Home ] content ] jan_tetstone_poetry_diary ] rainbow_site ] heartsongs_art_gallery ] patriotic_poetry ] dead_bankers ] terri the real truth ] Fairy Land ] poetry by year ]
Text: S.Hrg. 115-106 FOSTERING ECONOMIC GROWTH: MIDSIZED, REGIONAL, AND LARGE INSTITUTION PERSPECTIVE
Text available as: PDF (8MB)
[Senate Hearing 115-106]
[From the U.S. Government Publishing Office]




S. Hrg. 115-106


FOSTERING ECONOMIC GROWTH: MIDSIZED, REGIONAL, AND LARGE INSTITUTION
PERSPECTIVE

=======================================================================

HEARING

before the

COMMITTEE ON
BANKING,HOUSING,AND URBAN AFFAIRS
UNITED STATES SENATE

ONE HUNDRED FIFTEENTH CONGRESS

FIRST SESSION

ON

EXAMINING THE CURRENT STATE OF MIDSIZED, REGIONAL, AND LARGE
INSTITUTIONS, INCLUDING THEIR REGULATORY REQUIREMENTS, IMPACT ON
CLIENTS, AND THEIR ROLE IN PROMOTING ECONOMIC GROWTH

__________

JUNE 15, 2017

__________

Printed for the use of the Committee on Banking, Housing, and Urban
Affairs
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]


Available at: http: //www.fdsys.gov/
______

U.S. GOVERNMENT PUBLISHING OFFICE

27-425 PDF WASHINGTON : 2017
-----------------------------------------------------------------------
For sale by the Superintendent of Documents, U.S. Government Publishing
Office Internet: bookstore.gpo.gov Phone: toll free (866) 512-1800;
DC area (202) 512-1800 Fax: (202) 512-2104 Mail: Stop IDCC,
Washington, DC 20402-0001

COMMITTEE ON BANKING, HOUSING, AND URBAN AFFAIRS

MIKE CRAPO, Idaho, Chairman

RICHARD C. SHELBY, Alabama SHERROD BROWN, Ohio
BOB CORKER, Tennessee JACK REED, Rhode Island
PATRICK J. TOOMEY, Pennsylvania ROBERT MENENDEZ, New Jersey
DEAN HELLER, Nevada JON TESTER, Montana
TIM SCOTT, South Carolina MARK R. WARNER, Virginia
BEN SASSE, Nebraska ELIZABETH WARREN, Massachusetts
TOM COTTON, Arkansas HEIDI HEITKAMP, North Dakota
MIKE ROUNDS, South Dakota JOE DONNELLY, Indiana
DAVID PERDUE, Georgia BRIAN SCHATZ, Hawaii
THOM TILLIS, North Carolina CHRIS VAN HOLLEN, Maryland
JOHN KENNEDY, Louisiana CATHERINE CORTEZ MASTO, Nevada

Gregg Richard, Staff Director

Mark Powden, Democratic Staff Director

Elad Roisman, Chief Counsel

Jared Sawyer, Senior Counsel

Travis Hill, Senior Counsel

Graham Steele, Democratic Chief Counsel

Laura Swanson, Democratic Deputy Staff Director

Elisha Tuku, Democratic Senior Counsel

Dawn Ratliff, Chief Clerk

Cameron Ricker, Hearing Clerk

Shelvin Simmons, IT Director

Jim Crowell, Editor

(ii)



C O N T E N T S

----------

THURSDAY, JUNE 15, 2017

Page

Opening statement of Chairman Crapo.............................. 1

Opening statements, comments, or prepared statements of:
Senator Brown................................................ 2

WITNESSES

Harris H. Simmons, Chairman and Chief Executive Officer, Zions
Bancorporation, on behalf of the Regional Bank Coalition....... 4
Prepared statement........................................... 27
Greg Baer, President, The Clearing House Association............. 6
Prepared statement........................................... 32
Robert R. Hill, Jr., Chief Executive Officer, South State
Corporation, on behalf of Mid-Size Bank Coalition of America... 7
Prepared statement........................................... 51
Responses to written questions of:
Senator Brown............................................ 96
Senator Sasse............................................ 99
Senator Tillis........................................... 100
Saule T. Omarova, Professor of Law, Cornell University........... 9
Prepared statement........................................... 52
Responses to written questions of:
Senator Brown............................................ 102
Senator Sasse............................................ 104
Senator Reed............................................. 108

Additional Material Supplied for the Record

Statement submitted by the Independent Community Bankers of
America........................................................ 111
Statement submitted by the American Bankers Association.......... 116
Statement submitted by Fifth Third Bank.......................... 123
Statement submitted by the Institute of International Bankers.... 127
Statement submitted by Texas Capital Bank........................ 130
GSIB scores...................................................... 133

(iii)


FOSTERING ECONOMIC GROWTH: MIDSIZED, REGIONAL, AND LARGE INSTITUTION
PERSPECTIVE

----------


THURSDAY, JUNE 15, 2017

U.S. Senate,
Committee on Banking, Housing, and Urban Affairs,
Washington, DC.
The Committee met at 9:49 a.m., in room SD-538, Dirksen
Senate Office Building, Hon. Mike Crapo, Chairman of the
Committee, presiding.

OPENING STATEMENT OF CHAIRMAN MIKE CRAPO

Chairman Crapo. This hearing will come to order.
Last week, we received testimony on the role financial
institutions play in fostering economic growth in local
communities. The hearing focused on community lenders, and we
heard about the need for additional tailoring, the increasing
cost of compliance, and common-sense reforms to rules such as
QM, TRID, HMDA, and Volcker.
Today we will hear about the regulatory framework that
midsize banks, regional banks, and larger financial
institutions face. Midsize and regional banks are often
subjected to costly post-crisis rules designed for the most
systemically important banks. Many of these rules are applied
based on asset thresholds that do not reflect the underlying
systemic risk of financial institutions.
While the size of a bank is one factor in measuring
systemic importance, there are many other aspects of an
institution that are relevant to how difficult the company
would be to resolve and how consequential its distress or
failure would be to financial markets.
The result is a regulatory regime that is insufficiently
tailored for many of the firms subject to it, for example,
stress testing. Former Fed Governor Tarullo is among those who
have stated that the $10 billion threshold for company-run
stress tests is too low. Additionally, CCAR is a very costly,
time-consuming process that is overly burdensome, especially
for noncomplex regional banks. Another example is the Volcker
rule, which has proven far too complicated to implement and
incredibly difficult to comply with.
One of my key priorities this Congress is passing
legislation on a bipartisan basis to improve the bank
regulatory framework and stimulate economic growth. In March,
Senator Brown and I began our process to receive and consider
proposals to help foster economic growth, and I appreciate all
the valuable insights and recommendations that we have
received. Also in March, the Federal banking agencies issued
their EGRPRA report to Congress with their recommendations.
And earlier this week, the Treasury Department issued the
first of its reports examining how best to improve our
regulatory framework. The report focused on banks and credit
unions and provided a substantial number of helpful regulatory
and legislative suggestions corresponding to the President's
Executive Order on ``Core Principles for Regulating the
Financial System''.
I commend Secretary Mnuchin and his staff at Treasury for
all the work that went into this report and for the thoughtful
recommendations they have provided.
I am particularly encouraged by a number of specific
recommendations for midsize and regional banks, including
changing the $50 billion SIFI threshold; exempting midsize
banks from company-run stress tests; exempting banks without
significant trading activity from the proprietary trading
prohibition of the Volcker rule; and improving the transparency
and process of CCAR and living wills.
With the hundreds of recommendations that we have received
through our economic growth submission process, the testimony
we are receiving at these hearings, the EGRPRA report, and this
Treasury report, the Committee has no shortage of ideas to
consider as we work to improve our regulatory framework.
As this process continues, I look forward to working with
all Members of the Committee from both sides of the aisle to
bring strong, robust bipartisan legislation forward.
Senator Brown.

STATEMENT OF SENATOR SHERROD BROWN

Senator Brown. Thank you, Mr. Chairman. Thanks for holding
today's hearing. Thank you to the four witnesses for joining us
today.
Based on the current Wall Street reform debate, the
activity in the House and the report put out by Treasury
earlier this week, I am concerned that some seem to have
forgotten that we even had a financial crisis a decade ago. I
can assure you that families in my home State still remember
and continue to live with the consequences of the crisis. I
have told this Committee before, but my Zip code in Cleveland
where my wife and I live, 44105, in the year 2007, the first
half of that year, had more foreclosures than any Zip code in
the United States, and I see what a foreclosure on a home means
to my neighbors.
That crisis of a decade ago, when the unemployment rate for
our Nation reached more than 10.5 percent, we had 14
consecutive years of increasing foreclosures, some 400,000
homes lost in 5 years at the height of the housing crisis in my
State. That is to say nothing of the psychological damage
caused by lost jobs and children being forced to move away from
their friends and families.
Just one example, 17 percent of Ohio homeowners still owe
more on their mortgages than their home is worth--the second
highest rate in the Nation behind the States of two Members of
this Committee, Senator Cortez Masto and Senator Heller.
With these experiences fresh in our mind, Congress passed
Wall Street reform, including some of the most sweeping changes
to financial regulation in 70 years. Wall Street and its allies
are attacking rules like living wills and orderly liquidation
that are meant to ensure that a $1 trillion megabank can fail
without bringing down the economy with it. At the same time,
they attack the capital rules that are meant to reduce the
likelihood that banks will fail in the first place.
Let me be clear. Proposals to weaken oversight of the
biggest banks have no place in this Committee's process. Wall
Street banks caused a financial crisis that cost our economy up
to $14 trillion and took $160 billion in taxpayer bailouts.
They made the market for the late and unlamented predatory
lenders that have left more than one-fifth of Cleveland
homeowners still underwater. One-fifth. Letting them run wild
again will not help economic growth. It will just put our
economy at risk once again.
Having said that, I am optimistic there is room for
agreement on a modified regime for overseeing regional banks.
This will be the fifth hearing dedicated to the issue of these
enhanced prudential standards since July of 2014. The last 3
years, I have been encouraged by steps that the agencies have
taken to better tailor standards like stress tests and living
wills. We have heard that these two rules plus liquidity
requirements may impose the most burdens with the least amount
of benefits to financial stability when they are applied to
regional banks. I think we all understand that.
We have heard from both midsize banks and their regulators,
as the Chairman cited, that changes should be considered to the
Dodd-Frank-required stress tests. I look forward to working
with the Chairman and our colleagues to explore what might make
the oversight regime work better for both midsize and regional
banks as long as financial stability and safety and soundness
and consumer protections are not compromised.
Let me close with a different topic, if I could, Mr.
Chairman, in the last couple of minutes. Wall Street reforms'
opponents accuse the law of being too partisan despite the fact
that it received Republican votes in both the House and Senate
and that it included 15 Republican-sponsored floor amendments.
They say it was not well conceived even though we held more
than 30 Committee hearings and Chairman Dodd spent months
discussing the bill with Republicans on this Committee, some of
them still on this Committee. The bill spent more than a month
on the Senate floor.
Contrast that, Mr. Chairman, and I want to point out that
right now a small group of Republican Senators--maybe a dozen,
we read; we do not really know--is crafting a health care bill
behind closed doors. They are doing it with no participation
from Democrats. As the case with Dodd-Frank, contrast how
Democrats passed the Affordable Care Act a number of years ago.
It took more than a year, dozens and dozens of hearings in my
Committee alone, then the Health, Education, Labor, and Pension
Committee. We accepted 150 Republican amendments, open process,
lots of debate. But look at what has happened now with the
Affordable Care Act.
The Chairman of the Finance Committee--Senator Crapo and I
both sit on that Committee, as does Senator Scott. That
Committee has no plans to hold even a single hearing on the
bill. Both Dodd-Frank and the Affordable Care Act were the
product of painstaking legislative work. We were doing our job
bipartisanly. We put those together. It put money back in the
pockets of American families through lower health care costs
and lower credit card and mortgage fees. We owe it to these
families on that issue and other issues to have an open and
honest debate about the Republican health care bill. There is
no sign that Senator McConnell is going to do that. No sign at
all.
Nothing could be more important to economic growth than
safeguarding the health and the financial well-being of working
families. This Committee can play a role in that. I am hopeful
that we will.
Thank you.
Chairman Crapo. Thank you.
At this point we will move to the testimony of our
witnesses. Before we do so, however, I want to alert both the
witnesses and the Members that we have three votes scheduled at
11 o'clock, so we are going to need to wrap this hearing up by
11 o'clock, which means to my colleagues the 5-minute rule
really must be honored, and I will honor it strictly with you.
And to our witnesses, sometimes a question comes in right in
the last few seconds of the 5 minutes. Please keep your answers
brief to that so we can move on and let every Senator have an
opportunity to ask questions.
As each of you know, you have been allocated 5 minutes for
your oral remarks, and we welcome them. You will also have
plenty of opportunities to enhance your testimony in response
to questions.
Finally, I wanted to indicate to everyone, I have to
testify in the Judiciary Committee in just a few minutes, so I
will step out. But I will be back, and I look forward to
reviewing and listening to all of your comments.
With that, first we will receive testimony from Mr. Harris
Simmons, chief executive officer and chairman of Zions
Bancorporation, on behalf of the Regional Bank Coalition.
Following him we will hear from Mr. Greg Baer, president of
The Clearing House Association.
Then we will hear from Mr. Robert Hill, chief executive
officer of South State Corporation, on behalf of Mid-Size Bank
Coalition of America.
And, finally, we will hear from Ms. Saule Omarova--did I
get that right?
Ms. Omarova. Yes, you did.
Chairman Crapo. Thank you. Professor of law at Cornell
University.
Mr. Simmons, you may proceed.

STATEMENT OF HARRIS H. SIMMONS, CHAIRMAN AND CHIEF EXECUTIVE
OFFICER, ZIONS BANCORPORATION, ON BEHALF OF THE REGIONAL BANK
COALITION

Mr. Simmons. Thank you very much, Mr. Chairman. I
appreciate the opportunity to present some views to the
Committee this morning. I am Harris Simmons. I am the chairman
and CEO of Zions Bancorporation. We are a $65 billion in assets
regional bank headquartered in Salt Lake City. We operate
across the Western United States, including in the Chairman's
home State of Idaho. We particularly focus on small, medium,
and mid-market kinds of businesses. We do a lot of commercial
lending, the kind of lending that we believe creates a lot of
jobs and supports a great deal of economic activity in the
Western United States. We serve these businesses, their
employees, their owners, and a number of municipalities across
the West, many in rural locations, particularly in the
Intermountain West.
For the past several years, Zions Bancorporation has been
the smallest of the systemically important financial
institutions as defined by the Dodd-Frank Act. Sometimes we
joke about ourselves as being an ``itty-bitty SIFI.'' And we
have felt, I think, perhaps disproportionately, the brunt of
the burden of complying with the enhanced prudential standards
and other requirements of Dodd-Frank that are applicable to the
SIFI group. But that is the common lot of the regional banks
generally.
It has forced us to hold capital that is at the north end
of our peer group. It has retarded our growth and has hampered
the ability of relationship managers, those on the front lines,
to serve customers in the manner that they used to be able to
do.
In particular, the Comprehensive Capital Analysis and
Review process, or CCAR as it is known, has been costly, it has
been frustrating, and perhaps most of all, it has been
incredibly opaque, and challenging to understand what the
evolving requirements have been and what kind of standard is
required of us as we think about the economics of our own
business. It creates uncertainty, which in turn stifles
planning and prudent risk taking.
The liquidity coverage ratio is another feature that is
applicable to the SIFI group. In the case of banks under $250
billion but over $50 billion in size, a modified liquidity
coverage ratio will become a constraint. I think it has not
been particularly constraining. Over the past few years, the
industry has been awash in liquidity. But it will become a
constraint, and it will become a constraint at the worst
possible time. And at a time when liquidity is at a premium, we
will find ourselves as an industry withdrawing, and that will
not be good for the economy.
Finally, the banking regulatory apparatus has become a Rube
Goldberg contraption with overlapping regulators, redundant
regulations, such as the various capital regimes, scores of
compliance trip wires that cumulatively are overly expensive,
sometimes conflicting in their objectives, and they consume an
enormous amount of management and board time and resources.
They also have produced a lot of growth in the shadow
banking system outside of the regulated banking system. They
have, I believe, slowed the housing recovery and resulted in
slower capital formation and small business credit
availability.
The Treasury Department's outline for reform provides, I
think, a great blueprint for beginning to tackle many of these
issues, and we look forward to working with other financial
institutions and with members of Congress and Members of this
Committee in trying to come with solutions that will leave the
industry safer and sounder but that will foster economic
growth.
Thank you very much.
Senator Scott [presiding]. Thank you, Mr. Simmons.
Mr. Baer, you may proceed.

STATEMENT OF GREG BAER, PRESIDENT, THE CLEARING HOUSE
ASSOCIATION

Mr. Baer. Thank you, Senator Scott, Ranking Member Brown. I
am pleased to appear before the Committee today to discuss how
regulatory reform could stimulate economic growth.
I should emphasize at the outset that if the goal is
economic growth, it cannot be achieved while excluding large
and regional banks from that effort. Community bank relief is
warranted, but as the Treasury Department noted in its report
this week, community banks hold only 13 percent of U.S. banking
assets. The 25 banks that own The Clearing House fund more than
40 percent of the Nation's business loans held by banks and
more than 75 percent of loans to households. Large banks
originated 54 percent of small business loans in 2015 by dollar
amount and about 86 percent by number.
The starting point for any review of the American banking
system is one that is extremely resilient. Regulatory changes
have helped to increase the quality and quantity of capital.
Post-crisis, the aggregate Tier 1 common equity ratio for The
Clearing House's 25 banks has nearly tripled and increased from
$331 billion to over $1 trillion in capital.
Similarly, U.S. banks now hold unprecedented amounts of
cash and cash equivalents to protect against a run. Today such
assets compose nearly a quarter of U.S. large bank balance
sheets.
While much has been gained in fortifying the Nation's
largest banks, some overly stringent capital liquidity, and
other rules have diminished their ability to lend and
intermediate in financial markets. For example, Fed data show
that approval rates for small businesses were just 45 percent
at large banks subject to heightened capital requirements for
such loans, but 77 and 60 percent, respectively, at CDFIs and
small banks.
Similarly, with respect to mortgage lending, our research
demonstrates that the Federal Reserve CCAR test is imposing
dramatically higher capital requirements on residential
mortgage loans than bank internal stress test models or the
standardized approach to capital developed by the Basel
Committee. As a result, over the past 6 years, residential real
estate loans declined 0.5 percent per year at banks subject to
the CCAR stress test while they have risen 4.0 percent at banks
not subject to the test.
Capital markets have also been affected. Indeed, post-
crisis regulation by banking regulators has affected securities
markets more than regulation by securities regulators. Bank
regulations have made it significantly more expensive for
broker-dealers affiliated with banks, which now include all the
largest broker-dealers, to hold, fund, and hedge securities
positions. And the Volcker rulemakes the holding of market-
making inventory a potential legal violation.
The greatest impact has been felt by smaller companies.
Issuance of corporate bonds by small and midsize firms has
fallen over the past few years even as issuance by larger firms
has increased.
Supervision is also playing as large a role as regulation
in constraining credit as examiners increasingly dictate how
bank resources are to be allocated. For example, leveraged
lending is an important type of financing for growing
companies, which carry a lot of debt. Based on no empirical
evidence, the Federal banking agencies have issued guidance
setting arbitrary limits on such lending. Some of the guidance
makes little sense. For example, regulators require banks, in
evaluating whether a company is leveraged, to assume that all
its lines of credit are drawn, which is akin to lowering a
consumer's credit score because their credit line has been
increased for good payment history. As a result, some Fortune
500 companies with investment grade debt are now deemed by the
regulators to be highly leveraged and, therefore, risky.
More broadly, we believe that bank supervision has lost its
way post-crisis and requires a comprehensive reexamination.
Even as banks have dramatically improved their financial
condition, supervisors have transformed supervisory grades from
a measurement of financial condition to a measurement of
compliance. They have created unwritten rules that lead
isolated compliance problems serving as a barrier to expansion,
in some cases for years, and particularly for midsize and
regional banks.
Another result is simply a massive cost, which must be
passed along to consumers, as described in a recent CEO letter
to shareholders: ``At M&T, our own estimated cost of complying
with regulation has increased from $90 million in 2010 to $440
million in 2016, representing nearly 15 percent of our total
operating expenses. During 2016 alone, M&T faced 27 different
examinations from six regulatory agencies. Examinations were
ongoing during 50 of the 52 weeks of the year.''
Much of this burden comes as rules that make sense for
large complex firms are applied to firms that present few of
the same risks. Certainly the answer to a bad rule is not to
apply it to fewer people, and there are many rules that fit
that description. But there are other rules that make sense for
some but not all. My written testimony sets forth numerous
rules in both categories and ideas for how they could be
reformed.
There is no need for fundamental changes to post-crisis
regulation, but there is certainly room for improvement, and
particularly if the goal is stronger economic growth.
Thank you.
Senator Scott. Thank you, Mr. Baer.
Now, from the great State of South Carolina, Mr. Hill.

STATEMENT OF ROBERT R. HILL, JR., CHIEF EXECUTIVE OFFICER,
SOUTH STATE CORPORATION, ON BEHALF OF MID-SIZE BANK COALITION
OF AMERICA

Mr. Hill. Chairman Crapo, Ranking Member Brown, Senator
Scott, and Members of the Committee, I am Robert Hill, CEO of
South State Corporation. I am grateful that your leadership has
provided a nonpartisan platform to hear from bankers like
myself.
Today I represent my company, South State Bank, and also
the Mid-Size Bank Coalition, which is the voice of 78 midsize
banks in the U.S. with headquarters in 29 States. Our member
banks are primarily between $10 billion and $50 billion in
assets and serve customers and communities through more than
10,000 branches in all 50 States, the District of Columbia, and
three U.S. territories. Midsize banks most often are the
largest local bank serving their community, many for more than
a century.
South State specifically was founded in 1933 on the heels
of the Great Depression in rural South Carolina. Today we serve
communities both large and small in North Carolina, South
Carolina, and in Georgia.
In January of this year, our bank crossed the $10 billion
in asset threshold, and I have had the opportunity to see
firsthand how this threshold is having a negative impact on
economic growth.
At South State, not unlike my peers in the coalition, we
operate a very simple business model. We offer depository
services, and we lend money in local communities. This is very
similar to the business model we operated when we were a very
small community bank. We have stable deposit funding from our
customers, revenues that are driven from traditional banking
services, and that are well understood by both our regulators
and our management team. And we have never done any proprietary
trading.
South State and other midsize banks have prudent business
models that contribute to economic growth and support financial
stability. Our company never lost money during the financial
crisis. We never did subprime lending. We never stopped lending
to our customers during the crisis.
As the largest banks in many of our States, midsize banks
can have a significant impact on our communities, but some
banks are choosing not to cross this huge threshold or cross it
and shift their focus from investing in the businesses that
they have to investing in the preparation for large bank
regulation.
Under Dodd-Frank, crossing the $10 billion in asset
threshold has had very harsh implications for midsize banks.
This is happening to a segment of our industry not based on
risk but based purely on asset size. And I assure you that
adding $1 in incremental assets as we cross $10 billion did
little to change the risk profile of our company.
While we value many parts of Dodd-Frank and we like the way
our industry has been strengthened since the crisis, I have yet
to see the value to the public in any appreciable way of the
arbitrary $10 billion threshold. These requirements drain
resources of midsize banks, divert dollars from investment in
our customers to investment in large bank regulation. For
example, South State was impacted by over $20 million per year,
a significant sum for a bank our size. What impacts did this
have on our local communities? For us, that equates to 300
jobs. Approximately 10 percent of our branches were closed, and
even more jobs were diverted from lending to regulatory
compliance.
As banks that support Main Street and not Wall Street, we
need our communities and our communities deserve regulation
that encourages prudent behavior and also protects our
customer. But we also need to have common-sense regulation that
does not impose burdens or slow economic growth in our
communities. In our view, we must move away from the Dodd-Frank
$10 billion regulatory threshold.
Again, I appreciate the opportunity the Committee has given
the Mid-Size Bank Coalition and myself to express these views
today.
Senator Scott. Thank you, Mr. Hill.
Professor Omarova.

STATEMENT OF SAULE T. OMAROVA, PROFESSOR OF LAW, CORNELL
UNIVERSITY

Ms. Omarova. Senators, thank you for the opportunity to
testify on this important issue.
We are all here today because 9 years after the worst
financial crisis in generations, the banking industry is now
waging a massive campaign to roll back the Dodd-Frank Act and
the entire regime of post-crisis systemic risk regulation. The
banks claim that regulation is what directly prevents them from
lending to small businesses and struggling families and, thus,
prevents them from fostering America's economic growth. You
should take these claims with extreme skepticism.
First, it is important to understand what the banking
industry really means by growth. What America needs is real
economic growth--sustainable, socially inclusive, long-term
growth of the real, nonfinancial, sector of the American
economy. We need to restore the Nation's eroding industrial
base, rebuild and modernize our infrastructure, and create
sustainable, well-paying jobs. What we do not need is to have
another stock market or real estate bubble fed by cheap credit
and speculation in secondary markets.
Yet if Congress, you, deregulate big banks, that is
precisely what will grow. Big Wall Street banks derive the bulk
of their profits not from small business lending but from
massive high-risk trading and dealing in secondary markets.
They feed speculation, not real economic growth. That
speculation is precisely what caused the latest financial
crisis, and it will inevitably cause another one on your watch.
In a strategically savvy move, big banks are aligning
themselves with the smaller and midsized and regional banks as
far more sympathetic petitioners, almost the Jimmy Stewart bank
types. Almost but not quite. Even the tiniest among them have
more than $10 billion in assets, and the bigger ones, well over
$300 and sometimes $400 billion in assets. True, they are
smaller and less dependent on speculative trading than Wall
Street megabanks, and perhaps they do deserve a lighter
regulatory load. But if trying to help these smaller banks you
grant their request for a massive regulatory rollback, the
principal beneficiaries of the deregulation will be Wall Street
megabanks. Deregulation will reduce smaller banks' compliance
costs, but it will also enable megabanks to expand the high-
risk speculative trading, which is at the core of financial
instability and crisis.
To guard against that, you should require banks to provide
concrete evidence that they will actually use their savings
from specific deregulatory measures primarily, if not
exclusively, to increase lending to productive economic
enterprise. At the very least, banks should give you the exact
amounts of prudent, productive loans that they were ready to
make but were forced to decline solely because they did not
have enough money left after paying for regulatory compliance.
I doubt that such evidence exists. Yet there is plenty of
evidence that all banks, regardless of their size, generate
healthy net profits, in fact, to the tune of $175 billion only
in 2016 and choose to return the bulk of those profits to their
shareholders through dividend payments and share buybacks. Only
last year, insured banks paid out $103 billion in cash
dividends, an amount second only to the record high of $110
billion in bank dividends paid in 2007, the last pre-crisis
bubble year.
High dividends increase banks' stock price and management
bonuses, so that is where most of their money seems to go, not
to lending or regulatory compliance. That is a far cry from
rebuilding America's industrial base or helping struggling
American families to get out of poverty.
Unless banks put their own money where their very loud and
very well paid mouths are, you should not read their claims
about fostering growth as anything more than convenient
rhetoric. Indeed, it is much more likely that big banks'
massive push for deregulation is driven by their desire to
generate high speculative trading profits, increase their
executives' bonuses, and return more dividend cash to their
shareholders. Right now, all of these things are significantly
limited by the Dodd-Frank regime of enhanced prudential
supervision, including heightened capital ratios, supervisory
stress testing, and living will requirements applicable to
large systemically important financial institutions, or SIFIs.
No wonder that the banking industry attacks the key
elements of this post-crisis regulatory regime as supposedly
arbitrary and insufficiently ``tailored'' to their unique
circumstances. For example, banks are demanding that SIFI
designation is conducted on a strict case-by-case basis with
every bank getting the same tailored process as MetLife got,
and we all know how perfectly efficient and problem-free that
process turned out to be. If you allow this to happen, it will
effectively kill the entire regime of enhanced SIFI oversight.
The same goes for banks' demands to force the Federal
Reserve to publish its stress test scenarios for public notice
and comment and to restrict the Fed's ability to conduct and
use its own test models. Yes, that will make stress tests fully
transparent, just like giving the students exam questions
before the exam will make that exam transparent. And it will
also make it absolutely useless for its intended purposes. I
know better than to give my students such a special gift, and
you should know better than to do the same for the banks.
In conclusion, I urge you to keep focus not on what banks
want for the sake of their own profitability, but on what the
American economy and the American people need: not another
speculative frenzy but sustainable, employment-generating
growth of the real economy. Financial deregulation will hinder,
not foster, such growth.
Thank you.
Senator Scott. Thank you, Professor.
And I will just remind all of us that we have votes at 11
o'clock, and our goal is to be out by 11 a.m., so keeping our
questions to 5 minutes would be very helpful. I will start off
our questions.
Mr. Hill, thank you for being here today. Certainly it is
always good to have a homegrown South Carolina product like
yourself and your company do very well, and thank you for being
here to testify before us today.
I think it is critical that the Committee hears from voices
representing all parts of our country. It ensures we get a
holistic perspective on how to grow our economy for everyone.
That is why I am so glad that Mr. Hill is testifying today. His
company, South State Bank, is the largest financial institution
headquartered in South Carolina. It has less than $12 billion
in assets, but States like mine rely heavily on midsize banks
and regional banks to provide small business loans, mortgages,
and consumer financial services.
The Dodd-Frank Act created a lot of hoops for companies
like South State to jump through. Much of the added regulatory
burden is triggered by specific asset thresholds. It seems to
me that if you tell someone that they will get hammered by the
Federal Government if they hit XYZ number, everyone is going to
do all that they can to avoid hitting that number. The
collateral damage is to economic growth.
Mr. Hill, can you speak to the distortions in the market
and business behavior as institutions approach these
thresholds?
Mr. Hill. Yes, Senator Scott. First, I want to thank you
for the nonpartisan support of the CLEAR Act, as well as
Senators Heitkamp and Moran. It is this type of bipartisan
sensible legislation that I do think is moving the ball forward
to help deal with some of these issues.
Senator Scott. Thank you.
Mr. Hill. The changes as you approach this threshold is
versus--I think of a threshold as a small step forward. This is
a meaningful leap for a bank of $10 billion. I have been in the
business for 30 years. I have dealt with a lot of regulation. I
do not ever remember even fathoming the fact that a customer
could add one more dollar to a checking account or savings
account to take you to $10 billion and you would be saddled
with a $20 million burden and be treated as a large bank.
So it is doing two things. It is, one, driving banks out of
our industry. Senator Perdue certainly sees it in his State.
Senator Tillis, Senator Scott, we certainly see it in ours.
They cannot compete, and as you get larger, they realize they
do not want to go over that $10 billion hurdle and they elect
to exit.
The other is for those that elect to stay in like our
company, we are an 80-plus-year-old company. We are vitally
important to many of the communities we serve. We wanted to
stay in. But try to take the Dodd-Frank Act and put it over a
bank with a little bit more than 100 offices compared to a
national bank with 5,000 or 6,000. The burden is huge. So it
ends up resulting on different forms of behavior. Companies
decide to sell, or they have to cut expenses. In our case, we
closed 10 percent of our branches. All of that money went
toward regulatory reform, and it has taken the attention of our
board and our management team and a lot of our employees over 3
years to be able to make this journey. And now we have just
crossed it. Now we begin to be treated as a large company.
Senator Scott. Thank you, Mr. Hill. Facts and figures
aside, at the end of the day I am worried about the Aiken
family that is trying to buy their first home or the Nichols
small business owner who just survived the flood and is now
trying to get back on their feet.
Mr. Hill, what is the impact of enforcing these arbitrary
thresholds on economic growth and on the people of South
Carolina? Specifically, you mentioned--my words, not yours--
$9.9 billion in assets versus $10 billion in assets, 10 percent
closing of your locations, maybe up to $20 million of
additional regulatory burden. How does that impact the average
person in our State looking to borrow money for a home or
restore their business after a major flood?
Mr. Hill. Well, the midsize banks fill a very important gap
between the smallest banks and the largest banks in our
country. Because we are the community bank for South Carolina,
we are very focused on communities like Aiken. Many of the
large national banks would not know where that community would
be. It does two things. One, it drives costs up. That is very
simple, very clear. It is 15 to 20 percent. It takes
flexibility away. It does not allow us to treat customers
uniquely based on their needs, both financially and also their
situation. And it paints in a one-size-fits-all regulatory
environment.
Senator Scott. Thank you, sir.
Ranking Member Brown.
Senator Brown. Thank you, Senator Scott.
Professor Omarova, the Wall Street Journal recently said it
is hard to miss how much the Treasury report would benefit top
Wall Street banks. That comes from the paper of record, if you
will, for Wall Street. By my count, the report includes about
two dozen of The Clearing House's recommendations. You talked
in your testimony, Professor Omarova, about the argument that
rolling back rules for Wall Street banks would help lending.
What are the implications of rolling back the capital and the
leverage rules and the Volcker rule restrictions against
proprietary trading? Would that, in fact, lead to more lending
or economic growth?
Ms. Omarova. Well, there is absolutely no evidence that it
will, in fact, lead to more lending or economic growth. There
is a lot of confusion about what capital rules do, because
banks always tell us, oh, you know, capital just traps cash and
we cannot lend money out. And nothing could be further from
truth. Capital is not cash in the vault. It is not some kind of
gold that, you know, they have to put away. Capital is just an
accounting concept. It is basically shareholders' equity, and
banks are forced by regulation to hold these cushions of
shareholder equity to protect creditors from losses on their
assets. They can get away with much thinner cushions than, you
know, a normal company could get away with in the capitalist
market because the Government protects creditors of the banks
from banks' failure.
And so if you roll back capital requirements, what will
happen is that the banks will be able to take more risks. And
because banks are privately owned, profit-seeking enterprises,
quite legitimately they would look for investments in assets
that generate higher returns, which typically entails higher
risks. And that is what will happen. There is no evidence that
somehow Dodd-Frank Act is what prevents banks from lending. You
know, banks choose how to use their cash, and they choose
their--for example, they choose to declare dividends out of
their cash. And that directly takes away cash from lending.
In my view, basically if we roll back these regulations,
what we will have on our hands will be another crisis, and
everybody in this room should be warned about that.
Senator Brown. Thank you.
Let me ask everybody on the panel--and I would prefer a yes
or no, and I think you can answer this yes or no. This
Committee has talked for some time--and the sitting Chairman
was a leader in this issue 3 or 4 years ago--about housing
finance reform and its importance to economic growth. I will
not ask you for detailed thoughts because that would be a long,
long answer from each of you. But if you would answer yes or
no, do you think we should have hearings on the topic of
housing finance reform and have an open process where you can
weigh in and we can discuss it? Mr. Simmons.
Mr. Simmons. Absolutely.
Senator Brown. Mr. Baer.
Mr. Baer. Yes.
Senator Brown. Mr. Hill.
Mr. Hill. Yes.
Senator Brown. Professor Omarova.
Ms. Omarova. Yes.
Senator Brown. OK. Thank you. If that is the case for
housing reform--I will start again on the left--do you think it
should also be the case for the significant changes to health
care?
Mr. Simmons. It is not the purpose of the hearing, but yes.
Senator Brown. Mr. Baer.
Mr. Baer. I do not know. It is not my area of expertise.
Senator Brown. But you are a citizen.
Mr. Baer. I am a citizen who has saved a lot of time in my
life by not thinking about health care reform because it is so
difficult to understand, and I am sort of full up on bank
regulatory----
Senator Brown. But do you think we should have an open
process and discuss it?
Mr. Baer. I think in general open processes are better than
closed processes.
Senator Brown. Mr. Hill.
Mr. Hill. I tend to think health care is important to the
vital health of the local economies, and that the more
discussion we can have to move that along, the better.
Senator Brown. Professor Omarova.
Ms. Omarova. Absolutely. There must be an open, democratic,
and fully vetted process for deciding such an important issue.
We all have to know what is going on.
Senator Brown. OK. Thank you. And if we screw up either
effort, whether it is Dodd-Frank, whether it is housing finance
reform, whether it is health care, clearly the economy pays a
price. I mean, we know that.
Let me ask one last question, Professor Omarova, and give
your answer as short as you can in complying with the
President's request--the Chairman's request. You are not the
President.
[Laughter.]
Senator Brown. Professor, you said reasonable people may
disagree and argue about whether the current size threshold,
$50 billion in assets, is the right one or whether a higher or
lower number would be more socially beneficial. How do we
balance providing--as we talk about tailoring Section 165, as I
think we should, how do we balance providing appropriate relief
for regional banks against the intent of 165 to mitigate risk
to financial stability?
Ms. Omarova. Well, that is a complicated question, but the
one clear answer is that we should not just simply remove all
regulation from regional banks because they are less than $1
trillion in assets. As a group, they still present significant
risks. If they fail as a group in a correlated set of failures,
that will probably tank regional economies and maybe the
national economy.
Think about the S&L crisis in the 1980s. Those were also
very small and traditional lenders, and when they were
deregulated, those small traditional lenders almost brought
down the financial system. And that is what we should think
about today. We cannot look at these regionals in isolation. We
definitely should think about how to tailor regulatory burden
for them, but we cannot just blankly remove all the regulations
because they are smaller.
Senator Brown. Thank you, Mr. Chairman.
Chairman Crapo [presiding]. Thank you.
Senator Tillis.
Senator Tillis. Thank you, Mr. Chair, and thank you all for
being here. I cannot imagine anyone sitting up at this dais
would suggest just a blanket repeal of all regulations. It does
not make sense. Regulations exist for a reason. Some of them
are good. Some of them are awful. And I think that there are
some manifestations of regulations that have been promulgated
under Dodd-Frank that are absolutely awful.
Mr. Baer, I want to ask you a question about the concept of
tailoring the--instead of these simplistic, you know, $50
billion, $250 billion sorts of thresholds that we create to
determine the regulatory burden on the institution, I am more
of a proponent of tailoring. I am getting to a point to where
the regulatory burden is proportionate to the risk of the
target being regulated. Could you give me some thought with
respect to the regulatory environment that we have today--most
of it is driven by Dodd-Frank; some of it pre-dated it--that
you think lends itself to that kind of thought process and the
benefits that you think would accrue by doing that?
Mr. Baer. Yes, Senator, sure. I think when it comes to
tailoring, you need to think about what risks does the
institution present. You know, for example, does it have a
capital markets business? Does it have multiple subsidiaries or
only one? Is it primarily funded by deposits or is it funded in
other ways? Then you think about what rule are we talking about
here? Is it a living will? Is it capital requirements? Is it
liquidity requirements? And then marry those two up.
So, for example, the notion of having a living will for a
small deposit-funded firm which is going to be in any event
resolved, you know, under the pre-existing FDIC resolution
process does not seem to make a lot of sense.
There may be other rules. I mean, if you believe that there
should not be proprietary trading, perhaps some of those rules
should continue to apply. But, again, I think you need to make
the decision based on firms in categories in terms of the risks
they present and then the type of the rule you are talking
about.
Senator Tillis. The other thing that I find interesting, at
least in some of the prior committees--I am sorry I was not
here earlier; we have got four committees meeting at the same
time, so I did not get to hear the testimony. But with an eye
toward lean regulation, let us say that we go through that
stratification, and we come up with a more coherent way of
actually determining what regulatory burdens should be placed
on a financial services institution. What about other areas in
terms of executing-- and this is for anyone, but when I hear
the big banks who would be at the highest level and have the
highest amount of regs--and I guess in some cases if you get
the methodology right, appropriately so. But would it make--how
can it possibly make sense to have the stress test submissions
be in the hundreds of thousands of pages? I mean, isn't there
any thought given to how you create a leaner design around this
and get the paperwork and the time and the costs associated
with that out of it so that the consumers accrue a benefit, the
money is being spent on value to the consumer versus compliance
with the Government? Anyone have any opinion on specific things
that we could potentially do to reduce that burden?
Mr. Baer. I will just say briefly on CCAR, again, I think
the length of the submission should vary with the complexity of
the firm. There may be large firms that need very complex
submissions. In fact, our view is with respect to the stress
test that the DFAST models run by the banks, which are granular
down to loan level, are actually appropriate and a better
measure of capital perhaps than the opaque model that the
Federal Reserve is running. But certainly for less complex
firms, I think less burden would be appropriate.
Mr. Hill. Senator, I would just add, if you look at--I
could sit down and explain our balance sheet to you in about 5
minutes. It is pretty simple. It does not take complex
algorithms and quants to be able to figure out the sensitivity
of our company. I think it ultimately comes down to capital,
how much you hold--that is what is going to protect all of us
when we have the next downturn. And I think the Basel limits, I
think looking at capital rules, actually provide much more
sound banking practices than some theoretical analysis on
sensitivity.
Mr. Simmons. I might just add, you know, our CCAR
submission runs over 12,000 pages. You reach points of
diminishing return pretty quickly in a very straightforward
business model. We find the process itself to be useful, but
the incredible degree of precision to which the Federal Reserve
has pushed this does not yield benefits commensurate with the
cost.
Senator Tillis. We all know that has a disproportionate
impact on smaller banking institutions, but all of the
regulations that we are heaping on that I think have reached a
point of diminishing returns we have got to look at and right-
size. There was clearly a risk in 2008 that we had to produce
regulations to avoid in the future, but we have clearly gone
too far. And I have to take exception with anybody who thinks
that everybody who wants a loan can get it today. The reality
is a lot of people--it is sort of like people who leave the
labor market, and so they are just not searching for work
anymore. There are entire business enterprises that are not
looking for capital because they do not think they can get it
or the cost of getting it is just too great. And it is having a
chilling effect on our economic growth. That is one of the
reasons why we have such anemic economic growth. And unless we
start right-sizing some of these regulations and recognize
there is a lot of pent-up demand for capital and that the root
of that are a lot of regulations that overreached in Dodd-
Frank, then we are not going to get to the sort of economic
activity that we need to get to, to then dig ourselves out of
this $20 trillion in debt.
On the report, I had a lot of questions to ask you, but I
have gone over, and I normally do not go over, Mr. Chair.
Chairman Crapo. We will take it out of next time.
Senator Tillis. But there are a number of things that, if I
may in follow-up questions for the record, I would like to go
into the report itself, and some of the priorities and
objectives, we would like your input, because I think this is
critically--it is one of the most critical things we can do to
really get economic activity where we need it to be.
Thank you all for being here.
Chairman Crapo. Thank you.
Senator Donnelly.
Senator Donnelly. Thank you, Mr. Chairman. I actually only
have one question, and that would be for Mr. Hill.
Senator Toomey and I have previously introduced a bill to
increase CFPB examination thresholds from $10 billion to $50
billion. How would that make things better for your customers--
not so much the bank, but what does that do for your customers?
Mr. Hill. Senator, it is a great question. I think that you
can take the CFPB, but you can take numerous parts of Dodd-
Frank legislation and really kind of put them all under the
same umbrella. Removing that $10 billion threshold for the
company and also for a customer, it just makes things less
complex. If we want to talk about getting more money to Main
Street, more money to the individual, having one more regulator
is not a way to accomplish that. We have multiple regulators
already, and now we will have an additional one now that we
just crossed CFPB.
So I think a lot of this is about what makes sense and what
simplifies it. We have a very close relationship with our
regulators----
Senator Donnelly. And it would not reduce safety or
stability in your organization, would it?
Mr. Hill. No, sir. Our primary regulator is the FDIC. We
have a very close relationship. They are in----
Senator Donnelly. Those are obviously critical elements.
Mr. Hill. Absolutely. And I think they should be. I just
think they can be done by our existing regulatory bodies.
Senator Donnelly. All right. Thank you.
Thank you, Mr. Chairman.
Chairman Crapo. Thank you, Senator Donnelly.
Senator Donnelly. I made up for Mr. Tillis.
Chairman Crapo. You did, and I appreciate it very much.
As I indicated at the beginning of the hearing, we have a
hard stop at 11 because we have three votes, and so I
appreciate that time. You will get an extra credit.
Senator Kennedy.
Senator Kennedy. Thank you, Mr. Chairman. I, too, am sorry
I am late, but I was not watching ``The Price Is Right''. I was
in a committee, OK?
[Laughter.]
Senator Kennedy. Mr. Hill, I wanted to--you have been
there. Do you still run a bank? I notice you have been
president and chief operating officer.
Mr. Hill. Yes, sir. I am actively running the bank day to
day and have been for the last 22 years.
Senator Kennedy. About $8 billion in assets?
Mr. Hill. We just crossed $10 billion in January. We are
roughly $11.5 billion today.
Senator Kennedy. How many people do you have in your
compliance department?
Mr. Hill. This is an estimate, but it would be roughly 50
direct employees, but then there are numerous compliance people
embedded in our lines of business across our company.
Senator Kennedy. So some do it full-time, some do it part-
time.
Mr. Hill. And some are just in a full-time compliance role,
and some are in the active day-to-day administering of the
process. And so there is some overlap there. But compared to 5
or 6 years ago, that is probably tenfold.
Senator Kennedy. OK. That was my question. Tenfold.
Mr. Hill. And if you look at the cost of that--I look at
ultimately what does that mean to the customer. I think Senator
Donnelly's question was: What does it mean to the customer?
What this means to the customer for a mortgage loan is our cost
to deliver a mortgage loan is roughly $1,000 more today than it
was just a few years ago, mainly because of the increased
compliance costs.
Senator Kennedy. OK. Let me break that down, though. Today
you have roughly 50, plus a number of employees that are part-
time, if you will, and that is tenfold. OK?
Mr. Hill. Yes.
Senator Kennedy. How much are you spending on those 50
employees?
Mr. Hill. It would be, you know--I guess it would be in the
several million dollars range.
Senator Kennedy. OK. And do you believe that these
additional employees are a direct result of Dodd-Frank?
Mr. Hill. Oh, they are. Yes, sir. To comply with the rules
and regulations, two things happened: We had to close ten
offices. We saved almost $5 million from closing those ten
offices. All that money was spent directly on complying with
Dodd-Frank.
Now, there is a lot more than just that $5 million, but
that all was directly spent--a large part of that was in
process and compliance-related efforts.
Senator Kennedy. How much did your bank make last year?
Mr. Hill. Our company last year made--this is an estimate.
Senator Kennedy. Sure.
Mr. Hill. In the $65 million range.
Senator Kennedy. OK. Has Dodd-Frank helped at all?
Mr. Hill. Yes, sir. I do believe there have been parts of
Dodd-Frank that have been positive. I think the most important
piece is the capital. The banking industry as a whole is
holding more capital today than we did. We are less leveraged
as an industry. That is the ultimate safety net. Regulation is
not the ultimate safety net. Capital is the ultimate safety
net. And banks across the board today are holding more
capital--our bank included. But we never levered up like many
of the large companies did. We are still holding more capital
than we have.
So there have been things that have been done that have
been vitally important. But to go back to the $10 billion
threshold, we are choking out the most vitally important part
of our community banking system by having this arbitrary
threshold. And it is making people leave the industry or
significantly limiting their ability to impact their local
community because we are--while we are in the same industry as
the large banks, we are really significantly different
companies.
Senator Kennedy. You have been in this business--well, you
all have. Has there ever been a time when the Federal
Government and its regulation of your industry really did sit
down and say what are the costs and what are the benefits and
make the sort of calculation that normal people do every day in
their business or in their family? Or is that just lip service?
Have we ever done it right?
Mr. Hill. Senator, I have been doing this 30 years. I have
seen a lot of regulation come and go. Most of it has been
constructive. You figure out a way to deal with it. And I have
never felt the need to reach out to a Senator about that
regulation until now. But this arbitrary $10 billion threshold
is a painful process that is costing our consumer and our
communities and local economies, and we are overregulating a
systemically important part of our economy, which is our
community banks.
Senator Kennedy. Thank you, Mr. Chairman.
Chairman Crapo. Thank you.
Senator Cortez Masto.
Senator Cortez Masto. Thank you, Mr. Chairman.
I am actually over here in the corner. Good morning. Thank
you for joining us.
So let me start off with the first question, and I will
open it up to the panel. As a new Senator from the great State
of Nevada, I was not here for the debate on Wall Street reform,
but let me tell you, as the Attorney General there for 8 years
watching as the crisis unfolded and the impact to the State of
Nevada, I was paying close attention. If you do not know, it
was ground zero for the foreclosure crisis, highest
unemployment, more people in foreclosure than, I think, in the
rest of the country. At one point in time, 64 weeks, we had the
highest rate of foreclosure, highest loan-to-value ratio, 70
percent of homeowners underwater. Devastating.
And so I am curious your thoughts on this. One thing I am
concerned about what I have seen is that President Trump's
Executive order on financial regulation did not once mention
consumer or investor protection. And in looking at the 150-page
report that was released by the Treasury Department this week
in response to that Executive order, they did not offer one
single example where additional protections for students,
servicemembers, or seniors were needed.
If we are going to do a wholesale review of financial
rules, shouldn't we look at both additional needed protections
and regulatory relief? Isn't that the type of balance we should
be looking at? I open that up to all of you.
Mr. Baer. Senator, I agree. I think absolutely, you know,
any review of post-crisis regulation should include consumer
regulation. I think, you know, one issue that we focused on----
Senator Cortez Masto. Not consumer regulation. Consumer
protection and regulations to protect consumers.
Mr. Baer. Absolutely.
Senator Cortez Masto. Right.
Mr. Baer. I agree.
Mr. Simmons. In my view, a lot of the products--we operate
in Nevada. It is Nevada State Bank. And we suffered huge
losses, not from mortgages but from financing land improvements
and from businesses. And so we certainly experienced a lot of
that pain.
There have been a lot of consumer protections put in and
many of them probably necessary. We ought to be looking at
both. But the combination, the layering of all of this has made
this industry increasingly sclerotic and unable to meet the
legitimate needs of customers in a way that is sensible and
prudent and logical.
And so I think any of us would be open to looking at are
there needs for additional consumer protections. That should be
on the table. But that is not where the real problem is from
the point of view of us who are trying to deliver services to
consumers and businesses who are in need of them today.
Senator Cortez Masto. So one of the consumer protections
that I thought was very important and I fought for as Attorney
General were servicing standards. I think we need those, and I
am concerned that there is this idea that we need to do away
with them, roll those back somehow. I am curious your thoughts
on that.
Mr. Simmons. I am not aware of any argument being made by
the industry to roll back any protections for consumers in the
servicing standards. I do not think that is what we are here to
defend.
Senator Cortez Masto. Great. Thank you.
Anyone else have any other comments in general about
consumer protection and that should be a part of this
discussion?
Mr. Hill. Senator, we are a community bank. We have roughly
700,000 customers. I think the overarching thing for us is
doing away with the $10 billion threshold because we are
treating community banks like large banks. And I think that
consumer protection is a vitally important part of the role for
regulatory bodies and also for the banks. But I just think it
can be done by our existing regulator. We do not need a new
regulator for a community bank with 100 offices to be able to
do that. FDIC is in our offices almost year round, knows our
company very, very well, and I think it effectively enforces
that protection.
Senator Cortez Masto. Thank you.
Dr. Omarova, do you have any comments?
Ms. Omarova. I just have a general sort of observation that
banks will never publicly say anything against consumer
protection per se because that is bad PR, and yet they always
admit that the second that regulatory costs increase, they will
immediately pass those regulatory costs on to consumers. And
then they say, therefore, you should not regulate us because it
is costly.
To me, that is not consumer protection. Banks'
shareholders, banks' managers, those are the guys who should be
eating those additional regulatory costs. To me, that is the
essence of consumer protection in practice.
Senator Cortez Masto. Thank you. Go ahead.
Mr. Baer. Senator, not, you know, purely consumer
protection, but I do think there is a lot of research to be
done, and we have done some research, including a recent
edition of our quarterly magazine, around the question of
income inequality and bank regulation. I think the evidence
shows that one outcome of a lot of the post-crisis rules,
including even down to the level of stress testing, where the
assumption is that there is a very large rise in unemployment,
which tends to cause banks to, you know, more highly price
loans to people who are subject to that or are likely to be
affected by that spike in unemployment. There has definitely
been an increase in the price of credit to people at the lower-
income end of the spectrum, and a lot of that in very subtle
but very meaningful ways has to do with regulation. And so,
yes, consumer protection is really important, but we also think
access to credit for consumers, particularly low- and moderate-
income consumers, is really important. And that is the reason
that we have devoted increasing amounts of our research to it,
but we really think others should as well.
Senator Cortez Masto. Thank you. I notice my time is up.
Thank you very much for your comments.
Chairman Crapo. Thank you. And as I have indicated before,
we have a hard stop at 11 o'clock. I have not asked my
questions yet. That makes three of us if no one else comes. I
am going to have Senator Cotton and then Senator Warren go.
That will give me a couple of minutes at the end, and maybe I
can scoot it a little past 11 and get my 5 minutes in, too.
Senator Cotton.
Senator Cotton. Will I get my bills and amendments voted on
earlier if I give my time to you?
[Laughter.]
Chairman Crapo. Oh, yeah. There is extra credit here.
Senator Cotton. Mr. Hill, I know you have spoken about the
various threshold levels, $10 billion, $50 billion, and the
burdens those bring. I want to talk a little bit about the
arbitrariness of that. Obviously, anytime you pick a number, it
is somewhat arbitrary, you know, whether it is the designation
for a systemically important financial institution or a speed
limit or an age to vote or an age to drink alcohol. But in this
field, I mean, what are we talking about here in terms of an
institution that might hit a $10 or a $50 billion threshold?
Are we talking about thousands or hundreds of thousands of
institutions? Or are we talking of a scale of maybe a few
dozen?
Mr. Hill. Well, if you look at the 10 to 50 ranks, there
are 79 of us.
Senator Cotton. OK. That is what I thought. It would seem
that that could be done on a more discriminating basis than an
arbitrary threshold, wouldn't it?
Mr. Hill. Yes, sir. It is just not a risk-based approach,
and companies--customers are different and banks are different.
A risk-based approach versus a one-size-fits-all regulation
does not make sense. In my mind, it is quite simple. The role
here is risk management. We do not want another financial
crisis. And the risks come from the ``too big to fail'' banks.
So, to me, we need to regulate them as one industry, and then
outside of that, they are mostly community banks or large
community banks like our company. And I think those are very
different type banks.
Senator Cotton. So, in principle, you could have a $50
billion bank, or a $500 billion bank for that matter, that is
relatively plain vanilla, conservative, and, therefore, not all
that risky?
Mr. Hill. Well, I cannot speak to all the 500. I can speak
to the 10 to 50s, and I think I can speak to a bank somewhat
like Mr. Simmons', which is basically just a larger community
bank. But because we have done things right, because we have
attracted customers, because we help small businesses, we have
grown. And today we are penalized when we take that incremental
dollar over $10 billion.
Senator Cotton. And when you say ``community bank'' there,
to be exact, in this context, I think I understand you to mean
focusing on the functions that a bank performs.
Mr. Hill. Operating as a community bank. While we are $11.5
billion today, our operating model is the same as it was when
we were $400 million. We are just in more communities.
Senator Cotton. Isn't there an old joke about taking it in
at 3, lending it out at 6, and hitting the golf course by 3:00?
[Laughter.]
Mr. Hill. I think that was before I joined the industry.
Senator Cotton. But the point being that this risk analysis
is primarily--or this analysis should be primarily risk-based,
based on the nature of the institutions, or the nature of the
functions an institution performs, and, therefore, a relatively
large institution can be engaged in relatively low-risk
activities. But by the same token, an institution of less than
$50 or less than $10 billion, because of the nature of its
positions and interlocking counterparties could actually be
quite risky. Correct?
Mr. Hill. Correct. And I think just the opposite of that.
We want to incent less risk in our financial services industry.
So what better way to do that than hold the adequate amount of
capital, be in the less risky businesses. We have never done
any proprietary trading. So if you are in that business, hold
more capital. You are going to have more regulation. But if you
do operate a simple business model that positively impacts our
community, those are the ones that we have to be careful that
we do not go too far. Dodd-Frank did a lot of good things. It
overreached in this $10 to $50 billion sector and treated that
sector as it does many of the large banks in our country.
Senator Cotton. OK. Just to tie a bow on this part of the
conversation, I would say that I think the size of an
institution obviously needs to be a part of this analysis, but
a simple size-based approach does not seem to make much sense
to me. And given the number of institutions we are discussing
here, you would think that our financial regulatory agencies
could have a more discriminating approach. Again, we are not
dealing with, you know, millions of Americans who become 18
years old every year and, therefore, we just have to draw an
arbitrary line, even though we all know that some 17-year-olds
are very mature and exercise good judgment and plenty of 19-
year-olds do not, when you are talking about something on the
order of a few dozen institutions that we can take a more
sensible and case-by-case approach, with size being one factor
in that analysis.
I will yield 40 second back to the Chairman so I can get a
chit in the future.
Chairman Crapo. We will keep that record.
Senator Warren.
Senator Warren. Thank you, Mr. Chairman.
So after the financial crisis, Congress determined that
banks with more than $50 billion in assets--this is about
roughly the 40 biggest banks in the country--posed a greater
risk to the economy than community banks and credit unions. And
so we required the Federal Reserve to apply tougher rules and
more oversight to those banks. And now those banks want to
eliminate the $50 billion threshold. They want to cut all but
the very biggest banks loose from stricter oversight, and they
want to restrict the Federal regulators from applying tougher
rules except under somewhat more limited circumstances. So, in
other words, this is about rolling back a big part of Dodd-
Frank, and I just want to take a look at that.
Mr. Simmons, you are the CEO of Zions Bank, which is one of
the banks that would avoid tougher rules under the industry's
proposal. And in your written testimony, you argue that the
current approach covers too many banks given the minimal risks

posed by banks like yours, and that we are unnecessarily making
it harder for banks like yours to lend. Do I have that about
right?
Mr. Simmons. Yeah.
Senator Warren. Yeah, OK. Now, those arguments sounded very
familiar to me, so I went back and looked, and it turns out
that back in September of 2006, just 2 years before the
financial crisis, you were the head of Zions Bank, and you
testified before the House Financial Services Committee. Your
testimony strongly opposed guidance from Federal regulators
that increased the oversight for banks that had a high
concentration of commercial real estate loans. Regulators were
worried that the banks were overly exposed to one category of
lending and that might put them at greater risk of failing. You
thought, ``Ah, there is no problem.'' And in opposing the
guidance--I want to quote you on this--you said, ``The guidance
has been proposed at a time when the banking industry is
exceptionally healthy.''
Another one from you in this testimony: ``Commercial real
estate loans in particular have performed exceptionally well.''
Another one: ``By using blanket industry-wide guidance to
address concentrations, the regulators risk choking off the
flow of credit from banks that are engaging in commercial real
estate lending in a safe, sound, and profitable manner.''
Now, within 2 years of your testimony, the bank you led,
Zions, needed nearly $1.5 billion in taxpayer bailout money to
stay afloat. And here is the kicker: That was in part because
your bank was highly concentrated in commercial real estate
lending, the exact thing that you told Congress was not an
issue, nothing to worry about.
So, Mr. Simmons, when you say today that Congress can
safely roll back the rules on banks like yours and there will
not be any risks to taxpayers, why should anyone believe you?
Mr. Simmons. Well, listen, what we are saying is that the
enhanced prudential standards in Section 165 of Dodd-Frank are
industrial strength and intended, in my way of thinking about
this, for institutions that pose a----
Senator Warren. Mr. Simmons, let me stop you right there
because I know we are really trying to do this quickly. I am
not asking you to repeat your argument. We have already agreed
on what your argument is. The question I am asking, given your
previous testimony about how there is no problem here, and then
it turned out you needed $1.5 billion in bailout money on
exactly the thing you testified was not a problem, I am asking
why anybody should believe you when you come in here today and
say no problem in this area, let the $50 billion and above
banks go ahead. I am just trying to understand why you have any
credibility on this issue.
Mr. Simmons. Well, listen, because I deal with it every
day, and because----
Senator Warren. Well, you dealt with it every day back when
you testified in 2006, and the taxpayers had to pony up $1.5
billion to save your bank.
Mr. Simmons. Listen, every large bank took TARP money and
without----
Senator Warren. I am sorry. So your argument is that you
were right or wrong----
Mr. Simmons. The one large bank that did not receive it,
National City, was sold a week later. This was a matter of
preserving confidence across the industry. Our capital, our
equity capital, always remained above the regulatory minimums.
Senator Warren. I am sorry, Mr. Simmons. Are you trying to
make the argument that you did not have a problem? You know,
because actually----
Mr. Simmons. We incurred stress, but we never saw equity
capital, common equity capital----
Senator Warren. Let us just be clear about----
Mr. Simmons. --decline below the regulatory minimums.
Senator Warren. ----the problem. The regulators actually
went back in 2013 to reexamine their earlier guidance, the
guidance you had said was unnecessary, and they found, ``During
the 3-year economic downturn, banks with high commercial real
estate concentration levels proved to be far more susceptible
to failure. Specifically, 23 percent of the banks that were
highly concentrated in commercial real estate lending,'' what
you had testified about, ``failed compared with only one-half
of 1 percent of the banks that were not.''
So I understand we are out of time. I just want to say
here, you know, what I notice about this is whenever things are
going OK, the banks come in here and say, ``Yay, let us reduce
the rules, let us let everybody go out. What could possibly go
wrong?'' And then when things go wrong, banks like yours line
up and say to the taxpayers, ``Bail me out.''
Our job is to make sure that we do not permit the next
failure to happen because it helps short-term bank profits. Our
job is to watch out for the taxpayers and the security of this
economy.
Thank you, Mr. Chairman. I apologize for going over.
Chairman Crapo. We are going to put that on your record,
too.
[Laughter.]
Chairman Crapo. Just kidding.
Senator Warren. Well, it is not the first time.
Chairman Crapo. Thank you. And I will take my questions
now. I will try not to go 5 minutes because we do have a vote
starting in about 1 minute.
Mr. Simmons, I would like to ask you to finish the comment
that you were making just a moment ago with Senator Warren
about your equity capital back at the time when the stress
started to arise and the collapse in the housing market.
Mr. Simmons. Well, listen, capital across the industry has
increased dramatically. It has for us. It has more than
doubled. It is about 120 percent of what it was back in 2006.
Common equity capital, relative to risk-weighted assets, has
increased from about 5.5 percent to about 12.2 percent. So we
have not only a strong industry, but we have the strongest
banking industry in the world. So there is a lot of equity
capital in the industry today.
Chairman Crapo. All right. Thank you. And again, Mr.
Simmons, over the past few years, a number of financial
regulators have made comments before this Committee supporting
changes to the $50 billion SIFI threshold, including Federal
Reserve Chair Yellen, former Federal Reserve Governor Tarullo,
and former Comptroller Curry, and these are those who are the
regulators, in some cases were the regulators who are tasked
with getting it right to deal with the risk in our economy.
While there are different views on what to replace the
threshold with, it seems to me there is general bipartisan
agreement that a bank is not systemically important simply
because its assets exceed $50 billion. If the SIFI threshold
was amended so that noncomplex banks like Zions were no longer
subject to those enhanced standards, how would that impact the
broader economy?
Mr. Simmons. Well, it would--as I indicated, we have
become, I think, as an industry quite--just sclerotic in terms
of our ability to do business. I have a letter here from a
customer up in Seattle. It says, ``When Kerri''--I talked to
one of our people up there. ``When Kerri Knudsen informed me
last week the bank could not provide construction financing for
my upcoming development project, I was shocked. The concept
that a $3 million construction loan was not possible left me
dumbfounded. It leaves me incredulous that a multi-billion-
dollar institution is maxed out.'' It goes on to talk about
this. ``After 22 years of doing business together, I hit the
streets looking for a construction loan.''
We find ourselves trying to guess what is in the Federal
Reserve's models, in their CCAR models. We know that it is--I
mean, we have some vague outline of how--you know, what the
results are, but we do not know really how it is treating
individual loans. This lack of transparency is my major beef
with the CCAR regime. But the overlay of all of these
regulations has made it increasingly difficult to do business,
and for us, small businesses are sort of our forte, and we feel
kind of crippled in terms of our ability to serve them.
Chairman Crapo. Well, thank you. And I actually get letters
and visits from businessmen and women in Idaho who have the
same kind of concerns about the inability to get the kind of
financing that just seems so obviously appropriate. So I
understand the point you are making.
And one other point there quickly. If banks over $50
billion right now were no longer subject to the SIFI
thresholds, the $50 billion SIFI threshold, isn't it true that
they are still subject to very extensive safety and soundness
regulation across the system?
Mr. Simmons. Absolutely. Always have been, always will be.
And stress testing will remain a central part of what we do.
Chairman Crapo. And they will still conduct stress tests.
Mr. Simmons. Absolutely.
Chairman Crapo. I wanted to make it clear. Some make it
look like there is an exemption of regulation being discussed
here. It is a refinement and a tailoring of the type of
regulation that we are talking about.
Obviously, my time is up, and I think you heard the bells
go off, so this is going to have to be my last question, and
this is for you, Mr. Hill. I apologize to the other witnesses.
I do have questions for you, too. I will submit those.
But you mentioned in your opening testimony that your bank
recently crossed above the $10 threshold and as a result needs
to comply with numerous additional requirements, including the
Dodd-Frank Act stress test known as DFAST. Can you explain the
various steps your bank has had to take to comply with this
requirement?
Mr. Hill. Well, I think from a financial perspective, the
overall cost is several million dollars, and most of that is in
terms of buying very sophisticated models to stress-test our
bank under various different circumstances and add the
quantitative--the employees who have a quantitative background
to be able to do that. And so the ultimate impact is several
million dollars, more overhead, more complexity, for a balance
sheet that is relatively simple.
Chairman Crapo. And has that caused the cost of a mortgage
to your customers to go up?
Mr. Hill. Our costs of our mortgage loans have risen
multifaceted. I do not think you could put it all on DFAST or
QM or any others. But when you put all that regulation together
that comes with that $10 billion threshold, for every mortgage
loans we make it costs us $1,000 more to make it today than it
did just a few years ago.
Chairman Crapo. And who pays that $1,000?
Mr. Hill. It ends up out of the customer's pocket.
Chairman Crapo. So if we were talking about consumer
protection, if we could reduce the cost of that mortgage and
still maintain the safety and soundness, would that not be some
of the best consumer protection we could achieve?
Mr. Hill. It seems very logical to me, Senator.
Chairman Crapo. Well, thank you.
And to the others here, I apologize I did not get toy with
my questions. I apologize to everybody. Usually we have the
time to go on and have even a second round of questions. But
today we are wrapping up an Iran sanctions bill, and we are
going to be doing the final three votes on it starting right
now.
So before I close this hearing, I want to alert all
Senators that they should submit their further questions by
Thursday, and you will probably receive some further written
questions. I urge you to respond to those written questions as
promptly as you can.
Again, I want to thank all of you for coming and giving us
your time and your advice today. I assure you that both your
written and your oral testimony is very thoroughly reviewed and
utilized by us, and we are working together to try to build a
very strong package.
As you probably are aware, we are not calling it
``regulatory reform.'' We are calling it ``economic growth.''
And we are looking for statutory and regulatory reforms that
will help to grow the economy while still maintaining safety
and soundness in our financial institutions. I think that is
achievable. Thank you for being here to help us on that.
With that, this hearing is adjourned.
[Whereupon, at 11:07 a.m., the hearing was adjourned.]
[Prepared statements, responses to written questions, and
additional material supplied for the record follow:]
PREPARED STATEMENT OF HARRIS H. SIMMONS
Chairman and Chief Executive Officer, Zions Bancorporation, on behalf
of the Regional Bank Coalition
June 15, 2017
I. Introduction
Chairman Crapo, Ranking Member Brown, and Members of the Committee,
thank you for the opportunity to appear before you this morning. I am
Chairman and CEO of Zions Bancorporation, a $65 billion dollar (total
assets) bank holding company headquartered in Salt Lake City, Utah. We
primarily operate in 11 western States, with local management teams and
brand names, from Texas to the West Coast, including the Chairman's
home State of Idaho, where we are the third largest bank in the market,
and where we have consistently been the largest SBA lender. Indeed, we
have a particular focus on serving small and midsized businesses and
municipalities throughout the West. We believe we are very good at
serving such customers, and are proud to have been consistently
recognized by small and middle-market businesses as one of the best
banks in the Nation in providing banking services to such clients, as
measured by the number of Excellence Awards conferred through Greenwich
Research Associates' survey of approximately 30,000 small and middle
market businesses across the country each year. Virtually all our
banking activities are very traditional in nature, with a
straightforward business model that is highly focused on taking
deposits, making loans, and providing our customers with a high degree
of service. We are primarily a commercial lender, which is to say that
we are especially focused on lending to businesses. We provide
approximately one-third as much credit to businesses, in loan sizes
between $100,000 and $1,000,000, as Bank of America does in aggregate--
underscoring our focus on serving smaller businesses in the markets we
serve. And we do so without presenting the type of systemic risk that
is characteristic of the very largest banking organizations. Together
with other regional banks, we are highly focused on delivering credit
and depository services to the small and midsized businesses that have
been America's engine of economic growth.
Zions Bancorporation has the distinction of currently being the
smallest of the Systemically Important Financial Institutions--or
``SIFIs''--in accordance with the $50 billion asset threshold for the
determination of systemic importance as defined in section 165 of the
Dodd-Frank Act. And while we are proud of the services we provide to
our customers, and believe we incrementally make a real difference in
the local markets in which we operate, we certainly do not consider
ourselves to be systemically important to the United States economy. We
in fact half-jokingly refer to our company as an ``Itty Bitty SIFI,''
and we see evidence that an increasing number of thoughtful observers,
including our own regulators, are of the opinion that we, and other
regional banks, are of neither the size, complexity nor critical
importance to the workings of the U.S. economy to warrant the scope,
intensity and cost of additional regulation that the automatic
designation as a SIFI carries with it. \1\
---------------------------------------------------------------------------
\1\ See, e.g., remarks of Federal Reserve Board Governor Daniel K.
Tarullo in his testimony before the U.S. Senate Committee on Banking,
Housing, and Urban Affairs, March 19, 2015.
---------------------------------------------------------------------------
II. Regional Banks Have Simpler Business Models That Fundamentally Pose
Less Risk Than the Nation's Largest Money Center Banks
Regional banks overwhelmingly operate with straightforward,
traditional business models that focus on receiving deposits and making
loans. In my own bank's case, only 4.8 percent of our total assets are
financed with short-term nondeposit liabilities. And the great majority
of our loans are secured with various forms of collateral, providing a
secondary means of repayment. Like community banks, regional banks
focus on providing credit not only to consumers, but to small and
midsized businesses. For example, in the case of Zions Bancorporation,
business loans between $100,000 and $1 million in size comprise 19
percent of our entire commercial loan portfolio, as compared to
approximately 2 percent for Citigroup and 7 percent for JPMorgan Chase.
The revenue streams of regional banks are primarily generated
through lending spread income and the provision of ancillary services
to customers with long-term relationships with the bank. There is much
less focus on ``transactional'' income from trading and capital markets
activities. Indeed, approximately 90 percent of the banking industry's
total trading income last year was generated by five of the industry's
largest banks, each of which is considered by regulators to be a Global
Systemically Important Bank (G-SIB), and none of which was a regional
institution.
Using a more fulsome measure of risk than sheer asset size, 2 years
ago the Treasury Department's Office of Financial Research (OFR)
published a report on the relative systemic risk posed by 33 U.S. bank
holding companies. \2\ The methodology employed was a systemic risk
scorecard developed by the Basel Committee on Bank Supervision (Basel
Committee) and published by the Financial Stability Board (FSB), using
data provided by bank holding companies on Federal Reserve Form Y-15
with regard to an institution's size, interconnectedness,
substitutability, complexity and cross-jurisdictional activities. The
highest score, denoted as a percentage, belonged to JPMorgan Chase &
Co., with a score of 5.05 percent, followed by Citigroup at 4.27
percent. Applying the OFR/Basel Committee methodology to the two dozen
regional banks with assets of over $50 billion, and thus designated as
Systemically Important Financial Institutions (SIFIs) under provisions
of the Dodd-Frank Act, the aggregate risk score of the regionals as a
group (including banks as large as U.S. Bancorp and PNC Financial
Services Group, Inc.) is less than the score of either JPMorgan Chase
or Citigroup. The very largest banks, which pose the type of systemic
risk to the economy that Section 165 of the Dodd-Frank Act was meant to
circumscribe, are characterized by not only substantially larger
nominal asset exposures than those presented by regional banks, but
also by complex--and often global--organizational structures,
substantial off-balance sheet and market-making activities, and a high
degree of interconnectedness throughout the financial sector and in the
larger economy. For example, while JPMorgan Chase & Co.'s balance sheet
is 39 times the size of Zions Bancorporation's, it's total payments
activity last year was 616 times larger than Zions' levels, and its
total derivatives exposures are 5,253 times larger than ours. The same
general relative risk exposures characterize the entire regional bank
group.
---------------------------------------------------------------------------
\2\ Office of Financial Research Brief Series, 15-01, February 12,
2015.
---------------------------------------------------------------------------
III. The Dodd-Frank Act's Arbitrary Asset Thresholds Are Stifling Our
Ability To Serve Customers and Foster Economic Growth
a. Stress Testing and Capital Planning
As a covered institution, or SIFI, under section 165 of the Dodd-
Frank Act, Zions Bancorporation is subject not only to the Act's
rigorous stress testing (Dodd-Frank Act Stress Test, or ``DFAST'')
requirements, but to the annual Comprehensive Capital Analysis and
Review (CCAR) conducted in conjunction with the annual DFAST exercise.
The DFAST process is intensive, time-consuming and costly. It involves
the development and continual maintenance of sophisticated statistical
models designed to project a bank's performance over the course of a
hypothetical nine-quarter period of severe economic stress, using
scenarios incorporating a variety of macroeconomic variables supplied
annually by the Federal Reserve, and supplemented by a bank holding
company's own variables and assumptions reflecting any of its
idiosyncratic risk exposures. These statistical models are expected to
be capable of projecting the likely outcomes and interrelated effects
of each line item on a bank holding company's income statement and
balance sheet, and the resulting impact on capital levels, based on a
granular analysis of a bank's individual assets and liabilities. They
must be developed based on historical performance, back-tested,
validated, audited, and documented. So-called ``challenger'' models
must also be developed to identify potential weaknesses inherent in the
more material primary models. And the entire process must be conducted
under a rigorous governance process involving both the bank's
management and board of directors.
Each of the bank holding companies required to participate in the
Federal Reserve's supervisory stress test exercise furnishes the
Federal Reserve with millions of data elements derived from individual
loans and other balance sheet items on Form FR Y-14. This data is used
both in the banks' internal stress tests and in the Federal Reserve's
own models to project risk-weighted assets and capital levels during,
and at the conclusion of, the hypothetical period of severe stress in
an attempt to ensure that capital levels under stress will not breach
minimum regulatory standards. The CCAR exercise builds on the DFAST
process by incorporating a firm's projected capital actions over the
nine-quarter projection period. The objective is to determine that a
bank holding company's projected capital actions would not, during a
period of stress such as that reflected in the stress test, impair
capital levels below required regulatory capital thresholds.
After evaluating the results of its own and the banks' stress tests
and capital plans, the Federal Reserve provides each covered
institution with a quantitative assessment of its capital levels. \3\
Zions Bancorporation has been a participant in the CCAR process for the
past several years. We have spent well over $25 million in outside
consulting feels, and many thousands of hours of management and board
time focused on CCAR. We annually submit the equivalent of
approximately 12,500 pages of detailed mathematical models, analysis
and narrative to the Federal Reserve incorporating our CCAR projections
and capital plans. We also complete a mid-year stress test exercise to
complement the more intensive annual submission.
---------------------------------------------------------------------------
\3\ The Federal Reserve also provides large, complex banking
organizations (which it generally defines as those with over $250
billion in assets) with a qualitative assessment of stress testing and
capital planning processes. Regional banks have previously been given
such qualitative assessments; however, the Federal Reserve announced on
January 30, 2017, that it would discontinue that practice, while at the
same time tightening regulations regarding capital distributions to
shareholders without seeking Federal Reserve Board approval.
---------------------------------------------------------------------------
I view stress testing as a fundamentally important tool in the
management of a bank's risk and the assessment of its capital adequacy.
The value of the insights it yields, however, does not increase in
linear proportion to the investment made in the exercise, and this is
particularly true for less complex regional banking institutions. There
are diminishing returns from this exercise for both the banking
institutions and the regulators. Former Federal Reserve Governor Daniel
K. Tarullo has noted that `` . . . the basic requirements for the
aggregation and reporting of data conforming to our supervisory model
and for firms to run our scenarios through their own models do entail
substantial expenditures of out-of-pocket and human resources. This can
be a considerable challenge for a $60 billion or $70 billion bank. On
the other side of the ledger, while we do derive some supervisory
benefits from inclusion of these banks toward the lower end of the
range in the supervisory stress tests, those benefits are relatively
modest, and we believe we could probably realize them through other
supervisory means.'' \4\
---------------------------------------------------------------------------
\4\ Former Federal Reserve Board Governor Daniel K. Tarullo, in
remarks to the U.S. Senate Committee on Banking, Housing, and Urban
Affairs, March 19, 2015.
---------------------------------------------------------------------------
Ideally, the stress testing process should inform management's and
the board's thinking about managing credit concentrations, interest
rate risk, underwriting standards, pricing, and maintaining an
appropriate balance of risks in its portfolio. In our own experience,
these objectives are largely thwarted by the reality that the results
of the Federal Reserve's internal models trump our own internally
modeled results. Although the Federal Reserve has posed no material
objection to Zions Bancorporation's qualitative processes in recent
CCAR cycles, its own modeled measures of my firm's capital ratios after
nine quarters of severely adverse economic conditions have been
consistently and materially below our own projected outcomes. Such
variances in outcomes beg a reconciliation of the models used by each
organization if the results are to be truly useful in the management of
the company. And while Federal Reserve officials argue that
``transparency around the stress testing exercise improves the
credibility of the exercise and creates accountability both for firms
and supervisors,'' \5\ they continue to maintain that it is important
not to disclose details of their models, lest firms ``manage to the
test.'' Certainly it is not difficult to understand a regulator's
perspective about this, but the notion that the rules--which are
effectively incorporated into those models' algorithms--governing
banks' capital distributions to the firms' owners should be kept secret
finds little if any parallel in our legal and regulatory system.
---------------------------------------------------------------------------
\5\ Federal Reserve Board Vice Chairman Stanley Fischer, speaking
at the Riksbank Macroprudential Conference, June 24, 2015.
---------------------------------------------------------------------------
This lack of transparency has the effect of creating uncertainty,
and because the Federal Reserve's modeled capital results become the
``binding constraint'' for capital planning by most banks, including my
own, we are necessarily led to attempt to ``manage to the test''--even
if it's not clear how the test works. This uncertainty echoes recent
comments by former Federal Reserve Governor Daniel K. Tarullo, who
noted that ``while enhanced prudential standards are important to
ensure that larger banks can continue to provide credit even in periods
of stress, some of those same enhancements could actually inhibit
credit extension by rendering the reasonable business models of middle-
sized and smaller banks unprofitable.'' \6\ Federal Reserve Governor
Jerome Powell, Chairman of the Fed's Committee on Supervision and
Regulation, recently indicated in a televised interview his desire to
have the Fed provide ``much more granular information about our
expectations for loss rates on particular portfolios, of corporate
loans and other types of loans.'' \7\ While any improvement in
communicating outcomes is welcomed, real transparency will only be
attained when the Federal Reserve publishes details about the actual
content and mechanics of the models it uses to effectively govern
banks' capital levels, opening them to the kind of outside scrutiny and
debate which would inevitably result in stronger modeling processes.
---------------------------------------------------------------------------
\6\ Former Federal Reserve Governor Daniel K. Tarullo--before the
U.S. Senate Committee on Banking, Housing, and Urban Affairs, March 19,
2015.
\7\ CNBC, June 1, 2017, Steve Liesman interview with Governor
Jerome Powell.
---------------------------------------------------------------------------
In the absence of such transparency, banks are left to guess what
level of capital is required for each type of loan, and indeed for each
individual loan, since every loan has a unique blend of borrower
strength, collateral support and other characteristics that define
risk.
The uncertainty surrounding the Fed's modeling processes in CCAR
can cause banks to withdraw or limit certain types of lending. In our
own case, we've in particular established limits on construction and
term commercial real estate lending that are significantly more
conservative than those incorporated in current interagency guidelines
on commercial real estate risk management. \8\ Another example of the
uncertainty around the Federal Reserve's models involves small business
loans. The detailed FR Y-14 data templates used for the Federal
Reserve's models to capture granular data on collateral values and
other factors useful in evaluating potential loss exposures for
commercial loans expressly exclude loans of less than $1 million and
credit-scored owner-occupied commercial real estate loans, the
combination of which comprises a substantial portion of our total loan
portfolio. Rather, such loans are reported on a supplemental schedule
that includes only the loan balances. We can therefore only suppose
that such loans are treated relatively more harshly in the Federal
Reserve's models, resulting in uncertainty in terms of how much credit
of this type we can afford to grant, and at what price, in order to
reduce the risk of a quantitative ``miss'' in the Federal Reserve's
calculation of our required capital.
---------------------------------------------------------------------------
\8\ 5 Office of the Comptroller of the Currency, FDIC and Board of
Governors of the Federal Reserve System: Concentrations in Commercial
Real Estate Lending, Sound Risk Management Practices, December, 2006.
---------------------------------------------------------------------------
b. Liquidity Management
Having been designated as a Systemically Important Financial
Institution, Zions Bancorporation is also subject to the Modified
Liquidity Coverage Ratio. The three primary Federal banking regulatory
agencies, in implementing the Basel III liquidity framework, jointly
adopted the Liquidity Coverage Ratio (LCR) rule in September, 2014. The
rule is applicable to internationally active banking organizations,
generally those with $250 billion or more in total consolidated assets
or $10 billion or more in on-balance-sheet foreign exposure. At the
same time, the Federal Reserve went beyond the Basel Committee's LCR
framework, and adopted a somewhat less stringent rule, the Modified
Liquidity Coverage Ratio (MLCR), applicable to bank holding companies
with $50 billion or more in consolidated assets but that are not
internationally active. This quantitative measurement supplements a
qualitative liquidity management framework introduced in early 2014 to
fulfill Enhanced Prudential Standards requirements, including liquidity
standards, required by section 165 of the Dodd-Frank Act. The MLCR
requires a bank holding company to hold a narrowly defined portfolio of
``High Quality Liquid Assets'' (HQLA) equal to or greater than expected
net cash outflows over a 21-day period, in accordance with a prescribed
set of run-off calculations established in the rule. The qualitative
liquidity management framework requires, among other things, monthly
internal liquidity stress tests to supplement the prescriptive MLCR in
determining the size of the institution's required minimum liquidity
buffer. The full extent of the impact of the liquidity rules on SIFIs
is almost certainly not fully apparent in the current economic
environment. We have experienced a prolonged period of low interest
rates without precedent, and liquidity in the banking system has been
abundant by virtually any historical measure. But liquidity comes at a
cost, and the true cost of these rules will become manifest as interest
rates and liquidity levels eventually normalize. While it is important
for every depository institution to maintain appropriate levels of
reserves to deal with normal fluctuations in cash flows, maintaining
additional liquidity buffers as an insurance policy against times of
extreme stress is a costly exercise for banks and for the economy at
large. Every dollar invested in high quality liquid assets is a dollar
that cannot be loaned out and put to more productive use. In times of
liquidity stress, the impact will likely be most particularly acute for
smaller and middle-market businesses that do not have ready access to
the capital markets, and for whom bank credit is their financial
lifeblood. As noted earlier. regional banks subject to the MLCR and the
additional enhanced prudential liquidity standards imposed by the Dodd-
Frank Act provide a disproportionate share of credit to such
businesses.
c. Other Consequences of SIFI Designation
Since the financial crisis, Zions Bancorporation has more than
doubled its staffing in areas such as compliance, internal audit,
credit administration and enterprise risk management. In an effort to
manage costs, these increases have been accompanied by offsetting
reductions in other areas of the organization, including many customer-
facing functions. Many, though not all, of these increases in risk
management staffing are directly attributable to the Enhanced
Prudential Standards requirements of the Dodd-Frank Act and other
regulatory requirements that have arisen in the wake of the financial
crisis. We have also embarked on an ambitious program to replace core
software systems, revamp our chart of accounts and establish a data
governance framework and organization in order to ensure our ability to
meet the substantial data requirements necessary to fully comply with
the stress testing and liquidity management protocols applied to SIFIs.
While we will derive ancillary benefits from modernizing our systems,
ensuring regulatory compliance has been a significant factor in our
decision to make these investments which are in the hundreds of
millions of dollars in size. Additional investments have been made in
software systems directly related to compliance with the Enhanced
Prudential Standards. An example is the expenditure of approximately $3
million for software that facilitates compliance with incentive
compensation governance requirements. In addition to the software
investment, thousands of hours have been spent redesigning incentive
plans and validating their compliance with regulatory requirements. We
have also spent millions of dollars on the annual production of
resolution plans, or ``living wills,'' in accordance with requirements
of the Dodd-Frank Act. This is despite the fact that, like other
regional banks, we have a simple organizational structure, with a total
of 20 (mostly very small) subsidiaries, as compared to an average of
1,670 subsidiaries for each of the Nation's six largest banks.
IV. Alternative Means of Designating Systemic Importance
There is no apparent analytical foundation for the Dodd-Frank Act's
establishment of a $50 billion asset size threshold for the
determination of an institution's systemic risk. Indeed, there is a
lack of consistency in applying the Enhanced Prudential Standards of
Section 165 to all insured depository institutions with over $50
billion in assets, with the result that some federally insured
depository institutions with total assets greater than those of my own
bank holding company are not automatically subject to these rules. For
example, USAA, a diversified financial services company with $147
billion in assets, and whose federally insured USAA Federal Savings
Bank subsidiary has over $70 billion in assets, is not subject to the
requirements of section 165, since USAA is not a bank holding company.
Likewise, the Nation's largest credit union, Navy Federal Credit Union,
with $81 billion in assets, is not subject to these requirements.
We are supportive of an approach to the determination of systemic
importance that removes the hard-coded $50 billion asset threshold
currently incorporated in the Dodd-Frank Act, and that substitutes
banking regulators' thoughtful and transparent analysis, consistently
applied, taking into account not only an institution's size, but its
complexity, interconnectedness with the domestic and international
financial system, substitutability, cross-jurisdictional activities and
any other factors the Congress or regulators may deem relevant. We
believe that any such analysis would find that Zions Bancorporation and
most, if not all, other regional banking institutions would not be
found to be systemically important using such an approach, and that the
net benefit to the U.S. economy from redirecting the resources these
institutions currently expend on compliance with section 165
requirements to the prudent extension of credit and other banking
services to customers would be significant.
V. Other Regulations That Retard the Ability of Regional Banks To Serve
Customers and Foster Economic Growth
There are numerous other regulations as well as instances of
regulatory guidance, that hamper (or threaten to impair) the ability of
regional banks to serve the credit and depository needs of their
customers. These include greatly heightened requirements for compliance
with Bank Secrecy Act/Anti-Money Laundering regulations and the
policing of ``our customers' customers''; ambiguous and ever-changing
rules with respect to Fair Lending and other anti-discrimination laws;
and, highly prescriptive and evolving rules with respect to the
governance and oversight of third-party vendor relationships. Two areas
seem to me to be especially worthy of concern.
The first pertains to the incredible thicket of regulations that
has developed around the issuance of residential mortgages. Mortgage
lending has long been subject to a host of laws and regulations. But
the additional layers of regulation emanating from the Secure and Fair
Enforcement for Mortgage Licensing Act of 2008 (SAFE Act), tighter
appraisal standards at a time when there is a nationwide shortage of
qualified appraisers, Dodd-Frank's Ability to Repay and Qualified
Mortgage Standards, and others, has stifled the ability of many banks
to conduct straightforward mortgage operations with traditional
mortgage products--even when the resulting mortgage is held in a bank's
loan portfolio. These issues have been particularly challenging for
self-employed borrowers. In our own case, the cumulative effect of
these many rules has dramatically retarded our ability to originate
mortgage loans in our smaller branches, resulting in a substantial
reduction in the origination of straightforward fixed rate, fully
amortizing mortgages in our branch network in recent years.
A second prospective issue which I believe is deserving of
Congressional focus arises from outside the traditional bank regulatory
establishment, in the form of a new accounting standard on the horizon.
Under the Financial Accounting Standards Board's ``Current Expected
Credit Loss'' impairment standard, slated to take effect in 2020, banks
and other SEC registrants will be required to set aside loss reserves
not only for incurred losses inherent in a loan portfolio, but for all
expected future losses, as well. This will be a challenging accounting
standard for all lenders to implement, not least because it requires
well-documented prognostication about an uncertain future. But the
impact on the economy, and on borrowers in particular, is likely to
arise from the fact that this accounting standard may be expected to
produce the result that lenders will be incentivized to shorten the
tenor of loans, such that the period over which losses must be
estimated is shortened, and required reserves are accordingly reduced.
This would, I believe, provide banks with incrementally more liquid
balance sheets, and lower reserve requirements. But this will not be a
good outcome for borrowers, who will become less liquid with shorter
maturities or face the alternative of higher borrowing costs for
longer-duration loans. This will not be a positive outcome for capital
formation, which is critical to economic growth.
Thank you very much for allowing me the opportunity to present my
institution's views on these important subjects.
______

PREPARED STATEMENT OF GREG BAER
President, The Clearing House Association
June 15, 2017
Chairman Crapo, Ranking Member Brown, and Members of the Committee,
my name is Greg Baer and I am the President of The Clearing House
Association and General Counsel of The Clearing House Payments Company.
Established in 1853, we are the oldest banking association and payments
company in the United States. The Clearing House Association is a
nonpartisan advocacy organization dedicated to contributing quality
research, analysis and data to the public policy debate.
The Clearing House is owned by 25 banks which provide commercial
banking services on a regional or national basis, and in some cases are
also active participants in global capital markets as broker-dealers
and custodians. Our owners fund more than 40 percent of the Nation's
business loans held by banks, which include almost $200 billion in
small business loans, and more than 75 percent of loans to households.
Reflecting the composition of our membership, throughout my testimony,
I will focus on the effects of regulation on U.S. global systemically
important banks, U.S. regional banks of all sizes, and the U.S.
operations of foreign banking organizations with a major U.S. presence.
After nearly a decade of fundamental and continuing changes to
financial regulation, now is an opportune time to review the efficacy
of our current bank regulatory framework. My testimony will focus on
reforms that could directly and immediately enhance economic growth.
Certainly, there are many other areas where reform is urgently needed--
for example, the regulatory regimes for anti-money laundering,
cybersecurity, the Community Reinvestment Act, and corporate
governance, as well as a general breakdown in transparent
administrative procedure at the regulatory agencies--but those involve
other priorities, and have a more indirect effect on the economy.
I should emphasize at the outset that if the goal of regulatory
reform is to prompt economic growth, that goal cannot be achieved while
excluding regulation of large and regional banks from that effort. As
the Treasury Department noted in its report this week, community banks
hold only 13 percent of U.S. banking assets, so reform limited to those
firms will not have a significant economic impact. And large banks--
defined as those in holding companies with at least $50 billion in
assets--originated 54 percent small business loans in 2015 by dollar
amount and 86 percent by number.
I. The Case for Reform of Bank Regulation
Room for reform. The starting point for any review of post-crisis
regulation is an American banking system that is extraordinarily
resilient. U.S. banks now hold substantial amounts of high-quality

capital; since the crisis, the aggregate tier 1 common equity ratio of
TCH's 25 owner banks nearly tripled to 12.2 percent at the end of last
year. In absolute, dollar terms, that is an increase in tier 1 common
equity from $331 billion to over $1 trillion. Similarly, U.S. banks now
hold unprecedented amounts of high-quality liquid assets (HQLA) to
ensure that they can survive a period of persistent liquidity stress (a
run, in other words): today, nearly a quarter of U.S. large bank
balance sheets consists of cash, U.S. Treasury bonds, and similarly
low-risk and highly liquid assets.
Moreover, we now have in place a comprehensive legal and
operational framework that ensures that even the largest and most
complex banks can go bankrupt like any other company, without taxpayer
support and without risk to the broader financial system, ending too-
big-to-fail and replacing moral hazard with market discipline. Markets
clearly have recognized as much, as bank holding company debt is now
priced on the assumption that bondholders will not be bailed out, and
rather will be bailed in in order to recapitalize the institution. \1\
---------------------------------------------------------------------------
\1\ See The Clearing House eighteen53 Blog, ``The Canard That
Won't Go Away: Correcting the Record (Again)'' (April 21, 2017),
available at https://www.theclearinghouse.org/eighteen53-blog/2017/
april/21%20-%20icba.
---------------------------------------------------------------------------
As discussed below, there is considerable evidence that bank
capital and liquidity levels have now been pushed beyond what is
reasonably necessary for safety and soundness and financial stability
purposes. And other restrictions on banking activity have been imposed
without sufficient analysis or evidence--and without regard to current
capital and liquidity levels. Here, and generally, when I refer to
``banks,'' I am including their nonbank affiliates, which increasingly
are now subject to the same restrictions (while providers of financial
services that are not affiliated with banks are effectively
unregulated).
Need for reform. While much has been gained in fortifying the
Nation's largest banks, it is also clear that the banking system is
playing an unnecessarily diminished role in fostering economic growth
and vibrant capital markets, and that systemic risk is building up
outside of the banking system, which has been the sole focus of many
post-crisis reforms. A key driver here is the recent sea change in
banking whereby large and regional banks generally no longer allocate
capital and make business decisions based on their own assessment of
economic risk, with regulatory capital as a backstop; rather, because
regulatory capital requirements are so high and prescriptive,
regulation often dictates how capital--and therefore credit to the
economy--is allocated. \2\ A similar phenomenon is occurring with
respect to post-crisis liquidity requirements.
---------------------------------------------------------------------------
\2\ For example, the recent research by Viral Acharya, et al.,
finds that banks subject to stress tests have reduced the supply of
credit to relatively risky borrowers. In particular, the supply of
credit is reduced to large corporate borrowers that exhibit high risk,
commercial real estate, credit card, and small business borrowers who
also tend to be relatively risky. See Acharya, Viral V., and Berger,
Allen N., and Roman, Raluca A., ``Lending Implications of U.S. Bank
Stress Tests: Costs or Benefits?'' (May 23, 2017). Available at SSRN:
https://ssrn.com/abstract=2972919.
---------------------------------------------------------------------------
As described in detail below, there are numerous opportunities to
better align existing capital and liquidity requirements with the goal
of economic growth--without jeopardizing, and likely enhancing, the
strength and resiliency of the financial system. Three areas of
regulatory impact highlight the significant potential for reform.
Small business lending. As demonstrated in the chart below from the
recent Treasury report, bank lending has lagged significantly in the
current recovery.




[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

Much of the lag is attributable to small business lending. In April
2017, the Federal Reserve published an inaugural nationwide survey of
small business credit conditions, the Small Business Credit Survey
(SBCS), which reports widespread evidence of tight credit conditions
for small businesses. \3\ In particular, according to the results of
the SBCS, approximately 36 percent of small businesses reported not
having all of their borrowing needs satisfied. More specifically:
---------------------------------------------------------------------------
\3\ See Federal Reserve Banks of Atlanta, Boston, Chicago,
Cleveland, Dallas, Kansas City, Minneapolis, New York, Philadelphia,
Richmond, Saint Louis, and San Francisco, ``Small Business Credit
Survey'' (April 2017), available at www.newyorkfed.org/medialibrary/
media/smallbusiness/2016/SBCS-Report-EmployerFirms-2016.pdf.

About 60 percent of small businesses reported having faced
---------------------------------------------------------------------------
financial challenges over the past 12 months.

Of those, approximately 45 percent cited lack of credit
availability or ability to secure funds for expansion as a
reason.

About 75 percent of those firms facing financial
challenges said they used owners' personal funds to address
this problem.

About 45 percent of small businesses applied for financing
over the past 12 months. Of those that applied for credit, 24
percent received none of the funds requested and 36 percent
received only some portion of what they requested.

Notably, credit availability for small businesses is tighter at
large banks that are subject to the highest capital and liquidity
regulations. At these banks, approval rates were just 45 percent for
small businesses with less than $1 million in revenues. In contrast,
community development financial institutions and small banks reported
approval rates of 77 percent and 60 percent, respectively. This fact is
significant because, as noted, large banks originate a sizable share of
small business loans that cannot realistically be replaced by smaller
banks: 54 percent by dollar amount and 86 percent by number of loans.
\4\
---------------------------------------------------------------------------
\4\ See The Clearing House eighteen53 Blog, ``Myth Versus Reality
on Small Business Lending'' (March 24, 2017), available at
www.theclearinghouse.org/eighteen53-blog/2017/march/24-small-business-
lending.
---------------------------------------------------------------------------
Moreover, our own research has shown that the U.S. stress tests are
constraining the availability of small business loans secured by
nonfarm nonresidential properties, which accounts for approximately
half of small business loans on banks' books. Our analysis indicates
that subjecting a bank to the U.S. supervisory stress tests leads to a
reduction of more than 4 percentage points in the annual growth rate of
its small business loans secured by such properties, which translates
to a $2.7 billion decrease in the aggregate holdings of these small
business loans each year on average.
Mortgage lending. Another example of an asset class unnecessarily
burdened by post-crisis regulation is home mortgage lending, and here
again, capital regulation is a major driver. As demonstrated in our own
research, the Federal Reserve's Comprehensive Capital Analysis and
Review (CCAR) stress test is imposing dramatically higher capital
requirements on residential mortgage loans than bank internal (Federal
Reserve-approved) models and the standardized approach to risk-based
capital developed by the Basel Committee on Banking Supervision. \5\
Indeed, for first-lien mortgage loans, CCAR capital requirements are 45
percent higher than under banks' own projections and 95 percent higher
than under the Basel III standardized approach. Because smaller banks
are subject to less stringent capital requirements, they can act as a
control group in assessing the impact of new regulations on the supply
of credit. Between the fourth quarter of 2010 and the end of 2016,
residential real estate loans declined 0.5 percent on average on an
annual basis at banks subject to the CCAR stress test, while they rose
4.0 percent on average over the past 6 years on an annual basis at
banks not subject to that test.
---------------------------------------------------------------------------
\5\ See The Clearing House, ``The Capital Allocation Inherent in
the Federal Reserve's Capital Stress Test'', https://
www.theclearinghouse.org//media/TCH/Documents/TCH%20WEEKLY/2017/
20170130_WP_Implicit_Risk_Weights_in_CCAR.pdf.
---------------------------------------------------------------------------
Capital markets. In the United States, much of the lending to the
private nonfinancial sector and most of the borrowing by the Government
sector occurs outside the banking system, in capital or money markets.
Indeed, banks provide only about one-third of credit in the United
States. Large bank holding companies facilitate financial market
intermediation both by making markets in securities traded in those
markets and by providing funding to other market participants who
transact in those markets.
Interestingly, post-crisis regulation by banking regulators has
affected securities markets more than regulation by securities
regulators. In particular, bank regulations have made it significantly
more expensive for broker-dealers affiliated with banks--which
includes, post-crisis, all of the largest dealers--to hold, fund, and
hedge securities positions. Higher capital charges make holding of
inventory more expensive, and the Volcker Rule makes holding such
inventory a potential legal violation. The surcharge for global
systemically important banks (GSIBs) and liquidity rules make
securities financing more expensive. It has become more difficult for
dealers to hedge the risk associated with holding the inventories of
the bonds using credit default swaps. \6\
---------------------------------------------------------------------------
\6\ Recent research by economists at the Federal Reserve Bank of
New York has found that CDS have become much more costly to hold in
large part because of the capital that dealers are required to hold
against the transaction. Boyarchenko, Nina, Pooja Gupta, Nick Steel,
Jacqueline Yen, (2016) ``Trends in Credit Market Arbitrage'', Federal
Reserve Bank of New York Staff Reports No. 784, July 2016, p. 18.
---------------------------------------------------------------------------
The greatest impact has been felt by smaller companies, as the
capital rules impose lower capital charges on more liquid securities,
which tend to be issued by larger companies; broker-dealers, forced to
ration their balance sheets, are serving their largest customers first.
As shown in the chart below, issuance of corporate bonds by small and
midsized nonfinancial firms has fallen over the past few years while
issuance by larger firms has risen.



[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

As another example, the efficiency and liquidity of financial
markets are maintained by the ability of asset managers to take
leveraged positions in mispriced assets to earn a profit when the asset
price returns to normal. Such positions are financed in the market for
repurchase agreements. Broker dealers are often the intermediary
between two financial institutions, engaging in a repo with one and an
identical matched repo with another. While such matched transactions
are nearly riskless, the leverage ratio requirement forces banks to
hold considerable capital against their reverse repos. Moreover, if the
net stable funding ratio were adopted as proposed, banks would be
required to finance the loans with a material amount of longer-term
funding rather than a matched repo borrowing. As explained in a recent
TCH research note, these types of requirements make such transactions
more expensive, and dealers are passing those costs along. \7\
---------------------------------------------------------------------------
\7\ See The Clearing House, ``Shortcomings of Leverage Ratio
Requirements'', (Aug. 2016), www.theclearinghouse.org/-/media/tch/
documents/20160809_tch_research_note
_leverage_ratio.pdf?la=en.
---------------------------------------------------------------------------
Thus, more than four-fifths of the respondents to the Federal
Reserve's Senior Credit Officer Opinion Survey in June 2015 indicated
that liquidity and market functioning in Treasury markets had
deteriorated. Over 80 percent of those respondents reporting a
deterioration indicated that the most important cause was a decreased
willingness of securities dealers to expand their balance sheet for
market-making purposes as a result of regulatory change.
II. Reforming Capital Regulation
A. Stress Testing
When enough should be enough. Certainly, a key lesson of the
financial crisis is the critical importance of maintaining capital
levels sufficient to absorb outsized losses that typically accompany
periods of financial stress. Responding to that lesson, banks have
significantly increased the amounts of high-quality capital they
maintain, and regulators have enacted a range of reforms that require
these heightened levels of capital to remain in place over time.
Implementation of Basel III changes has increased the quality of
capital, focusing on common equity as opposed to hybrid debt/equity
instruments. In the United States, there is an increased and wise
emphasis on stressed rather than static measures of capital adequacy--
in particular, the Federal Reserve's CCAR exercise and the banks' own
internal stress tests. These are important improvements to the bank
capital framework that resolve key shortcomings revealed by the
financial crisis, and we support them.
Unfortunately, these sensible reforms have been accompanied by
other changes to the U.S. capital framework which have introduced a
significant degree of unnecessary opacity, subjectivity and uncertainty
to capital regulation in the United States. Large U.S. banks today are
subject to dozens of different capital requirements. Of those, the
Federal Reserve's CCAR stress test and the enhanced supplementary
leverage ratio (eSLR) are set at such high levels that they most
frequently dictate bank's decision making. In addition, U.S. regulators
have consistently implemented capital reforms in a manner that both
significantly exceeds agreed-upon international standards and is much
more stringent than necessary to support safety, soundness, and
financial stability.
Of course, a crucial question is how much capital is enough. TCH's
25 owners hold roughly triple the amount of capital they did pre-
crisis, but should it be quadruple, or double? We believe three
benchmarks are useful here. First, consider the results of the Federal
Reserve's severely adverse scenario under CCAR, which presents for
large banks a greater economic and market shock than was present in the
global financial crisis. Then, compare the losses projected under that
stress scenario to the loss absorbency currently held by those banks,
as detailed in the following chart.









[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

In sum, CCAR imposes a stress scenario significantly harsher than
the previous financial crisis. \8\ Yet as of 2016 tangible common
equity was five times the losses implied under that scenario. Total
loss absorbency--which includes debt holders who would be ``bailed in''
as part of a bankruptcy under the Title I living will process or the
Title II Orderly Liquidation Process, was ten times those losses.
---------------------------------------------------------------------------
\8\ It is worth noting that neither banks nor their regulators
place exclusive focus on a single scenario; rather, banks run, and the
Federal Reserve monitors, numerous stress scenarios, including ones
chosen by each bank to focus on its unique vulnerabilities.
---------------------------------------------------------------------------
Consider as a second benchmark JPMorgan Chase, which is universally
considered to have had sufficient capital to have weathered the past
financial crisis without need for taxpayer assistance, while making two
acquisitions and continuing to lend and make markets. Thus, one could
reasonably suppose that the amount of capital it held pre-crisis was
sufficient (and would have been all the more sufficient if all other
firms had held that amount as well). Today, JPMorgan Chase holds double
the capital it did pre-crisis. More importantly, all large banks are
now required to hold similar levels of capital (with some variation
based on the size of any GSIB surcharge). And the firms subject to
those capital rules today include the largest broker dealers--which is
significant, because pre-crisis, monoline investment banks like Bear
Stearns and Lehman Brothers were not subject to bank-like capital
requirements and operated with a fraction of the capital of large
banks.
And consider as a third benchmark long-term debt spreads and CDS
spreads of large U.S. banks, which have remained stable over the past
five years. While we have not seen a significant financial crisis
during this period, we observed a large trading loss at one large U.S.
bank in mid-2012, volatility around the Brexit vote in the United
Kingdom in the middle of last year, a significant consumer scandal at
another large U.S. bank in the second half of last year, and more
generally, a fair amount of international political instability in
recent months.
There is also reason to believe that higher capital standards have
reached levels at which they are having a counterproductive effect. In
a recent paper, Sarin and Summers (2016) point out that by several
capital markets-based measures, including stock price volatility and
CDS spreads, banks appear to be riskier now than they were before the
crisis, even as bank capital and liquidity standards have been
substantially raised over that same period of time. \9\ The authors
conclude that the most likely explanation is that banks' franchise
values have declined. Specifically, a bank's franchise value depends on
its ability to generate earnings and increase those earnings over time.
The tightening of regulations that has occurred since the crisis, while
increasing loss absorbency, has also reduced the profitability of
banks.
---------------------------------------------------------------------------
\9\ See Sarin, Natasha, and Lawrence H. Summers; ``Have Big Banks
Gotten Safer?'', Brooking Papers on Economic Activity, Fall 2016.
---------------------------------------------------------------------------
While no one would recommend a return to the low and uneven capital
levels that existed pre-crisis, or to treating as capital hybrid
instruments that did not prove to be loss absorbing, the largest U.S.
banks are now overcapitalized by any objective measure. Hundreds of
billions of trapped capital is not necessary to meet any quantifiable
safety and soundness need, and could be redeployed to furthering
economic growth--either through more lending or returning excess
capital to shareholders for reinvestment elsewhere.
Potential for reform. The Federal Reserve's stress testing
framework attempts to measure the ability of banks to withstand a very
severe economic downturn (and, where relevant, a market shock). Under
CCAR, the Federal Reserve runs its own proprietary models to determine
the effect of various supervisory scenarios on banks' capital
adequacy--that is, the estimated net losses and resulting reduction in
capital under those scenarios. After this stress, a large bank must
meet a series of capital requirements, including a 4.5 percent common
equity tier 1 ratio. And it must do so assuming that it does nothing to
shrink its balance sheet, reduce its dividend, or postpone planned
share repurchases under severely adverse economic conditions--almost
certainly deeply counterfactual assumptions. Thus, a large bank that
passes the CCAR exercise not only has sufficient capital to avoid
failure under historically unprecedented adverse conditions--it has
enough capital to emerge from such an event doing business as usual,
and without taking actions that would be normal (or even compelled)
under the circumstances.
Stress testing is an important tool for assessing the health of the
banking system because it incorporates a forward looking, dynamic
assessment of capital adequacy, and is therefore less reliant on recent
historical performance. However, the Federal Reserve's CCAR stress
tests are highly and unnecessarily opaque, relying upon macroeconomic
scenarios that are never published for public comment and a series of
unidentified models (combined in unspecified ways) that have never been
subject to peer review or public comment.
To the best of our knowledge, which is necessarily limited by the
opacity of the CCAR process, the accuracy of the Federal Reserve's
models, individually and collectively, has never been back-tested. The
results of this nonpublic process continue to differ markedly from the
results of the banks' own, more robust earnings forecasting models--
models that the Federal Reserve itself subjects to rigorous review.
(The bank process is part of what is known as DFAST, short for ``Dodd-
Frank Act Stress Test''.) At this point, there is no basis to conclude
that the Federal Reserve's models do a better job of projecting losses
than the banks' own (Federal Reserve-approved) models.
Both the quantitative test of CCAR and the qualitative test
described below also are needlessly complex and consume enormous
resources at the largest banks, which resources could be more
effectively redeployed; the CCAR annual submissions for the largest
banks average in excess of 50,000 pages.
Effects on economic activity. Collectively, the opacity,
subjectivity, and overall stringency of the CCAR framework act as a
significant constraint on lending, economic growth, and liquid capital
markets. As we have demonstrated in detailed empirical research, this
is largely the result of the excessively high amounts of capital banks
are required to hold against their small business lending, mortgage
lending, and trading book assets to pass the test. \10\ Under banks'
own DFAST projections, capital requirements for small business loans
and home mortgage loans are 80 percent and 45 percent higher than under
the Basel III standardized approach, respectively. For trading assets,
the higher capital requirements under CCAR are driven by the Federal
Reserve's prescribed global market shock that is part of the CCAR
scenarios for banks with large trading operations. However, the market
shock also applies to the DFAST stress tests that are calculated using
the banks' own models, and the capital requirement for the trading book
under CCAR is 20 percent higher than DFAST. \11\
---------------------------------------------------------------------------
\10\ Capital Allocation in CCAR, supra note 5.
\11\ See id.
---------------------------------------------------------------------------
CCAR's excessively high capital requirements for small business
loans and home mortgages likely reflect in large part the severity of
the stress scenario used in the test. The stress test includes
increases in unemployment that are more sudden in some cases more
severe than seen in the global financial crisis, and other parameters
that go beyond any historical experience.
The inevitable result is that banks are shifting away from
cyclically sensitive sectors (where loss of employment is likely to
trigger default) like small businesses and households with less-than-
pristine credit. Bank behavior is consistent with this set of
incentives:

Small commercial real estate loans, which account for
approximately half of small business loans outstanding on
banks' books, declined about 2 percent on average over the past
5 years. \12\
---------------------------------------------------------------------------
\12\ Capital Allocation in CCAR, supra note 5.

On the residential real estate lending side, home equity
lines of credit declined more than 6 percent per year over the
past 5 years, despite the significant appreciation in housing
prices, and are about 110 basis points more expensive than they
were pre-crisis. \13\
---------------------------------------------------------------------------
\13\ Capital Allocation in CCAR, supra note 5.

The declines in these categories of lending have been
larger at banks subject to CCAR than at banks not subject to
---------------------------------------------------------------------------
CCAR.

Substantial benefits to economic growth could be achieved through
three limited reforms to CCAR, all of which would increase banks'
capacity and propensity to make these types of loans.
First, banks' more robust, Federal Reserve-approved models should
be used to estimate stress losses for purposes of the CCAR quantitative
assessment. The Federal Reserve should use its own, more simplified
models as a check on the bank models. With the Federal Reserve models
no longer binding in the first instance, no concentration of risk or
``gaming'' concern would prevent their being made transparent. Notice
and comment or other peer review would doubtless improve their
accuracy.
We note that such an approach is not a theoretical construct, but
current practice at the Bank of England where banks' own models play
the central role in the United Kingdom's supervisory stress tests. \14\
Banks use of their own models motivates them to better develop their
own stress scenarios, which are than more tailored to their business
models. That said, the Bank of England does not rely entirely on banks
own models and has its own suite of models for peer-benchmarking and to
ensure consistency of results across participating banks. In adopting
the system, the Bank of England has noted that it does not want its own
models to drive capital requirements at the risk of stifling stifle
innovation in risk management at banks. More generally, if a set of
unique models being used is overly conservative, the efficiency of the
financial system would be reduced. Conversely, if those models are
vulnerable to a particular source of risk, the entire system could be
undercapitalized during a period of financial stress. \15\
---------------------------------------------------------------------------
\14\ See Bank of England (2015) ``The Bank of England's Approach
to Stress Testing the U.K. Banking System'' (October 2015). Available
at http://www.bankofengland.co.uk/financialstability/Documents/
stresstesting/2015/approach.pdf.
\15\ See Gallardo, German, Til Schuermann, and Michael Duane (May
2016), ``Stress Testing Convergence'', Journal of Risk Management in
Financial Institutions 9, pp. 32-45.
---------------------------------------------------------------------------
Second, annual stress test scenarios should be subject to a 30-day
public notice and comment period to ensure that they meet the Federal
Reserve's identified standard--consistency with post-war U.S.
recessions. While we believe that standard is sensible, it should be
subjected to notice and comment rulemaking.
Third, counterfactual and incorrect assumptions about how banks
would behave in a crisis (e.g., continuing share repurchases and
balance sheet growth under severe stress) should be corrected.
B. Leverage Ratio
A leverage ratio measures the capital adequacy of a bank by
dividing its capital by its total assets (and, in some cases, off-
balance-sheet exposures) without taking into account the risk of any
particular asset or exposure. Requiring the same amount of capital to
be held against every asset makes the holding of low-risk, low-return
assets relatively more costly when compared with the holding of higher-
risk assets, higher-return assets. Put another way, if a capital
regulation requires a bank to hold the same amount of capital against
each asset, the bank will by necessity gravitate to relatively higher-
risk, higher-return assets.
A leverage ratio can still be a useful tool as a backup measure
when banks collectively misunderstand the risk of a certain asset class
(as they did with mortgages and mortgage-related securities in the past
crisis), but serious problems have emerged for U.S. banks because U.S.
regulators have set the minimum leverage ratio for the largest U.S.
banks at nearly double the international standard, without adequate
analysis of (i) whether such a high leverage ratio is necessary to
prevent excessive risk taking or (ii) the impact of such a high
leverage ratio on lending, market activity and economic growth. These
are the very same banks that provide support to U.S. capital markets
and ensure the safekeeping of investor assets, and in the course of
doing so hold large amounts of low-risk, liquid assets like central
bank placements and Treasury securities.
More specifically Basel III introduced a 3 percent supplementary
leverage ratio for internationally active banks, which includes both
on-and off-balance-sheet assets. \16\ U.S. regulators have not only
applied this 3 percent supplementary leverage ratio requirement to all
larger banks, but have also imposed a still-higher requirement for U.S.
GSIBs--an eSLR of 5 percent at the holding company level and 6 percent
at depository institution subsidiaries. \17\ Consequently, for several
of the largest U.S. banks, the eSLR, as opposed to a risk-based
requirement, that acts as a current or potential future binding
constraint and therefore affects bank capital and business planning.
\18\
---------------------------------------------------------------------------
\16\ See Basel Committee on Banking Supervision, ``Basel III: A
Global Regulatory Framework for More Resilient Banks and Banking
Systems'' (June 2011), available at www.bis.org/publ/
bcbs189_dec2010.pdf.
\17\ See 79 FR 187.
\18\ See Federal Financial Analytics, Inc., ``Mutual-Assured
Destruction: The Arms Race Between Risk-Based and Leverage Capital
Regulation'' (Oct. 13, 2016).
---------------------------------------------------------------------------
The overall impact of the leverage ratio as a measure of capital
adequacy, and the resulting misallocation of capital, have increased
dramatically in recent years as a result of other regulatory mandates.
As noted, large banks presently are required by liquidity regulations
to hold about a quarter of their balance sheets in high quality liquid
assets (HQLA)--predominantly cash, Treasury securities and other
Government securities. Large banks now hold approximately three times
as much of these assets as they did pre-crisis. Those assets rightly
receive a zero or low risk weight in risk-based capital measures, but
the leverage ratio completely ignores their actual risk and requires
banks to hold capital against these assets.
Banks with sizeable custody, treasury services or other businesses
that employ a servicing business model or take sizeable corporate
deposits are particularly affected. In practice, this means that, under
the liquidity rules, these banks must hold cash or Treasury securities
against these deposits, on the assumption that up to 100 percent of
them will run in a crisis (although the outflow rate during the
financial crisis was substantially lower) and then hold 6 percent
capital against the same cash and U.S. Treasury bills that the
regulators require they hold for liquidity purposes. Of systemic
concern, these problems are likely to become more pronounced in periods
of financial market uncertainty, as institutional investors seek to
lower their risk exposure by raising cash and banks must manage the
resulting deposit inflows in the most conservative way possible, via
placements at the Federal Reserve and other national central banks.
Another issue that has received recent notice is how the
supplementary leverage ratio makes it more costly for U.S. banking
organizations to provide clearing member services to clients on
centrally cleared derivatives. While risk-based capital rules allows
banking organizations to exclude from their denominator any initial
margin posted by their clients on derivatives transaction--and rightly
so, as the bank bears no risk of loss on such margin--the leverage
ratio does not. As a result, the leverage ratio exaggerates the
exposure amount of these derivatives and effectively requires banks to
hold un-economic amounts of capital when providing clearing services to
clients. Because of this, at least three major dealers have exited the
business. Accordingly, former CFTC Chairman Massad called for the U.S.
leverage ratio to be amended to take account of segregated margin. \19\
---------------------------------------------------------------------------
\19\ See Timothy Massad, Keynote Address by Chairman Timothy G.
Massad before the Institute of International Bankers (March 2, 2015),
available at www.cftc.gov/PressRoom/SpeechesTestimony/opamassad-13.
---------------------------------------------------------------------------
In sum, under the eSLR, U.S. GSIBs are currently required to trap
approximately $53 billion in capital against cash reserve balances
deposited at the Federal Reserve, and an additional $15 billion against
U.S. Treasury securities. These are assets whose value banks are at no
risk of misjudging; capital allocated to them could be far better
deployed to lending or supporting market liquidity. Thus, the answer is
not to dispense with the leverage ratio but rather to eliminate the
enhanced supplementary leverage ratio, and to deduct from the
denominator of the supplementary leverage ratio high-quality liquid
assets like central bank reserves and Treasury securities, as well as
segregated client margin.
It is sometimes said that deducting these assets would begin a
``slippery slope.'' This worry is difficult to understand--bank
regulation is replete with line drawing. For example, the liquidity
coverage ratio gives 100 percent credit for a central bank reserve or
U.S. Treasury security as a liquid asset; this has not created a
``slippery slope'' whereby loans have been given 100 percent credit as
a liquid asset. The Bank of England, on July 25, 2016, began deducting
central bank reserves from the leverage ratio denominator for U.K.
banks--and no ``slippery slope'' has emerged whereby it has felt the
need to do so for, say, subprime loans. \20\
---------------------------------------------------------------------------
\20\ See Bank of England, ``Financial Policy Committee Statement
and Record From Its Policy Meeting'', July 25, 2016 (August 2016),
available at www.bankofengland.co.uk/publications/Pages/news/2016/
062.aspx.
---------------------------------------------------------------------------
C. Operational Risk
Large institutions currently are required to build and maintain
models to measure operational risk for capital purposes based on a
Federal Reserve-approved Advanced Measurement Approach. Because it is
exceedingly difficult to base a capital charge on a subjective
assessment of the risk inherent in a bank's current operations, these
models generally look at past large litigation losses and treat them as
a proxy for the risk of something going wrong in the future.

In contrast to international peers, the U.S. banks are often
prohibited from excluding losses from their models even when the bank
has exited the business line that caused the loss, or sold such
business to another institution. (The acquirer also assumes the capital
charge associated with the past event, effectively doubling the capital
requirement on an aggregate basis.) U.S. banks are prohibited from
using expert judgment to lower the output of their model even when
factors make certain operational losses less likely in the future,
while non-U.S. banks are permitted to make such adjustments. Similarly,
banks may put in place a range of other risk mitigants, such as
insurance or hedges, but none of these are meaningfully recognized or
reflected in the current operational risk capital framework. Finally,
for some banks, the regulators add to any modeled results a
``supervisory overlay,'' which is a completely arbitrary add-on
presented with no analytical or evidentiary basis.
As a result of all these factors, operational risk capital charges
are inflated and extremely sensitive to any data anomalies or extreme
events. At least one bank was reported in 2014 to be holding over $30
billion in operational risk capital, \21\ and as a general matter, U.S.
banks currently hold significantly more operational risk capital than
their international counterparts. It is also worth noting that
operational risk losses tend to be idiosyncratic and thus uncorrelated,
so extraordinarily high capital is being held against a risk that is
unlikely to be systemic. (Clearly, some operational risks of the
mortgage business did prove correlated during the financial crisis;
however, even here, those losses were experienced years after the
associated credit and market losses.)
---------------------------------------------------------------------------
\21\ See Peter Rudegeair, Reuters, ``Trading Scandals, Legal Fines
May Ramp Up U.S. Banks' Capital Needs'' (June 9, 2014), available at
www.reuters.com/article/us-banks-oprisk-idUSKBN0EK1M520140609.
---------------------------------------------------------------------------
An ongoing Basel Committee review of operational risk capital could
rectify these problems if an improved regime could be constructed
appropriately and--importantly--adopted by U.S. regulators without
``gold plating.'' The result would be to free up billions of dollars of
capital for more productive uses.
III. Reforming Liquidity Regulation
A. Liquidity Coverage Ratio
A key lesson of the financial crisis was the need for banks to
maintain sufficient liquidity to survive periods of financial stress.
The regulatory response includes Basel III's liquidity coverage ratio
(LCR), which requires banks to maintain a sufficient stock of liquid
assets to cover a 30-day run on the bank with no access to additional
funding, plus a Dodd-Frank Act requirement that large banks conduct
liquidity stress tests on a monthly basis across at least overnight,
30-day, 90-day, and 1-year time horizons, and maintain a sufficient
``liquidity buffer'' based on their expected liquidity needs under
these stress tests. These are concrete improvements to the bank
liquidity framework, which we generally support.
These regulations have dramatically increased the ratio of HQLA to
total assets in the U.S. banking sector. The largest 33 banks held 12
percent of their assets in HQLA in 2008; today they hold 24 percent of
their balance sheets in these assets. Compared to the onset of the
crisis, this improvement is even more pronounced, with the proportion
of HQLA increasing nearly five times since the end of 2006 (i.e., from
5.75 to 24 percent). \22\ The question, of course, is whether this
large an expansion of bank balance sheets is necessary, and whether it
is having unintended effects.
---------------------------------------------------------------------------
\22\ See The Clearing House, State of American Banking (2016) at
exhibit 3 (updated).
---------------------------------------------------------------------------
Although the LCR is conceptually sound, in practice it makes
assumptions about which liabilities will run, and which assets can be
sold, that a TCH study shows have no empirical bases and appear
inconsistent with even crisis-era experience. \23\ For example, while
the LCR assumes that 30 percent of liquidity lines of credit provided
to nonfinancial corporations in a future 30-day period of systemic and
idiosyncratic stress would be drawn, the highest draw on such lines at
large commercial banks (including several that failed or nearly failed)
over any month in the financial crisis was 10 percent. While the LCR
assumes that 100 percent of the nonoperational deposits of financial
institutions would be drawn, the worst experience during the crisis was
38 percent. \24\
---------------------------------------------------------------------------
\23\ See The Clearing House, ``The Basel III Liquidity Framework:
Impacts and Recommendations'' (Nov. 2, 2011), available at
www.theclearinghouse.org/-/media/files/association
%20documents/20111102_liquid_liquidity%20coverage%20ratio.pdf.
\24\ Id.
---------------------------------------------------------------------------
These seemingly arcane calibration errors have major real-world
consequences. In recent years, U.S. companies of all sizes have
complained that standby letters of credit are unavailable, or more
expensive and difficult to obtain. A major reason is because banks must
assume that in crisis those lines will be drawn in amounts three times
greater than even the worst historical experience would indicate, and
therefore hold cash or cash-equivalent assets to fund those draws. And,
in turn, under the leverage ratio, they must hold significant amounts
of capital against those riskless or low-risk assets.
B. Net Stable Funding Ratio
The net stable funding ratio (NSFR) is intended to establish a
maximum safe amount of liquidity transformation that a bank can engage
in by ensuring that banks have sufficient ``sticky'' liabilities to
fund assets that would be unable to liquidate easily over a 1-year
horizon. When the NSFR was first proposed by the Basel Committee in
2009, the metric was designed to ensure that a bank with an NSFR
greater than 100 percent would be able to weather a 1-year episode of
idiosyncratic liquidity stress. The NSFR thereby was meant to be a
complement to the LCR requirement, which was designed to ensure that a
banking organization could weather a shorter (30-day) but more severe
period of stress.
In those initial formulations of the NSFR, the ``extended stress''
was defined by specific characteristics--for example, ``a potential
downgrade in a debt, counterparty credit or deposit rating by any
nationally recognised credit rating organisation.'' That benchmark was
not included in the final NSFR standard released by the Basel
Committee, or in the proposed rule to implement the NSFR in the United
States. Nor was any other benchmark included, making it unclear what
goal(s) the NSFR is intended to achieve and how it was calibrated.
Moreover, for U.S. banks already subject to the LCR, uniquely
stringent liquidity stress testing under the Dodd-Frank Act
requirements, a Comprehensive Liquidity Analysis and Review and a U.S.-
only short-term-wholesale funding surcharge as part of the GSIB
surcharge, it is unclear what additional risk the NSFR would mitigate
that is not sufficiently addressed by these requirements.
The NSFR, if implemented, would significantly inhibit economic
growth and liquid financial markets due to its flawed design and lack
of transparency with respect to its calibration to ensure its
efficiency and effectiveness. As demonstrated in our research, The Net
Stable Funding Ratio: Neither Necessary nor Harmless, over time, the
NSFR, if implemented in the United States, could be expected to
significantly limit lending and capital markets activity. \25\ If
central bank reserve balances and retail deposits shrink in line with
the Fed's forecast for policy normalization, and banks shift their
funding toward wholesale deposits in line with historical experience
many individual banks would not comply unless they took some
compensating action. In particular, we show that the annual growth in
bank lending would have to be cut by about 3.5 percentage points, to
near zero, even to offset only half of the projected decline in the
NSFR.
---------------------------------------------------------------------------
\25\ See The Clearing House eighteen53 Blog, ``The Net Stable
Funding Ratio: Neither Necessary Nor Harmless'' (July 5, 2016),
available at www.theclearinghouse.org/eighteen53-blog/2016/july/05-
research-note-nsfr.
---------------------------------------------------------------------------
IV. Reforming the Bank Living Will Process
Title I of the Dodd-Frank Act requires each large bank holding
company to construct a plan for its rapid and orderly resolution, and
requires regulators to review the credibility of that plan.
Regulators have required bank holding companies to file living
wills on an annual basis, against ever shifting, often nonpublic
standards, even though the regulators have been generally unable to
review them and provide feedback within that timeframe. Recognizing
that section 165(d) of the Dodd-Frank Act requires the submission of
livings wills on a ``periodic,'' not annual, basis, an appropriate and
sensible approach is to eliminate the formal requirement for an annual
submission in favor of submission cycle that is better tailored to the
objectives of the living will process.
The Federal Reserve and FDIC also have required, through the living
will process, substantial amounts of liquidity and capital to be pre-
positioned--and therefore, trapped--at numerous subsidiaries. The most
recent living will guidance issued in April 2016 states that bank
holding companies must assume, counterfactually, that a net liquidity
surplus in one material entity cannot be transferred to meet liquidity
deficits at another material entity (even between branches of the same
banking legal entity). Further, the guidance also requires bank holding
companies to assume that cash balances held by material entities
(including branches of the bank) within their primary nostro accounts
with the main bank entity of the firm are unavailable in a stress prior
to, and during resolution. The guidance imposes similar requirements
with respect to pre-positioning of loss absorbing capital resources at
material entities. None of this guidance has been published for notice
and comment. Reform here could take the form of a statement that for
any firm using the single-point-of-entry resolution strategy and in
compliance with the TLAC requirement for holding company loss
absorbency, the living will process should not include any incremental
liquidity requirement at the operating subsidiary level; for all firms,
we would recommend withdrawing the presumption that liquidity cannot be
transferred among subsidiaries.
Currently, each Federal (and State) banking agency is authorized to
impose its own set of recovery and resolution planning requirements on
different parts of a banking organization, leading to an unnecessary
amount of duplicative and at times contradictory requirements. Many of
these requirements were not subject to a rigorous impact analysis, and
are not appropriately tailored. This may also reinforce ring fencing of
entities as bank regulators focus only on the entities for which they
are responsible.
We would recommend eliminating the separate insured depository
institution-level resolution and recovery planning regimes. At a
minimum, the agencies should be required to coordinate among themselves
to establish a single set of consistent recovery and resolution
planning requirements.
V. Reforming Activity Limitations
Post-crisis regulation has included not only capital and liquidity
regulation to reduce the risk of bank failure to the taxpayer and the
broader system, but also direct limits on bank activities--however well
capitalized and funded they are. In some cases, these limits are
unjustified.
A. Leveraged Lending
Leveraged lending is an important type of financing for growing
companies, which tend to carry a lot of debt. Although these companies
therefore represent a greater repayment risk than more established
firms, this risk is one that banks have considerable experience
managing. Banking organizations have long played a critical role in
arranging, originating, and administering funding for leveraged loans
as part of their larger role as credit intermediaries. Following the
financial crisis, capital requirements have increased significantly for
such loans, as have requirements for modeling their risks.
Nonetheless, the Federal banking agencies have issued guidance
setting arbitrary limits on such lending, based on no empirical
evidence--in particular, any evidence that the capital supporting such
activity is somehow inadequate. It is a classic example of bank
regulators substituting their judgment for lenders and markets without
any meaningful analysis or evidence. For example, regulators require
banks, in evaluating whether a company is leveraged for purpose of the
new restrictions, to assume that all lines of credit are drawn, and to
ignore cash held by the company. As a result, some Fortune 500
companies with investment grade debt are now deemed by the regulators
to be highly leveraged, and thus subject to limits on their bank
borrowing.
It also appears that this guidance has been supplemented by further
direction, from examiners to banks, to limit lending activities in the
area--although the precise details of such direction are unknown as
they are deemed by the agencies to be ``confidential supervisory
information,'' and therefore immune to public scrutiny.
Of course, these loans are subject to capital requirements, and the
regulators have not identified any flaws in those standards (including,
in the case of the Federal Reserve, its own CCAR models) that would
cause these types of loans to be uniquely undercapitalized. Nor have
the agencies presented any data to show that there is an unhealthy
concentration of these loans in the banking system. Also, consistent
with post-crisis behavior in a range of areas, the banking agencies
have implemented these substantial new limits on bank lending through
guidance and ``frequently asked questions,'' rather than formal notice-
and-comment rulemaking and regulation. They have nonetheless deemed the
guidance binding, and enforced it just as if it were a rule.
As a result of the leveraged lending guidance and examiner
pressure, banks have been forced to turn away hundreds of millions,
perhaps billions, of dollars of loans to growing businesses.
Furthermore, there is scant evidence that leveraged lending guidance
and subsequent direction have constrained the risk perceived by the
Federal banking agencies. Despite any potential concerns regarding
poorly underwritten or low-quality loans, a bank-centric approach is
simply shifting risk rather than limiting it, and increasing the cost
to borrowers, as banks tend to be lower cost providers of credit.
Recent research by a team of Federal Reserve Bank of New York
economists illustrates that the guidance has had the effect of reducing
bank activity in this area, but has also increased nonbank activities,
demonstrating limited effectiveness from a macroprudential view. \26\
Notably, those nonbanks do not appear to be experiencing the outsized
losses that the bank regulators implicitly predicted in forcing banks
to abandon much of this lending.
---------------------------------------------------------------------------
\26\ Sooji Kim et al., Liberty Street Economics, ``Did the
Supervisory Guidance on Leveraged Lending Work?'' (May 2016), available
at http://libertystreeteconomics.newyorkfed.org/2016/05/did-the-
supervisory-guidance-on-leveraged-lending-work.html.
---------------------------------------------------------------------------
We recommend that the guidance be rescinded immediately and in its
entirety, which would provide an immediate boost to economic growth as
a large number of growing companies once again became eligible for bank
credit.
B. The Volcker Rule
Section 619 of the Dodd-Frank Act (commonly referred to as the
``Volcker Rule'') generally prohibits U.S. insured depository
institutions, U.S. operations of foreign banks and their affiliates
from engaging in ``proprietary trading'' and sponsoring or investing in
hedge and private equity funds subject to some limited exceptions,
including exceptions for customer-related activities such as market-
making. Prior to the enactment of the Volcker Rule, very few of the
firms now subject to the Rule engaged in proprietary trading
activities. Of those that did, many of them were in the process of
divesting or ceasing their proprietary trading activities. Today,
trading businesses of covered financial institutions are focused solely
on serving client needs and hedging the attendant risk.
The final regulations implementing and interpreting the Volcker
Rule are voluminous and complex, contained in 964 pages, including an
893 page preamble. Under these rules, the five U.S. Federal financial
agencies charged with implementing and enforcing the Volcker Rule have
interpreted it in a highly restrictive way, with a broad spectrum of
trading activity (i.e., not only short-term, speculative activities
that the Volcker Rule was intended to target) presumed to be prohibited
proprietary trading unless proven otherwise.
Market-making. Under the rules, a covered banking entity is
required to go to extraordinary lengths to prove that its routine
market making and underwriting activities (included related hedging) do
not constitute ``proprietary trading.'' The agencies have adopted a
broad definition of proprietary trading with strict requirements for
permissible activities that could potentially captures legitimate
market making in less liquid securities, particularly when markets are
under stress and there is less demand. For example, banking
organizations are required to strictly limit their inventory to the
reasonably expected near-term demand of customers or counterparties.
For debt of smaller companies, which may trade only weekly or even
monthly (especially during times of stress), banking organizations may
be required to unduly limit their positions, thus prohibiting them from
taking any action to stabilize markets.
Recent research has begun to bear out longstanding reports from
market participants that the regulations that implement the Volcker
Rule are inhibiting economic growth and reducing market liquidity by
constraining the ability of banking groups to buy, sell and underwrite
securities, including corporate bonds that could help finance the
operations of corporate customers. A Federal Reserve staff study
released in December 2016, The Volcker Rule and Market-Making in Times
of Stress, finds that the illiquidity of stressed bonds has increased
after the Volcker Rule, as dealers regulated by the rule have decreased
their market-making activities. \27\ Other research indicates that with
many brokers constrained in their ability to hold inventory as a result
of the Volcker Rule and other post-crisis regulations, the secondary
market for smaller issuers' debt has tightened. The impact is that
since the enactment of the Dodd-Frank Act in 2010, new debt issuances
by smaller firms has generally declined. When lower liquidity puts debt
markets out of reach of smaller firms, it impedes their ability and the
economy at large to grow.
---------------------------------------------------------------------------
\27\ See Jack Bao, Maureen O'Hara, and Alex Zhou (2016), ``The
Volcker Rule and Market-Making in Times of Stress, Finance and
Economics Discussion Series 2016-102''. Washington: Board of Governors
of the Federal Reserve System, available at https://doi.org/10.17016/
FEDS.2016.102.
---------------------------------------------------------------------------
Notably, Congress specifically exempted market making from the
Volcker rule. Fault here thus lies not with the statute but the regime
chosen to implement it.
Funds. The Volcker Rule also prohibits banks from engaging in
proprietary or speculative trading by investing in private equity or
hedge funds, notwithstanding the absence of evidence that such
investments contributed to the financial crisis or have otherwise
caused outsized losses. While the agencies must implement the statute
as Congress has enacted it, they have extended its reach to numerous
other types of funds that bear little in relation to either private
equity or hedge funds. This has created enormous and unnecessary
compliance challenges for institutions with asset management businesses
serving customers seeking to save and build wealth, as well as for
market making in a number of asset classes, including securitized
products, covered bonds, and non-U.S. public funds.
More specifically, the regulation's overly broad definitions of
``hedge fund'' and ``private equity fund'' (so-called ``covered funds''
under the regulations) include vehicles that are not traditionally
considered to be hedge funds or private equity funds and require
extensive analysis and documentation of a banking entity's
determination of whether a particular vehicle is a covered fund or
qualifies for an exclusion or exemption. Moreover, the Volcker Rule
regulations restricts banking entities from engaging in activities that
could promote lending, capital formation and job creation, through
investing in vehicles such as certain types of credit funds,
infrastructure funds, energy funds, real estate funds and REITs. In
addition, the Volcker Rule, as implemented, makes it difficult for a
bank-owned asset manager to seed and test new asset management
strategies for customers as a result of the 3 percent statutory limits
on ultimate bank ownership after an initial 1-year seeding period, the
unduly burdensome process for extending the temporary seeding period,
and the lack of clarity on use of bank assets to fund separate account
seeding structures under the proprietary trading rules. Making the rule
more rational through appropriately tailored definitions of ``hedge
fund'' and ``private equity fund'' and more reasonable ancillary
requirements would lead to more efficient regulations, promote lending,
capital formation and job creation, and enhance customer offerings and
financing opportunities.
Asset-liability management. Firms use portfolios of liquid assets
to hedge firm-wide risk. These positions are managed by the corporate
treasury, not traders in the investment bank; are not short-term
trading positions; and are not engaged in to benefit from short-term
price movements. Nonetheless, the regulators have imposed the same
compliance obligations on this activity.
Enforcement. Five U.S. Federal financial agencies are tasked with
examining and enforcing compliance with the Volcker Rule, thereby
complicating efforts by financial institutions to comply with its
requirements on an enterprise-wide basis and to receive interpretive
guidance relating to its restrictions.
The whole approach to Volcker Rule compliance differs radically
from the standard supervisory paradigm, whereby firms are charged with
compliance and subject to enforcement action if they fail to comply.
Only with the Volcker Rule have the agencies set themselves on a ``pre-
crime'' mission, performing constant monitoring for compliance.
This ``pre-crime'' approach is even odder and more unnecessary
given a GAO report on proprietary trading during the financial crisis,
which demonstrated that proprietary trading was not a cause of the
financial crisis. \28\ Given the idiosyncratic nature of proprietary
trading's losses (and gains), it does not represent a systemic risk,
and the prudential risks that both trading and fund activities pose are
now subject to significantly higher capital charges under the Basel 2.5
and Basel III reforms. Notwithstanding the relatively low demonstrable
risk profile of the activities in question, the regulations have
nonetheless implemented a wide-ranging and highly complex set of
requirements that have and will continue to impair markets and slow the
real economy. These consequences are only exacerbated by the extra-
territorial manner with which the agencies have implemented the Volcker
Rule's restrictions.
---------------------------------------------------------------------------
\28\ Government Accountability Office, ``Proprietary Trading:
Regulators Will Need More Comprehensive Information to Fully Monitor
Compliance With New Restrictions When Implemented'' (July 2011),
available at www.gao.gov/assets/330/321014.pdf.
---------------------------------------------------------------------------
Impact on asset-liability management. The risk of proprietary
trading in a corporate treasury function, where assets are held in
available-for-sale or held-to-maturity accounts, is remote at best. But
the Volcker compliance regime is not tailored to the risk that
proprietary trading will actually occur, and therefore corporate
treasuries face high burdens in defending business-as-usual activity.
Needed reforms. Given the breadth and scope of these problems, the
financial regulators should revise their Volcker Rule regulations to
establish simpler criteria for identifying what trading and fund
activities are impermissible and a simpler and more reasonable process
for conducting examinations and issuing interpretive guidance. They
should eliminate the regulation's odd presumption that all trading
activity is illegal unless it can be proven to supervisory
satisfaction, through detailed analysis and continuous monitoring, to
meet a laundry list of specific criteria. Proprietary trading can be
easily distinguished by just a few key features. These regulatory
changes should be complemented with a shift in the compliance regime
from real-time enforcement to traditional reliance on bank compliance
and internal audit functions, with examiners reviewing their results;
specialized compliance program requirements and unnecessary metrics
should be eliminated. The result would be a substantial improvement in
market liquidity and investment in funds that promote capital formation
and job creation.
VI. Reforming Supervision, Examination, and Enforcement
We believe that bank supervision (as opposed to regulation) has
lost its way post-crisis, and requires a comprehensive reexamination.
While the link to economic growth in this area is less direct than in
the others cited, it is very real.
In sum, three developments have converged to restrict or even halt
the ability of many banks to open branches, invest, or merge to better
meet the needs of their customers. First, even as banks have
dramatically improved their financial condition by increasing their
capital, liquidity, and asset quality positions, supervisors have
transformed the supervisory scorecard (the CAMELS rating system) from a
measurement of financial condition to a measurement of compliance.
Second, supervisors have adopted a series of unwritten rules that
produce lower CAMELS ratings. Third, supervisors have adopted another
series of unwritten, or in some cases written, rules (albeit none with
any basis in statute) that translate those low ratings and other
supervisory issues into a bar on expansion. The result is a regime,
effectively invented by bank supervisors without notice and comment or
Congressional input, that makes an examiner's expectations regarding
bank compliance matters a fundamental determinant of whether banks can
invest and grow.
For perspective, consider that we routinely see serious compliance
violations across a wide range of American industries. Those companies
are subjected to enforcement proceedings and are required to pay fines
and remediate their practices, but no one ever suggests that while
those proceedings are pending they should be stopped from opening new
franchises, building new plants, developing new drugs, designing new
cars, or launching new apps. Yet somehow we have reached the point in
banking where the punishment for a compliance problem routinely
includes, in addition to a vast array of civil and criminal liabilities
imposed by a wide array of Federal and State authorities (often by
multiple authorities for the same underlying conduct), a prohibition on
any type of expansion by the bank. The opportunity lost is not just for
the bank but for its customers, and ultimately an economy that relies
on its banking system for financing.
Of course, banking is different in the sense that bank deposits are
insured by the FDIC. But that gives Government a special interest in
the financial condition of banks. As a result, Congress has in limited
instances linked expansion to financial condition. As we will see,
though, financial condition is no longer what banks are being graded
on, and the penalties for a bad grade now vastly exceed what Congress
has authorized.
Another result is simply a massive cost, which must be passed along
to consumers, as described in M&T Bank's most recent annual report
message to shareholders:

At M&T, our own estimated cost of complying with regulation has
increased from $90 million in 2010 to $440 million in 2016,
representing nearly 15 percent of our total operating expenses.
These monetary costs are exacerbated by the toll they take on
our human capital. Hundreds of M&T colleagues have logged tens
of thousands of hours navigating an ever more entangled web of
concurrent examinations from an expanding roster of regulators.
During 2016 alone, M&T faced 27 different examinations from six
regulatory agencies. Examinations were ongoing during 50 of the
52 weeks of the year, with as many as six exams occurring
simultaneously. In advance of these reviews, M&T received more
than 1,200 distinct requests for information, and provided more
than 225,000 pages of documentation in response. The onsite
visits themselves were accompanied by an additional, often
duplicative, 2,500 requests that required more than 100,000
pages to fulfill--a level of industry that, beyond being
exhausting, inhibits our ability to invest in our franchise and
meet the needs of our customers. \29\
---------------------------------------------------------------------------
\29\ M&T Bank Chief Executive Office Robert G. Wilmers, 2016
Annual Report Message to Shareholders (Mar. 9, 2017), available at
https://newsroom.mtb.com/document-archive/annual-report-letters/2016-
annual-report-message-to-shareholders.htm.
---------------------------------------------------------------------------
A. CAMELS Ratings
The centerpiece of bank supervision is the CAMELS rating system. It
was created by examiners in 1979 as a scorecard to evaluate an
institution's ``financial condition and operations''--in other words,
its safety and soundness. (Interestingly, the creation of the CAMELS
system was not specifically mandated by any statute or regulation.) The
CAMELS system evaluates a bank across six categories--Capital, Asset
quality, Management, Earnings, Liquidity, and Sensitivity to market
risk, especially interest rate risk--and assigns a composite rating,
all on a scale of 1 (best) to 5. With the sole exception of a few small
changes in 1996 (most notably, the addition of the ``S'' component),
the CAMELS standards have not been materially updated in the almost 40
years since their adoption--not after adoption of the original Basel
Accord on capital in 1988, the Basel III regime in 2010, the
Comprehensive Liquidity Analysis and Review in 2012, or the Liquidity
Coverage Ratio in 2014.
The result is a system that is hopelessly out of date. Detailed
capital, liquidity, and other rules have been expressly designed and
carefully calibrated to evaluate the key components of the CAMELS
ratings: capital, liquidity, and, less obviously, earnings and asset
quality, which are now evaluated through stress testing for certain
banks. Thus, for example, the published standards that examiners apply
in deciding the capital component of the rating do not include
consideration of any post-1978 regulatory capital standards--or any
market indicators, which also have grown in sophistication over the
past 40 years. There is no mention of CCAR, the self-described
Comprehensive Capital Analysis and Review. Rather, the published
standards speak vaguely of factors like ``the ability of management to
address emerging needs for additional capital.'' This is not to say
that there cannot be cases where a bank that is deemed well-capitalized
under the current 35-plus different capital tests could not, in theory,
still require more capital. It is, however, pretty unlikely.
It is worth examining the predictive ability of CAMELS ratings.
Consider the number of banks rated as weak (CAMELS 3, 4, or 5 in
Exhibit 1).




[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

This chart seems to demonstrate little predictive ability for
CAMELS ratings, even when they were focused on financial condition. In
2007, a small percentage was rated as weak, but hundreds failed.
Changing the subject: The move from financial condition to
compliance. The appropriate response to the diminished value of CAMELS
as a measure of financial condition would have been to decrease its
importance in the supervisory process or incorporate better measures of
financial condition. To their credit, the Federal Reserve and other
banking agencies adopted several crucial post-crisis reforms to improve
bank resiliency: most notably, CCAR stress testing, Basel III, and the
LCR. However, exactly because more objective, analytically sound
standards have overtaken the CAMELS system as a gauge of financial
condition, examiners have shifted their emphasis to the one entirely
subjective component: management. And not management as viewed through
the lens of maintaining sound financial condition, but rather through
the lens of ``compliance''--not just with laws, but with examiner
guidance and criticisms too.
Various ``unwritten rules'' reportedly have been adopted as part of
this shift:

All components do not count equally toward the composite
rating; the management rating counts the most, and it
increasingly appears that the composite rating cannot be higher
than the management rating. This elevation of management as the
``super component'' has never been subject to public comment. A
1996 update to the CAMELS standards stated that ``the
management component is given special consideration when
assigning a composite rating.'' Over time, it has become the
dominant consideration.

The management rating does not depend primarily on the
financial condition of the bank (because, if it did, it would
track the other ratings), but rather on compliance with banking
agency rules and guidance. In practice, any compliance problem
resulting in enforcement action or penalty, regardless of its
materiality, can result in a downgrade of management; so, too,
can unresolved ``Matters Requiring Attention'' (a confidential
examiner criticism).

Management ratings increasingly are driven by the results
of a consumer compliance rating that was adopted as an
independent evaluation.

Thus, the examination system has changed from primarily an
evaluation of the safety and soundness of an institution to,
increasingly, an evaluation of routine compliance matters and the
readiness with which management accedes to examiner criticism. And this
change has been accompanied by a substantial increase in the
consequences of a low rating, with supervisors raising the stakes
dramatically. While compliance matters are important, they are not
uniquely and exclusively important, and should not pollute a system
designed for an altogether different, and vital, safety and soundness
purpose.
Consequences. Bankers now routinely refer to being in the ``penalty
box,'' where they cannot expand through investment, merger, or adding a
branch. Mid-size and regional banks are particularly affected. There
are various ways into the penalty box:

As described above, a ``3'' rating for management operates
as a halt on expansion. Under section 4(k) of the Bank Holding
Company Act, a financial holding company whose bank receives a
``3'' rating for management must receive Federal Reserve
approval to expand certain nonbanking activities. Regulators
now extend that to almost any type of expansion, or at least to
any expedited review of branching or other applications.

Any AML consent order operates as a multiyear ban on
expansion for any purpose, regardless of the seriousness of the
conduct motivating the order or the progress made by the firm
in remediating it. While consent orders bring to mind large
banks in highly publicized cases, small and midsized banks
routinely receive such orders.

A ``Needs Improvement'' CRA rating also operates as a
multiyear ban, regardless of what triggered it or how it is
being remediated. While some statutes governing expansion
require an assessment of management (for example, the Bank
Holding Company Act, governing bank acquisitions), many do not.
And of those that do, each speak in particular to ``management
resources''--presumably, the ability of management to oversee
an integration--and not compliance issues. Large banks have
sufficient resources to remediate problems in one area while
expanding in another area--for example, to remediate an AML
issue at an overseas subsidiary while opening a new branch in
the Midwest United States.

The results of this new supervisory regime are significant:

Many banks--of all sizes, but particularly midsized banks--
have been blocked from branching, investing, or merging to meet
their customers' needs.

Bank technology budgets often are devoted primarily not to
innovation but to redressing frequently immaterial compliance
concerns.

Board and management time is diverted from strategy or real
risk management and instead spent remediating frequently
immaterial compliance concerns and engaging in frequent
meetings with examiners to ensure that they are fully
satisfied. Numerous banks report that their boards now spend a
majority of their time on regulation and compliance.

Of course, for examiners interested in having their compliance
criticisms acknowledged and immediately remediated, this system works
well. But as we note, it is not a tool that regulators in any other
industry feel they need, and it has important economic consequences.
Recommendations. A few core reforms are necessary. The first is an
unequivocal statement that the purpose of a CAMELS rating is to assess
the financial condition of the bank from the perspective of its
potential risk to the Deposit Insurance Fund. The second is the
withdrawal of the Federal Reserve's SR Letter 14-02 and all other
restrictions on bank expansion that do not have a basis in statute or a
regulation adopted pursuant to the Administrative Procedure Act. The
third is a complete overhaul of the CAMELS regime (including its
potential replacement) that emphasizes clear, cogent, and objective
measures of financial condition over vague, arbitrary, and subjective
ones.
B. Tailoring of Enhanced Prudential Standards
As we have described above, post-crisis regulatory reforms have
established a myriad of new prudential requirements for banking
organizations. The scope of application varies by requirement, but in
many cases new regulations have been applied in a uniform fashion to
large and diverse cohorts of banks of differing sizes, business models
and risk profiles. For example, the Federal Reserve has implemented a
number of so-called ``enhanced prudential standards'' under section 165
of the Dodd-Frank Act (including capital, liquidity, and other
requirements) on the basis of asset size thresholds.
While the Federal Reserve has made some effort to tailor its
enhanced prudential standards (e.g., by providing for a modified LCR
for some firms and recently eliminating the CCAR qualitative assessment
for others), it has generally done so based on arbitrary size
thresholds rather than careful consideration of the scope and type of
regulation warranted by different business models, risk profiles and
other more meaningful criteria. And in many cases, the Federal Reserve
has established one-size-fits-all rules that are not tailored at all.
The result is insufficiently tailored regulatory regime for many banks
that imposes unnecessary burdens and unduly limits their ability to
lend to and otherwise support businesses and consumers. There is
therefore a clear need to review all enhanced prudential standards
established under section 165 of the Dodd-Frank Act in order to
identify and implement more appropriate and robust tailoring of their
scope and extent of application; doing so would better enable banks
unduly and unnecessarily burdened by the current regime to lend and
otherwise serve customers and the economy.
C. Regulation of Foreign Banking Organizations
Generally outside the notice of policymakers, foreign bank
operations in the United States have decreased somewhat in recent
years.
Certainly, one cause of this retrenchment has been a delay in
foreign banks' recapitalization post-crisis, global economic
instability, and a general need to reduce balance sheet size. However,
U.S. requirements have been another key driver, as recent years have
also seen extensive revision to the rules and regulations governing
foreign banking organizations (FBOs) that have U.S. operations. In
particular, the Federal Reserve has required that, on the basis of
asset size thresholds, many FBOs operating in the United States
establish intermediate holding companies (IHCs) through which their
U.S. activities must be operated and managed. These IHCs in turn are
now subject to a wide range of new prudential requirements, including
capital, liquidity, stress-testing, resolution planning and other
rules. The resulting regime is one that is simply not appropriately
tailored to the varying sizes, business models, and risk profiles of
different types of FBOs and the inherently sub-consolidated nature of
their U.S. operations.
For example, the stand-alone capital and liquidity requirements
applied to the U.S. IHC of an FBO effectively hinder the foreign
parent's ability to allocate capital and liquidity across its entire
global business. All internationally active banks (whether foreign or
domestic) manage their capital and liquidity on a consolidated, global
basis, oftentimes acting nimbly to allocate financial resources to
geographic locations or business operations where it is needed in a
time of stress. Stand-alone U.S. capital and liquidity requirements
effectively trap those resources in the FBO's U.S. IHC, making them
unavailable for use elsewhere. The U.S. regime, which also mandates
stress testing at the IHCs, ignores (and duplicates) similar
consolidated requirements imposed on the FBOs by their home country
authorities. And of course, for the various U.S. capital and liquidity
rules that are applied to the U.S. IHC of FBOs, the general concerns
and recommendations that we have highlighted earlier in this paper are
just as relevant, and apply equally.
On top of these capital, liquidity, and stress test requirements,
the Federal Reserve also has required foreign GSIBs to ``pre-position''
extraordinarily high amounts of internal TLAC. This stands in contrast
to the process described in the FSB's term sheet on TLAC, which the
Federal Reserve developed in coordination with foreign supervisors;
that process identifies a range of potential internal TLAC
requirements, with the precise requirement to be established on an
institution-specific basis through collective dialogue among that
institution's home and host country supervisors. Instead, the Federal
Reserve imposed internal TLAC requirements at the top end of the FSB
range, unilaterally. Here, too, the result is insufficiently tailored
regulatory regime for many FBOs that imposes unnecessary burdens and
unduly limits their ability to lend to and otherwise support businesses
and consumers.
VII. Conclusion
Our banking system now stands on a solid foundation of capital and
liquidity. That foundation affords us the opportunity to consider
whether particular components of the regulatory and supervisory regime
are unnecessary, duplicative or more stringent than necessary to
achieve safety and soundness and financial stability goals.
Considerable economic benefits can be achieved through such a
reexamination.
______

PREPARED STATEMENT OF ROBERT R. HILL, JR.
Chief Executive Officer, South State Corporation, on behalf of Mid-Size
Bank Coalition of America
June 15, 2017
Chairman Crapo, Ranking Member Brown, and Members of the Committee,
I am Robert Hill, CEO of South State Corporation, which is the holding
company of South State Bank. South State, founded in 1933, is
headquartered in Columbia, South Carolina, and serves communities in
South Carolina, North Carolina and Georgia. In January of this year,
our bank passed the $10 billion in assets threshold, which subjects
South State to unduly burdensome requirements under the Dodd-Frank Act.
In light of this experience, I appreciate the opportunity to present
the views of the Mid-Size Bank Coalition of America (MBCA) on the
significant compliance burden placed on midsize banks as a result of
Dodd-Frank.
The MBCA is the voice of 78 midsize banks in the United States with
headquarters in 29 States. MBCA member banks are primarily between $10
billion and $50 billion in asset size, average less than $20 billion,
and serve customers and communities through more than 10,000 branches
in all 50 States, the District of Columbia, and three U.S. territories.
Midsize banks most often are the largest, local bank serving
communities, many for more than a century.
Unlike the largest banks in the country, for whom lending is
largely automated, midsize banks are run by people who are focused on
establishing long-term relationships with our communities and our
customers on a daily basis. As a result, we are able to use actual
knowledge of our customers and base our credit decisions on intangible
factors, such as character and local economic conditions. We have also
made the necessary risk and compliance investments that support our
business models, which are uniformly based on stable deposit funding,
revenues driven by traditional banking activities well-understood by
bank management and regulators and limited or no trading operations or
market-making activity. In sum, midsized banks have prudent business
models that contribute to economic growth and support financial
stability. To the extent restraints can be reduced, midsized banks can
provide even more credit and support to small businesses and Main
Street.
Under Dodd-Frank, crossing the $10 billion in assets threshold has
harsh implications for midsize banks. When banks cross $10 billion,
they are considered midsized institutions--a designation that
introduces an enhanced supervisory approach from regulators. These
banks can expect more frequent compliance requirements, which may
include full-scope examinations coupled with regular, targeted reviews.
In connection with these additional burdens, midsize banks must
allocate further resources to compliance, from business units to senior
management and the board of directors.
The imposition of these demands does not benefit the public in any
appreciable way. These requirements drain resources of midsize banks,
and less money is thus available to provide credit to individuals and
small businesses in our communities. For example, as a result of this
threshold, South State incurs costs over $20 million per year.
In April of this year, the MBCA submitted a letter to the Chairman
and Ranking Member urging the Committee to revisit the $10 billion
number, an arbitrary figure that does not meaningfully capture systemic
risk. In addition to the unfair consequences of using this number that
already exists, the MBCA is deeply concerned the figure could become
the default threshold for even more rules and regulations in the
future.
The MBCA's highest priority would be to eliminate the $10 billion
threshold and replace the number with an activities-based standard,
which would focus regulation more closely on systemic risk, or, at a
minimum, to raise the threshold to an appropriate level. The key
sections of Dodd-Frank, which will need to be amended in this regard,
are Sections 165, 1025, 1026, and 1075.
As an example, Section 165 imposes a mandatory stress testing
burden on banks between $10 billion and $50 billion, known as the Dodd-
Frank Annual Stress Test (DFAST). Former Federal Reserve Governor
Daniel Tarullo testified before Congress that such testing is not
necessary, and, in fact, it is actually a burden on the regulators with
no commensurate regulatory benefit. As CEO of an institution that
recently passed the $10 billion threshold, I can personally attest to
the significant compliance burden that follows and the cost that it
entails.
Independent Bank of Texas, an MBCA member bank with assets of just
under $10 billion, estimates the cost of implementing the mandatory
stress testing required under Section 165, in the event it crosses the
$10 billion threshold, would be $5-6 million in the year of
implementation and $2-3 million per year thereafter. Independent Bank
has stated they will have to add a team of three to four people to
manage this process.
All of this cost would be for something Governor Tarullo has
testified provides no regulatory benefit. In addition, the current
regulatory regime imposed by Section 165 forces midsize banks to divert
capital away from the products we offer and the lending that drives
growth and development in our communities. The MBCA believes freeing
midsize banks from the unreasonable burdens posed by Section 165 should
be one of your highest priorities.
To this end, we applaud Senators Moran, Tester, and Heitkamp for
sponsoring S.1139, the Main Street Regulatory Fairness Act, which would
remove the DFAST mandate currently imposed on banks between $10 billion
and $50 billion in assets. As Senator Tester noted, ``This bill cuts
red tape and makes it easier for Main Street lenders to invest in
entrepreneurs, families buying their first home and parents sending
their kids to college.''
The MBCA also applauds Senators Tester and Moran for introducing
the CLEAR Relief Act of 2017, which would provide the Qualified
Mortgage protections to loans originated and held in portfolio by banks
under $10 billion. The MBCA, however, strongly urges the Committee, as
it moves this legislation forward, to not limit this important relief
to banks under $10 billion. The rationale for the Qualified Mortgage
protections relates to the fact the banks with the status are holding
the mortgage loans in portfolio. It has nothing to do with the size of
the institution holding the mortgage. Using the arbitrary $10 billion
figure once again reinforces this number with no rational basis.
In our market, we have a lot of retirees, who do not have jobs. As
a result, they do not meet the QM status requirements. If we are
keeping the mortgages on our books, we believe we should be given QM
status. Otherwise, it is not just the bank that is impacted, but our
consumers are unfairly limited in their choices.
Former House Financial Services Chairman Barney Frank, one of the
principal authors of Dodd-Frank, has testified that he supports giving
a safe harbor status on loans where the lender retains the risk by
holding the loan in portfolio. A loan made by a bank and held to
maturity is the strongest possible statement of confidence in the
ability of the borrower to repay regardless of the size of the bank.
We have only raised two examples where the $10 billion figure
currently imposes--or may impose--an unnecessary burden on midsized
banks. But there are a variety of thresholds that need to be
eliminated, replaced by an activities-based standard or, at a minimum,
raised substantially to capture systemic risk. This is not simply about
fairness to midsize banks. It is fundamental to growing our economy.
Recently, the MBCA asked its member banks to submit examples from
their customers of specific, real-world customer impacts from the
current regulatory system. The examples received included everything
from mortgages rejected because of the ability to repay/qualified
mortgages requirement to business loans not made. These examples have
one thing in common--the absence of economic activity due to
unnecessary regulatory requirements, which results in limited to no job
creation and growth.
As Main Street banks, we support a regulatory regime that
encourages prudent behavior and protects our customers. But we also
need common-sense regulation that does not unnecessarily impose burdens
and impede the banking services communities need to create jobs and
drive economic growth--and this, in our view, requires a move away from
the Dodd-Frank $10 billion regulatory threshold. I am happy to answer
any questions the Committee may have, and again appreciate the
opportunity the Committee has given the MBCA to express its views.
______

PREPARED STATEMENT OF SAULE T. OMAROVA
Professor of Law, Cornell University
June 15, 2017
Dear Chairman Crapo, Ranking Member Brown, Members of the
Committee: Thank you for inviting me to testify at this hearing. My
name is Saule Omarova. I am Professor of Law at Cornell University,
where I teach subjects related to U.S. and international banking law
and financial sector regulation. Since entering the legal academy in
2007, I have written numerous articles examining various aspects of
U.S. financial sector regulation, with a special focus on systemic risk
containment and structural aspects of U.S. bank regulation. For 6 years
prior to becoming a law professor, I practiced law in the Financial
Institutions Group of Davis Polk & Wardwell and served as a Special
Advisor on Regulatory Policy to the U.S. Treasury's Under Secretary for
Domestic Finance. I am here today solely in my academic capacity and am
not testifying on behalf of any entity. I have not received any Federal
grants or any compensation in connection with my testimony, and the
views expressed here are entirely my own.
The global financial crisis of 2007-09 has left a deep, crippling
mark both on the American economy and on the lives of millions of
Americans who lost their homes, their jobs, their savings, and their
hopes for a better future. It threw the country into a prolonged
economic recession, accompanied by growing levels of poverty,
inequality, and political discord. According to an estimate by the
Federal Reserve Bank of Dallas, the crisis resulted in an economy-wide
output loss of up to $14 trillion, or $120,000 per single U.S.
household--both of which amounts will easily double if the broader
economic and societal effects of the crisis are permanent. \1\ These
output losses triggered a familiar vicious circle of economic
stagnation and instability. Faced with steadily declining real incomes,
Americans are forced to finance their consumption with an increasingly
unsustainable debt, which already reached a record $13 trillion mark.
\2\ Debt overhang depresses consumer spending, which in turn leads to
further contraction in production and employment. \3\
---------------------------------------------------------------------------
\1\ David Luttrell, Tyler Atkinson, Harvey Rosenblum, ``Accessing
the Costs and Consequences of the 2007-09 Financial Crisis and Its
Aftermath'', DallasFed Economic Letter, Vol. 8, No. 7 (Sept. 2013),
available at https://www.dallasfed.org/research/eclett/2013/el1307.cfm,
at 1-2.
\2\ See, Matt Scully, ``Trumps' America Is Facing a $13 Trillion
Consumer Debt Hangover'', Bloomberg.com (6 June 2017), available at
https://www.bloomberg.com/news/articles/2017-06-06/trump-s-america-is-
facing-a-13-trillion-consumer-debt-hangover?srnd=157391092.
\3\ See, Robert Hockett and Richard Vague, ``Debt, Deflation, and
Debacle: Of Private Debt Write-Down and Public Recovery'' (2013),
available at https://www.interdependence.org/wp-content/uploads/2013/
04/Debt-Deflation-and-Debacle-RV-and-RH1.pdf.
---------------------------------------------------------------------------
Furthermore, neither economists' estimates nor the actual
statistics capture the enormous nonpecuniary--or human--costs of the
crisis, including the lasting psychological effects of unemployment,
underemployment, diminished job security and reduced opportunity. While
difficult to quantify, these ``hidden'' costs of the financial crisis
will be borne by the American people for years to come.
Yet, the very same financial institutions whose reckless profit-
seeking created the crisis in the first place were largely protected
from the downside of their own excessive risk-taking, because the
Federal Government was compelled to bail them out. Not only did the
2008-09 bailouts effectively exact ``an unfair and nontransparent tax
upon the American people'' but they also significantly undermined
public trust in the American capitalist system, thus undermining the
system itself. \4\
---------------------------------------------------------------------------
\4\ Luttrell et al., supra note 1, at 3.
---------------------------------------------------------------------------
In this context, the Committee's current efforts to evaluate the
role of financial institutions in fostering America's economic growth
acquire particular significance. Promoting sustainable, stable long-
term growth is an issue of enormous political as well as economic
importance. It is the only way of remedying pervasive socially
destructive consequences of the financial crisis: only by deliberately
and systematically channeling public and private efforts toward the
expansion of productive capacity and employment in the real (i.e.,
nonfinancial) sector of the national economy can we reverse the
crippling effects of the extraordinary wealth transfer from the
American taxpayers to the financial industry that the latest crisis
laid bare for all to see. There is hardly a greater task facing
Congress today, and the Committee's decision to tackle it is a much
needed act of public-minded statecraft.
Ironically, however, the financial industry is using this
opportunity to mount a massive lobbying effort to achieve the opposite
goal: to reverse key post-crisis regulatory reforms enacted with an
explicit goal of curbing financial institutions' ability to generate--
and then socialize--excessive levels of risk in the financial system.
The industry explicitly targets the Dodd-Frank Wall Street Reform and
Consumer Protection Act of 2010 (the ``Dodd-Frank Act'' or the
``Act''), the centerpiece of post-crisis financial regulation reform.
\5\ In effect, banks are trying to put the Dodd-Frank Act on trial
under the rhetorical guise of ``fostering economic growth.'' It is,
however, a dangerous misconception to equate economic growth with
financial sector deregulation: not only are the two phenomena
fundamentally different but, as explained below, they are often
mutually exclusive.
---------------------------------------------------------------------------
\5\ Dodd-Frank Wall Street Reform and Consumer Protection Act of
2010, Pub. L. No. 111-203, 124 Stat. 1376 (2010) (codified at 12 U.S.C.
5301 et seq. (2012)).
---------------------------------------------------------------------------
It is especially ironic--indeed, astounding--to watch the financial
industry complain about its supposedly unbearable regulatory compliance
costs, when the industry is doing exceptionally well for itself. All
banks and their parent-companies, regardless of size, saw their profits
increase steadily during the entire time after the Dodd-Frank Act was
passed in 2010. In 2016 alone, banking institutions earned a total of
$175 billion in net profits. \6\ This is not a sign of a struggling
industry.
---------------------------------------------------------------------------
\6\ The total number for 2016 is the sum of the revenant quarterly
numbers. See, Federal Reserve Bank of New York, ``Quarterly Trends for
Consolidated U.S. Banking Organizations'', available at https://
www.newyorkfed.org/medialibrary/media/research/banking--research/
quarterlytrends2016q4.pdf?la=en.
---------------------------------------------------------------------------
Of course, it may be reasonable, and even desirable from the public
policy perspective, to revisit the continuing efficacy of certain post-
crisis laws and regulations or to recalibrate their application to
certain small, low-risk financial institutions. If carefully designed
and thoughtfully implemented, such recalibration may help to increase
the availability of affordable local financing to small businesses in
some rural and small-town areas. However, even under the best of
circumstances, that possibility alone is not sufficient to support a
wide-ranging repeal or rollback of the existing financial laws and
regulations. Any such measure would directly and disproportionately
benefit large, complex, systemically risky megabanks by removing all
meaningful constraints on their ability to destabilize the Nation's
financial system and, once again, jeopardize American taxpayers' long-
term (and even short-term) well-being. It is against this background
that the Committee should evaluate every proposal for regulatory reform
submitted to it by self-interested industry players.
I. Financial Deregulation Is Likely To Hinder, Not Foster, Real
Economic Growth
The starting point of all deregulatory proposals and arguments
advanced by the financial industry is a blanket assertion to the effect
that financial institutions' core, if not sole, business purpose is to
finance America's economic growth. Accordingly, the argument goes, any
regulatory constraint on financial institutions' business activities,
by definition, restricts their ability ``to serve customers, grow the
economy and create jobs.'' \7\ Therefore, an implicit conclusion
follows, removing such regulatory constraint will necessarily and
automatically improve customers' lives, boost the economy, and create
jobs.
---------------------------------------------------------------------------
\7\ Letter to The Honorable Michael D. Crapo and The Honorable
Sherrod Brown from Francis Creighton, Executive Vice President of
Government Affairs of Financial Services Roundtable (Apr. 14, 2017)
[the ``FSR Letter''], at 1.
---------------------------------------------------------------------------
It is an inherently faulty argument, insofar as these financial
institutions--regardless of their size or other attributes--are
privately owned firms whose overarching business priority is to
maximize their own profits and shareholder returns, not the Nation's
macroeconomic goals. It becomes a deliberately misleading and dangerous
argument, however, when used by large financial institutions the bulk
of whose profits comes from massive secondary-market trading and
dealing operations.
The financial industry's argument either inadvertently conflates or
deliberately confounds two very different things: (1) actual growth in
the real economy, and (2) mere speculation-driven asset price inflation
in the secondary markets.
It is indisputable that what America needs is real economic growth:
stable and sustainable long-term growth of the real--i.e.
nonfinancial--sector of the national economy. We urgently need to grow
our Nation's industrial output and capacity, to facilitate employment-
and wealth-generating technological advances, to rebuild and modernize
the country's physical and social infrastructure. We also need to
ensure that the benefits of these real growth-promoting activities are
distributed more equally and fairly, so as to restore the lost strength
of America's middle class and to enable lower-income American families
to move up the ladder of economic and social success. It is this kind
of real, sustainable, structurally balanced, and socially inclusive
economic growth that is necessary in order to help the country recover
from the post-crisis economic recession.
It is also fundamentally different from the mere asset price
inflation, or growth in prices at which various already existent
assets--stocks, bonds, commodities, real estate, etc.--are traded in
secondary markets. Because increases in market value of such tradable
assets at least temporarily increase their owners' individual wealth,
the aggregate growth in the market value of all such assets is
routinely and erroneously taken as a direct indicator of the aggregate
economic ``wealth'' or national economic ``growth.'' Of course, an
increase in the current market price of a particular company's stock
may reflect, at least in part, an increase in that company's real-life
productivity. But it may also reflect merely the generalized
expectations of today's stock buyers that those prices will be even
higher tomorrow. In that sense, all asset price inflation is inherently
speculative: it is not directly or necessarily linked to actual
productive gains in the real economy. \8\
---------------------------------------------------------------------------
\8\ ``The Fundamental Logic of Speculative Manias and Crashes Has
Been Explained and Documented Many Times''. See, Charles P.
Kindleberger and Robert Aliber, ``Manias, Panics, and Crashes: A
History of Financial Crises'' (2005); Erik Gerding, ``Law, Bubbles, and
Financial Regulation'' (2013).
---------------------------------------------------------------------------
Moreover, speculative asset price inflation, in fact, significantly
impedes real economic growth. There are two reasons why that is the
case.
First, by making purely speculative investments financially
attractive, asset price inflation effectively diverts investment flows
away from the primary markets in which companies raise new capital for
expanding their productive capacity. Put simply, investors looking to
put their money to use in financial markets face two competing choices:
asset price speculation (an easy short-term commitment of capital with
virtually no ``hard'' constraints on the upside) or productive
investment in the real economy (a long-term commitment of capital with
various real-life constraints on potential returns). In that sense,
asset price inflation actively undermines the real economy's potential
for productive, employment-generating growth that America so
desperately needs.
Second, asset price inflation creates instability that directly
threatens the economy's ability to operate and grow. When speculation-
induced asset price inflation reaches its peak, the inevitable market
crash tends to be fast and furious. During these dramatic moments,
markets tend to over-correct, sending asset prices far below the levels
supported by the ``fundamentals.'' Thus, the ultimate bursting of an
unsustainable speculative bubble wipes out not only the artificial,
purely speculative gains in asset values but also a lot of real
economic wealth. Massive defaults, bankruptcies, business closures,
worker layoffs, and other familiar symptoms of a severe ``market
correction'' extinguish both the fruits of the Nation's past and the
foundations of its future economic growth and prosperity.
To appreciate these dynamics, one need not go as far back as the
Great Depression and the Roaring Twenties that led to it. More recent
history provides plenty of evidence to the same effect.
In the era of massive financial sector deregulation, throughout the
1990s and all the way until 2008, America's economic boom was based
largely on secondary-market asset price inflation. This trend is
particularly visible in the period after the enactment of the Gramm-
Leach-Bliley Act of 1999 (the ``GLB Act''), which repealed the Glass-
Steagall Act's prohibition on combining, under the same corporate roof,
traditional banking activities and full-blown dealing and trading in
securities and other financial (and even nonfinancial) markets. \9\
Throughout the 1990s, large financial institutions--both commercial
banks and investment banks--lobbied for this ``regulatory relief'' from
the supposedly outdated Glass-Steagall rules, using the familiar
rhetoric of ``facilitating economic growth'' and providing ``more
choices'' and ``better services'' to their customers.
---------------------------------------------------------------------------
\9\ Financial Services Modernization Act (Gramm-Leach-Bliley Act),
Pub. L. No. 106-102, 113 Stat. 1338 (1999).
---------------------------------------------------------------------------
Once enacted, the GLB Act unleashed an unprecedented consolidation
in the financial services industry and the emergence of a handful of
extremely large FHCs that began aggressively growing their large-scale
trading and dealing operations in securities, derivatives, short-term
money-like instruments, and physical commodities. \10\ Their core
banking operations, while still a critical point of access to public
subsidy, quickly lost their status as the ``core'' source of
profitability. \11\ In the short 9 years between the enactment of the
GLB Act and the near-collapse of the financial markets in the fall of
2008, these universal megabanks have effectively turned into universal
dealers making secondary markets in everything and anything that could
be quantified and turned into a trading asset. A result of this
unprecedented growth of secondary market speculation was an equally
unprecedented asset price inflation--a story aptly told by many an
expert already.
---------------------------------------------------------------------------
\10\ See, Arthur E. Wilmarth, Jr., ``The Dark Side of Universal
Banking: Financial Conglomerates and the Origins of the Subprime
Financial Crisis'', 41 Conn. L. Rev. 963, 1002-46 (2009).
\11\ See, Federal Reserve Bank of New York, ``Quarterly Trends for
Consolidated U.S. Banking Organizations'', Fourth Quarter 2016,
available at https://www.newyorkfed.org/medialibrary/media/research/
banking-research/quarterlytrends2016q4.pdf?la=en, at 13.
---------------------------------------------------------------------------
For present purposes, however, one point deserves special emphasis:
As secondary market trading volume, stock price indices, and financial
firms' profits were all going up, domestic industrial production
declined, manufacturing jobs were massively outsourced overseas, wages
stagnated, and consumer debt (itself converted to a ``securitized''
trading asset) ballooned. In effect, this ``financialization'' of the
American economy represented an unprecedented transfer of wealth from
the real economy to the increasingly speculation-oriented financial
sector. \12\ The systematic redistribution of wealth from the Main
Street makers to Wall Street takers was starkly exposed when the
speculative craze--particularly, in mortgage-backed securities,
underlying mortgage loans and houses--finally triggered the world's
first truly systemic financial crisis. \13\ To protect the economy from
collapse, American taxpayers were forced to bail out the same megabanks
that fueled--and profited from--the crisis-inducing asset price
inflation. Today, financial institutions are doing very well, in terms
of profits and returns on their shareholders' equity. \14\ But the
middle class and poor Americans, whose livelihood is tied to the real
economy, continue to bear the full burden of the sluggish post-crisis
recovery.
---------------------------------------------------------------------------
\12\ See, Thomas Philippon, ``Finance Versus Wal-Mart: Why Are
Financial Services So Expensive?'' Rethinking the Financial Crisis 235
(Alan Blinder et al., Eds. 2012).
\13\ See, Saule T. Omarova, ``The New Crisis for the New Century:
Some Observations on the `Big-Picture' Lessons of the Global Financial
Crisis of 2008'', 13 N.C. Banking Inst. 157 (2009).
\14\ According to official statistics, the top 50 largest BHCs'
total net income for the last quarter of 2016 was over $32B, while
their average annualized return on equity was above 7 percent.
Tellingly, the total quarterly income of just the top six BHCs--
JPMorgan Chase & Co., Bank of America Corp., Wells Fargo & Co.,
Citigroup, The Goldman Sachs Group, and Morgan Stanley--was over $24B.
Three of these largest institutions had annualized return on equity of
almost 11 percent. Federal Reserve Bank of New York, Quarterly Trends
for Consolidated U.S. Banking Organizations, Fourth Quarter 2016,
available at https://www.newyorkfed.org/medialibrary/media/research/
banking_research/quarterlytrends2016q4.pdf?la=en, at 36.
---------------------------------------------------------------------------
The centerpiece of post-crisis regulatory reform, the Dodd-Frank
Act, aims to minimize the likelihood of recurring speculative asset
price inflation. Thus, the Dodd-Frank Act established a new systemic
oversight body, the Financial Stability Oversight Council (FSOC). \15\
The Act mandated enhanced prudential supervision of so-called
``systemically important financial institutions'' (SIFIs), which
includes large bank holding companies (BHCs) with at least $50 billion
in assets and certain nonbank financial institutions designated as
SIFIs by FSOC. \16\ As part of such enhanced prudential supervision,
SIFIs are required to maintain higher capital and liquidity buffers,
conduct regular stress tests, and prepare and submit to regulators
comprehensive resolution plans (living wills).
---------------------------------------------------------------------------
\15\ 12 U.S.C. 5321.
\16\ 12 U.S.C. 5323; 5325.
---------------------------------------------------------------------------
Among other things, these heightened requirements are designed to
limit the ability of large financial conglomerates to create dangerous
levels of risk through their massive dealing and trading operations in
secondary markets. By restricting SIFIs' ability to fuel destabilizing
asset price inflation, the Dodd-Frank regime of enhanced prudential
supervision also helps to channel investment away from socially
destructive speculation (secondary markets) and toward productive
investment in the real economy (primary markets). While it is not
commonly perceived or discussed in these terms, this potential
``channeling'' effect of the Dodd-Frank Act on real economic growth
should not be underestimated. Put simply, if investors find fewer
lucrative opportunities in speculative assets trading, they will direct
more of their money into nonspeculative investments.
It is, therefore, no surprise that large financial conglomerates--
Wall Street megabanks that dominate and profit from secondary market
trading and dealing activities--are now asking Congress to reverse all
of the major post-crisis regulatory reforms that threaten their ability
to promote speculative asset price inflation. Rhetorically, these
financial institutions are deliberately using the language of
``fostering economic growth'' and ``creating jobs,'' ostensibly through
lending to ``small businesses'' and ``American families.''
Strategically, they are taking advantage of the fact that many smaller
BHCs, regional lenders without meaningful trading operations, voice
their own, qualitatively different, concerns about the unintended
consequences of applying SIFI regulation to their more traditional
banking-based business models. Neither of these clever tactics should
sidetrack the Committee in its deliberations on how to foster
sustainable, stable growth in the real economy, as opposed to mere
speculation in secondary markets.
II. The Financial Industry's Deregulatory Proposals Will Not Foster
Real Economic Growth
The financial industry has submitted numerous letters and proposals
for deregulatory reforms that would ostensibly promote economic growth.
A comprehensive or detailed analysis of all such letters and proposals
would make my testimony unwieldy. Instead, I will focus on the
industry's key deregulatory proposals targeting the Dodd-Frank's regime
of enhanced prudential supervision, including the process of SIFI
designation and supervisory stress testing. While these proposals offer
clear potential benefits from the standpoint of financial institutions'
own profitability and stock price (at least in the short run), the
financial industry failed to establish how the proposed deregulatory
measures would promote sustainable long-term growth of the American
economy.
A. Rolling Back Enhanced Prudential Regulation Will Promote
Speculation-Driven Asset Price Inflation, Not Real Economic
Growth
The financial industry's proposals nearly uniformly try to make a
case that one of the key impediments to creating American jobs and
fostering economic growth is the Dodd-Frank Act's explicit focus on
systemic risk prevention. This argument targets the FSOC's general
authority to designate SIFIs, the process and criteria for application
of enhanced prudential standards, the substance of such standards, and
the Federal regulators' ability to exercise discretion in implementing
their statutory oversight responsibilities. The principal justification
for this sweeping attack on the core features of the post-crisis
regulatory regime is that it increases individual financial
institutions' costs of compliance, compared to their pre-crisis
regulatory compliance costs.
This line of argument lacks merit.
Every regulation, by definition, increases regulated firms' costs
of doing business: such costs may include both the direct expenses of
complying with regulations and the foregone profits from the prohibited
or restricted activities. Child labor laws, environmental regulations,
anti-fraud rules all raise costs of doing business for those private
firms that stand to profit from activities the society deems
undesirable. The mere imposition, via regulation, of additional private
costs is not an ``unintended consequence'' that must be avoided: it is
the principal mechanism of protecting the public from potential harm
caused by profit-seeking private actors.
The appropriateness of additional private costs of regulation,
therefore, must be weighted not against pre-regulation private costs
but against potential public costs likely to accrue in the absence of
regulation. None of the financial industry's proposals offer any
discussion, let alone quantification, of the full public costs of
rolling back the Dodd-Frank regime of systemic oversight. In that
sense, while styled as public policy proposals, these are merely
requests for special private benefits.
The rhetoric of ``promoting economic growth'' is meant to mask this
fundamentally self-interested nature of the financial industry's
requests for deregulation. As discussed above, removing prudential
restrictions on large financial institutions' risk-taking will hinder,
not promote, the kind of real economic growth that the American people
so urgently need. It will spur precisely the kind of secondary market
speculation and asset price inflation that enriches Wall Street
megabanks and further decimates America's real productive capacity.
Notably, all of the financial industry's proposals to roll back
Dodd-Frank's enhanced prudential regulation use the same basic
rhetorical device: they frame the issue as a clear binary choice
between ``arbitrary'' and ``tailored'' rules. They claim that existing
SIFI determination criteria (in particular, the $50B asset size
threshold for treating BHCs as SIFIs), the Federal Reserve's
supervisory stress tests, and even the long-standing CAMELS rating
system are ``arbitrary'' and should be either repealed or replaced with
something that is ``appropriately tailored'' to each financial
institution's ``unique'' business and risk profile. \17\
---------------------------------------------------------------------------
\17\ See, e.g., Letter to The Honorable Michael D. Crapo and The
Honorable Sherrod Brown from The Clearing House (Apr. 14, 2017) [the
``TCH Letter''].
---------------------------------------------------------------------------
It is not my goal in this testimony to engage in any technical
disputes regarding any specific capital requirements or stress test
methodologies. My comments go to the overarching misconception that the
industry actors' concerted (possibly coordinated?) use of the false
dichotomy--``arbitrariness'' vs. ``tailoring''--engenders.
``Tailored'' SIFI Determination Is a Path to Eliminating SIFI
Oversight
Generally, setting a specific numeric threshold as a jurisdictional
device--i.e., a criterion for subjecting a particular person to a
particular set of legal rules--is not ``arbitrary,'' per se. We all
live with a myriad of such fundamentally ``arbitrary'' but practically
necessary threshold-based rules every day: the legal age for voting is
18, the legal age for drinking is 21, the individual income tax rates
are drawn on the basis of specified income thresholds, and so on. What
would happen if we removed all such numerical thresholds as
``arbitrary'' and replaced them with ``tailored'' determinations
seeking to establish with complete precision every single person's
``unique'' individual ability to exercise voting rights, consume
alcohol, or pay income taxes? In theory, it could make everything
better. In practice, however, it would create a far more arbitrary,
unpredictable, and chaotic world in which nobody will be able to
anticipate--or assess the fairness of--the ``uniquely tailored''
treatment they receive under voting, drinking, or tax laws. It would
also require an enormous amount of Government resources to provide
sufficiently individualized and ``appropriately tailored''
determination of every person's many legal rights and obligations. The
sheer cost to the public of giving everyone their own ``tailored'' law
will far outweigh any private costs of having to live with
``arbitrary'' but universally applicable and clearly drawn boundaries.
This simple common-sense logic should be applied to evaluating the
financial industry's request to replace the Dodd-Frank's $50B size
threshold for treating BHCs as SIFIs with an ``indicator-based'' regime
of specific case-by-case designation. Midsize banks worried about
approaching the $50B threshold in the future and regional banks that
already qualify as SIFIs based on their asset size are especially keen
to see this part of the Dodd-Frank repealed and replaced.
Reasonable people may disagree and argue about whether the current
size threshold--$50 billion in assets--is the right one, or whether a
higher or a lower number would be more socially beneficial. On the one
hand, as midsize and regional banks argue, their traditional lending-
based business model does set them qualitatively apart from Wall Street
megabanks with massive and systemically risky trading and dealing
activities. The fact that the top six megabanks' size is measured in
trillions of dollars further underscores that difference. On the other
hand, $50 billion is by no means an insignificant number. Only 38 BHCs
currently exceed that threshold. It is also instructive to remember
that, in the 1980s, savings banks and thrifts--small, local traditional
lenders squeezed by competitors--made very similar pleas for regulatory
relief. The resulting S&L crisis showed that hasty deregulation of
small lending-oriented financial institutions may create significant
risks to the system. \18\
---------------------------------------------------------------------------
\18\ See, Federal Deposit Insurance Corporation, ``The S&L Crisis:
A Chrono-Bibliography'', available at https://www.fdic.gov/bank/
historical/sandl/.
---------------------------------------------------------------------------
These complexities notwithstanding, the existing regime should not
be simply replaced with an ``indicator-based'' system under which the
FSOC would be forced to go through a tedious and inevitably contentious
exercise of determining whether any particular BHC's scope, scale,
nature, and mix of activities warrants a SIFI designation. While this
``flexible'' and ``individually tailored'' approach may sound good in
theory, it will significantly undermine the entire post-crisis
regulatory framework for safeguarding systemic stability. There are
three main reasons why that is the case.
First, mandating an individualized SIFI determination procedure for
each potentially systemically significant BHC will impose an enormous,
and completely unnecessary, financial and organizational burden on
FSOC, Federal Reserve, and other regulators. The sheer costs of the
``tailored'' designation process--hiring and training dedicated
personnel and devoting countless amounts of regulators' time and energy
to gathering and processing huge amounts of information, most likely
over BHCs' constant objections and complaints--will threaten to derail
the entire regime of SIFI oversight. It is nearly a certainty that this
costly and time-consuming process will effectively preclude both FSOC
and the Federal Reserve from exercising their statutory
responsibilities as systemic regulators. Notably, financial
institutions advocating this measure deliberately ignore these public
costs of giving them their own, individually ``tailored'' supervisory
regime. Nor do they undertake to cover all of the additional regulatory
costs of the ``tailored'' SIFI designation process.
Second, individually ``tailored'' SIFI designation will immediately
become vulnerable to the financial industry's other favorite line of
attack as being inherently unpredictable, unclear, and nontransparent:
in other words, ``arbitrary.'' The very nature of the complex inquiry
into various qualitative indicators of systemic riskiness of an
individual financial institution is bound to open FSOC to potential
allegations of misjudgment, misinterpretation, and misbehavior. The
MetLife saga provides a vivid example of that tactic. Given what is at
stake for large BHCs, the odds of FSOC being constantly embattled and
ultimately incapacitated are unacceptably high. Instituting
individually ``tailored'' SIFI designation process will virtually
ensure the next round of industry lobbying, aiming to eradicate the
very notion of enhanced SIFI supervision as ostensibly nonadministrable
in practice.
Third, there is significant danger that loosening the SIFI
designation process, primarily to accommodate the demands of midsize
and regional banks and BHCs, will pave the way for large Wall Street
megabanks to seek additional deregulatory measures specially
``tailored'' to enable them to expand their lucrative secondary-market
trading and dealing operations. While such activities are precisely
what creates unsustainable levels of risk in the financial system, it
will be much more difficult for lawmakers and (already significantly
weakened) regulators to resist what will likely be framed as simply
``further tailoring'' of supervisory rules.
Attacks on ``Arbitrary'' Stress Testing Are Attacks on Supervisory
Discretion
Ironically, even as the financial industry ostensibly wants FSOC to
exercise individualized judgment instead of applying ``arbitrary''
generalized rules in the context of SIFI designation, the same industry
vehemently attacks the exercise of individualized judgment--or
supervisory discretion--by the Federal Reserve in the context of
supervising SIFIs. The key target of this attack is the post-crisis
regime of supervisory stress testing.
The largest megabanks appear particularly determined to limit the
Federal Reserve's ability to conduct meaningful Comprehensive Capital
Analysis and Review (CCAR), mandated by the Dodd-Frank Act. Among other
things, they seek to

subject the Federal Reserve's annual stress test scenarios
to a 30-day notice and comment period under the Administrative
Procedure Act;

restrict the Federal Reserve's ability to use its own
independent assumptions in constructing test models;

mandate advance publication of the Federal Reserve's stress
test models for ``peer review;''

restrict the use of the Federal Reserve's own models merely
to a ``supervisory assessment'' of banks' own models; and

eliminate CCAR's qualitative assessment for all banks. \19\
---------------------------------------------------------------------------
\19\ See, TCH Letter, supra note 17.

If implemented, these industry-advocated changes will effectively
nullify the CCAR regime. The principal reason for subjecting SIFIs to
both internal and supervisory stress tests is to create a reliable
early-warning mechanism for identifying potential weaknesses in the
firm's capital planning and management. Forcing the Federal Reserve to
disclose ahead of time its stress test models will enable financial
institutions ``to manage to the test,'' thus defeating the whole
purpose of stress testing. Moreover, subjecting its test scenarios or
modeling methodologies to public notice and comment will inevitably
create unnecessary delays in the implementation of stress tests. It
will impose potentially prohibitive additional costs on the Federal
Reserve, burdened with the duty to respond to numerous industry
comments and criticisms. Finally, relegating the Federal Reserve's
models to a mere back-up reference function will render the entire
exercise inherently unreliable.
In sum, what the financial industry is advocating here is not a
``more robust and transparent'' stress testing process, but a de facto
sidelining of the Federal Reserve by forcing it to surrender its key
supervisory function to SIFIs themselves. Doing so will significantly
endanger the country's financial stability and increase the likelihood
of another systemic crisis. Accordingly, it will hinder, not promote,
America's long-term economic growth.
B. There Is No Evidence That Financial Deregulation Is Necessary To
Foster Economic Growth
Financial institutions' deregulatory proposals claim that the post-
crisis regime of enhanced prudential oversight directly prevents them
from extending more loans to small businesses and struggling American
families. They routinely assert that the more stringent capital
requirements and the higher costs of regulatory compliance are the
principal, if not the sole, reason why banks cannot increase their
financing of productive economic enterprise.
Constant repetitions of this blanket assertion are intended to
condition the audience--including the Members of this Committee--to
associate the rollback of Dodd-Frank (something the financial industry
wants) with the creation of domestic manufacturing jobs (something the
American people need). It is calculated to propagate dangerous
confusion about the real causes of financialization, ongoing erosion of
America's industrial base, rising poverty and inequality, and other
social and economic ills of the last several decades. In essence, the
industry wants us to believe that forcing Citigroup and Bank of America
to finance just 5 percent of their multi-trillion-dollar high-risk
assets with common shareholder equity is the root of all of the
Nation's economic woes.
This is an incredible claim. There are three main reasons why it is
fundamentally false:

First, capital regulation does not reduce banks' cash
available for lending: that notion is based on a fundamental
misunderstanding of what bank capital is.

Second, banks are not short of cash necessary to expand
their lending: banks' soaring profits and record dividend
payments in recent years show there is plenty of cash they
could, but choose not to, lend out.

Third, there is no evidence that banks are striving to
increase lending that would foster the economic growth: the
industry offers no proof (beyond simple assertions) that
individual banks' asset allocation decisions are driven, in any
meaningful way, by their desire to raise the rate of growth in
the real economy. In the absence of such evidence, banks'
deregulatory demands should not be taken as bona fide proposals
to foster America's long-term economic growth.
Enhanced Capital Levels Do Not Restrain Availability of Credit
Higher capital requirements have nothing to do with reducing money
available to banks for lending and productive investment. Capital is
not cash in the vault. It is merely an accounting concept, the amount
of shareholder equity on a bank's balance sheet: i.e., amount
contributed by the bank's shareholders and not borrowed from depositors
and other creditors. \20\ Banks do not ``hold'' capital in the same way
as they ``hold'' cash or gold--and in the exact same way as Exxon-Mobil
or Microsoft are never said to ``hold'' their shareholder equity.
Capital is simply what the owners of the corporation would receive if
the corporation liquidated all of its assets and repaid all of its
debts. In that sense, capital is a critical equity cushion that
protects corporations'--including banks' and BHCs'--creditors from
losses. It is only because financial institutions' creditors are
explicitly or implicitly protected from such losses by the Federal
Government that banks and BHCs are allowed to operate with much thinner
equity cushions than would be sustainable in our free capitalist
market. \21\
---------------------------------------------------------------------------
\20\ For a discussion of bank capital and a compelling argument
that capital requirements should be much higher than they are
presently, see Anat Admati and Martin Hellwig, ``The Banker's New
Clothes'' (2013).
\21\ See id.
---------------------------------------------------------------------------
It is therefore nonsensical to claim that reducing this creditor-
protecting, loss-absorbing equity cushion will somehow ``free up cash''
for bank lending--and, specifically, for lending to small businesses
and credit-needy Americans. How much a bank is willing or able to lend
is a complex asset allocation decision that is driven primarily by
considerations of bank's own profitability: ``Should we extend a long-
term loan to a risky small startup, or should we use that money to
increase our fee revenues from short-term derivatives trading?'' The
impact of this decision on the bank's regulatory capital ratios or
stress test results may be an important factor in its choice between
lending and trading, but only insofar as it affects--indirectly and in
combination with many other factors--that bank's overall profits.
In other words, capital regulation constrains banks' (individually
rational) propensity to choose ``high risk, high return'' assets and
limits their ability to maximize shareholder profits by jeopardizing
creditors. These regulatory capital constraints, however, still leave
plenty of room for banks to choose whether to finance productive
economic enterprise or to channel money into secondary market
speculation. To the extent the latter increases short-term shareholder
returns, it remains a potentially more attractive choice. The largest
banks' massive shift into secondary market trading and dealing,
especially after the passage of the GLB Act, aptly illustrates that
dynamic.
If Congress grants these largest banks' demands to weaken existing
capital requirements, supervisory stress testing, and other elements of
enhanced prudential regulation and supervision, it will affirmatively
sanction virtually unconstrained growth in the volume and speculative
riskiness of banks' trading and dealing activities, not traditional
``small-business'' lending. That, in turn, will spur precisely the kind
of speculation-driven asset price inflation that threatens the
stability of the American financial and economic system and undermines
the country's long-term economic growth.
Banks Are Not ``Short of Cash'' for Lending
According to the FDIC statistics, the U.S. banking industry has
fully recovered from the crisis and is doing exceedingly well. Thus, in
the first quarter of this year, nearly 96 percent of all U.S. insured
depository institutions were profitable; their average return on equity
stood at a healthy 9.37 percent; and their total quarterly income
reached $44 billion, which is 12.5 percent higher than a year earlier.
\22\ Insured banks' total net income in 2016 exceeded $171 billion.
\23\ BHCs are also turning handsome profits. For example, in the last
quarter of 2016, the total quarterly income of just the top six BHCs--
JPMorgan Chase & Co., Bank of America Corp., Wells Fargo & Co.,
Citigroup, The Goldman Sachs Group, and Morgan Stanley--exceeded $24
billion. \24\
---------------------------------------------------------------------------
\22\ Federal Deposit Insurance Corporation, ``Statistics at a
Glance'', As of March 31, 2017, available at https://www.fdic.gov/bank/
statistical/stats/.
\23\ FDIC Quarterly Banking Profile, Fourth Quarter 2016,
available at https://www.fdic.gov/bank/analytical/qbp/2016dec/qbp.pdf,
at 7.
\24\ Federal Reserve Bank of New York, Quarterly Trends for
Consolidated U.S. Banking Organizations, Fourth Quarter 2016, available
at https://www.newyorkfed.org/medialibrary/media/research/
banking_research/quarterlytrends2016q4.pdf?la=en, at 36.
---------------------------------------------------------------------------
These profits directly increase banks' and BHCs' shareholder
equity--the capital cushion that is a subject of so many of the
industry's complaints--and are easily available for use in their
lending or other growth-promoting activities. However, it appears that
a big chunk of these profits is instead being distributed to the
banking institutions' shareholders, in the form of cash dividends and
share repurchases. Thus, in 2016, federally insured banks alone
returned to their shareholders $103 billion in cash dividends, \25\ a
number second only to the record high of $110 billion in cash dividends
they paid in 2007, the last full pre-crisis year. \26\
---------------------------------------------------------------------------
\25\ FDIC Quarterly Banking Profile, Fourth Quarter 2016,
available at https://www.fdic.gov/bank/analytical/qbp/2016dec/qbp.pdf,
at 7.
\26\ FDIC Quarterly Banking Profile, Fourth Quarter 2007,
available at https://www5.fdic.gov/qbp/2007dec/qbp.pdf, at 5.
---------------------------------------------------------------------------
By any measure, $103 billion is an enormous amount of money that
could be used both (1) to increase banks' total loss-absorbing and
risk-reducing regulatory capital cushion, and (2) to finance small
family-owned businesses, entrepreneurial startups, medium-size
industrials, aspiring students, and struggling families. In other
words, using $103 billion of banks' profits to increase lending to
productive economic enterprise would advance both (1) the public
interest in having a safer and more efficient system of credit
allocation, and (2) the banks' self-professed interest in fostering
economic growth and creating American jobs. Yet, banks chose not to go
that socially beneficial route.
It is astonishing to see that, after voluntarily sending all that
money to shareholders, the banking industry complains that the
additional cost of complying with post-crisis regulations ``takes
capital away from small business loans, home purchases and other
productive uses.'' \27\ ``Every dollar spent on hiring compliance
attorneys,'' the argument goes, ``is potentially $10 dollars of loans
that could be made to improve someone's economic opportunity.'' \28\ As
the dividend numbers cited above show, every dollar diverted away from
banks' regulatory compliance would most likely improve only bank
shareholders' and managers' ``economic opportunity.'' In fact, using
the same mathematical logic, it follows that, in 2016 alone, banks have
willingly deprived the real economy of a whopping $1.03 trillion in
``small business loans, home purchases and other productive uses.''
---------------------------------------------------------------------------
\27\ FSR Letter, supra note 7, at 2.
\28\ Id.
---------------------------------------------------------------------------
It is not my contention that banks should never declare shareholder
dividends. The key point here is that bank dividend payouts expose the
fundamental falsity of the industry's claims to the effect that
excessive regulatory costs deplete banks' resources and prevent them
from financing real economic growth. Banks have plenty of extra money
for expanding their lending. They choose not to lend that money,
instead ``returning capital'' to their shareholders. That is because
stable and high dividends increase individual banks' stock prices,
which directly benefits not only bank shareholders but also their
executives and managers. Higher stock price translates directly into
higher bonuses. Higher volume of small-business lending does not.
This basic fact about banks' use of available capital may explain
why these institutions--particularly, the largest Wall Street
megabanks--are waging such an adamant campaign against CCAR, ``living
wills,'' and other key elements of the Dodd-Frank's systemic risk
prevention regime. Under the current regime, SIFIs' ability to pay
shareholder dividends or repurchase their own shares is expressly
conditioned on supervisory approval, based in part on the results of
the latest stress tests. Not meeting supervisors' expectations,
therefore, limits their ability to pay dividends and depresses their
stock price. It is telling, for example, that securities analysts and
investment advisers have been buzzing about Citigroup's and Bank of
America's recent and expected future hikes in dividend payouts after
both of these firms performed better in the CCAR tests. \29\ As one
expert put it, ``The ability to pay dividends is currently a hallmark
of strength in the sector.'' \30\ Accordingly, deregulatory rollback of
the Federal Reserve's stress testing and other prudential regulations
is expected to enable Citigroup, Bank of America, and other SIFIs to
raise their dividends and share repurchases, thus lifting the value of
their stock. \31\
---------------------------------------------------------------------------
\29\ See, Gemma Acton, ``These Three U.S. Bank Stocks Are My Top
Sector Picks'', CNBC.COM (22 Mar. 2017), available at http://
www.cnbc.com/2017/03/22/these-three-us-bank-stocks-are-my-top-sector-
picks-analyst.html; Rebecca Keats, Bank of America Declares a Quarterly
Dividend, Marketrealist.com (31 Jan. 2017), available at http://
marketrealist.com/2017/01/bank-of-america-declares-a-quarterly-
dividend/; Stone Fox Capital, Citigroup: ``Expect Another Big Dividend
Hike To Move Stock'', Seeking Alpha (4 Jan. 2017), available at https:/
/seekingalpha.com/article/4078558-citigroup-expect-another-big-
dividend-hike-move-stock.
\30\ Stone Fox Capital, supra note 29.
\31\ Id.; Keats, supra note 29.
---------------------------------------------------------------------------
Thus, it appears that market experts have no confusion about the
real benefits of massive financial deregulation: it will increase
banks' profits, dividend payouts, and stock prices. Whether or not it
will also promote the country's long-term economic growth does not seem
to be part of the conversation.
There Is no Evidence That ``Fostering America's Economic Growth''
Is a Meaningful Factor in Banks' Business Decisions
It is, of course, possible that, while returning massive amounts of
capital to shareholders, banks and BHCs are nevertheless genuinely
dedicated to their self-declared mission of financing America's real
economic growth. Rather than take their word for it, however, the
Committee should require specific, robustly documented and empirically
supported, evidence that that is indeed the case.
For example, the Committee should ask each financial institution
asking for regulatory relief to provide specific, quantified, and fully
documented answers to the following questions:

What was your institution's specific (i.e., quantified)
annual contribution to the growth of your local, regional, and/
or national economy, in the period between 2010 and 2017 (after
the enactment of Dodd-Frank)? What was it in the pre-Dodd-Frank
period between 2000 and 2009?

How much additional annual contribution to the growth of
your local, regional, and/or national economy would your
institution have made, but was prevented from making directly
as a result of [insert a specific regulatory provision of the
Dodd-Frank regime]?

In each year since 2010, what was the aggregate amount of
commercial and industrial (C&I) loans that your institution
refused to extend, solely because of [insert a specific
regulatory provision of the Dodd-Frank regime]?

Does your institution have an enterprise-wide strategy for
facilitating domestic job-creation and promoting the growth of
your local, regional, and national economy? What are the core
elements of that strategy?

How often does your institution's Board of Directors and
top management discuss the institution's performance in
implementing that strategy?

This type of targeted inquiry would help to (1) establish the
credibility of the financial industry's claims; and (2) discover the
real link, if any, between financial institutions' deregulatory agenda
and the country's real economic growth. Presently, none of the
financial industry's numerous deregulatory proposals establish that
link, relying instead on purely declarative rhetoric. They then demand
effective removal of key regulatory safeguards against systemic
financial crises, solely on the strength of that rhetoric. Assessing
the public costs and benefits of any such deregulatory proposals,
however, requires ascertaining that such steps are actually--and not
just rhetorically--going to generate substantial growth in the
country's real economy.
It is unlikely that any financial institution will be able to
produce satisfactory answers to any of these questions. As private
shareholder-owned firms, these institutions' primary concern is their
own profitability, not the overall performance of the American economy.
They simply do not track, and are not equipped to track, the relevant
macroeconomic data: measuring and worrying about such data is the
Government's responsibility, an inherently public task. That means that
determining what should be done to spur America's economic growth--and
whether relaxing any of the existing financial regulations should be a
part of that endeavor--is also an inherently public responsibility. The
financial industry's attempts to usurp or sidetrack the process of
public deliberation on such an important matter should therefore be
subjected to intense scrutiny.
III. Broader Structural Solutions Are Needed To Channel Capital Into
Productive Economic Activity and Sustainable Growth
The Dodd-Frank regime of systemic oversight and enhanced prudential
supervision of SIFIs correctly aims to limit potentially destabilizing
speculation and asset price inflation in secondary markets. To the
extent it strengthens the resilience and stability of our financial
system, it lays down an important foundation for the Nation's long-term
economic growth. However, simply limiting the opportunities for
diverting capital into speculative trading is not sufficient to spur
and sustain such growth in practice. It is equally important to ensure
that a significantly greater share of available financial capital is
actively and consistently flowing into long-term productive investment.
In other words, it entails cultivating new sources of, and creating new
avenues for profitably deploying, truly ``patient'' capital.
Abundant patient capital is what enables construction of large-
scale physical and social infrastructures, supports transformative R&D
projects, generates productivity gains, and creates sustainable well-
paying jobs throughout the Nation. By the same token, the chronic
shortage of patient capital--and persistent glut of speculative
capital--is the key reason for the sluggishness of America's real
economy today.
Incentivizing investor ``patience'' on the scale needed to spur the
Nation's long-term economic growth, however, is a difficult task that
cannot be reduced simply to a few deregulatory or tax-relief measures.
Nor can it be left to the private sector alone. Private investors' time
horizons and risk tolerance levels are inherently limited by the finite
nature of their economic resources and their biological lifespan. It is
fundamentally rational for private investors to prefer shorter-term
investments, which entail less unforeseeable future risk and promise
returns within such investors' reasonable lifetime horizons. Private
investors' short-term bias, therefore, is not a deviation from market
rationality; it is a built-in feature of such rationality.
Overcoming investor short-termism and facilitating the formation of
``patient'' capital requires new, more effective forms of public-
private partnership. The public component of such partnership will
bring into the investment process a number of unique advantages that
public instrumentalities enjoy when they act as market participants:
their vast scale, high risk tolerance, lengthy investment horizons, and
direct backing by the full faith and credit of the United States.
Combining these unique capacities of a public investor with private
investors' informational agility, superior knowledge of local
conditions, and market expertise will help to channel both private and
public resources into the critical growth-inducing projects. This new,
patient public-private capital will finance the building of new roads,
bridges, high-speed train lines, clean energy networks, and next-
generation industrial plants. It will also create new well-paying jobs,
offer new educational opportunities, and unleash new entrepreneurial
energy of the American people.
My colleague, Professor Robert Hockett, and I have developed a
specific proposal for creating this new kind of infrastructural growth-
oriented public-private partnership. A White Paper detailing our
proposal is attached as an Appendix to this testimony.
It is this kind of programmatic reform--not a massive rollback of
the Dodd-Frank Act--that the American economy and the American people
need. I urge the Committee to rise to this challenge and not allow
banks' self-serving deregulatory demands to distract its attention from
what really matters to the American economy and the American public.
Conclusion
Promoting sustainable, socially inclusive long-term growth in
America's real economy is a task of enormous public significance.
Massive financial deregulation urged by the banking industry, however,
will not foster such real economic growth: it will merely spur
speculation-driven asset price inflation in secondary markets. That is
what generates Wall Street's greatest short-term profits and causes
wealth- and growth-destroying systemic crises. Weakening regulatory
standards and effectively incapacitating FSOC, the Federal Reserve, the
Consumer Financial Protection Bureau, and other Federal regulators will
put us much closer to another financial disaster.
Contrary to the financial industry's assertions, massive
dismantling of the Dodd-Frank regime of systemic regulation will not
benefit small businesses and struggling families across America. There
is no evidence that the additional cost of banks' regulatory compliance
with post-crisis regulations actually depletes banks' resources and/or
diverts them away from productive uses. Despite their complaints about
regulatory costs, American banking institutions are highly profitable
and awash in cash, which they could use for lending to small businesses
but instead choose to return to their shareholders in the form of
dividends and share repurchases. Against that background, it is
impossible to take the industry's attacks on Dodd-Frank seriously.
The Committee should, therefore, reject the financial industry's
unsubstantiated claims and requests for massive deregulation and
demolition of the Dodd-Frank Act. The Committee should focus its
attention on finding real solutions to the real problems associated
with speculative short-termism and persistent misallocation of capital,
which impede economic growth. Devising such solutions is challenging
but necessary in order to make finance serve the Nation's long-term
economic goals.


[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]



RESPONSES TO WRITTEN QUESTIONS OF SENATOR BROWN
FROM ROBERT R. HILL, JR.

Q.1. Your testimonies cited certain metrics for measuring your
banks' compliance costs. Of course, there are a variety of ways
to measure a bank's costs and compliance burden. To help the
Committee better understand the overall compliance burdens on
your institutions:
Please provide the ratio of total employees in your bank's
workforce to employees.

A.1. In 2009 and 2010, our compliance employees were housed in
the Compliance Department. After the enactment of Dodd-Frank,
the number of employees dealing with compliance grew
dramatically; however, the growth was through employees
throughout the bank and company, and not just in the Compliance
Department. At the time, we did not endeavor to keep track of
the number of compliance-related employees. In the last several
years, however, we have endeavored to do so. The data below is
a good faith effort to provide the requested numbers, but we
would caution that they are conservatively stated in that we
have employees in virtually every area of the bank working on
compliance issues. As you will see, even on a conservatively
stated basis, the number of compliance employees has gone up
over eightfold.


----------------------------------------------------------------------------------------------------------------
2009 2010 2011 2012 2013 2014 2015 2016
----------------------------------------------------------------------------------------------------------------
Full time 700 1,015 1,071 1,324 2,106 2,081 2,058 2,055
equivalents
(FTEs)
Compliance FTEs 6 6 N/A N/A N/A N/A 40 51
Ratio 0.86% 0.59% N/A N/A N/A N/A 1.94% 2.53%
----------------------------------------------------------------------------------------------------------------


Q.2. Please provide your bank's compliance costs for each year
from 2009 to the present.

A.2. In reviewing the data below, please refer to the response
above. As stated above, the data is difficult to capture with
total precision. As noted above, the compliance costs in 2009
or 2010 were largely contained in the Compliance Department.
Today, they are spread over various areas of the bank and
company. Again, this is our good faith effort to be responsive.


----------------------------------------------------------------------------------------------------------------
2009 2010 2011 2012 2013 2014 2015 2016
----------------------------------------------------------------------------------------------------------------
Estimated $780,000 $840,000 N/A N/A N/A N/A $5,800,000 $7,500,000
Total
Compliance
Costs
----------------------------------------------------------------------------------------------------------------


Q.3. Please provide the combined dollar value of your bank's
stock dividend payments and share repurchases for each year
from 2009 to the present.

A.3. Below is the stock dividend information you requested. In
reviewing these numbers, however, there are several factors you
should bear in mind so you will know that the growth of the
bank and the number of its shareholders has changed
dramatically over the course of the years for which you have
asked us to provide information. For example, in 2013, the bank
almost doubled in size and its number of shareholders, and
capital, increased by roughly 40 percent. As a result, the
dollar amount of dividends went up significantly. The change
did not reflect an equivalent increase in dividend income to
individual shareholders; it was simply the function of the
bank's growth through acquisition.


----------------------------------------------------------------------------------------------------------------
2009 2010 2011 2012 2013 2014 2015 2016
----------------------------------------------------------------------------------------------------------------
Combined value $8,623 $8,935 $9,856 $11,080 $16,207 $20,702 $25,072 $33,136
of stock
dividend
payments and
share
repurchases (in
thousands)
----------------------------------------------------------------------------------------------------------------


Q.4. Please provide your bank's efficiency ratio for each year
from 2009 to the present.

A.4. Below is the data on the bank's efficiency ratios for the
period in question. As you review this data, you should bear in
mind that efficiency ratios are a function not just expenses
but also revenue. Thus, as the bank grew through acquisitions,
its revenue increased. In other words, it makes comparisons
difficult. Further, in 2010, the bank recorded a nonrecurring
gain. Absent that gain, the 2010 efficiency ratio would have
been 69.89 percent, not 46.68 percent.


----------------------------------------------------------------------------------------------------------------
2009 2010 2011 2012 2013 2014 2015 2016
----------------------------------------------------------------------------------------------------------------
Efficiency 61.17% 46.68% 68.77% 72.20% 75.85% 71.41% 64.19% 64.16%
Ratio
----------------------------------------------------------------------------------------------------------------


Q.5. Please provide your bank's revenue from interchange fees
linked to debit cards for each year from 2009 to the present.

A.5. Below is the requested data. As I testified, South State
Bank, for the relevant years, was under $10 billion. Thus, the
Durbin Amendment interchange fee caps did not affect the bank's
interchange fee revenue. Also, as noted above, the bank grew
significantly during the time period in question. As a result,
the revenue from interchange grew significantly for that
reason.


----------------------------------------------------------------------------------------------------------------
2009 2010 2011 2012 2013 2014 2015 2016
----------------------------------------------------------------------------------------------------------------
Interchange $3,717 $7,261 $9,467 $11,178 $18,143 $25,192 $27,939 $31,801
Fee Revenue
(in
thousands)
----------------------------------------------------------------------------------------------------------------


Q.6. During your testimony, you said ``Under Dodd-Frank,
crossing the $10 billion in asset threshold has had very harsh
implications for midsize banks For example, South State was
impacted by over $20 million per year, a significant sum for a
bank our size. What impacts does this have on our local
communities? For us, that equates to 300 jobs. Approximately 10
percent of our branches were closed, and even more jobs were
diverted from lending to regulatory compliance.''
Your bank appears to have grown from roughly 50 branches
before the passage of Dodd-Frank, to about 180 this year,
assuming completion of your proposed merger with Park Sterling.
The Park Sterling transaction will be the ninth merger or
acquisition by your bank of another bank or its branches since
the beginning of 2010.
As you know, branch closures can occur for a variety of
reasons not related to regulatory costs, for example, in 2010,
your bank closed 10 of the 36 branches that it acquired through
the purchase and assumption agreement related to Community Bank
& Trust, well before you passed the $10 billion threshold. (In
another example, you closed two branches in Orangeburg, SC,
after acquiring 12 Bank of America branches, reportedly because
they were in close proximity to existing South State branches.)
Finally, it appears that you closed 13 of 140 branches in
2015, or about 9 percent, before you crossed the $10 billion
threshold, and nine of your 127 branches last year and one so
far this year, or about 8 percent of your branches. This
compares to 2010, when you closed 10 of 86 branches, a rate of
nearly 12 percent.
Please explain the rationale behind the 10 percent closure
number that you cited, and please provide supporting evidence
that explains why these closures were caused by the various $10
billion thresholds contained in Dodd-Frank, as opposed to other
potential factors.

A.6. As it relates to our branch closing strategy, well before
officially crossing the $10 billion asset threshold in early
2017, we began estimating the costs associated with, and
resources required, to operate a bank with assets over $10
billion. Beginning approximately 3 years ago, we started
examining the magnitude of these expenses and thinking through
steps to pay for them. In my testimony, I stated that we closed
about 10 percent of our branches in order to cover these
increased costs of growing to over $10 billion. Closing
branches does reduce cost and can help pay for this burden, but
this step alone does not pay for the costs of crossing $10
billion. As you can see in the data above, the incremental
compliance costs are approximately $6-8 million per year for
our company plus we will lose approximately $17 million dollars
in interchange income beginning in 2018 due to impact from the
Durbin amendment. In total, we will realize an approximate $25
million pre-tax negative impact to earnings by crossing this
threshold. On average, it costs us approximately $500,000
annually to operate a branch. If we were to accomplish all of
the savings through branch closures alone, we would have to
close roughly 50 branches and eliminate over 250 jobs. Whether
we close branches or find other ways to pay for these expenses,
the burden is considerable. The $25 million represents
approximately 15 percent of the net income of our company in
2016.
------


RESPONSES TO WRITTEN QUESTIONS OF SENATOR SASSE
FROM ROBERT R. HILL, JR.

[From Robert R. Hill, Jr.--We have prepared responses to
many of the follow up questions submitted by Senators Brown,
Sasse, and Tillis. Some of the questions pertained to banks
larger than South State or touched on macro-economic issues
that we did not feel equipped to answer and therefor
respectfully did not supply an answer. What follows are the
questions to which we do offer an answer.]
Q.3. As you know, Dodd-Frank imposed new stress test
requirements on banks above $10 billion.
How should policymakers balance providing more transparency
and guidance to regulated entities about passing stress tests,
without enabling regulated entities to, as some have suggested,
``game'' these processes?

A.3. We believe that, in general, policymakers should focus on
minimum acceptable capital ratios as opposed to arbitrary
stress tests.

Q.4. Do stress tests accurately depict how a firm would perform
during a financial crisis? If not, what should be done, if
anything, to improve their accuracy?

A.4. To help improve the accuracy of stress tests, focus on the
amount of risk weighted assets each firm holds would, in our
opinion, be beneficial.

Q.5. Are stress tests properly tailored to match the unique
risk profile of smaller financial institutions?

A.5. Stress test are currently structured based on the asset
size of financial institutions and do not focus on the risk
profile of the institutions. As stated above, focusing on the
amount of risk weighted assets each firm holds would more
properly align with the institution's risk profile.

Q.6. As you know, House Financial Services Chairman
Hensarling's legislation, the Financial CHOICE act--in part--
would allow banks to opt-out of various regulatory
requirements, in exchange for meeting a 10 percent leverage
ratio.
What are the most persuasive arguments for and against
relying upon a leverage ratio as a significant means of
reducing systemic risk in the financial system?

A.6. The most persuasive argument for relying upon a leverage
ratio as a significant means of reducing systemic risk in the
financial system is the 10 percent leverage ratio, which is a
simple approach and represents a significant amount of capital.
The most persuasive argument against relying on the ratio is
that the leverage ratio does not take into consideration the
risk rating of the assets and the loan loss reserve.

Q.7. Under this legislation, is the 10 percent leverage ratio
the right level? If not, where should policymakers set the
level at?

A.7. A 10 percent leverage ratio is the right level.

Q.8. What evidence do you find or would you find to be the most
persuasive in discerning the proper capital levels under this
proposal?

A.8. Different asset categories carrying different degrees of
risk, so focusing on risk rating the assets, rather than cash
and loans being equal, would seem logical.

Q.9. I'm concerned that our Federal banking regulatory regime
relies upon arbitrary asset thresholds to impose prudential
regulations, instead of an analysis of a financial
institution's unique risk profile.
Should a bank's asset size be dispositive in evaluating its
risk profile in order to impose appropriate prudential
regulations?

A.9. No, a bank's asset size should not be dispositive in
evaluating its risk profile in order to impose appropriate
prudential regulations.

Q.10. If not, what replacement test should regulators follow
instead of an asset-based test?

A.10. As a replacement test, regulators should focus on the
levels of risk based capital and the amount of debt the
financial institution has, as better tests for evaluating a
bank's Risk Profile.
------


RESPONSES TO WRITTEN QUESTIONS OF SENATOR TILLIS
FROM ROBERT R. HILL, JR.

Q.1. I'm a proponent of tailoring regulations based off of the
risk profiles of financial institutions, as opposed to having
strict asset thresholds that do not represent what I believe is
the smart way to regulate. But, my question here is really
about the importance of ensuring that we have a system that is
rooted in fundamental, analytical, thoughtful regulation so
that we can achieve and execute on goals, whether balancing
safety and soundness with lending and growth, or encouraging
more private capital in the mortgage market to protect
taxpayers and reform the GSEs.
On the latter point, I would like for you to comment on the
down-stream consequences that the wave of new mandates have on
housing finance reform--whether Dodd-Frank, regulatory rules
and/or Basel prudential requirements--and, specifically, the
ability to bring private capital as we look to reform the GSEs?

A.1. Recent regulatory reform which specifically impacts the
mortgage industry has placed a wave of burden and uncertainty
on us as originators of mortgage loans. The Dodd-Frank reforms
specifically limited access to credit for many borrowers.
Borrowers who fit into the credit box of the GSEs are exempt
from some of the QM requirements as long as the GSEs are in
conservatorship or until 2021. Due to the unknown liability and
litigation expenses, our bank has tightened our credit box in
our portfolio lending, specifically around the 43 percent DTI.
Basel Requirements have burdened many institutions in their
ability to originate new loans which they service themselves.
This has forced them to sell loans to aggregators that they
would desire to retain as customers and earning assets on the
books. These are not large institutions and their holding of
MSRs is not a driver of the performance of their institutions,
however it is restricting how they do business and forcing them
to send more loans to what are likely SIFI.
Other regulations such as TRID, HMDA, and other CFPB
enforcement has been implemented with many questions still
unanswered. This causes disparities in practices in our
industry which put institutions that are playing by the rules
at a competitive disadvantage. This lack of clarity has also
increased the cost to originate a loan, which today is now
above $7,000 where it was close to $4,000 6 years ago. This is
increasing the cost to the customer and limiting competition in
the mortgage space as it has driven consolidation.
Regulations should be based more on the type of lending
that is done and should focus on firms who are purposely
defying regulations rather than institutions afraid of clerical
errors which can cost exorbitant enforcement fees and personnel
costs.
GSE reform is absolutely necessary as we continue to
rebuild a sound secondary mortgage market. The GSEs perform a
critical task in offering a 30-year fixed option that is liquid
and with clear guidelines. The GSEs have done a great job in
the past few years of focusing on their customers and what can
enable us to lend to more borrowers. The ability of the
benefits the GSEs bring the market is imperative to continue
into the future. With that said, taxpayers should not be
exposed to the risk present in this market and credit risk
should be transferred to the private sector. We would hope that
the GSEs or their future state can continue with some mandate
to offer affordable housing support with low-down payment
options and less restrictive credit standards than private
entities would likely offer.

Q.2. Can you give me your opinion on where the current credit
box is, how consumers get loans both for personal, business,
etc., how about in the mortgage space? Is it harder to get a
mortgage today versus in years past?
How does this impact underserved and underbanked
populations in the U.S.?

A.2. Underbanked and underserved borrowers are not being fully
served today. Borrowers with little or no credit are
immediately not within the credit parameters set forth by
programs offered by the GSEs, FHA, and USDA. The current FICO
models reward borrowers with very established credit that
underbanked populations do not necessarily possess. Buyback and
compare ratio risks from the agencies have forced lenders to
place overlays on the agency credit parameters which limit an
underserved borrower's ability to find lending options. It is
critical for the FHFA to continue to promote access to these
programs and find ways to encourage responsible but broad
lending across the credit spectrum.

Q.3. Can you tell me how, in addition to other macroeconomic
variables that affect lending, the regulatory environment has
played a role in lenders' willingness/ability to extend
mortgage credit? What needs to change?
I see evidence that suggests that nonbanks now make up the
predominate percentage in the mortgage market, what is your
view of how this affects the stability of servicing mortgages
in a crisis environment and how does this affect access to
financing?

A.3. Nonbanks now make up more than 50 percent of originations
based on 2016 HMDA data. These loans are typically sold to SIFA
institutions and large national servicers. These servicers will
continue to be slow to react in a crisis and not able to meet
the needs of all customers.
Nonbank mortgage lenders are typically regulated by States
and do not receive the scrutiny as a federally regulated bank
and enforcement for noncompliance is rare. A smaller nonbank
lender offering riskier products that may be taking advantage
of borrowers may never see the types of audits a bank does. A
bank will err on the side of caution in order to remain in
compliance while making decisions that hurts its ability to
compete fairly in the marketplace.

Q.4. Just to be clear--from my perspective I'm not advocating
to return to the lending standards of the sub-prime/pre-crisis
era, but I do believe that we need to evaluate some the
regulations such as the use of the False Claims Act, having
unified servicing standards, TILA-RESPA, etc., so that we can
help expand credit safely to both first-time buyers and
refinancers, and I would like your perspective on how we (1)
address larger regulatory reform in the banking ecosystem; and
(2) how your institutions and similarly situated ones play a
role in providing credit to consumers.

A.4. A bank like ours has shown through years of prudent
lending, low complaints, and positive financial performance our
ability to properly run our business while meeting the lending
needs of the communities we serve. We agree that the lending
standards pre-crisis were well below where they should have
been, however the pendulum has swung too far alongside
regulation by enforcement that quite frankly scares prudent
lenders like us from offering anything but the basic products
for our customers. The uncertainty in recent regulations has
forced us to take a competitively disadvantaged position to our
nonbank peers. Regulation uncertainty has also put our teams in
a position to constantly question what is right or the intent
of the law. Loans are taking longer to close, at a higher cost,
and with more uncertainty sometimes to the borrower due to
recent regulations, namely TRID.
Regulatory reform in the mortgage market should be based
not just on size, but on the types of lending. The biggest help
that our regulators can give us is clarity. Confusion leads us
to always take the most conservative route which has downstream
effects such as limited access to borrowers and higher costs to
lenders.

Q.14. Are assets alone the single most important factor for
determining systemic risk? If not, why use that as a threshold
at all?

A.14. Yes, assets are the single most important factor for
determining systematic risk, but the focus should be on the
type and risk of assets rather than just the total amount of
assets.
------


RESPONSES TO WRITTEN QUESTIONS OF SENATOR BROWN
FROM SAULE T. OMAROVA

Q.1. Can you please describe the nature of banking agencies'
so-called ``safety and soundness'' authority? Would you be
concerned by any proposals to interfere with the existing
safety and soundness regime?

A.1. Ensuring ``safety and soundness'' of the banking system
is, and has always been, the fundamental substantive goal of
U.S. bank regulation. In addition to performing vital functions
of taking deposits and facilitating payments, banks play a
critical role in the transmission of monetary policy. Yet, the
very nature of their business makes banks especially vulnerable
to runs and other shocks that can quickly bring them down.
Traditionally, banks' primary liabilities are short-term (e.g.,
demand deposits), while their assets are long-term and illiquid
(e.g., loans). The concept of ``safety and soundness'' reflects
the long-standing recognition of this built-in vulnerability of
the banking business model and the importance of preserving
banks' solvency and ability to operate uninterrupted.
Accordingly, Federal and State banking agencies' authority
to continuously monitor, evaluate, and act to enhance
individual banks'--and the entire banking system's--safety and
soundness is very broad. The agencies' mandate to maintain and
ensure safety and soundness permeates and underlies the entire
regime of U.S. bank regulation and supervision. In that sense,
it is difficult to describe precisely the scope or the specific
nature of that authority. In effect, every specific rule
governing banking institutions' activities represents a
particular articulation of the safety and soundness
requirement. Moreover, it is impossible to reduce the safety
and soundness mandate to any specific, limited, quantifiable
factor or rule. It empowers and obligates the regulatory and
supervisory agencies to exercise an inherently context-specific
judgment as to whether any particular activity, transaction, or
business practice potentially threatens stable and reliable
operation either of an individual bank or of the banking/
financial system more generally. Most, if not all, legal rules
applicable to U.S. banks expressly provide for the relevant
regulators' authority to make a particularized substantive
determination of permissibility or legality of an otherwise
permissible action based on its potential safety and soundness
consequences.
Put simply, this authority may be analogized to the medical
professionals' maxim of ``Do no harm.'' And just like any
attempt to limit or qualify that principle would undermine not
only the integrity but the very efficacy of medical care, so
would any attempt to limit or qualify bank regulators'
authority to ensure safety and soundness of the banking system
undermine that system's integrity and efficacy.
Reflecting the dramatic lessons of the latest financial
crisis, the Dodd-Frank Act strengthened Federal bank
regulators' mandate to ensure safety and soundness of the U.S.
financial system. The creation of the Financial Stability
Oversight Council (FSOC), heightened prudential oversight of
certain large bank holding companies (BHCs) and nonbank
financial firms designated by FSOC as systemically important
financial institutions (SIFIs), mandatory periodic stress
testing and submission of ``living wills'' by large BHCs and
SIFIs--all of these post-crisis regulatory innovations are
merely an updated version of the centuries-old safety and
soundness regime. They constitute a coherent framework for
ensuring vital stability of today's complex and dynamic
financial system. Regulators need all of these tools to be able
to monitor the levels of risk in the financial system and to
prevent potentially destructive systemic shocks. Taking away or
in any way limiting regulators' and supervisors' flexibility
and ability to exercise discretion in determining how and when
to act in the name of the safety and soundness of the U.S.
financial system would be unacceptably reckless. It will
effectively guarantee another systemic financial crisis.
Any legislative reform that seeks to eliminate, limit, or
weaken the Dodd-Frank Act's regime of system-wide prudential
oversight, therefore, goes directly against the most important
public interest in preserving the safety and soundness of the
American financial system. The list of such dangerous reforms
includes proposals to limit FSOC's power to designate SIFIs and
the Federal Reserve's power to supervise them, to lower
existing capital adequacy requirements, to make stress testing
more easily predictable and thus subject to ``gaming,' to
replace the Orderly Liquidation Authority, and so forth. None
of these proposed reforms will strengthen our financial system.
To the contrary, they will make it a lot more vulnerable and
dysfunctional.
------


RESPONSES TO WRITTEN QUESTIONS OF SENATOR SASSE
FROM SAULE T. OMAROVA

Q.1. Federal Reserve Governor Powell testified at the April 14,
2016, Senate Banking Committee hearing entitled ``Examining
Current Trends and Changes in the Fixed-Income Markets''. He
said that ``some reduction in market liquidity is a cost worth
paying in helping to make the overall financial system
significantly safer.''
Is there also a risk that reducing liquidity in the
marketplace also makes the marketplace unsafe?
If so, how should regulators discern the difference between
an unsafe reduction in liquidity and a safe reduction in
liquidity?

A.1. There are trade-offs in every regulatory choice. For
example, prohibiting distribution of marijuana protects people
from serious drug addiction, but it also reduces availability
of marijuana for medical or recreational purposes. How should
we discern the difference between an unsafe reduction in
availability of marijuana and a safe reduction in such
availability? In that context, lawmakers often don't seem to
spend too much time on finding the precise scientifically
proven level of ``safeness'' and take a principled normative
stand: one public policy goal is more important than the other.
The same logic should apply to this theoretical debate on
``safety vs. liquidity'' in financial markets. There is no
objective, scientific mark for an absolute optimal level of
market liquidity. It all depends on the context and the
consequences. There was plenty of liquidity in the years before
the latest crisis, and then that liquidity evaporated
overnight. ``Liquidity'' was merely another word for
unproductive churning and excessive speculation. It was bad for
the market and for the American economy. So, clearly, we cannot
use the pre-crisis ``liquidity'' measures as our reference
point for ``safe vs. unsafe'' determinations. That choice is
not scientific; it is fundamentally normative. We as a society
should make a normative, political choice whether it is more
important for us to enable uninhibited financial speculation
(for which ``market liquidity'' is often a euphemism) or to
prevent financial crises.

Q.2. As you know, regulators imposed numerous capital
requirements after the 2008 financial crisis.
Have Federal regulators sufficiently studied the cumulative
impact--including on liquidity in the marketplace--of these
various changes?
If not, how should Federal regulators resolve this issue?
For example, some have called to delay the imposition of new
financial rules and regulations in order to facilitate a
broader study of these issues.

A.2. Industry actors' calls for delaying the application of
capital rules, ostensibly for the purposes of ``studying''
their impact, are merely a strategy to avoid higher capital
requirements--and thus to continue to operate with high levels
of leverage. There is no absolute, scientifically precise level
of capital that is perfectly optimal under any circumstance.
How much equity (i.e., capital) a particular bank should have
at any single point depends on how volatile the value of its
assets is--and that can change quickly in response to various
internal business decisions and external forces in the markets.
That is why, for example, countercyclical capital buffer
requirements are especially important: they provide the
necessary cushion for absorbing sudden losses when an asset
price boom turns to the inevitable bust.
Thus, financial regulators and supervisors should have
sufficient flexibility and discretionary authority to
determine, on the ground and in the context of a particular
institution, whether that institution has sufficient capital to
withstand the loss of asset value without hurting its
creditors. Regulators and supervisors are best equipped to
render such determinations. They should be given a broader
authority to do just that. Saddling regulators with the
inherently meaningless task of studying ``cumulative effects''
of all capital requirements on all financial firms will
effectively render them incapable of doing their job. The only
real-world effect of that strategy will be more leverage, more
risk, and more instability in the financial system. No real new
knowledge produced as a result of any such studies will be even
remotely worth that risk.

Q.3. As you know, Dodd-Frank imposed new stress test
requirements on banks above $10 billion.
How should policymakers balance providing more transparency
and guidance to regulated entities about passing stress tests,
without enabling regulated entities to, as some have suggested,
``game'' these processes?
Do stress tests accurately depict how a firm would perform
during a financial crisis? If not, what should be done, if
anything, to improve their accuracy?
Are stress tests properly tailored to match the unique risk
profile of smaller financial institutions?

A.3. I discuss the issue of stress tests' transparency at
length in my written statement.
To add to that discussion, I would like to emphasize how
misguided it is to judge the efficacy or public benefits of
stress tests by reference to how ``accurately'' they depict the
actual behavior of any specific firm in an actual crisis. No
such predictions or assessments can be made, and that is not
the purpose of stress testing. Stress tests play out a variety
of scenarios, including extreme ones, as a way of identifying
serious weaknesses in the institution's financial condition and
risk management. Going through the process of stress testing is
just as important as passing the tests. It is that aspect of
stress testing--the dynamic, learning, procedural aspect--that
is critical for preserving financial stability more broadly.
Forcing the Federal Reserve to be more ``transparent'' about
its methodologies and assumptions will significantly weaken, or
even eliminate, that beneficial effect of stress testing. It
will only benefit financial institutions, but not the financial
system--and not the American economy.

Q.4. As you know, House Financial Services Chairman
Hensarling's legislation, the Financial CHOICE act--in part--
would allow banks to opt-out of various regulatory
requirements, in exchange for meeting a 10 percent leverage
ratio.
What are the most persuasive arguments for and against
relying upon a leverage ratio as a significant means of
reducing systemic risk in the financial system?
Under this legislation, is the 10 percent leverage ratio
the right level? If not, where should policymakers set the
level at?
What evidence do you find or would you find to be the most
persuasive in discerning the proper capital levels under this
proposal?
If the leverage ratio was set at the right level, do you
find merit in eliminating a significant portion of other
regulatory requirements, as with the Financial CHOICE Act? Are
there any regulations that you would omit beyond those covered
by the Financial CHOICE Act?
What impact would this proposal have on liquidity in the
marketplace?

A.4. A straight leverage ratio is important as a useful
baseline for judging the level of capital adequacy of financial
institutions. Unlike risk-based capital ratios, the leverage
ratio does not allow for risk-weighting of asset values, which
can be easily and dangerously miscalculated (either
intentionally or unintentionally). In that sense, it provides a
critical corrective to the malleable risk-based capital ratios.
As explained above, there is no scientifically derived,
theoretically ``perfect'' level of capital for all banks at all
times. True, the leverage ratio of 10 percent would be a
significant improvement over the current requirement of 5
percent-6 percent at maximum. However, it is not a magical
number that will somehow eliminate the need for close
regulatory and supervisory oversight, stress tests, and other
systemic risk-reducing measures.
To the extent we do have any sort of an ``objective''
benchmark for judging the functional leverage ratio levels, the
market gives us a number much higher than 10 percent. A 10
percent leverage ratio is woefully low compared to the level of
equity at nonfinancial firms. Thus, publicly traded
nonfinancial firms typically have 30 percent-40 percent in
shareholder equity on their balance sheets: i.e., their
leverage ratio is in the 30 percent-40 percent range. If a
company has less equity, it has to pay higher price for its
debt. This is how the free capitalist market operates in the
absence of a public subsidy.
Banks enjoy such a subsidy, which is precisely why they
prosper with such extremely low levels of true equity. The
multifaceted system of bank regulation and supervision exists
as the necessary substitute for the missing market discipline.
Any attempt to weaken such system of oversight will only free
banks to incur more leverage and risk, and to shift the risk
and the ultimate losses onto the American taxpayer. A 10
percent leverage ratio will not prevent that result. Perhaps a
30 percent-40 percent leverage ratio would, but even that is
debatable.

Q.5. I'm concerned that our Federal banking regulatory regime
relies upon arbitrary asset thresholds to impose prudential
regulations, instead of an analysis of a financial
institution's unique risk profile.
Should a bank's asset size be dispositive in evaluating its
risk profile in order to impose appropriate prudential
regulations?
If not, what replacement test should regulators follow
instead of an asset-based test?

A.5. In my written statement, I discussed at length the reasons
why it is fundamentally misleading to characterize asset size-
based thresholds for enhanced prudential supervision as
``arbitrary.'' We live with a myriad of ``arbitrary'' but
practically necessary threshold-based rules every day: the
legal age for voting is 18, the legal age for drinking is 21,
and so forth. If all such numerical thresholds were deemed
unacceptably ``arbitrary'' and replaced with ``tailored''
determinations of every single person's individual ability to
exercise voting rights or consume alcohol, it would create a
far more arbitrary, unpredictable, and chaotic world. Nobody
would ever seriously propose such a ``reform'' in the name of
``tailoring'' law to every person's ``unique'' circumstances.
Similarly, replacing clear bright-line rules by a requirement
that Federal regulators assess each financial firm's unique
risk profile and circumstances would be impractical and
ineffective. Doing so will impose unbearable costs on the
public and essentially eliminate the entire regime of enhanced
oversight of systemically important financial institutions.
That would be an extremely dangerous result.
A better, more pragmatic way to accommodate the inevitable
differences between megabanks and smaller institutions, which
currently fall into the same SIFI category, could be to allow
for a discretionary downward adjustment of the intensity of the
enhanced supervision regime for certain institutions, based on
a combination of their size and business activities. In other
words, it may be desirable to have a formalized process
whereby, e.g., a traditional midsized bank with assets above
$50 billion could petition the regulators for a lighter
regulatory or supervisory treatment. The petitioning bank would
have the burden of proving to the regulators why such special
dispensation would be reasonable in its case--and why it would
not create any unreasonable risks to the safety and soundness
of the U.S. financial system.
This approach would properly place the burden of securing a
special private benefit (lower compliance costs) on those
private entities that seek it, and not on the public that
finances Federal regulatory agencies. Moreover, this system
would be in line with the fundamental free-market principles:
the firms that want to have their own, uniquely tailored law
will have to pay for it and bear responsibility for its
outcomes.
------


RESPONSES TO WRITTEN QUESTIONS OF SENATOR REED
FROM SAULE T. OMAROVA

Q.1. This Committee heard testimony last week from Professor
Adam Levitin that since the Wall Street Reform and Consumer
Protection Act, the cumulative pre-tax return on equity in the
banking sector has been 225 percent for community banks and 320
percent for megabanks. At the same time, since 2010, real
income has actually fallen by 0.6 percent for the average
American family. These statistics highlight that banks, and
especially large banks, have done very well in the years since
the Wall Street Reform Act, while the typical American family's
real income is now less than before the bill. In order to grow
the economy, wouldn't pursuing policies that increase
productivity and lift middle class wages be better than
deregulating banks?

A.1. This is absolutely correct. To stimulate healthy long-term
growth of the national economy, it is critical to ensure that
the maximum number of Americans actually receive wages that are
regular, stable, and sufficiently high to enable them to
increase their spending--without incurring potentially
dangerous amounts of debt. Higher incomes for more ordinary
Americans will translate directly into higher demand for
consumer goods and services, which will stimulate expansion in
production of such goods and services. That, in turn, will
translate into broader industrial growth, further job creation,
technological innovation, and research and development. It is
in that fundamental sense that, ultimately, consumer demand is
the key catalyst of the Nation's economic growth.
Importantly, however, financing demand through consumer
debt is an unsustainable and extremely dangerous strategy. As
the financial crisis of 2007-09 demonstrated, cheap credit
booms inevitably lead to wealth-destroying economic crashes.
Only by lifting Americans' real income levels can we create the
conditions necessary for spurring sustainable long-term growth
of the American economy. Currently, Americans are forced to
resort to borrowing to make up for their steadily declining
real incomes, so that the aggregate consumer debt now stands at
a record high of about $13 trillion. Such high debt burden,
combined with stagnating wages and continuing erosion of
America's manufacturing base, is bound to depress consumer
demand and, as a result, hold back economic growth.
Deregulating banks will not remedy this underlying dynamic.
Banks currently have plenty of cash available for lending. The
real problem here is not some alleged contraction of banks'
lending capacity because of regulatory compliance costs: it is
the lack of effective and sound demand for credit from over-
extended consumers and struggling businesses in the sluggish
real economy. Deregulating banks will only enable them to use
more of their money for speculative trading, shareholder
dividends, and executive bonuses. It will not magically create
a robust real economy that can put that cash to productive use.

Q.2. Many of us have come to recognize that the Orderly
Liquidation Authority is an incredibly important part of the
Wall Street Reform and Consumer Protection Act. Could you
please explain why OLA is so important to our constituents,
especially those who may be working multiple jobs just to make
ends meet?

A.2. The Orderly Liquidation Authority (OLA) is a special new
regime for handling the failure of a major financial
institution. It was created in response to the financial crisis
of 2007-09, which made it clear that large financial firms--
banks, securities firms, insurance conglomerates, etc.--are
fundamentally different from any regular private company in a
free-market economy. When a nonfinancial company fails, the
ordinary bankruptcy process--governed by the general Bankruptcy
Code and administered by bankruptcy courts--aims to use the
company's remaining assets to satisfy claims of its direct
creditors: lenders, employees, suppliers, utility providers,
etc.
When a large financial institution like JPMorgan or Goldman
Sachs fails, however, this time-consuming court-administered
process does not work as well. Large financial institutions are
highly leveraged, much more so than nonfinancial firms. Their
debt obligations are much more varied, complex, and difficult
to value. Many of these institutions' creditors trade in
financial markets and, therefore, cannot afford to wait for the
bankruptcy court to decide how much money they will get from
the bankrupt firm. If these trading counterparties don't get
paid on time, they may default on their own obligations or try
to avoid that by rapidly selling their own financial assets
(such as stocks, bonds, etc.): in either case, these creditors'
behavior may trigger a dangerous chain reaction. Financial
institutions provide services of critical public importance:
they are central to the smooth operation of the payments and
clearing systems, they manage ordinary Americans' savings and
investments, they insure against various risks, etc. The
bankruptcy of a large financial institution threatens to
disrupt performance of these functions and thus cause much
unanticipated distress in the broader economy.
OLA is designed to provide a tailored approach to handling
the failure of such an institution. It is simply a special
version of corporate bankruptcy, which shifts the primary focus
toward minimizing systemic disruptions that are likely to occur
when a large, systemically important financial firm goes down.
Among other things, OLA enhances the ability of the relevant
Federal regulatory agencies to (1) monitor the financial
condition, solvency and liquidity of all big banks, investment
banks, and other financial firms; (2) mandate that all such
financial firms have in place reliable plans for raising money
and addressing any sudden shocks to their solvency or
liquidity, before such shocks actually hit them; and (3) put in
place the necessary ``safety net'' (i.e., access to last-resort
emergency funding, international agreements with other
countries' regulators, etc.) to ensure that the failure of a
systemically important financial institution does not cause a
major breakdown in the provision of financial services to the
public.
Put simply, the OLA regime is designed to ensure that, even
if a major bank is about to collapse, ordinary Americans can
keep going about their daily business without worrying about
accessing their cash at ATMs, having their checks cleared,
getting their wages deposited and their credit card payments go
through. By strengthening the resilience of the financial
system to the failure of any single firm, the OLA regime is
critical for preventing full-on financial crises and subsequent
economic depressions. It is a well-established fact that
working-class and middle-class people get hit the hardest by
such crises. That means that every hard-working American
directly benefits from the existence and proper functioning of
the OLA regime and other complementary elements of the Dodd-
Frank regulatory system. Repealing or weakening the Dodd-
Frank's OLA provisions, by contrast, will weaken and expose the
American economy to increasingly frequent and devastating
financial crises. Therefore, it is in the interest of every
hard-working American to keep the OLA regime in place.


Additional Material Supplied for the Record


STATEMENT SUBMITTED BY THE INDEPENDENT COMMUNITY BANKERS OF AMERICA
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

STATEMENT SUBMITTED BY THE AMERICAN BAMKERS ASSOCIATION
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]


STATEMENT SUBMITTED BY FIFTH THIRD BANK
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

STATEMENT SUBMITTED BY THE INSTITUTE OF INTERNATIONAL BANKERS
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

STATEMENT SUBMITTED BY TEXAS CAPITAL BANK
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

GSIB SCORES
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]

https://www.congress.gov/event/115th-congress/senate-event/LC51023/text