs currently classified as SIFIs. They also
satisfy most of the other criteria suggested for SIFI designation such
as a high degree of interconnectedness. \2\ Federal mortgage guarantors
were at ground zero of the financial crisis. Those considerations
support the idea that such institutions represent an important source
of systemic risk and hence should fall under FSOC's mandate.
---------------------------------------------------------------------------
\2\ Other examples of governmental activities that could pose
systemic risk include the student loan programs of the U.S. Department
of Education and the many pension-related activities of State and local
governments.
---------------------------------------------------------------------------
The Dodd-Frank Wall Street Reform and Consumer Protection Act was
passed in the wake of the most severe financial crisis and subsequent
economic downturn since the Great Depression. Those events revealed the
vulnerability of the global financial system and the real economy to
cascading failures of complex, highly interconnected financial
institutions, and were the impetus for the enhanced regulatory
framework established. At this 5-year anniversary of the Act, and with
the benefit of experience and new data, it makes sense to consider ways
to improve its implementation so as to more effectively reduce systemic
risk while minimizing the associated regulatory burden.
SIFI Designation for Bank Holding Companies
BHCs deemed to be SIFIs are subject to a higher level of oversight
and additional restrictions, such as increased capital requirements and
stress testing. Those provisions reduce the likelihood of spillovers of
financial distress to the broader market, but entail costs for the
affected institutions. The cost-benefit tradeoffs are difficult to
quantify. Major systemic risk events are rare but the potential private
and social costs are enormous. There is little data to assess
probabilities or likely costs, and history may be a poor guide to the
future. There also is considerable disagreement about magnitude of the
costs imposed by SIFI status.
Despite the measurement challenges, recent analyses of newly
collected data suggest that the current criteria used for SIFI
designation could be improved upon in several ways.
Asset Size Threshold
A growing body of evidence suggests that the asset size threshold
of $50 billion for BHCs to be automatically deemed as SIFIs is much
lower than is necessary to protect financial stability. That conclusion
rests on the findings of several studies that employ a variety of
approaches to identifying SIFIs. It is also supported by the
commonsense observation that however one measures it, the very largest
BHCs are enormously more complex and interconnected than their midsized
peers.
The OFR recently released a policy brief showing that a
multidimensional measure of systemic risk only identifies the very
largest U.S. banks as SIFI candidates. \3\ That analysis identifies the
eight BHCs listed in Table 1 as standing out for their systemic
importance. The smallest of those, State Street, had assets of $279
billion as of March 2015.
---------------------------------------------------------------------------
\3\ ``Systemic Importance Indicators for 33 U.S. Bank Holding
Companies: An Overview of Recent Data'', by Meraj Allahrakha, Paul
Glasserman, and H. Peyton Young, Office of Financial Research Brief,
February 12, 2015.
A very different approach to identifying systemically important
banks has been proposed and implemented by Professor Robert Engle of
NYU and his colleagues. \4\ Their method relies on statistical analysis
of stock price dynamics and bank leverage. It currently identifies five
of the eight institutions listed in Table 1 as being in the top 10 of
systemically risky U.S. financial institutions. I mention this study
primarily because it demonstrates that very different methodologies
seem to come to similar conclusions on which BHCs are most systemically
important.
---------------------------------------------------------------------------
\4\ Those statistics and a description of the methodology are
available at: http://vlab.stern.nyu.edu/.
---------------------------------------------------------------------------
Just last week, the Federal Reserve issued a White Paper that
discusses replacing the $50 billion asset size threshold with one of
three alternatives that effectively would increase the cutoff to at
least $250 billion. \5\ They consider two related formulas, one
developed by the Bank for International Settlements (based on size,
interconnectedness, complexity, cross-jurisdictional activity, and
substitutability). The second replaces substitutability with reliance
on short-term wholesale funding. Both formulas identify the group of
banks shown in Table 1 as having the highest systemic risk. The White
Paper also suggests the possibility of setting a threshold for the
determining globally systemically important BHCs based on relative
systemic risk scores rather than setting a dollar size cutoff. Such an
approach has the advantage of automatically adjusting over time, and
certainly deserves further consideration.
---------------------------------------------------------------------------
\5\ ``Calibrating the GSIB Surcharge'', Board of Governors of the
Federal Reserve System, July 20, 2015.
---------------------------------------------------------------------------
Criteria for SIFI Designation
There is general agreement that size alone is not the best proxy
for an institution's contribution to systemic risk. Financial
regulators in the U.S. and abroad have identified five broad categories
of factors to consider. Those include size, interconnectedness,
substitutability, complexity, and cross-jurisdictional activity. The
OFR and Federal Reserve analyses described above incorporate those
criteria into the risk scores used to identify the most systemically
risky BHCs.
Incorporating those multiple criteria involve two sets of
challenges: (1) creating well-defined metrics for each criterion; and
(2) laying out a weighting scheme that determines the relative
importance of each in an overall risk score. Broad considerations in
making those choices include data availability, stability of outcomes,
avoiding excessive complexity, and preserving transparency.
To illustrate the complexity of constructing a risk score based on
multiple characteristics, it is telling that even the definition of
size is not straightforward to determine. For example, the OFR and
other regulators measure size in the risk scores they report by
including total assets plus the net value of certain securities
financing transactions plus credit derivatives and commitments as well
as counterparty risk exposures.
Choosing a weighting scheme is especially difficult. There isn't a
precise definition or complete agreement about what makes a financial
institution systemically risky, and there is little evidence about the
relative importance of the different criteria or their predictive
accuracy.
It is promising that the various approaches now under consideration
point to a consistent set of BHCs as SIFIs, and that size is highly
correlated with all of the leading measures. However, the metrics that
regulators are beginning to adopt are still new and evolving. Hence it
seems prudent to allow some latitude for revising the methodology used
as new data become available and as market practices and perceived
risks change over time.
SIFI Designation for Nonbank Financial Institutions
It is beyond the scope of this testimony to discuss in detail the
criteria for SIFI designation of nonbank financial institutions.
However, similar issues regarding size cutoffs and what other criteria
to include will certainly arise. In making those rules, a caution is
that the relevance and relative importance of various criteria will
differ considerably across different types of institutions. For
example, major exchanges such as the CBOT are likely to be deemed
systemic because of their centrality in certain derivatives markets,
but the overall size of their balance sheets is largely irrelevant to
their contribution to systemic risk. Therefore it will be important to
think carefully about the specific mechanisms that generate systemic
risk in each instance, and to avoid using a one-size-fits-all approach.
Government Financial Institutions as SIFIs
Several factors support the contention that the Government is a
significant source of systemic risk. The most obvious is its sheer size
in its role as a financial institution (or more accurately, a
collection of loosely affiliated financial institutions). My
calculations show that just through its traditional credit programs,
the Government comprised a $3 trillion financial institution in 2013,
and that figure increases to over $18 trillion when Fannie Mae, Freddie
Mac, the Federal Home Loan Banks, deposit insurance, and the Pension
Benefit Guarantee Corporation are included. \6\ Figure 1 illustrates
the size of those Government institutions relative to the largest BHCs.
---------------------------------------------------------------------------
\6\ ``Evaluating the Government as a Source of Systemic Risk'',
Deborah Lucas, Journal of Financial Perspectives, November, 2014.
---------------------------------------------------------------------------
Many of the other criteria identified as important for BHCs,
including interconnectedness, substitutability, and complexity, also
apply to these Government financial institutions. Lack of transparency
and light supervision also contribute to the likelihood that they are a
source of systemic risk.
However, probably more important for systemic risk than the
Government's direct effect on the allocation and riskiness of credit is
its influence on the incentives facing private individuals and
institutions through its regulatory, tax and other policies. The
Government's policies reflect a variety of sometimes competing
political objectives, and there is no ``invisible hand'' guiding the
Government toward adopting policies that foster efficiency and avoid
the buildup of systemic risks. In fact, systemic risks arising from
Government actions may be relatively hard for policymakers and the
public to identify because of the lack of transparency surrounding
Government activities.
For those reasons, bringing large Government financial institutions
under the oversight of FSOC would have important benefits for the
stability of the financial system. Actions that FSOC could consider
include initiating a regulatory audit, whereby the OFR would be
directed to undertake a systematic evaluation of Federal financial
regulations across agencies to identify unintended consequences that
could give rise to systemic risk. It could also require the improvement
and standardization of certain financial disclosures by those
institutions.
Thank you for the opportunity to share these ideas. I look forward
to your questions.
______
PREPARED STATEMENT OF JONATHAN R. MACEY
Sam Harris Professor of Corporate Law, Corporate Finance, and
Securities Law, Yale Law School
July 23, 2015
Chairman Shelby, Ranking Member Brown, Members of the Committee,
and panel colleagues, I am grateful for the opportunity to talk to you
today. My name is Jon Macey, and I am a professor of law at Yale Law
School. I am here only in that capacity. I represent no firm, industry,
organization, or party. It is a pleasure to be here. Thank you giving
me the opportunity to address your Committee on the important topic of
measuring systemic risk in U.S. Bank Holding Companies.
The central question for today is whether it makes sense to
continue to assume that all banking companies with more than $50
billion of assets are systemically important and therefore subject to a
heightened level of prudential regulation. Currently under
consideration is Senator Shelby's proposal to reduce the central
reliance on a bright line test by moving the automatic threshold to
$500 billion in assets, and authorize the Federal Reserve and the
Financial Stability Oversight Council to evaluate the systemic
importance of banking companies below this asset size threshold.
The Shelby bill would reduce from 36 to 6 the number of financial
institutions subject to the automatic cutoff. I support this new
approach for five reasons.
First, the bill will reduce the distortive effect of the current
regulatory regime, which provides incentives for midsize banks to stop
growing to avoid the SIFI designation, provides incentives for
institutions above the threshold to grow at least until they approach
the size of the so called ``big six'' financial institutions in order
to be able to amortize the additional costs of regulation placed on
institutions designated as systemically important.
Second, the proposal in the bill under consideration would inject a
degree of intellectual rigor into the SIFI designation process that is
currently lacking. Regulators would have to pay more attention to
factors besides asset size. The role played by other factors, such as
operational complexity, balance between the liquidity characteristics
and maturity dates of assets and liabilities, off-balance sheet
positions, earnings volatility, interconnectedness, and cross-country
exposures would receive attention. While supporters of Dodd-Frank
initially marketed the legislation as eliminating the long-standing
practice of treating certain financial institutions as ``too big to
fail,'' nobody seriously asserts that financial institutions designated
as SIFIS would be allowed to disappear. In my view the flawed process
by which MetLife was designated as a SIFI illustrates the need to
impose more intellectual rigor on the SIFI designation process. The
MetLife designation process ignored basic principles of risk
regulation, failed to distinguish plausible risks from implausible
risks, and failed to appreciate the differences between MetLife's
business and balance sheet and the business and balance sheets of bank
holding companies. Requiring regulators to rely less on the $50 billion
Maginot Line would incentivize regulators to be more analytically
rigorous in the designation process.
A third reason to support this bill is that the new approach to
SIFI designation reflected in the statute would make the regulatory
system more fair by reducing reliance on an arbitrary line of
demarcation that nobody has been able to support or defend either
empirically or theoretically.
Fourth the change would reduce some of the current pathologies in
bank regulation that Dodd-Frank created. The financial system is more
concentrated, more interconnected and more opaque than it was before
the financial crisis. Almost all of this increase occurred during the
crisis as regulators encouraged big distressed financial firms to
acquire other even more distressed financial firms. Bank of America
acquired Countrywide and Merrill Lynch, JPMorgan acquired Washington
Mutual and Bear Stearns, and Wells Fargo acquired Wachovia. Now the six
largest financial institutions hold over 60 percent of all of the
assets in the financial system and hold a near-100 percent market share
of shadow banking sector activities.
For people who, like me, believe that the administrative State
should be subject to the rule of law, Dodd-Frank poses significant
challenges. Never has so much rulemaking authority and regulatory
discretion been granted so broadly. As I previously observed in the
Economist Magazine, ``Laws classically provide people with rules. Dodd-
Frank is not directed at people. It is an outline directed at
bureaucrats and it instructs them to make still more regulations and to
create more bureaucracies.''
The key term ``systemically important financial institution'' is
not defined, other than with reference to the fact that financial firms
that are designated as systemically important are systemically
important. Since systemic failure is, by definition, catastrophic,
regulators feel justified in acting aggressively to reduce the
likelihood that such failure will occur.
The efforts of regulators to have money market funds and mutual
funds designated as SIFIs is a prime example of the regulatory over-
reaching that is not merely enabled but encouraged by Dodd-Frank.
Simply by recognizing the primordial fact that the assets in these
funds belongs to the investors and not to the funds themselves, so that
losses in the value of the assets held by these funds is not a loss for
the entity, but rather for the investors who hold shares in the entity.
Recently we have seen bespoke regulations imposed on General
Electrical Capital Corporation (GECC), as well as with the recent
imposition of customized capital requirements on JPMorgan Chase,
Citigroup, Bank of America, and the five other largest U.S. banks that
are tailored to the perceived riskiness of each of these financial
institutions. My point is not that such firm-by-firm regulation is bad.
My point is that such regulation is inevitable, and that it inevitably
creates an uneven competitive playing field among institutions. It is
only modestly comforting that these financial institutions are so
complex that it is not possible to tell, a priori which institutions
advantaged and which are disadvantaged by the Federal Reserve's new
rules. We are clearly not living in a first or even second best
regulatory environment as we pass the fifth anniversary of Dodd-Frank.
From a policy perspective, as regulations increasingly are tailored to
reflect regulators' views of banks' riskiness as measured by the
formulas they themselves develop, exposure to the risk of favoritism,
capture and other symptoms of a runaway regulatory State multiply
exponentially.
Fifth, for those who, like me, believe that the best way to avoid
having financial institutions that are too big to fail is to reduce to
zero the number of institutions that are too big to fail, the proposed
legislation provides positive incentives for banks to be smaller and
negative incentives on banks to become larger. Like the Fed's new
capital requirements for the eight largest financial institutions, the
proposed statute imposes some costs on the very biggest financial
institutions.
On the bright side, it is worth noting that the proposed statute
would require the FSOC to provide any BHC under review for possible
designation as a SIFI with (1) a ``detailed explanation'' for any
proposed or final designation as a SIFI, (2) opportunities to meet with
FSOC members and staff, and (3) the opportunity to submit a ``remedial
plan'' prior to final designation to avoid a SIFI designation. Further,
the FSOC must reevaluate existing BHC SIFIs with assets of less than
$500 billion at the request of the Federal Reserve and at least every 5
years. These aspects of the legislation seem modest and
uncontroversial, but in my view they are an important first step in
restoring a measure of the regulatory accountability that was lost with
the passage of Dodd-Frank.
Regulators have incentives to increase the list of systemically
important financial institutions and to regulate those institutions
expansively. These incentives are unfortunate. Regulators should be
given incentives to reduce, not to expand the list of SIFIs. Unless our
regulators have truly lost their way, it must be the case that reducing
and indeed eliminating the number of financial institutions designated
as SIFIs is a key goal of our public servants. The probability of
failure of every firm in the private sector except for those that are
too big to fail is above zero. For financial institutions the
probability of failure can change dramatically in a very short period
of time. The more systemically risky firms there are in the economy,
the more risky the economy will be. If the concept of systemic risk has
any meaning whatsoever it must be the true that reducing systemic risk
by reducing the number of firms that pose such risk is an important
goal of any regulator worth her salt.
______
PREPARED STATEMENT OF MICHAEL S. BARR
Roy F. and Jean Humphrey Proffitt Professor of Law, University of
Michigan Law School
July 23, 2015
Chairman Shelby, Ranking Member Brown, distinguished Members of the
Committee, it is my pleasure to appear before you today, 5 years after
enactment of the Dodd-Frank Wall Street Reform and Consumer Protection
Act.
That Act was passed in response to the worst financial crisis since
the Great Depression. In 2008, the United States plunged into a severe
financial crisis that shuttered American businesses, and cost millions
of households their jobs, their homes and their livelihoods. The crisis
was rooted in years of unconstrained excesses and prolonged complacency
in major financial capitals around the globe. The crisis demanded a
strong regulatory response in the U.S. and globally as well as
fundamental changes in financial institution management and oversight
worldwide. The U.S. has led these reforms, both domestically and
internationally.
In the U.S., the Dodd-Frank Act created the authority to regulate
Wall Street firms that pose a threat to financial stability, without
regard to their corporate form, and to bring shadow banking into the
daylight; to wind down major firms in the event of a crisis, without
feeding a panic or putting taxpayers on the hook; to attack regulatory
arbitrage, restrict risky activities through the Volcker Rule and other
measures, regulate repo and other short-term funding markets, and beef
up banking supervision and increase capital; to require central
clearing and exchange trading of standardized derivatives, and capital,
margin and transparency throughout the derivatives market; to regulate
payments, settlement, clearance, and other systemic activities; to
improve investor protections; and to establish a new Consumer Financial
Protection Bureau to look out for the interests of American households.
I want to focus today on aspects of the system of prudential
oversight established in the Act.
Supervision of Bank Holding Companies
The Federal Reserve has supervisory authority, as it has long had,
over bank holding companies. The Fed is directed under section 165 of
the Act to provide for a graduated system of regulation, with
increasing stringency, depending on the risk that the firm poses to
financial stability, based on its nature, scope, size, scale,
concentration, interconnectedness, or other factors. The Fed may tailor
these more stringent prudential standards for individual firms or
categories of firms, based on a similar set of factors regarding risk.
These enhanced prudential measures include risk-based capital
requirements and leverage limits, liquidity requirements, risk
management, resolution planning, credit exposure reporting,
concentration limits, and annual stress tests.
The Fed is not required under this provision to apply these more
stringent standards to bank holding companies with assets under $50
billion. Annual firm-led stress tests, however, are required for firms
between $10 and $50 billion in size, and the Fed must itself stress
tests firms over $50 billion in size, in addition to such firms semi-
annual firm-led stress tests. Publicly traded bank holding companies
$10 billion in asset size and above must establish risk committees. (I
should also note that under the Act, the Federal Reserve may, upon
recommendation of the Financial Stability Oversight Council, raise the
threshold above $50 billion for certain prudential standards, those
involving contingent capital, resolution planning, concentration
limits, enhanced public disclosures and short-term debt limits.)
None of these enhanced measures apply to about 95 percent of banks,
the category commonly described as community banks, those under $10
billion in assets--more than 6,000 banks in communities all across the
country.
Graduated standards are already at work. Fed stress testing applies
to the largest firms in the country, the 31 firms with assets of $50
billion and above. Such firms represent a wide variety of risk
profiles, business strategies, sizes, specializations, and include both
foreign and domestic firms. The largest, most complex financial
institutions face the most stringent standards, as provided for under
the Act. The Fed, for example, imposes a supplementary leverage ratio,
a countercyclical capital buffer, and detailed liquidity coverage rules
only on 14 firms with over $250 billion in assets. The very largest
U.S. banks on a global basis, currently eight bank holding companies,
are subject to even tougher standards, including capital surcharges,
more stringent leverage ratios, and long-term debt requirements.
In my view, this graduated approach to supervision and regulation
makes sense. Some have argued that the size threshold for heightened
prudential standards should be substantially increased, while others
have argued that banks should not be subject to any heightened
standards unless they are specially designated as systemic. Both
approaches, in my judgment, are mistaken.
First, as to size, some have mistakenly said that the Act describes
firms with only $50 billion in assets as systemic. But that is simply
not the case. Congress set the $50 billion threshold, and another
threshold for other measures at $10 billion, to provide a floor under
which smaller firms would know that they are not subject to the new
sets of rules. But the rules were not meant to only apply to the very
few largest firms in the country. They are not intended to apply only
to systemically important firms.
They are designed to work in a graduated, tailored way to increase
the resiliency of the financial system as a whole. Risks aggregate
across the financial system, including from institutions of a variety
of sizes and types. It is the very antithesis of macroprudential
supervision to focus only on the very largest handful of financial
firms and to ignore risks elsewhere in the system. Moreover, smaller
financial institutions themselves face risk from larger institutions
and from activities across the system as a whole. Understanding those
risks is essential if we are to have a safer financial system than the
one we had before the financial crisis. We must not intentionally blind
regulators to these risks in advance.
Second, as to the idea of designation, others have argued that bank
holding companies should have to be designated for heightened
supervision by the same process the FSOC uses for nonbank firms. But
that runs counter to the purpose of nonbank designation. Bank holding
companies should not be required to be designated for heightened
supervision. Bank holding companies are already supervised by the Fed,
and the Fed already has authority to impose heightened prudential
supervision on such firms, on a graduated basis, as they increase in
size and complexity.
The reason for the designation process, under section 113 of the
Act, for nonbank financial institutions is that such institutions were
not subject to meaningful, consolidated supervision by the Fed at all.
Firms such as Lehman Brothers and AIG could operate with less
oversight, more leverage and riskier practices. Recognizing that
policing the boundaries of financial regulation is critical to making
the financial system safer, fighting regulatory arbitrage, and
providing oversight of shadow banking, the Dodd-Frank Act established a
process for bringing such nonbank financial institutions into the
system of regulatory oversight.
It makes little sense to require designation of firms that are
already supervised by the Fed, and it will dramatically slow down and
disrupt the Fed's existing oversight system. It will make the financial
system weaker, not stronger.
None of these changes will help truly small, hometown banks. There
is undoubtedly much that could be done to reduce regulatory burden on
the smallest banks. Small banks could benefit from clear safe harbor
rules and short, plain-language versions of regulations that do apply
to them. The Fed can continue to improve its tailored and graduated
approach to supervision. Strong, compliant small banks should have
longer examination cycles and streamlined reporting requirements.
Regulators and the industry should come together in a task force to
come up with better ways to implement the goals of the Bank Secrecy Act
and related rules to make it more likely that we catch terrorists and
criminals, with lower regulatory burden. And we need a level playing
field for small business lending, so community banks can compete with
nonbank providers to provide safe, transparency, consumer-friendly
loans to small businesses and entrepreneurs.
Nonbank Designations and the Financial Stability Oversight Council
Critics have also attacked the work of the Financial Stability
Oversight Council, or FSOC. FSOC has authority to designate
systemically important firms and financial market utilities for
heightened prudential oversight by the Federal Reserve; to recommend
that member agencies put in place higher prudential standards when
warranted; and to look out for and respond to risks across the
financial system.
One of the major problems in the lead up to the financial crisis
was that there was not a single, uniform system of supervision and
capital rules for major financial institutions. The Federal financial
regulatory system that existed prior to the Dodd-Frank Act developed in
the context of the banking system of the 1930s. Major financial firms
were regulated according to their formal labels--as banks, thrifts,
investment banks, insurance companies, and the like--rather than
according to what they actually did. An entity that called itself a
``bank,'' for example, faced tougher regulation, more stringent capital
requirements, and more robust supervision than one that called itself
an ``investment bank.'' Risk migrated to the less well-regulated parts
of the system, and leverage grew to dangerous levels.
The designation of systemically important nonbank financial
institutions is a cornerstone of the Dodd-Frank Act. A key goal of
reform was to create a system of supervision that ensured that if an
institution posed a risk to the financial system, it would be
regulated, supervised, and have capital requirements that reflected its
risk, regardless of its corporate form. To do this, the Dodd-Frank Act
established a process through which the largest, riskiest, and most
interconnected financial firms could be designated as systemically
important financial institutions and then supervised regulated by the
Federal Reserve. The Council has developed detailed interpretive
guidance and a hearing process that goes beyond the procedural
requirements of the Act, including extensive engagement with the
affected firms, to implement the designation process outlined in Dodd-
Frank. The approach provides for a sound deliberative process;
protection of confidential and proprietary information; and meaningful
and timely participation by affected firms. The Council has already
designated a number of firms under this authority.
Critics of designation contend that it fosters ``too big to fail,''
but the opposite is true. Regulating systemically important firms
reduces the risk that failure of such a firm could destabilize the
financial system and harm the real economy. It provides for robust
supervision and capital requirements, to reduce the risks of failure,
and it provides for a mechanism to wind down such a firm in the event
of crisis, without exposing taxpayers or the real economy to the risks
of their failure. The FDIC is developing a ``single point of entry''
model for resolution that would allow it to wind down a complex
financial conglomerate through its holding company with ``resolution-
ready'' debt and equity, while permitting solvent subsidiaries to
continue to operate. Similar approaches are being developed globally.
Other critics argue that the FSOC should be more beholden to the
regulatory agencies that are its members, but again, the opposite is
true: Congress wisely provided for its voting members, all of whom are
confirmed by the Senate, to participate based on their individual
expertise and their own assessments of risks in the financial system,
not based on the position of their individual agencies, however
comprised. Members must individually attest to their assessments in the
FSOC's annual reports. The FSOC has the duty to call on member agencies
to raise their prudential standards when appropriate, and member
agencies must respond publicly and report to Congress if they fail to
act. This system of checks and balances requires that FSOC members
leave their agency's ``turf'' at the door, and focus on systemwide
risks and responses. If anything, the FSOC's powers should be
strengthened, so that fragmentation in the financial regulatory system
does not expose the United States to enormous risk, as it did in the
past.
Some critics contend that certain types of firms in certain
industries or over certain sizes should be categorically walled off
from heightened prudential supervision, but such steps will expose the
United States to the very risks we faced in the lead up to the last
devastating crisis. The failure of firms of diverse types and diverse
sizes at many points in even very recent memory--from Lehman and AIG to
Long Term Capital Management--suggest that blindspots in the system
should at the very least not be intentionally chosen in advance by the
Congress. The way to deal with the diversity of sizes and types of
institutions that might be subject to supervision by the Federal
Reserve is to develop regulation, oversight and capital requirements
that are graduated and tailored to the types of risks that such firms
might pose to the financial system, as the agencies have been doing.
FSOC and member agencies also have other regulatory tools available
with respect to risks in the system for firms not designated for Fed
supervision, including increased data collection and transparency,
collateral and margin rules for transactions, operational and client
safeguards, risk management standards, capital requirements, or other
measures.
Some critics complain that the FSOC's work is too tied to global
reforms by bodies such as the Financial Stability Board (FSB). But
global coordination is essential to making the financial system safe
for the United States, as well as the global economy. The United States
has led the way on global reforms, including robust capital rules,
regulation of derivatives, and effective resolution authorities. These
global efforts, including designations by the FSB, are not binding on
the United States. Rather, the FSOC, and U.S. regulators, make
independent regulatory judgments about domestic implementation based on
U.S. law. And U.S. regulators follow the normal notice and comment
process when developing financial regulations. The FSB itself has
become more transparent over time, adopting notice and comment
procedures, for example, but it could do more to put in the place the
kind of protections that the FSOC has established domestically. \1\
---------------------------------------------------------------------------
\1\ See Michael S. Barr, ``Who's in Charge of Global Finance?'',
Georgetown Journal of International Law 45, no. 4 (2014): 971-1027.
---------------------------------------------------------------------------
As with designation, global coordination--and independent
regulatory judgment--is essential to capital rules. Strong capital
rules are one key to a safer system. Before the crisis, the financial
system was woefully undercapitalized, and that the system was saved
only with a massive infusion of taxpayer-funded capital, and a wide
variety of unprecedented guarantees, liquidity provision and other
backstops by the FDIC, the Federal Reserve, and Treasury. There's
already double the amount of capital in the major U.S. firms than there
was in the lead up to the financial crisis. Globally, regulators are
developing more stringent risk-based standards and leverage caps for
all financial institutions, and tougher rules for the biggest players.
In the U.S., regulators have proposed even stronger leverage and
capital requirements for the largest U.S. firms, and other countries
are putting in place stricter approaches when warranted by their local
circumstances.
In my judgment, the local variation based on a strong minimum
standard is healthy for the system, taking into account the different
relative size of financial sectors and differing local economic
circumstances. There's been progress on the quality of capital--
focusing on common equity--and on better and more comparable measures
of the riskiness of assets, but more could be done to improve
transparency of capital requirements across different countries and to
make them stronger buffers against both asset implosions and liquidity
runs. We need to continue to insist that European capital standards and
derivatives regulations are strong--and enforced even-handedly across
the board.
The United States has taken a strong lead in pursuing global
reforms, galvanizing the G20, pushing for the creation of the global
Financial Stability Board, and pursuing strong global reforms on
capital, derivatives, resolution, and other matters.
The G20 has been driving financial reforms at a global level; the
Financial Stability Board pursues agreement among regulators; and
technical teams at the Basel Committee on Banking Supervision, the
International Organization of Securities Commission, and the
International Association of Insurance Supervisors hash out industry-
relevant reforms. While the process of reaching global agreement has at
times been quite messy, divisive, and incomplete, the last thing we
need is to hamstring global cooperation or U.S. regulation. These
mechanisms should be strengthened and improved, not ignored or
weakened.
Strong U.S. financial rules are good for the U.S. economy, American
households and businesses, and we also need a stronger, harder push to
reach global agreement on core reforms. In fact, such an approach is
essential in order to reduce the chances of another devastating global
financial crisis that crushes the U.S. economy.
Measuring Risk
The 2007-2009 financial crisis revealed the pressing need to
develop better methods to understand and manage risk in the financial
system. Since the crisis, financial regulators, scholars, and the
financial industry have turned their attention to these issues, and
made progress, but our ability to identify, monitor, and mitigate risk
in the financial system remains far behind where we need to be. This is
particularly challenging because many of the risks that are of central
concern are low probability events with unacceptably high costs for the
real economy.
Stress testing is a central and innovative risk management tool
used since the financial crisis by both regulators and practitioners.
Stress testing attempts to capture the effects of macro-shocks on the
balance sheets and activities of firms. Unlike fixed capital ratios, of
either the risk-based or leverage ratio type, stress testing seeks to
understand how macro-shocks would deplete capital. Moreover, the stress
tests are not as easy as fixed capital rules for firms to game. Despite
these advantages, stress testing remains crude and static with respect
to systemic effects, and is focused on the risks facing each individual
firm. Although the goal of stress testing is to analyze and measure
systemic risk, it is in many ways still stuck measuring the static
effects of macro-shocks on units of capital at individual firms.
While there have been significant recent advancements, our current
stress tests fail to account for the increased interconnectedness and
complexity of the financial system. The models do not yet capture the
complex network of financial transactions that connect firms and that
can spread and magnify risk in the event of a crisis. The models are
not dynamic--meaning, they do not account for market participants'
responses to stressful events. Such responses themselves may change the
nature of the events in question. Moreover, even if these more robust
models existed today, regulators do not, at least as of yet, have full
access to the financial data needed to use the models to measure
systemic risk.
We need to continue to develop new ways of thinking about how to
identify, measure, and mitigate systemic risks by drawing on methods
from other disciplines and experience from other sectors that face
systemic risks. We should explore how methods from other disciplines--
such as system analysis, agent-based modeling, machine-based learning,
behavioral finance, and data visualization and security--can be used to
improve stress testing and financial risk management practices and
regulation. We should also examine how risk is measured, monitored, and
mitigated in other sectors and contexts, such as in supply chains and
electrical grids, and in the context of climate change; how
stakeholders in these contexts make tradeoffs between stability,
efficiency, and innovation; and how lessons from these contexts should
be applied or adapted to understand risks in the financial system.
At the end of the day, no one model will be adequate to
understanding and measuring risk in the financial system. We will need
to improve stress testing, early warning, macro asset, equity, and
credit price models, and other ex ante measures of risk. We will need
to do better at crisis monitoring and response, including resolution of
failing firms during a crisis. We will also need to develop better ex
post analytics to understand the crisis that have occurred.
The Path of Reform
The Dodd-Frank Act laid a firm foundation for a more resilient
financial sector, one that works for American families, instead of
exposing us all to needless risk and harm. Since enactment, a new
Consumer Financial Protection Bureau has been built from scratch. New
rules governing derivatives have been implemented to bring trading out
of the shadows and reduce risk through central clearing, capital and
margin requirements. A resolution authority has been put in place to
deal with failing firms so we are no longer faced with the devastating
consequences of the failure of a firm like Lehman Brothers or the
untenable bailouts of firms like AIG. Regulators have the ability to
designate large firms for supervision by the Fed, so the financial
sector can no longer avoid stringent regulation just by altering their
corporate form. The largest firms have to hold a lot more equity
capital as a buffer against losses, and the Volcker Rule, heightened
prudential supervision, stress tests, and other measures are reining in
risk.
The U.S. financial system is more resilient than it was in 2008.
But there's still much work to do.
We need to keep pushing for stronger reforms of the largest, most
complex banks and other financial institutions. Stress testing and new
capital rules have dramatically increased the levels of capital at the
largest firms, but we do not yet know whether these levels are
sufficiently robust to withstand a severe financial crisis. A bank
liability tax could help further reduce incentives to take on risky
short-term debt. And shadow banking activities, repo and securities
financing transactions, and other activities need to be made safer with
strong margin and collateral rules. We need to better align manager's
incentives with financial stability, by putting banker bonuses at risk
when a firm's capital level drops below specified levels or when the
firm is hit with fines or sanctions.
More broadly, Fannie Mae and Freddie Mac remain in conservatorship
without a decision about long-term housing finance; money market mutual
funds remain susceptible to runs; certain high-frequency trading
strategies and market structure problems threaten financial stability
and undermine the fairness of our markets; and critical investor
protection authorities have gone unused.
To be clear: the financial system is safer, consumers and investors
better protected, and taxpayers more insulated, than they were in
2008--by a lot. But that is not enough. We need to stay on the path of
reform to make the financial system safer, fairer, and better harnessed
to the needs of the real economy. We need to keep pushing for a
financial system that works for us.
RESPONSES TO WRITTEN QUESTIONS OF CHAIRMAN SHELBY
FROM ROBERT DEYOUNG
Thank you for inviting me to appear before the Committee on
July 23, 2015, at the hearing on ``Measuring the Systemic
Importance of U.S. Bank Holding Companies''. It is my pleasure
to provide written answers to these additional questions.
Please note that I have added one additional question to
this list. Question (11) below is a question that was asked at
the hearing by Senator Warren, but was not directed to me.
Q.1. Enhanced prudential standards pursuant to Section 165 of
Dodd-Frank impose additional costs and burdens on financial
institutions and on the broader economy. Please identify what
you believe to be the costs attributable to the current
regulatory regime for bank holding companies above $50 billion
because of the Section 165 requirements. In your opinion, has
the Federal Reserve done an adequate analysis to determine how
these burdens affect both the banks subject to Section 165 and
the economy as a whole?
A.1. All banks with more than $50 billion in assets should
regularly perform some type of macroeconomic stress testing,
regardless of whether it is mandated by Government regulators.
Prudent risk management requires these banks to understand
their vulnerabilities to potential changes in macroeconomic
conditions.
Stress testing requires increased spending on internal
labor and/or external consultants. When these expenses result
from a bank's internal risk management practices, they cannot
be characterized as ``burden.'' However, any additional
expenses beyond these--that is, expenses incurred by the bank
to perform additional layers of testing mandated by Government
regulators--by definition constitute regulatory burden. The
expenses associated with preparing and submitting the Federal
Reserve's annual CCAR fall largely into this category.
Banks do not report these expenses publicly. However, in a
May 18, 2015, American Banker article (``Banks Keep Mum About
Stress-Test Costs, Clouding Reg Debate''), Chris Cumming
reports that Wells Fargo allocated 128,000 labor hours to the
CCAR task in the fourth quarter of 2014. Assuming a relatively
low figure for salaries and benefits of $30 per hour, this
amounts to $3.84 million in expenses for one quarter, or just
over $15 million on an annual basis. This is a very rough
estimate of the CCAR burden. It might be too high (e.g., the
fourth quarter may have been peak time for the CCAR exercise)
or too low (e.g., it does not include expenditures on external
consulting, nor the lost output from diverting these workers
from other tasks). In any case, this rough estimate
demonstrates that the regulatory burden imposed on banks by the
CCAR is nontrivial.
It is important to note that the costs of complying with
CCAR cannot simply be scaled up or down based on a bank's size,
because much of the CCAR exercise entails fixed costs. For
example, if a bank is only one-tenth the size of Wells Fargo
($160 billion in assets, versus $1.6 billion for Wells), the
burden associated with CCAR will be substantially more than 10
percent of the burden on Wells Fargo.
Q.2. Is it possible that a very large bank could fail without
causing widespread damage to the financial system? Please
explain.
A.2. This is surely possible. If the FDIC is allowed to
exercise its Orderly Liquidation Authority (OLA), a large
insolvent bank will be able to continue its operations. That
is, the bank will be able to provide payments services for its
depositors, make its insured depositors fully liquid, fulfill
all of the credit commitments it has made to its borrowers,
honor all of its short-term credit market contracts (e.g.,
commercial paper, Treasury repos, purchased Fed funds), and
honor all of its derivatives counterparty obligations.
Under this scenario, there may be short-run spikes in
financial markets, but these will be temporary and will
dissipate as market participants observe that the bank is
honoring all of its contracts and obligations. There should not
be any widespread damage to financial markets. Indeed, as the
FDIC establishes its reputation by exercising its OLA authority
on multiple occasions, even these temporary disruptions should
lessen.
This is not to say that losses will not be taken. The
bank's equity holders will take a 100 percent loss. Some or
perhaps all of the bank's bondholders will take partial or full
losses. And some of the bank's uninsured depositors may take
partial losses. It is likely that the FDIC will also take
losses in the short run, as it injects the funds necessary to
recapitalize the bridge bank, as well as to offset any ongoing
operational losses of the bridge bank and its subsidiaries.
However, in the long run, the FDIC should be able to recover
these losses with increased charges to the banking industry.
Q.3. Is it possible that regulating all banks with $50 billion
in assets as systemically important might actually encourage
systemic risk rather than reduce it? Please explain.
A.3. This is a novel theory. It is based on the presumption
that any bank declared to be systemically important (a SIFI)
will change its risk-taking behavior and begin to act like it
is too big to fail (TBTF). But this is a false presumption. If
a bank will not be bailed out, it cannot be a TBTF bank, and
hence it will not take the additional risks typically
associated with TBTF banks. Indeed, the FDIC has already
established its ability to resolve banks in the general size
range of $50 billion (e.g., IndyMac) and beyond (e.g.,
Washington Mutual). For banking companies that are larger or
more complex than these examples, the FDIC can now use its OLA
powers of seizure and resolution. So if the FDIC is allowed to
exercise its new resolution authority, then SIFI designation by
itself will not encourage banks to take or create systemic
risks.
Q.4. At the hearing, you did not get an opportunity to respond
to certain questions. I would be interested in your response to
the following questions:
Is it a good thing that large banks have more capital and
less leverage?
A.4. The increased equity capital requirements in Basel III for
the most part represent an improvement. The inclusion of a
plain vanilla minimum leverage ratio for all banks (which the
U.S. has required for many years) was an important step. Even
more important is the adoption of procyclical capital
minimums--a macroprudential tool that will help reduce the
buildup of excess bank credit during economic expansions, and
will help prevent harmful reductions in bank credit during
economic recessions.
Q.5. Is it a good thing for large banks to have more liquidity
than they did before the crisis?
A.5. We have very little understanding of how mandatory
liquidity minimums, such as Basel III's LCR and the NSFR, will
influence bank risk taking in general or bank insolvency in
specific.
By itself, establishing higher capital minimums should
prevent banks experiencing liquidity problems from failing. A
clearly solvent bank can always access short run liquidity from
the interbank market or from its central bank. This allows the
bank to honor its short-term financial obligations, thus
eliminating fire sales that drive down asset prices and
investor flight from short-term credit markets.
My fear is that the main effect of regulatory liquidity
mandates, when placed on top of higher regulatory capital
mandates, will be to reduce the creation of bank credit.
Q.6. Comptroller of the Currency Tom Curry has made it his
mission, in part, to install more enhanced prestige and stature
with higher compensation for Chief Risk Officers at medium-
sized and large banking institutions. Is that a good idea?
A.6. It is a great idea, but it is unlikely to be effective. To
the extent that Chief Risk Officers at large banking companies
have too little prestige and stature, this is because of faulty
corporate cultures. Regulators have a poor track record of
affecting changes to corporate cultures.
Q.7. Should large banks have strong risk management structures
in place?
A.7. Of course they should. For example, see my answer to
question (1) above, regarding internal stress testing. But the
most effective way to encourage strong risk management at banks
is to credibly ensure that failed banks are never bailed out.
Q.8. Should large banks be able to detail how they could fail
safely?
A.8. Again, we have very little understanding of how a
resolution plan or ``living will'' will make it easier for
either the FDIC or a bankruptcy court to efficiently resolve a
failed complex banking company. These efforts may end up being
helpful . . . or they may end up being 100 percent burden,
imposed partially on bank shareholders and partially on
taxpayers (who are paying for this new regulatory effort). All
we really have at this point is a hope for the former.
Q.9. Is it appropriate for large banks to conduct regular
stress tests?
A.9. Yes. See my answer to question (1) above.
Q.10. Are capital structure, riskiness, complexity, financial
activities, size, and other risk-related factors appropriate
criteria to use as a basis for crafting prudential standards?
A.10. Prudential standards start with capital structure. Banks
should hold enough capital to (i) absorb 100 percent of the
losses that a bank expects to incur under a historical worst
case scenario and (ii) allow Government supervisors to observe
large losses occurring in real time, well before the bank
actually becomes insolvent.
Banks' riskiness and banks' financial activities stem from
banks' business models. They are of secondary importance for
prudential regulation; regulators should set bank-specific
capital minimums high enough to reflect these risks. Moreover,
regulatory interference with banks' business models has the
potential to cause more harm than good.
Complexity for complexity's sake is not desirable and
should be discouraged. Complexity that arises naturally from a
bank's business model should be allowed. Complexity that arises
due to compliance with Government regulation (e.g., the
multibank holding company structures necessary to legally
operate an interstate bank prior to 1996) calls for a
reexamination of that regulation.
Bank size should influence prudential regulation only for
banks that are too large to fail. This would also hold for
banks that are too complex to fail. In both of these cases,
systemic risk is the underlying worry. But given the FDIC's new
OLA powers, I believe that neither of these cases will be
operative going forward, so long as the FDIC is permitted to
seize and resolve large and complex insolvent banking
companies.
Q.11. At the hearing, I did not have the opportunity to answer
the following question, which Senator Warren directed to just
one of the other panel members. I paraphrase: ``Could the
simultaneous failure of multiple banks, each of which holds
assets of $100 billion, cause systemic risk?''
A.11. Again, I point out that the FDIC has already established
its ability to resolve banks in this general size range
(IndyMac, Washington Mutual) during quite difficult
macroeconomic conditions. The FDIC's ability to perform large
bank resolutions has only been strengthened by its new OLA
powers.
There is nothing about ``simultaneous'' large bank failures
that changes this assessment. If an OLA-based seizure and
resolution works as designed, the failed bank will not default
on any contracts or obligations with its insured depositors,
with its line of credit customers, with its counterparties in
credit markets, or with its counterparties in derivatives
markets. As it becomes clear that all of these contracts are
being honored, any disruptions in credit markets (e.g.,
commercial paper) and asset markets (e.g., loan-backed
securities) will be minimal.
One potentially limiting factor is the length of time that
the FDIC needs operate these banks prior to selling their
assets (in whole or in parts) back into private hands. If these
multiple bank failures occur during a recession, the FDIC will
likely need to operate these banks for several years before
they are stabilized. In this case, many commentators will argue
that ``we have nationalized these banks.'' This is obviously an
incorrect statement, as the goal is stabilization, not
ownership. But such an argument could create political pressure
for the FDIC to sell the insolvent banks too quickly--or
perhaps even create pressure to bail out these banks rather
than invoke OLA powers.
Thank you again for soliciting my opinion on these issues
of importance to the U.S. banking industry. Please do not
hesitate to contact me regarding clarification or additional
questions.