[Senate Hearing 111-58]
[From the U.S. Government Publishing Office]
S. Hrg. 111-58
ENHANCING INVESTOR PROTECTION AND THE REGULATION OF SECURITIES MARKETS
=======================================================================
HEARING
before the
COMMITTEE ON
BANKING,HOUSING,AND URBAN AFFAIRS
UNITED STATES SENATE
ONE HUNDRED ELEVENTH CONGRESS
FIRST SESSION
ON
EXAMINING WHAT WENT WRONG IN THE SECURITIES MARKETS, HOW WE CAN PREVENT
THE PRACTICES THAT LED TO OUR FINANCIAL SYSTEM PROBLEMS, AND HOW TO
PROTECT INVESTORS
__________
MARCH 10, 2009
__________
Printed for the use of the Committee on Banking, Housing, and Urban
Affairs
Available at: http: //www.access.gpo.gov /congress /senate/
senate05sh.html
U.S. GOVERNMENT PRINTING OFFICE
51-395 WASHINGTON : 2009
-----------------------------------------------------------------------
For sale by the Superintendent of Documents, U.S. Government Printing
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
CHRISTOPHER J. DODD, Connecticut, Chairman
TIM JOHNSON, South Dakota RICHARD C. SHELBY, Alabama
JACK REED, Rhode Island ROBERT F. BENNETT, Utah
CHARLES E. SCHUMER, New York JIM BUNNING, Kentucky
EVAN BAYH, Indiana MIKE CRAPO, Idaho
ROBERT MENENDEZ, New Jersey MEL MARTINEZ, Florida
DANIEL K. AKAKA, Hawaii BOB CORKER, Tennessee
SHERROD BROWN, Ohio JIM DeMINT, South Carolina
JON TESTER, Montana DAVID VITTER, Louisiana
HERB KOHL, Wisconsin MIKE JOHANNS, Nebraska
MARK R. WARNER, Virginia KAY BAILEY HUTCHISON, Texas
JEFF MERKLEY, Oregon
MICHAEL F. BENNET, Colorado
Colin McGinnis, Acting Staff Director
William D. Duhnke, Republican Staff Director
Dean Shahinian, Senior Counsel
Brian Filipowich, Legislative Assistant
Julie Chon, Senior Policy Advisor
Drew Colbert, Legislative Assistant
Mark Oesterle, Republican Deputy Staff Director
Andrew Olmem, Republican Counsel
Hester Peirce, Republican Counsel
Jim Johnson, Republican Counsel
Dawn Ratliff, Chief Clerk
Devin Hartley, Hearing Clerk
Shelvin Simmons, IT Director
Jim Crowell, Editor
(ii)
C O N T E N T S
----------
TUESDAY, MARCH 10, 2009
Page
Opening statement of Chairman Dodd............................... 1
Opening statements, comments, or prepared statements of:
Senator Shelby............................................... 4
Senator Crapo................................................ 5
WITNESSES
John C. Coffee, Jr., Adolf A. Berle Professor of Law, Columbia
University Law School.......................................... 6
Prepared statement........................................... 51
Response to written questions of:
Senator Dodd............................................. 231
Senator Crapo............................................ 235
Senator Vitter........................................... 236
T. Timothy Ryan, Jr., President and Chief Executive Officer,
Securities
Industry and Financial Markets Association..................... 9
Prepared statement........................................... 79
Response to written questions of:
Senator Dodd............................................. 237
Senator Crapo............................................ 269
Paul Schott Stevens, President and Chief Executive Officer,
Investment
Company Institute.............................................. 11
Prepared statement........................................... 83
Response to written questions of:
Senator Dodd............................................. 271
Senator Crapo............................................ 301
Mercer E. Bullard, Associate Professor, University of Mississippi
School of Law, and President, Fund Democracy, Inc.............. 13
Prepared statement........................................... 104
Senator Dodd............................................. 309
Senator Crapo............................................ 312
Robert Pickel, Executive Director and Chief Executive Officer,
International Swaps and Derivatives Association................ 15
Prepared statement........................................... 129
Response to written questions of:
Senator Dodd............................................. 312
Senator Crapo............................................ 313
Damon A. Silvers, Associate General Counsel, AFL-CIO............. 18
Prepared statement........................................... 133
Senator Dodd............................................. 314
Senator Crapo............................................ 321
Thomas Doe, Chief Executive Officer, Municipal Market Advisors... 20
Prepared statement........................................... 144
Response to written questions of:
Senator Dodd............................................. 328
Senator Crapo............................................ 337
Lynn E. Turner, Former Chief Accountant, Securities and Exchange
Commission..................................................... 22
Prepared statement........................................... 220
Senator Dodd............................................. 338
Senator Crapo............................................ 341
Senator Vitter........................................... 341
(iii)
ENHANCING INVESTOR PROTECTION AND THE REGULATION OF SECURITIES MARKETS
----------
TUESDAY, MARCH 10, 2009
U.S. Senate,
Committee on Banking, Housing, and Urban Affairs,
Washington, DC.
The Committee met at 10:38 a.m., in room SD-538, Dirksen
Senate Office Building, Senator Christopher J. Dodd (Chairman
of the Committee) presiding.
OPENING STATEMENT OF CHAIRMAN CHRISTOPHER J. DODD
Chairman Dodd. The Committee will come to order. Let me
thank our witnesses for being here this morning, and colleagues
as well, and just to notify the room how we will proceed.
Again, there are only a handful of us here, but we have eight
witnesses, and so we have got a long morning in front of us to
go through these issues. And what I would like to do is I will
make some opening remarks, turn to Senator Shelby, and then as
long as the room does not all of a sudden get crowded with a
lot of Members here, I will ask Senator Reed and Senator Bennet
if you would like to make a couple of opening comments, and we
will get right to our witnesses, who have supplied very
thorough testimony. And if they each read all of their
testimony, we are going to be here until Friday, in a sense.
But it is very, very good and very helpful to us. So we will
proceed along those lines and hopefully have a good, engaging
morning here on a very, very critical issue.
So I welcome all of you to the hearing this morning
entitled ``Enhancing Investor Protection and the Regulation of
Securities Markets.'' The purpose of today's hearing is to
examine what went wrong in the securities markets and to
discuss how we can prevent irresponsible practices that led to
our financial system seizing up from ever happening again and
how to protect investors, including small investors, from
getting burned by the kinds of serious abuses and irresponsible
behavior that we have seen in certain quarters of the markets
in recent years.
We are going to hear about proposals to regulate the
securities market so that it supports economic growth and
protects investors rather than threatens economic stability. As
important, today we will begin to chart a course forward--a
course that acknowledges how complex products and risky
practices can do enormous damage to the heart of our financial
system, the American people as well, absent a strong foundation
of consumer and investor protections.
Half of all U.S. households are invested in some way in
securities, meaning the path we choose for regulating this
growth and growing segment of our financial system will
determine the futures not only of traders on Wall Street but of
families, of course, across the country. A year ago this coming
Saturday, the collapse of Bear Stearns underscored the
importance role that securities play in our financial system
today.
When I was elected to the Senate in 1980, bank deposits
represented 45 percent of the financial assets of the United
States and securities represented 55 percent. Today, the
securities sector dominates our financial system, representing
80 percent of financial assets, with bank deposits a mere 20
percent.
As the securities market has expanded, so, too, has its
influence on the lives of average citizens. Much of that
expansion has been driven by the process known as
``securitization,'' in which everyday household debt is pooled
into sophisticated structures, from mortgages and auto loans to
credit cards and student loans.
In time, however, Wall Street not only traded that debt, it
began to pressure others into making riskier and riskier loans
to consumers. And lenders, brokers, and credit card companies
were all too willing to comply, pushing the middle-class family
in my State of Connecticut and elsewhere across the country who
would have qualified for a traditional secure product into a
riskier subprime mortgage or giving that 17-year-old college
student, who never should have qualified in the first place, a
credit card with teaser rates that were irresistible but terms
that were suffocating.
As one trader said of the notorious subprime lender, they
were moving money out of the door to Wall Street so fast, with
so few questions asked, these loans were not merely risky, they
were, in fact, built to self-destruct.
As we knew it, securitization did not reallocate risk. It
spread risk throughout our financial system, passing it on to
others like a high-stakes game of hot potato. With no incentive
to make sure these risky loans paid off down the road, each
link in the securitization chain--the loan originators, Wall
Street firms and fund managers, with the help of credit rating
agencies--generated more risk. They piled on layers of loans
into mortgage-backed securities, which were piled into
collateralized debt obligations, which were in turn piled into
CDO squared and cubed, severing the relationship between the
underlying consumer and their financial institutions.
Like a top-heavy structure built on shoddy foundations, it
all, of course, came crashing down. I firmly believe that had
the Fed simply regulated the mortgage lending industry, as
Congress directed with the law passed in 1994, much of this
could have been averted. But despite the efforts of my
predecessor on this Committee, myself, and others over many
years, the Fed refused to act.
But the failure of regulators was not limited to mortgage-
backed securities. As many constituents in Connecticut and
elsewhere have told me, auction rate securities, misleadingly
marketed as cash equivalents, left countless investors and city
pension funds across the country with nothing when the actions
failed and the securities could not be redeemed.
As this Committee uncovered at a hearing about AIG last
week, the unregulated credit derivatives market contributed to
the largest quarterly loss in history.
In recent months, we have unearthed two massive Ponzi
schemes, bilking consumers, investors, charities, and municipal
pension funds out of tens of billions of dollars that two
separate regulators failed to detect in their examinations. In
January, I asked Dr. Henry Backe of Fairfield, Connecticut, to
address this Committee about the losses suffered by the
employees at his medical practice in the Bernard Madoff fraud.
His testimony prompted Senator Menendez and me to urge the IRS
to dedicate serious resources to helping victims like Linda
Alexander, a 62-year-old telephone operator from Bridgeport,
Connecticut, who makes less than $480 a week and lost every
penny of her retirement savings. In an instant, the $10,000 she
had saved over a lifetime evaporated because regulators has no
idea a massive fraud was occurring right under their noses.
This crisis is the result of what may have been the
greatest regulatory failure in human history. If you need any
further evidence, consider this: At the beginning of the credit
crisis in 2008, the SEC regulated five investment banks under
the Consolidated Supervised Entity Program: Lehman Brothers,
Bear Stearns, Merrill Lynch, Goldman Sachs, and Morgan
Stanley--names synonymous with America's financial strength,
having survived world wars and the Great Depression. And though
the seeds of their destruction have been planted nearly a
decade ago, each was sold, converted to a bank holding company,
or failed outright inside of 6 months--every single one of
them.
Our task today is to continue our examination of how to
begin rebuilding a 21st century financial structure. We do so
not from the top down, focusing solely on the soundness of the
largest institutions, with the hope that it trickles down to
the consumer but, rather, from the bottom up, ensuring a new
responsibility in financial services and a tough new set of
protections for regular investors who thought these protections
were already in place.
The bottom-up approach will create a new way of regulating
Wall Street. For the securities markets, that means examining
everything, from the regulated broker-dealers and their sales
practices, to unregulated credit default swaps. It means
ensuring that the creators of financial products have as much
skin in the game when they package these products as the
consumers do when they buy them, so that instead of passing on
risk, everyone shares responsibility. And that means we need
more transparency from public companies, credit rating
agencies, municipalities, and banks.
We are going to send a very clear message that these
modernization efforts, the era of ``don't ask''--in these
modernization efforts, the era of ``don't ask, don't tell'' on
Wall Street and elsewhere is over. For decades, vitality,
innovation, and creativity have been a source of genius of our
system, and I want to see that come back. It is time we
recognized transparency and responsibility are every bit as
paramount, that whether we are homebuyers, city managers,
entrepreneurs, we can only make responsible decisions if we
have the accurate and proper information. We want the American
people to know that this Committee will do everything in its
power to get us out of this crisis by putting the needs of
people first, from constituents like Linda Alexander, who I
mentioned a moment ago, to millions more whose hard-earned
dollars are tied up in our securities markets.
Today's hearing will provide an opportunity to hear ideas
and build a record upon which this Committee can legislate a
way forward for the American people to rebuild confidence in
these securities markets, and to put our country back on a
sound economic footing.
With that, I thank our witnesses again for being here, and
let me turn to my colleague, former Chairman, Senator Shelby.
STATEMENT OF SENATOR RICHARD C. SHELBY
Senator Shelby. Thank you, Chairman Dodd.
I think the greatest challenge in dealing with this
financial crisis is understanding its multiple, complex, and
interrelated causes. This hearing provides us an opportunity to
examine some of the causes that relate to our securities
markets and securities regulation.
Without presupposing the specific causes of the financial
crisis, I think it is appropriate to conclude that a broad
failure of risk management in the financial system led us to
where we are today. It appears that everybody assumed that
someone else was monitoring the risk. Regulators assumed that
financial institutions had properly assessed the risk of their
own activities or assumed that other regulators were watching
what those entities were doing.
Financial institutions failed to adequately monitor risks
across business units and failed to thoroughly understand the
risks associated with new financial products. They did not
adequately assess either their exposures to or the health of
their counterparties.
Very sophisticated investors assumed that someone else had
done their due diligence. Less sophisticated investors assumed,
unreasonably, that asset prices would only climb. The excessive
reliance on credit ratings and the failure of the market to
develop a clearinghouse for credit default swaps are just two
examples of this widespread market failure. The disastrous
consequences of this nearly universal passing of the buck
should serve as the guidepost for us and the SEC as we consider
reforms.
I think there should be clear lines of responsibility for
regulators. Only then can Congress hold regulators accountable
for their performance. It is also important not to make changes
to the statutory and regulatory framework that would further
lull market participants into believing that regulators or
other market participants are doing their work for them.
We cannot build a regulator big enough to be everywhere at
all times. Market participants need to do their own due
diligence before and after they make an investment decision.
They need to bear the costs of an unwise investment, just as
they reap the benefit of a wise investment. In the end, I
believe our markets will be best served by the combined efforts
of diligent regulators and responsible market participants
working under rules that are clear and consistent.
Uncertainty about the rules impedes the market from working
as it should. Ad hoc Government actions lead private capital to
sit on the sidelines because a change in rules can radically
change a market participant's expected return. A consistent
legal framework is an essential component of a competitive
capital market. Investors will avoid a market if they believe
the rules may change in the middle of the game. A clear example
of this dynamic is the world of accounting where many are
calling for the suspensions of mark-to-market because of the
adverse impact that it is presently having on the books of so
many companies.
Accounting rules should be designed to ensure that a firm's
disclosures reflect economic reality, however ugly that reality
may be. Changing the accounting rules now will simply compound
investors' wariness about investing in a market where many
firms have bad or illiquid assets on their books.
I will be interested in hearing from today's witnesses on
this topic and how the SEC can improve its efforts to protect
our securities markets while also facilitating continued
innovation and responsible risk taking.
Chairman Dodd, I thank you for calling this hearing. I
think you are on to something here.
Chairman Dodd. Well, Senator, thank you very, very much. We
have got quite a row here of witnesses to testify. Let me turn
to Jack Reed or Michael Bennet. Any opening quick comments,
Jack?
Senator Reed. Mr. Chairman, I am very interested in hearing
the witnesses assembled. We have an impressive panel. Thank
you.
Chairman Dodd. Senator Bennet.
Senator Bennet. Mr. Chairman, thanks for holding the
hearing. It is from this perspective truly a row of witnesses.
[Laughter.]
Senator Bennet. So I will wait until we are done.
Chairman Dodd. Let me invite you as well, Senator, unless
others show up, we are more than willing to have you--I am
sorry I did not see you. If you want to move on up and join us
here.
[Laughter.]
Chairman Dodd. If not, I am fearful I may call on you as a
witness here.
Senator Bennet. Well, I have a point of view. I would be
happy to----
Chairman Dodd. I am sure you do--I know you do.
Senator Shelby. We have all sat there.
Chairman Dodd. Senator Shelby just pointed out, we have
been in that chair before. In fact, I think I was a chair
further back in the room a long time ago along the way.
Senator Crapo.
STATEMENT OF SENATOR MIKE CRAPO
Senator Crapo. I was in that chair on this side. I know
very much what it is like.
Thank you very much, Mr. Chairman. I will be very brief. I
really appreciate your holding this and the other hearings that
you have scheduled. As you know, I am very interested in this
issue, and I look forward to working with you. And let's get on
with the witnesses.
Chairman Dodd. You bet. And I appreciate it. Senator Crapo
has had a longstanding interest in reg reform issues, and he
has expressed to me on numerous occasions his desire to be
involved in this discussion, as has Senator Bennet and others.
So we have some very interested Members on the Committee who
want to work together on this issue as we move forward in the
coming weeks to put together a bill.
I was listening this morning to the speech by Ben Bernanke
talking about his ideas--and maybe some of our witnesses--I
know you have prepared statements, but certainly feel free in
your comments to react to some of his thoughts this morning.
That would be welcomed as well, since he made the speech this
morning--where was it? Brookings?
Senator Shelby. Foreign Relations.
Chairman Dodd. Oh, Council on Foreign Relations.
With that, let me briefly--again, I think most of our folks
here know our witnesses. Very briefly, we have Professor John
Coffee, who has been before us many times. He is the Adolf A.
Berle Professor of Law at Columbia's Law School; Timothy Ryan,
President and CEO of Securities Industry and Financial Markets
Association; Paul Schott Stevens, President and CEO of
Investment Company Institute; Professor Mercer Bullard,
Associate Professors, University of Mississippi School of Law;
Robert Pickel, who is the Executive Director and CEO of
International Swaps and Derivatives Association; Damon Silvers,
the Associate General Counsel of the AFL-CIO; and Thomas Doe,
CEO of Municipal Market Advisors; Lynn Turner, the former Chief
Accountant, U.S. Securities and Exchange Commission.
So a good row of witnesses here to testify, and we will
begin in the order that I have introduced you. Dr. Coffee, you
seem to occupy that chair every time you come here.
STATEMENT OF JOHN C. COFFEE, JR., ADOLF A. BERLE PROFESSOR OF
LAW, COLUMBIA UNIVERSITY LAW SCHOOL
Mr. Coffee. Well, good morning, and thank you, Chairman
Dodd, Ranking Member Shelby, and fellow Senators. I have
prepared an overly long, bulky, 70-page memorandum for which I
apologize for inflicting on you. It attempts to synthesize a
good deal of recent empirical research by business school
scholars, finance scholars, and even law professors, about just
what went wrong and what can be done about it.
I cannot summarize all that, but I would add the following
two sentences to what Senators Dodd and Shelby very accurately
said at the outset. The current financial crisis is unlike
others. This was not a bubble caused by investor mania, which
is the typical cause of bubbles. It was not a demand-driven
bubble; rather, it was more a supply driven bubble. It was the
product of a particular business model, a model known as the
``originate-and-distribute model,'' under which financial
institutions, including loan originators, mortgage lenders, and
investment banks, all behaved similarly and went to the brink
of insolvency and beyond, pursuing a model.
What is the key element of this originate-and-distribute
model? You make lax loans. You make non-creditworthy loans
because--because you do not expect to hold those loans for long
enough to matter. You believe that you can transfer these loans
to the next link in the transmission chain before you will bear
the economic risk.
When everyone believes that--and they correctly believed
that for a few years--then all standards begin to become
relaxed, and we believe that as long as we can get that
investment grade rating from the credit rating agencies, we
will have no problem, and weak loans can always be marketed.
There is no time for statistics here, but let me add just one.
Between 2001 and 2006, a relatively short period, some of
the data that I cite shows you that low-document loans in these
portfolios went from being something like 28 percent in
mortgage-backed securities in 2001 to 51 percent in 2006--
doubling in 4 or 5 years. Investment banks and credit rating
agencies are not responding to that change. That is the
essential problem.
This gives rise to what I will call and economists call a
``moral hazard problem,'' and this moral hazard problem was
compounded by deregulatory policies that the SEC and other
institutions followed that permitted investment banks to
increase their leverage dramatically between 2004 and 2006,
which is only just a few years ago. This is yesterday we are
talking about. They did this pursuant to the Consolidated
Supervised Entity Program that you have already been
discussing, and it led to the downfall of our five largest,
most important investment banks.
All right. Essentially, the SEC deferred to self-
regulation, by which these five largest banks constructed their
own credit risk models, and the SEC deferred to them. The 2008
experience shows, if there ever was any doubt, that in an
environment of intense competition and under the pressure of
equity-based executive compensation systems that tend to be
very short-term oriented, self-regulation alone simply does not
work. The simplest way for a financial institution to increase
profitability was to increase its leverage, and it did so to
the point where they were leveraged to the eyeballs and could
not survive the predictable downturn in the economic weather.
So what should be done from a policy perspective? Well,
here is my first and most essential point: All financial
institutions that are too big to fail, which really means too
entangled to fail, need to be subjected to prudential financial
oversight, what I would call ``financial adult supervision,''
from a common regulator applying a basically common although
risk-adjusted standard to all these institutions, whether they
are insurance companies, banks, thrifts, hedge funds, money
market funds, or even pension plans, or the financial
subsidiaries of very large corporations, like GE Capital. In my
judgment, this can only be done by the Federal Reserve Board.
That is the only person in a position to serve as what is
called the ``systemic risk regulator.''
I think we need in this country a systemic risk regulator,
and specifically to define what this means, let me say there
are five areas where their authority should be established. The
Federal Reserve Board should be authorized and mandated to do
the following five things:
One, establish ceilings on debt-to-equity ratios and
otherwise restrict leverage for all major financial
institutions.
Two, supervise and restrict the design and trading of new
financial products, including, in particular, over-the-counter
derivatives and including the posting of margin and collateral
for such products.
Three, mandate the use of clearinghouses. The Federal
Reserve has already been doing this, formulating this, trying
to facilitate this, but mandating it is more important. And
they need the authority to supervise these clearing houses, and
also if they judge it to be wise and prudent, to require their
consolidation into a single clearinghouse.
Four, the Federal Reserve needs the authority to require
the writedown of risky assets by financial institutions,
regardless of whether accounting rules mandate it. The
accountants will always be the last to demand a writedown
because their clients do not want it. The regulator is going to
have to be more proactive than are the accounting firms.
Last, the Federal Reserve should be authorized to prevent
liquidity crises that come from the mismatch of assets and
liabilities. The simple truth is that financial institutions
hold long-term illiquid assets which they finance through
short-term paper that they have to roll over regularly, and
that mismatch regularly causes problems.
Now, under this ``Twin Peaks'' model that I am describing,
the systematic risk regulator--presumably, the Federal
Reserve--would have broad authority. But the power should not
be given to the Federal Reserve to override the consumer
protection and transparency policies of the SEC. And this is a
co-equal point with my first point, that we need a systemic
risk regulator. Too often, bank regulators and banks have
engaged in what I would term a ``conspiracy of silence'' to
hide problems, lest investors find out, become alarmed, and
create a run on the bank.
The culture of banking regulators and the culture of
securities regulators is entirely different. Bank regulators do
not want to alarm investors. Securities regulators understand
that sunlight is the best disinfectant. And for the long run,
just as Senator Shelby said, we need accounting policies that
reveal the ugly truth.
We could not be worse off now in terms of lack of public
confidence. This is precisely the moment to make everyone
recognize what the truth is and not to give any regulator the
authority to suppress the truth under the guise of systematic
risk regulation.
For that reason, I think SEC responsibilities for
disclosure, transparency, and accounting should be specially
spelled out and exempted from any power that the systematic
risk regulator has to overrule other policies.
Now, two last points. As a financial technology, asset-
based securitization, at least in the real estate field, has
decisively failed. I think two steps should be done by
legislation to mandate the one policies that I think will
restore credibility to this field.
First, to restore credibility, sponsors must abandon the
originate-and-distribute business model and instead commit to
retain at least a portion of the most subordinated tranche, the
riskiest assets. Some of them have to be held by the promoter
because that is the one signal of commitment that tells the
marketplace that someone has investigated these assets because
they are holding the weakest, most likely to fail. That would
be step one.
Step two, we need to reintroduce due diligence into the
process, into the securitization process, both for public
offerings and for Rule 144A offerings, which are private
offerings. Right now Regulation AB deregulated; it does not
really require adequately that the sponsor verify the loans,
have the loan documentation in its possession, or to have
examined the creditworthiness of the individual securities. I
think the SEC can be instructed by Congress that there needs to
be a reintroduction of stronger due diligence into both the
public and the private placement process.
Last point. Credit rating agencies are obviously the
gatekeeper who failed most in this current crisis. The one
thing they do not do that other gatekeepers do do is verify the
information they are relying on. Their have their rating
methodology, but they just assume what they are told; they do
not verify it. I think they should be instructed that there has
to be verification either by them or by responsible,
independent professionals who certify their results to them.
The only way to make that system work and to give it teeth
is to reframe a special standard of liability for the credit
rating agencies. I believe the Congress can do this, and I
believe that Senator Reed and his staff are already examining
closely the need for additional legislation for credit rating
agencies, and I think they are very much on the right track,
and I would encourage them.
What I am saying, in closing, is that a very painful period
of deleveraging is necessary. No one is going to like it. I
think some responsibility should be given to the Federal
Reserve as the overall systematic risk regulator, but they
should not have authority to in any way overrule the SEC's
policies on transparency.
Thank you.
Chairman Dodd. Thank you very much, Professor.
Mr. Ryan, welcome.
STATEMENT OF T. TIMOTHY RYAN, JR., PRESIDENT AND CHIEF
EXECUTIVE OFFICER, SECURITIES INDUSTRY AND
FINANCIAL MARKETS ASSOCIATION
Mr. Ryan. Good morning. Thank you for inviting me. I
appreciate being here.
Our current financial crisis, which has affected nearly
every American family, underscores the imperative to modernize
our financial regulatory system. Our regulatory structure and
the plethora of regulations applicable to financial
institutions are based on historical distinctions between
banks, securities firms, insurance companies, and other
financial institutions--distinctions that no longer conform to
the way business is conducted.
The negative consequences to the investing public of this
patchwork of regulatory oversight are real and pervasive.
Investors do not have comparable protections across the same or
similar financial products. Rather, the disclosures, standards
of care, and other key investor protections vary based on the
legal status of the intermediary or the product or service
being offered.
In light of these concerns, SIFMA advocates simplifying and
reforming the financial regulatory structure to maximize and
enhance investor protection and market integrity and
efficiency.
Systemic risk, as Professor Coffee noted, has been at the
heart of the current financial crisis. As I have previously
testified, we at SIFMA believe that a single, accountable
financial markets stability regulator, a systemic regulator,
will improve upon the current system. While our position on the
mission of the financial markets stability regulator is still
evolving, we currently believe that its mission should consist
of mitigating systemic risk, maintaining financial stability,
and addressing any financial crisis, all of which will benefit
the investing public. It should have authority over all markets
and market participants, regardless of charter, functional
regulator, or unregulated status. It should have the authority
to gather information from all financial institutions and
markets, adopt uniform regulations related to systemic risk,
and act as a lender of last resort. It should probably have a
more direct role in supervising systemically important
financial organizations, including the power to conduct
examinations, take prompt corrective action, and appoint or act
as the receiver or conservator of all or part of systemically
important organizations.
We also believe, as a second step, that we must work to
rationalize the broader regulatory framework to eliminate
regulatory gaps and imbalances that contribute to systemic risk
by regulating similar activities and firms in a similar manner
and by consolidating certain financial regulators.
SIFMA has long advocated the modernization and
harmonization of the disparate regulatory regimes for
investment advisory, brokerage, and other financial services in
order to promote investor protection. SIFMA recommends the
adoption of a ``universal standard of care'' that avoids the
use of labels that tend to confuse the investing public and
expresses, in plain English, the fundamental principles of fair
dealing that individual investors can expect from all of their
financial services providers. Such a standard could provide a
uniform code of conduct applicable to all financial
professionals. It would make clear to all individual investors
that their financial professionals are obligated to treat them
fairly by employing the same core standards whether the firm is
a financial planner, an investment adviser, a securities
dealer, a bank, an insurance agency, or any other type of
financial service provider.
The U.S. is the only jurisdiction that splits the oversight
of securities and futures activities between two separate
regulatory bodies. We support the merger of the SEC and the
CFTC.
We believe that the development of a clearinghouse for
credit derivatives is an effective way to reduce counterparty
credit risk, facilitate regulatory oversight, and, thus,
promote market stability. In particular, we strongly support
our members' initiative to establish a clearinghouse of CDS,
and we are pleased to note that ICE US Trust opened its doors
for clearing CDS transactions yesterday.
Finally, the current financial crisis reminds us that
markets are global in nature, and so are the risks of
contagion. To promote investor protection through effective
regulation and the elimination of disparate regulatory
treatment, we believe that common regulatory standards should
be applied consistently across markets. Accordingly, we urge
that steps be taken to foster greater cooperation and
coordination among regulators in all major markets.
Thank you.
Chairman Dodd. Thank you very much for that.
Paul, welcome. It is nice to see you and have you before
the Committee.
STATEMENT OF PAUL SCHOTT STEVENS, PRESIDENT AND CHIEF EXECUTIVE
OFFICER, INVESTMENT COMPANY
INSTITUTE
Mr. Stevens. Thank you very much, Mr. Chairman. On behalf
of the Institute and our member funds, I thank you, Chairman
Dodd, Senator Shelby, and all the Members of the Committee for
making it possible for me to appear today. We serve 93 million
American investors, as you know, and we strongly commend the
Committee for the attention you are devoting to improving our
system of financial regulation.
I believe the current financial crisis provides a very
strong public mandate for Congress and for regulators to take
bold steps to strengthen and modernize regulatory oversight.
Like other stakeholders, and there are many, of course, we have
been thinking hard about how to revamp the current system. Last
week, we published a white paper detailing a variety of
reforms, and in it we recommend changes to create a regulatory
framework that provides strong consumer and investor protection
while also enhancing regulatory efficiencies, limiting
duplication, closing conspicuous regulatory gaps, and frankly,
emphasizing the national character of our financial services
markets. I would like briefly to summarize the proposals.
First, we believe it is crucial to improve the government's
capability to monitor and mitigate risks across the financial
system, so ICI supports creation of a Systemic Risk Regulator.
This could be a new or an existing agency or interagency body,
and in our judgment should be responsible for monitoring the
financial markets broadly, analyzing changing conditions here
and overseas, evaluating and identifying risks that are so
significant that they implicate the health of the financial
system, and acting in coordination with other responsible
regulators to mitigate these risks.
In our paper, we stress the need to carefully define the
responsibilities of a Systemic Risk Regulator as well as its
relationships with other regulators, and I would say, Mr.
Chairman, that is one of the points that Chairman Bernanke made
in his speech today, to leverage the expertise and to work
closely with other responsible regulators in accomplishing that
mission.
In our judgment, addressing systemic risk effectively,
however, need not and should not mean stifling innovation,
retarding competition, or compromising market efficiency. You
can achieve all of these purposes, it seems to us, at the same
time.
Second, we urge the creation of a new Capital Markets
Regulator that would combine the functions of the SEC and the
CFTC. This Capital Markets Regulator's statutory mission should
focus sharply on investor protection and law enforcement. It
should also have a mandate, as the SEC does currently, to
consider whether its proposed regulations promote efficiency,
competition, and capital formation.
We suggest several ways to maximize the effectiveness of
the new Capital Markets Regulator. In particular, we would
suggest a need for a very high-level focus on management of the
agency, its resources, and its responsibilities, and also the
establishment of mechanisms to allow it to stay much more
effectively abreast of market and industry developments.
Third, as we discuss more fully in our white paper,
effective oversight of the financial system and mitigation of
systemic risk will require effective coordination and
information sharing among the Systemic Risk Regulator and
regulators responsible for other financial sectors.
Fourth, we have identified areas in which the Capital
Markets Regulator needs more specific legislative authority to
protect investors and the markets by closing regulatory gaps
and responding to changes in the marketplace. In my written
statement, I identify four such areas: hedge funds,
derivatives, municipal securities, particularly to improve
disclosure standards, and the inconsistent regulatory regimes
that exist today for investment advisors and broker dealers.
Now, as for mutual funds, they have not been immune from
the effects of the financial crisis, nor, for that matter, have
any other investors. But our regulatory structure, and this
bears emphasizing, which grew out of the New Deal as a result
of the last great financial crisis, has proven to be remarkably
resilient, even through the current one.
Under the Investment Company Act of 1940 and other
securities laws, fund investors enjoy a range of vital
protections: Daily pricing of fund shares with mark-to-market
valuation every business day; separate custody of all fund
assets; minimal or no use of leverage in our funds;
restrictions on affiliated transactions and other forms of
self-dealing; required diversification; and the most extensive
disclosure requirements faced by any financial products.
Funds have embraced this regulatory regime and they have
prospered under it. Indeed, I think recent experience suggests
that policymakers should consider extending some of these very
same disciplines that have worked so well for us since 1940 to
other marketplace participants in reaction to the crisis that
we are experiencing today.
Finally, let me comment, Mr. Chairman, briefly on money
market funds. Last September, immediately following the
bankruptcy of Lehman Brothers, a single money market fund was
unable to sustain its $1 per share net asset value. Coming hard
on the heels of a series of other extraordinary developments
that roiled global financial markets, these events worsened an
already severe credit squeeze. Investors wondered what other
major financial institution might fail next and how other money
market funds might be affected.
Concern that the short-term fixed income market was all but
frozen solid, the Federal Reserve and the Treasury Department
took a variety of initiatives, including the establishment of a
temporary guarantee program for money market funds. These steps
have proved highly successful. Over time, investors have
regained confidence. As of February, assets in money market
funds were at an all-time high, almost $3.9 trillion.
The Treasury Department's temporary guarantee program will
end no later than September 18. Funds have paid more than $800
million in premiums, yet no claims have been made and we do not
expect any claims to be made. We do not envision any future
role for Federal insurance of money market fund assets and we
look forward to an orderly transition out of the temporary
guarantee program.
The events of last fall were unprecedented, but it is only
responsible that we, the fund industry, look for lessons
learned. So in November 2008, ICI formed a working group of
senior fund industry leaders to study ways to minimize the risk
to money market funds of even the most extreme market
conditions. That group will issue a strong and comprehensive
set of recommendations designed, among other things, to enhance
the way money funds operate. We expect that report by the end
of the month. We hope to place the executive summary in the
record of this hearing, and Mr. Chairman, I would be delighted
to return to the Committee, if it is of interest to you, to
present those recommendations at a future date.
Thank you very much.
Chairman Dodd. Thank you very much for that. We will
welcome that addition to the Committee record, as well.
Professor Bullard, thank you very much for joining us.
STATEMENT OF MERCER E. BULLARD, ASSOCIATE PROFESSOR, UNIVERSITY
OF MISSISSIPPI SCHOOL OF LAW, AND PRESIDENT, FUND DEMOCRACY,
INC.
Mr. Bullard. Thank you, Chairman Dodd, Ranking Member
Shelby, and Members of the Committee for the opportunity to
appear here today. I congratulate the Committee for its
thorough and deliberate investigation into the causes of the
current financial crisis.
Recent events have provided useful lessons on the
management of systemic risk, prudential regulation, and
investor protection in the investment management industry. The
performance of stock and bond mutual funds, for example, has
demonstrated the remarkable resiliency of the investment
company structure in times of stress. As equity values have
plummeted, most shareholders and stock funds have stood their
ground, notwithstanding that they have the right to redeem
their shares at short notice at their NAVs.
There is no scientific explanation for the stability of
mutual funds during this crisis, but I believe it is related to
this redemption right, as Paul was describing a moment ago.
Mutual fund investors are confident that they will receive the
net asset value of their holdings upon redemption and they
appear to believe that the net asset value of those shares--the
net asset value will be fair and accurate. This confidence in
the valuation and redeemability of mutual fund shares reduces
the likelihood of the kind of panic selling that creates
systemic risk and may provide a useful lesson for the
regulation of other financial intermediaries.
The current crisis has exposed certain investor protection
issues, however. Many investors in target date and short-term
bond funds have experienced investment returns that are not
consistent with returns typical of that asset class. If a fund
uses the name Target Date 2010, for example, its equity
allocation should fall within the generally expected range for
someone on the brink of retirement. Similarly, a 529 plan
option that is touted as appropriate for a 16-year-old should
not lose 40 percent of its value 2 years before the money will
be needed for college.
Investors should be free to choose more aggressive asset
allocations than those normally considered most appropriate for
this situation. But funds that use a name that most investors
will assume reflects a particular strategy should be required
to invest consistent with that strategy.
In contrast with other types of mutual funds, the
performance of money market funds has raised systemic and
prudential regulation concerns. Money market funds constitute a
major linchpin in our payment system and therefore a run on
these funds poses significant systemic risk. The management of
this risk has been inadequate, as demonstrated by the recent
run on money market funds following the failure of the reserve
primary fund.
There are important lessons to be learned from this
experience, but not the lessons that some commentators have
found. The Group of 30, for example, has recommended that the
money market franchise be eliminated. Former Fed Chairman
Volcker explained that if money market funds are going to talk
like a bank and squawk like a bank, they ought to be regulated
like a bank. The problem with this argument is that money
market funds don't fail like banks.
Since 1980, more than 3,000 U.S. banks have failed, costing
taxpayers hundreds of billions of dollars. During the same
period, two money market funds have failed, costing taxpayers
zero dollars. I agree that a regulatory rearrangement is in
order, but it is banks that should be regulated more like money
market funds. Banks routinely fail because they invest in
risky, long-term assets while money market funds invest in
safe, short-term assets. Insuring bank accounts may be
necessary to protect against the systemic risk that a run on
banks poses to the payment system, but there is no good reason
why banks should be permitted to invest insured deposits in
anything other than the safest assets. And there is no good
reason why money market funds that pose the same systemic risk
to our payment system should be left uninsured.
I note, Chairman Dodd, you may have picked up this morning
on Chairman Bernanke's comments that some kind of interim
insurance program may be an appropriate response to the crisis,
and I have to disagree with Paul that the program will
necessarily end in September. I posted an article on SSRN that
deals with one thesis of how to approach money market fund
insurance and I hope the Committee and staff will take a look
at that.
The current crisis has also exposed significant weaknesses
in hedge fund and investment advisor regulation. For example,
hedge funds are permitted to sell investments to any person
with a net worth of at least $1 million, a minimum that has not
been adjusted since 1982. This means that a hedge fund is free
to sell interest to a recently retired couple that owns a
$250,000 house and has $750,000 in investments, notwithstanding
that their retirement income is likely to be around $35,000 a
year before Social Security.
Finally, the Madoff scandal has again reminded us of the
risks of the SEC's expansive interpretation of the broker
exclusion from the definition of investment advisor. It appears
that Madoff did not register as an investment advisor in
reliance on the SEC's position that managing discretionary
accounts could somehow be viewed as solely incidental to
related brokerage services. This over-broad exclusion left
Madoff subject only to broker regulation, which failed to
uncover this fraud.
The SEC has since rescinded its ill-advised position on
discretionary accounts, but it continues to read the ``solely
incidental'' exception so broadly as to leave thousands of
brokers who provide individualized investment advice subject
only to a broker's suitability obligation. These brokers should
be subject to the same fiduciary duties that other investment
advisors are subject to, including the duty to disclose revenue
sharing payments and other compensation that create a potential
conflict with their clients' interests.
And finally, I would just add to the comments on the
question of systemic or prudential regulator. I agree with
Professor Coffee's comments that there is something simply
fundamentally inconsistent with the SEC's investor protection
role and the prudential role that it has not served
particularly well in recent times and that those roles should
be separated. I agree that there should be a Federal Prudential
Regulator, which is what I would call it, that oversees all of
those similar characteristics, such as net capital rules, money
market fund rules, banking regulations that share those
prudential or systemic risk concerns. It is not clear to me,
however, that the particular types of liabilities that
insurance companies hold would be suitable for one common
prudential regulator, but that is something we don't
necessarily need to consider unless the Federalizing of
insurance regulation begins to make additional progress.
And I would also add that we need to keep in mind that
there is a significant difference between customer protection
and investor protection. I think when Paul was talking about a
Capital Markets Regulator, the way I would think of capital
markets as being a way of talking about a regulator as investor
protection regulator, which would serve fundamentally different
functions, especially in that it embraces risk and looks to the
full disclosure of that risk as its principal objective as
opposed to what might be viewed as customer protection, which
is really to ensure that promised services are what are
delivered in a fully disclosed and honest way.
These and other issues are addressed in greater detail in
my written submission. Thank you very much.
Chairman Dodd. Thank you very, very much.
Mr. Pickel, we thank you for joining the Committee.
STATEMENT OF ROBERT PICKEL, EXECUTIVE DIRECTOR AND CHIEF
EXECUTIVE OFFICER, INTERNATIONAL SWAPS AND DERIVATIVES
ASSOCIATION
Mr. Pickel. Thank you, Mr. Chairman and Ranking Member
Shelby and Members of the Committee. Thank you for inviting
ISDA to testify here today. We are grateful for the opportunity
to discuss the privately negotiated derivatives business, and
more specifically, the credit default swaps market.
I have submitted written testimony, and as you noted, Mr.
Chairman, it is a lengthy submission and so I would like to
summarize some of the key remarks that I included in that
testimony.
I think first and foremost, we need to understand that the
benefits of the OTC derivatives business are significant for
the American economy and American companies. They manage a
broad range of risk using these instruments that are not
central to their businesses, allowing them to manage these
financial risks typically so that they can focus on the
business that they run. So a borrower borrowing on a floating
basis can use an interest rate swap to manage its exposure to
exchange fixed for floating obligations.
Currency exposure--many companies have significant
operations around the world and have significant currency
exposure and use currency swaps, OTC currency swaps, to manage
that risk. ISDA itself uses currency swaps to manage its
overseas exposure.
Commodity exposure--airlines have significant exposure to
fuel costs and they typically look to utilize OTC derivatives
to manage that exposure.
And finally, credit exposure--using credit derivatives,
credit default swaps, exposure to suppliers or to customers
where credit is a significant concern, companies can use these
products to help manage that risk.
These products also create efficiencies in pricing and
wider availability of credit, particularly credit default
swaps. They facilitate lending at lower rates, and they are
critical to have available, and have them widely available, as
we come out of this recession. I think they will be an
important part of the ability of firms to manage credit risk as
they look at these important credit issues that we face in this
financial crisis.
As far as the role of the credit default swaps and OTC
derivatives generally in the financial crisis, first of all, I
think, and this Committee has certainly heard testimony, the
fundamental source of the crisis is imprudent lending,
particularly in the U.S. housing sector, but extending to other
markets, as well, credit cards and commercial lending as an
example. We must distinguish between the credit default swap
business and the collateralized debt obligation business. There
has been reference to the originate to distribute model. That
certainly applies to the securitization process and to the CDO
process.
In the credit default swap business, a company, a bank that
has lent money may use a credit default swap to hedge its
exposure on that credit. In that process, it will be
maintaining its lending relationship with the borrower and it
will also be taking on credit risk and paying a fee to the
company that is selling protection. So it is distinctly
different from the originate to distribute model.
We certainly have heard testimony, and this Committee heard
testimony last week on the AIG situation. I think we need to
spend some time and this organization needs to spend some time
talking about that. AIG obviously was significantly involved in
credit default swaps. It was the means by which it took risk.
But we must understand the poor choices, the adverse policies,
and the misunderstood risk that were involved there, and this
Committee heard a lot about that in the testimony last week,
particularly from Mr. Polakoff from the Office of Thrift
Supervision.
These were the source of their problems, these
misunderstood risks and poor decisions. They were contrary to
the best practices in the industry and to the experiences of
swap market participants for the past 20 years. The fundamental
decision that AIG made was to take on exposure to the housing
market. They did that, yes, via credit default swaps. They also
did that, as the Committee heard last week, in other means, as
well, through their securities lending business, in which they
actually continued to lend and take exposure to those markets
into 2006 and 2007, when the worst of these securities were
generated through the lending process.
They also had a very myopic view of loss. They were only
looking at the payout potential, the possibility they would
actually have to pay out on these transactions. There was no
consideration of the implications of the mark-to-market losses
that they could face and to the effects on their capital and
their liquidity. They seem to have completely ignored the
possible effects of that.
They relied on their AAA rating and refused to provide
collateral from the start of their trading relationships. It
takes away the discipline that collateral provides in that
trading relationship. Collateral is extensively used in the OTC
derivatives business to help manage risk and also introduce
discipline to the trading relationship. They agreed, on the
other hand, to provide collateral on the downgrade of their
credit rating. That led to a falling off of a cliff,
effectively, leading to substantial liquidity problems which
eventually led to the decision to intervene.
So yes, these decisions and policies are important to
understand and we need to take steps to make sure that this
does not happen again, but those relate to the decisions they
made and not to the products themselves. The products, in fact,
have performed as the parties intended. In fact, just
yesterday, the Senior Supervisors Group, which is a group of
senior supervisors from the G-7 countries, talked about how the
CDS product had performed multiple times over the course of
last fall and into this year in helping to settle transactions,
credit default swaps that parties had engaged in, and they
acknowledged that this process has been extremely effective.
And then finally, this is a very important week in the
credit default swap market. I believe Mr. Ryan referred earlier
to the fact that one of the clearinghouses that has been talked
about for many, many months has now actually begun to clear
transactions, and that is a very significant development. And
later this week, ISDA itself will introduce some changes to the
standard contract that will facilitate the settlement of these
trades in the future and will also facilitate moving more
transactions onto a clearinghouse. So that is a very important
development.
There is much to be done by ISDA, by the industry, in close
consultation with this Committee and other committees in
Congress as well as the regulators here in the United States
and globally and we are committed to be engaged in that
process. We look forward to working with you as you analyze the
causes of this financial crisis, and based on that analysis,
consider changes to our regulatory structure with a goal to
obtaining greater transparency, greater disclosure, and greater
coordination among regulators.
Thank you again for your time, and I look forward to your
questions.
Chairman Dodd. Thank you very, very much, Mr. Pickel.
Damon Silvers. Damon, it is good to have you with us. Thank
you.
STATEMENT OF DAMON A. SILVERS, ASSOCIATE GENERAL COUNSEL, AFL-
CIO
Mr. Silvers. Good morning, Chairman Dodd and Ranking Member
Shelby. Thank you for inviting me here today. Before I begin, I
would like to note that in addition to my role as Associate
General Counsel of the AFL-CIO, I am the Deputy Chair of the
Congressional Oversight Panel created by the Emergency Economic
Stabilization Act of 2008 to oversee the TARP. While I will
describe in my testimony aspects of the Congressional Oversight
Panel's report on regulatory reform, my testimony reflects my
views alone and the views of the AFL-CIO unless otherwise noted
and is not on behalf of the panel, its staff, or its chair.
The vast majority of American investors participate in the
markets as a means to secure a comfortable retirement and to
send their children to college, as you noted, Mr. Chairman, in
your opening remarks. While the spectacular frauds like the
Madoff Ponzi scheme have generated a great deal of publicity,
the bigger question is what changes must be made to make our
financial system a more reasonable place to invest the hard-
earned savings of America's working families.
Today, I will address this larger question at three levels:
Regulatory architecture, regulating the shadow markets and the
challenge of jurisdiction, and certain specific steps Congress
and regulators should take to address holes in the investor
protection scheme.
First, with respect to regulatory architecture, the
Congressional Oversight Panel in its special report on
regulatory reform observed that addressing issues of systemic
risk cannot be a substitute for a robust, comprehensive system
of routine financial regulation. Investor protection within
this system should be the focus of a single agency within the
broader regulatory framework. That agency needs to have the
stature and independence to protect the principles of full
disclosure by market participants and compliance with fiduciary
duties among market intermediaries. This has been noted by
several of the panelists prior to me.
This mission is in natural tension with bank regulators'
mission of safeguarding the safety and soundness of the banks
they regulate, and that natural tension would apply to a
Systemic Risk Regulator that was looking more broadly at safety
and soundness issues. Because of these dynamics, effective
investor protection requires that any solution to the problem
of systemic risk prevention should involve the agency charged
with investor protection and not supercede it.
I have a more detailed document on issues associated with
creating a Systemic Risk Regulator that I will provide the
Committee following the hearing. I should just note that in
relation to this, it is my belief that more of a group approach
to systemic risk regulation rather than designating the Fed as
the sole regulator would be preferable.
Among the reasons for this are the issues of information
sharing and coordination that other panelists raised, but most
importantly, the fact that the Federal Reserve in its
regulatory role fundamentally works through the regional Fed
banks, which are fundamentally self-regulatory in nature.
Several of the prior witnesses have mentioned some of the
problems with self-regulation on critical issues. Furthermore,
a Systemic Risk Regulator, as we have learned through the TARP
experience, is likely to have to expend public dollars in
extreme circumstances. It is completely inappropriate for that
function to be vested in a body that is at all self-regulatory.
While the Fed could be changed, its governance could be changed
to make it fully a public agency, that would have implications,
I believe, for the Fed's independence in its monetary policy
role.
Now, we have already in the Securities and Exchange
Commission a regulator focused on investor protection. Although
the Commission has suffered in recent years from diminished
jurisdiction and leadership failure, the Commission remains an
extraordinary government agency whose human capital and market
expertise needs to be built upon as part of a comprehensive
strategy for effective re-regulation of the capital markets.
This point flows right into the issue of jurisdiction and the
shadow markets.
The financial crisis we are currently experiencing is
directly connected to the degeneration of the New Deal system
of comprehensive financial regulation into a Swiss cheese
regulatory system where the holes, the shadow markets, grew to
dominate the regulated markets. The Congressional Oversight
Panel specifically observed that we need to regulate financial
products and institutions, in the words of President Obama,
``for what they do and not what they are.''
The Congressional Oversight Panel's report further stated
that shadow institutions should be regulated by the same
regulators that currently have jurisdiction over their
regulated counterparts. So, for example, the SEC should have
jurisdiction over derivatives that are written using public
debt or equity securities as their underlying asset. At a
minimum, the panel stated, hedge funds should also be regulated
by the SEC in their role as money managers.
There is a larger point here, though. Financial re-
regulation will be utterly ineffective if it turns into a
series of rifle shots at the particular mechanisms used to
evade regulatory structures in earlier boom and bust cycles.
What is needed is a return to the jurisdictional philosophy
that was embodied in the founding statutes of Federal
securities regulation: Very broad, flexible jurisdiction that
allowed the Commission to follow changing financial market
practices.
If you follow this principle, the SEC should have
jurisdiction over anyone over a certain size who manages public
securities and over any contract written that references
publicly traded securities. Applying this principle would
require at least shifting the CFTC's jurisdiction over
financial futures to the SEC, if not merging the two agencies
under the SEC's leadership, as I gather some of my fellow
panelists believe is necessary.
Moving on to substantive reforms, beyond regulating the
shadow markets, the Congress and the Securities and Exchange
Commission need to act to shape a corporate governance and
investor protection regime that is favorable to long-term
investors and to the channeling of capital to productive
purposes.
First, strong boards of publicly traded companies that the
public invests in--having strong boards requires meaningful
accountability to long-term investors. The AFL-CIO urges
Congress to work with the SEC to ensure that long-term
investors can nominate and elect psychologically independent
directors to company boards through access to the corporate
proxy.
Second, effective investor protection requires
comprehensive executive pay reform involving both disclosure
governance and tax policy around two concepts. Equity-linked
pay should be held significantly beyond retirement. And two,
pay packages as a whole should reflect a rough equality of
exposure to downside risk as well as to upside gain. Part of
this agenda must be a mechanism for long-term shareholders to
advise companies on their executive pay packages in the form of
an advisory vote.
Finally, Congress needs to address the glaring hole in the
fabric of investor protection created by the Central Bank of
Denver and Stoneridge cases. These cases effectively granted
immunity from civil liability to investors for parties such as
investment banks and law firms that are actual co-conspirators
in securities frauds.
Now, to address very briefly the international context, the
Bush administration fundamentally saw the internationalization
of financial markets as a pretext for weakening U.S. investor
protections. That needs to be replaced by a commitment on the
part of the Obama administration, the Congress, and the
regulators to building a strong global regulatory floor in
coordination with the world's other major economies. However,
Congress should not allow the need for global coordination to
be an impediment or a prerequisite to vigorous action to re-
regulate U.S. financial markets and institutions.
Obviously, this testimony simply sketches the outline of an
approach and notes some key substantive steps for Congress and
the administration to take. While I do not speak for the
Oversight Panel, I think I am safe in saying that the Panel is
honored to have been asked to assist Congress in this effort
and is prepared to assist this Committee in any manner the
Committee finds useful. I can certainly make that offer on
behalf of the AFL-CIO. Thank you.
Chairman Dodd. Thank you very much.
Mr. Doe, we thank you for joining us, the Municipal Market
Advisors.
STATEMENT OF THOMAS DOE, CHIEF EXECUTIVE OFFICER, MUNICIPAL
MARKET ADVISORS
Mr. Doe. Chairman Dodd, Senator Shelby, and Committee
Members, is a distinct pleasure that I come before you today to
share my perspective on the municipal bond market. My firm,
Municipal Market Advisors, has served for the past 15 years as
the leading independent research and data provider to the
industry. In addition, from 2003 to 2005, I served as a public
member of the Municipal Securities Rulemaking Board, the self-
regulatory organization of the industry established by Congress
in 1975.
There are nearly 65,000 borrowers in the municipal market
that are predominantly States and local governments. Recent
figures identify an estimated $2.7 trillion in outstanding
municipal debt. This is debt that aids our communities in
meeting budgets and financing society's essential needs,
whether it is building a hospital, constructing a school,
ensuring clean drinking water, or sustaining the safety of
America's infrastructure.
A distinctive characteristic of the municipal market is
that many of those who borrow funds--rural counties and small
towns--are only infrequently engaged in the capital markets. As
a result, there are many issuers of debt who are inexperienced
when entering a transaction and are unable to monitor deals
that may involve movement of interest rates of the value of
derivative products.
According to The Bond Buyer, the industry's trade
newspaper, annual municipal bond issuance was $29 billion in
1975; whereas, in 2007, issuance peaked at $430 billion. In the
past 10 years, derivatives have proliferated as a standard
liability management tool for many local governments. However,
because derivatives are not regulated, it is exceptionally
difficult, if not impossible, to identify the degree of
systemic as well as specific risk to small towns and counties
who have engaged in complex swaps and derivative transactions.
Municipal issuers themselves sought to reduce borrowing
costs in recent years by selling bonds with a floating rate of
interest, such as auction rate securities. Because States and
local governments do not themselves have revenues that vary
greatly with interest rates, these issuers employed interest
rate swaps to hedge their risk. Issuers use the instruments to
transform their floating risk for a fixed-rate obligation.
A key factor in the growth of the leverage and derivative
structures was the prolific use of bond insurance. Municipal
issuers are rated along a conservative rating scale, resulting
in much lower ratings for school districts and States than for
private sector financial and insurance companies. So although
most States and local governments represent very little default
risk to the investor, the penal ratings scale encouraged the
use of insurance for both cash and derivatives in order to
distribute products to investors and facilitate issuer
borrowing.
So instead of requiring more accurate ratings, the
municipal industry chose to use bond insurance to enhance the
issuer's lower credit rating to that of the higher insurance
company's rating.
The last 18 months have exposed the risks of this choice
when insurance company downgrades, and auction rate security
failures, forced numerous leveraged investors to unwind massive
amounts of debt into an illiquid secondary market. The
consequence was that issuers of new debt were forced to pay
extremely high interest rates and investors were confused by
volatile evaluations of their investments.
The 34-year era of the municipal industry's self-regulation
must come to an end. Today, the market would be in a much
better place if:
First, the regulator were independent of the financial
institutions that create the products and facilitate issuers'
borrowing. Municipal departments represent a relatively small
contribution to a firm's revenue, and this inhibits MSRB board
members from seeking regulatory innovation.
Second, if the regulator were integrated into the national
regime of regulation. Since the crisis began, we have
discovered a limited market knowledge here in D.C., in the
Federal Reserve, Treasury, Congress, and the SEC. I might add
that when the crisis began to emerge in August 2007, we were
immediately contacted by the New York Federal Reserve and the
Federal Reserve itself, and are quite impressed in the last 18
months with their vigilance and interest in this sector. So
integration, we believe, would speed market recovery by the
shared information.
Third, the regulator's reach and authority needs to be
extended to all financial tools and participants of the
municipal transaction. This meant ratings agencies, insurers,
evaluators, and investment and legal advisors for both the cash
and swaps transactions. This need has become more apparent as
we uncover the damaged issuers, and States such as Alabama,
Tennessee, and Pennsylvania are suffering relative to interest
rate swaps.
Fourth, if the regulator were charged with more
aggressively monitoring market data with consumers' interests
in mind. When I think of consumers, I think of both investors
and the issuers. In 2008, there were specific instances of
meaningful transactions and price irregularities that should
have prompted regulatory investigation to protect consumers.
The good news is that this new era of regulatory oversight
can be funded by the MSRB's annual revenue in 2008 of $20-plus
million, collected from the bond transactions themselves, and
can be staffed by the current MSRB policy and administrative
infrastructure.
I should be clear. The innovations of derivatives and swaps
have a useful application and have been beneficial for those
for which they are appropriate. However, it is also important
that these instruments become transparent and regulated with
the same care as the corresponding municipal cash market.
It is critical to get this right. There is simply too much
at stake.
Thank you for having me here today, and I look forward to
participating in the questions of the session.
Chairman Dodd. Thank you very much, Mr. Doe.
Lynn Turner is the former Chief Accountant at the
Securities and Exchange Commission. Mr. Turner, we thank you.
STATEMENT OF LYNN E. TURNER, FORMER CHIEF ACCOUNTANT,
SECURITIES AND EXCHANGE COMMISSION
Mr. Turner. Thank you, Chairman Dodd and Ranking Member
Shelby. It is always good to be here in this particular case,
so I must commend both of you for holding this hearing on an
issue that has not only impacted millions of investors but just
literally everyone that has been devastated from this economy.
I would ask that my written testimony be included in the
record.
Chairman Dodd. Yes, let me make that clear. There is a lot
of additional documentation, and some of you may want to add as
well, and so all of that will be included as part of the
record. We thank you for that.
Mr. Turner. It is only 17 pages, so it is a little bit
quicker read than Professor Coffee's.
Senator Dodd, I think you are right. There are really three
root causes of this problem: people made bad loans, gatekeepers
sold out, and a lack of regulation or regulators missing in
action, quite frankly. And it is not the first time. As an
auditor in the Southwest, in Denver, I lived through this with
the S&Ls, had to do restructurings, workouts at that point in
time. And those issues were all existent then, and we are back
to a repeat.
So, as Mr. Doe and Mr. Silvers indicated, I think it is
especially important that this go-round the Committee get it
right. I know that there is some push to try to get something
done by an August recess. I would say it is more important to
hit this target. We have seen the markets are serviced, a
trustee of two large institutional investors. We have seen
legislation come out that has not instilled that confidence to
date. And we need to get it right this time so we instill that
confidence and do not see a market of 5,000, quite frankly. So
I would ask you to take your time, whatever is necessary,
sooner better than late, but I am not sure this is one that can
be both fast and right.
Senator Shelby, you asked someone to comment on the mark-
to-market accounting. Being the one accountant, the one green
eyeshade on this, let me say I could not agree more with you.
The mark-to-market accounting that we are debating now is the
same issue we debated two decades ago during the S&L crisis,
and as the 1991 GAO report stated, the failure of the banks and
the S&Ls during that travesty, to turn around, take marks down
in a timely manner resulted in lax regulator action, people not
getting on top of managing the assets and problems quick
enough, and contributed to a significant increase in the cost
to the taxpayers of that bailout. And so I would again urge you
to push for transparency here, not step on those accounting
standards, and let us get the real numbers.
When you look at banks like Citigroup, who are trading at a
stock price for less than what you can buy a Happy Meal these
days at McDonald's, we know that the market clearly is not
viewing those financials as credible, and we need to get that
credibility back into the system.
Certainly, as my fellow panelists mentioned, there are also
gaps in regulation. Without a doubt, the credit derivatives
market--we all know about that. You certainly have all heard
about that as recently as this last week. But it was not so
much the failure of a regulatory system, although things need
to be fixed, as it was a failure of regulators to act. The
Office of the Comptroller of the Currency, the SEC, both had
risk management offices. The Federal Reserve had examiners day
in, day out at Citigroup, and this was not the first time
Citigroup became, for all practical purposes, insolvent and in
need of a bailout. When I was at the Commission two decades
ago, the exact same thing happened. And you ask, How can the
Fed turn around and allow that to happen?
I remember being in a meeting with banking regulators and
the Chairman of the Fed some time ago, and I was asked, along
with the Chairman of the SEC at the time, What is wrong if the
banks are allowed to fudge the numbers a little bit? Now I
think we know. If you turn around and ask me is that who you
are going to make our systemic regulator, I would turn around
and say, ``I would hope not.'' Rather, I think the notion that
Professor Goldsmith, the former SEC Commissioner, has advocated
as the council or commission--I think Damon talked on it as
well--is a much better approach. You have got to give us this
in investors, someone regulating that we can trust in. The
notion of prudential supervision needs to be a notion that
dies. What we want is actual regulators doing their job. That
is what we are turning around and paying them for.
Now, while certainly the SEC has fallen off the track here
recently, I must say that over the years it has been very
successful in its mission to protect investors and gain their
confidence. I think investors would be ill served and very
concerned if some other regulator with a mission other than
investor and consumer protection, first and foremost, was given
that leading role to protect them.
As the Committee crafts a solution, I simply believe a
focus on a systemic regulator in and of itself and doing
regulation around just a systemic regulator does not get the
job done. I think a more comprehensive single bill is the right
way to go after that. And in doing so, I think you should focus
on a few key principles that you need to ensure are
established: independence in the system, transparency,
accountability, enforcement of law, and making sure those
responsible for doing the job have adequate resources.
Following these key principles, as more specifically
spelled out in the written statement, I think there needs to be
a closure of the regulatory gaps such as with credit
derivatives; SEC oversight over the investment banks; certainly
the mortgage brokers who brought this problem upon us; greater
accountability established through governance and investor
rights, including private right of actions, as Damon has
mentioned, for credit rating agencies; assisting others in the
commission of fraud. Regulators simply cannot do it all and
will never have enough resources, so we have to give
institutional investors a chance to get justice and recover
money when there has been fraud involved.
We need to enhance transparency and disclosure, not only by
the issuers but also by the regulators. The testimony last week
where the Fed would not give us the names and the details
behind the credit derivatives and who are really getting bailed
out at AIG was most concerning and disappointing. There needs
to be improvements in self-regulation, and there obviously
needs to be better enforcement of the laws and regulation.
But in the end, no agency here, neither the CFTC, SEC, the
banking regulators, can do it without adequate resources. For
example, the Office of Compliance and Inspections at the SEC,
you are asking them to inspect 16,000 mutual funds, 11,000-plus
investment advisers with 440 people. It simply cannot be done.
At a minimum, the SEC needs $100 million to get the type of
technology that just brings them up to what we use on the
street in the market. If they do not even have those tools,
there is no way they can supervise and stay on top of it--$100
million in technology, and then they need about another $85,
$90 million, just to bring staffing up to the levels they were
4 to 5 years ago. And they need it now. They do not need it on
October 1 of 2010. That needs to go into the budget now, not a
year and a half from now.
So I would certainly urge--and I know this is not the
Appropriations Committee, but I would certainly urge the Senate
to find a way to get them the resources. Without that, you are
asking them to go into a gunfight with an empty gun, and we all
know what happens then.
So, with that, I will close and be happy to answer any
questions. Thank you.
Chairman Dodd. Well, thank you very much, Mr. Turner, and
what I will do in terms of time--and I will not rigidly hold
people to it, obviously, but try and do 5 or 6 minutes, and it
will give us a chance at least to get a round. If you go over
that, don't worry about it so much. We will just try to move
along, because, again, we want to get our witnesses involved.
The temptation here is to focus on sort of one aspect of
this. There are a lot of issues, obviously, across the spectrum
from obviously credit default swaps, transparency, corporate
governance issues, conflicts of interest, credit rating
agencies--just a lot of matters to pick out, so I am going to
try and ask a broader question and then ask each one of you to
comment on a broader question. And I think I know the answer to
this one, having listened to all of you and I looked at your
testimony. But I would like to ask, beginning with you, Dr.
Coffee--and I think you identified this. But if you could make
one recommendation as we are looking at this--and obviously we
have got our hands full here in the coming weeks.
By the way, I take the point that was raised, either by Mr.
Doe or Mr. Turner, of getting this right and, obviously, there
is a sense of urgency, but I think the Committee would agree
that we want to get it done, we want to work at this very hard,
but we do want to get it right. And so striking that balance
between moving with some haste but not to such a degree that
that becomes the goal rather than producing a product here that
has been well thought out.
But if you could make one recommendation that you feel is
the most important legislative or regulatory action that the
Committee could take to improve investment protection or the
quality of securities regulation in the light of the financial
crisis we are experiencing, what would it be? What is your one
recommendation for us? Just go down the line.
Mr. Coffee. I would tell you that you should endorse----
Chairman Dodd. I need that microphone on.
Mr. Coffee. I would tell you that is the twin peaks model
for securities regulation, that is, having separate peaks--one,
the systematic risk regulator, the prudential supervisor. I do
not think the SEC is the best agency for that. By training or
culture, it is a lawyer-dominated agency focused on enforcement
and disclosure, which it does well. It is not able to deal with
financial institutions, at least in terms of its first-line
responsibilities. Someone else can do that better, presumably
the Federal Reserve.
But as was said by other people today, I think there is
always the danger that the cultures of securities and banking
regulation are so different, and if you put them all in one
agency, the centralized regulator, much like the British model,
the FSA, you are going to have tensions and tradeoffs between
investor protection and the protection of bank solvency.
Thus, I think you have to have a separate investor
protection agency. You could merge the SEC and the CFTC, or you
could transfer financial futures to the SEC. That is going to
be costly in terms of the political process. I do not know
whether it is feasible. But the first step, I would say, is to
try to have a systematic risk regulator and to not give it the
authority to override the disclosure regulators on questions of
accounting or on investor protection. So that is the structural
issue from 40,000 feet, and I would start there.
Chairman Dodd. Gene Ludwig I have talked to, and he has
talked about a similar twin peaks structure that you have just
described. Are you familiar with his thoughts on this?
Mr. Coffee. Well, this term ``twin peaks'' has been used by
a number of people. Around the world we see two models: the
centralized regulator, which is what the U.K. has, and it has
not worked that well there, either; and we have the twin peaks
models which Australia, the Netherlands, and other countries
have.
We in the U.S. have a unique fragmented system that is
virtually Balkanized, with a different regulator for every
class of institution. We have got to move one of those two, and
I am telling you that the twin peaks model, I think, is vastly
superior.
Chairman Dodd. OK. I agree with you.
Tim Ryan.
Mr. Ryan. This is unique. We basically agree with the
professor. I mean, we have not worked out most of the details
on both aspects of the twin peak, but we know that the No. 1
recommendation we have today--and actually, the country needs
it. We need confidence in the system. There is no confidence
today. We need a systemic risk regulator for the major
institutions, and we would urge you to do that in a timely
fashion. And we would define ``timely'' by the end of this
year, and to do it right.
Thank you.
Chairman Dodd. Mr. Stevens.
Mr. Stevens. We, as you know, endorse with some cautions
the idea of a systemic risk regulator, but if you push me to
say what one thing, Mr. Chairman, I would say that we need a
capital markets regulator that is really at the top of its
game. This problem would not have grown unless the securities
markets made available, through packaging and reselling and all
the rest of it, a vast opportunity to take these mortgages and
distribute them to financial institutions globally.
I in 30 years have been a close observer of the SEC, and I
think it is a remarkable agency. But it has seen challenges in
keeping up with the growth and the complexity and the linkages,
the internationalization of our capital markets. What we need
to do is give it the right tools, give it the right range of
authority, and make sure there is a very strong management
focus there that keeps it on its mission. And I agree with you,
investor protection is mission one, but it also has a very
important regulatory role, and so it needs to understand and be
able to keep pace with these market changes in the way that
has, I think, proven to be very, very difficult.
So that would be my recommendation.
Chairman Dodd. OK. Professor Bullard.
Mr. Bullard. Mine would be to--probably to expand on what
Professor Coffee was saying, I think that from an investor
protection point of view, what is important to understand is
that investor protection actually assumes that we want
investors to take risk. And, therefore, investor protection is
about making sure that the risks that they take are consistent
with their expectations.
That is fundamentally inconsistent with a prudential
oversight role. Prudential oversight is what you want when
somebody buys life insurance and expects the money to be there
if their spouse dies. They invest in a banking account, and
they expect those assets to be there. They buy a money market
fund, and they expect that to be a safe investment.
That is antithetical to investor protection risk because
there sometimes the disclosure of the truth undermines the
confidence that you need that those people rely on to keep
their investments in banks--to keep their assets in banks and
money market funds. So I would say those have to be separated,
and investor protection needs to be, again, as I mentioned
earlier, kept distinct from customer protection, which again is
not something that needs to be regulated with an eye to
promoting risk taking--that is, risk taking based on high
expected value investments.
And then, finally, I would say that I am a little concerned
about mixing the systemic issue and the prudential issue. The
way I have always thought about prudential oversight is that
you are making sure that the promises made with respect to
generally liabilities on one side are matched by the kind of
assets that are created to support those liabilities.
Systemic oversight is where you assume that prudential
regulation, being necessarily imperfect, will sometimes lead to
a breakdown, and the question is: What is the role of the
Government going to be to prevent that breakdown under a
prudential system, which will happen sometimes, and what will
it do when it steps in? I think that that is a fundamentally
separate function from prudential regulation, and that is why--
and I am not sure what a systemic regulator is as apart from a
prudential regulator. But that is something, I think, that it
would help to have more clarity on.
Chairman Dodd. Yes, Mr. Pickel.
Mr. Pickel. I think the requirement for greater
coordination among the regulators is very important. I think
that one of the things we are looking at here in the financial
crisis is the ability to connect the dots across different
products and across different markets, both nationally and
globally. I think that is the root of some of the suggestions
for the systemic risk regulator. But I think it could also be
achieved, as I think Mr. Silvers and Mr. Turner suggested,
through greater coordination or some collection of supervisor
who would look at these issues and connect those dots.
Chairman Dodd. Mr. Silvers.
Mr. Silvers. I find myself in the unusual position of
having really nothing to disagree with in what I have heard so
far at the table. I would say, though, that the single item
that I would put to you, I would put differently than my co-
panelists have done so far.
I think that conceptually the thing you want to be most
focused on is ensuring that we no longer have a Swiss cheese
system, that we no longer have a system where you can do
something like insure a bond, either in a completely regulated
fashion, in which there are capital requirements and disclosure
requirements and pre-clearance, or in a completely unregulated
fashion through essentially a derivative and where you have
none of these things; that the content of what a financial
market actor does should determine the extent and type of that
regulation.
Closing regulatory loopholes, ending the notion that we
have shadow markets, I think is the most important conceptual
item for Congress to take up, because, otherwise, if it
continues to be possible to essentially undertake the same
types of activity with the same types of risk but to do so in
an unregulated fashion, we will replay these events with a fair
degree of certainty. And I believe that much of what the
discussion about structure here has been is all about how we do
that ending of shadow markets and regulatory gaps.
I think in certain respects, some of the how is less
important than actually getting it done. I would say, though,
that I really strongly endorse what Mercer said about the
different functions of regulation, that there is investor
protection, disclosure and fiduciary duty oriented; there is
consumer protection, although I think, Mercer, you had a
different phrase for it, but protection around the public
buying financial services which does not want to take risk; and
then there is safety and soundness regulation. Those things are
different, and it is dangerous to blend them.
Chairman Dodd. Yes. Mr. Doe.
Mr. Doe. As I have listened to my fellow panelists, I am
reminded of a book called ``Why Most Things Fail'' by a U.K.
economist named Paul Oremerod, where he draws comparisons
between species extinction models and those of corporate and
market failures. And in it, he cites two conditions where we
have failure of a species. And one is when it gets soft and is
not challenged. The second is when there is not incremental
learning that is constantly being done.
And I think what I have gathered in the last 18 months--
and, again, in our small niche in the municipal bond industry,
which is smaller than others that have been addressed here
today. But it is that the idea of a regulator that has an
inspired inquisitiveness and a sense of purpose so that they
are eager to pursue an understanding of the markets that they
regulate. If there has been one--and this is where I hope that
if we have a consolidated or a sharing of information across
the different asset classes or the different products, whether
they be cash, whether they be swaps, whether they are equity,
whether they are fixed income, that this provides an
opportunity of being able to identify the first hints of
failure that might occur in a system, and that way we might be
faster to act.
I think what is very interesting about, again, the industry
that I have been involved with in the municipal bond sector is
that when innovation of products finally makes its way into the
public sector, it is almost the last place, again, because the
revenue relatively is small compared to the other asset classes
in the taxable markets. But I think it becomes magnified
because you are starting to deal now with the public trust in
the most intimate form.
And I think so that when we start looking at regulation, it
is, again, how do you inspire that trust, but how do you
inspire that inquisitiveness of the regulator, and maybe it is
the pride that is associated with doing the job that they feel
that they are really able to accomplish something and make a
difference. And I think that is what we are all trying to do
here.
Thank you.
Chairman Dodd. Yes. Last, Mr. Turner.
Mr. Turner. I actually think if I could just pick up a
magic wand and do one single thing here, it would make sure
that inside the agencies, the regulators, we had competent
people who were in the right mind-set to go do their jobs. Bad
loans, the Fed had the law that you all passed in 1994 giving
them clear-cut authority to go eliminate those. They did not do
it. Enforcement agencies have not done enforcement.
The bottom line here is much of this could have been
prevented without a single additional piece of legislation
being done if people had just done their jobs back here. And I
would urge you, go back, let us make sure we get the right
people in, and then let us make sure, quite frankly, there is
more active, proactive oversight by the appropriate committees
of those responsible.
Aside from that, I would turn around and say the No. 1
thing in the system has to be independence. These agencies have
to clearly understand they are independent and free to go do
what they need to do to protect investors and consumers. There
should have been independence in the credit rating agencies.
They clearly sold out. The e-mails and all show it. That was
not there.
There needs to be independence in the compliance officers
in these businesses, in these banks. Clearly, that was not
there.
So aside from making sure you got people doing the job, we
can have one peak, we can have two peaks. We can have 52--the
14,000 peaks that we have got in Colorado. But if you have got
bad people sitting on the top of each of those peaks, it is not
going to matter what you legislate here.
Chairman Dodd. With that encouraging note, let me turn to
Senator Shelby.
[Laughter.]
Senator Shelby. Mr. Turner, we are glad you are here. Thank
you. Thank you very much.
Professor Coffee, I would like to direct this to you, if I
could. Thank you, and welcome to the Committee again. You spend
a lot of time here, and we welcome you, and you have added a
lot to us.
You recommend giving the Federal Reserve Board authority to
regulate capital adequacy, safety and soundness, and risk
management of all financial institutions that are ``too big to
fail.'' Is this suggestion based on a careful examination of
the Federal Reserve's track record as a prudential supervisor
up to this point, which I think is lacking? Did you take into
consideration the fact that the Fed already has responsibility
for monetary policy, bank supervision, and lender-of-last-
resort functions? And are you concerned about the implications
of the fact that, as you noted, the Fed is not politically
accountable in the way other agencies are?
I know that is a lot of question, but you are a
distinguished professor and you can handle it.
[Laughter.]
Mr. Coffee. Those are all good points.
Senator Shelby. Can you bring your microphone up a little
bit? It is not on yet.
Mr. Coffee. The problem is that the Fed is not perfect. It
is far from perfect. I think it is better positioned than
agencies like the SEC. The SEC is focused on transparency and
enforcement, not on prudential supervision, and the second-tier
functions of an agency are the functions that are most likely
to be captured by the industry.
Also, frankly, there no longer are any investment banks.
They have all moved someplace else. There is nothing left for
the SEC to exercise prudential supervision over. Therefore, I
have got to think we have to start, warts and all, with the
Federal Reserve as the only body that has this capacity and has
the orientation and has the competence. It may not always have
performed well.
Your point about political accountability is very
important, and that is why I keep insisting that investor
protection should not be subordinated and should be given to a
very independent agency, the SEC or the SEC/CFTC, because I do
not think you can count on the Fed with its orientation to ever
be a champion of the investor's rights. The culture is one of
secrecy, and you saw this all in the AIG. I think AIG is
representative of the problems you will have if you depend upon
the Fed for transparency. But I do not think the SEC is going
to do much better than it did in the Consolidated Supervised
Entity Program.
Senator Shelby. Professor Coffee, are you concerned that
when you identify institutions as too big to fail, that will
dull the market discipline of those firms which the market will
view as having a Federal guaranty? Is that a concern always in
the marketplace?
Mr. Coffee. I am not testifying that every organization
should be bailed out. I think the ones that most merit this are
the ones that are so entangled that you get the true problem of
systemic risk. Systemic risk is the danger of interconnected
failures, the chain of falling dominoes. I am not telling you
whether or not AIG should get more money. I am telling you only
that where we have companies that are too entangled to fail,
that is where we most need prudential supervision and a
Systemic Risk Regulator.
Senator Shelby. Mark-to-market accounting, Mr. Turner wrote
this up and I think he is right. Do you believe the current
attacks on mark-to-market accounting, Mr. Turner, are motivated
by a similar understandable desire to avoid taking painful
write-downs?
Mr. Turner. I think, without a doubt, Senator Shelby, they
are a problem here. You know, if you make a loan at $100 and
you are only going to get $70 back, that is OK once or twice,
but we did it millions and millions of times. The bottom line
is they just aren't worth what they were, and to report to the
public, to investors, regulators that you have got a balance
sheet that is substantially different than what it is really
worth is just flat out misleading, if not straightforward
fraudulent.
Senator Shelby. Mr. Ryan, your testimony recommends a
Financial Market Stability Regulator that, among other things,
would have a direct role in supervising, quote, your words,
``systemically important financial organizations.'' What are
the criteria, Mr. Ryan, that you would recommend for
identifying systemically important entities, and do you believe
that there would be any competitive implications for firms that
are not so designated?
Mr. Ryan. Thank you for your question. We have given a lot
of thought to a number of issues, and on some of these issues
we do not have final decisions. I am talking now within the
industry. For instance, we spent a lot of time talking about
should we recommend the Fed immediately as the systemic
regulator, and we have not come to that conclusion yet. If we
had to do it right away, they are probably the best qualified
to do it, but we think that the industry and the Congress, the
American people, deserve a really comprehensive view.
The same is true of who is systemically important. It is
pretty easy to identify the early entrants because they meet
the test that Professor Coffee has enunciated. They are too
interconnected. They are very large. They are providing
consolidated services to the citizens of this country and we
need a better understanding of their interconnected aggregated
risks.
So the first group will be easy. The second group will be
more difficult because they may not be so interconnected. They
may not even be that large. But they may be engaged in
practices which could have a very dramatic impact on our
health.
So our hope would be that we anoint a systemic regulator,
maybe it is a new entity, maybe it is within Treasury, maybe it
is the Fed, that we orient them in legislation toward
preselection of the people who are very obvious, and that we
give them the flexibility to include and actually to have
people move out of systemically important status going forward.
So once you are in it doesn't necessarily mean that you will
stay in it.
I think it is pretty clear, though, we all know the basic
early entrants and they are our larger financial institutions.
We, by the way, would not limit this by charter at all, so
there will be banks, there will be insurance companies, there
will be hedge funds, there could be private equity players. It
is people who could have a dramatic effect on our lives.
Senator Shelby. Professor Coffee, why should we continue to
prop up banks that are basically insolvent, some of our large
banks that are walking dead, so to speak, give them a
transfusion, and there is no end in sight? Why should we do
that rather than take over some of their--guarantee some of
their assets and whatever we have to do and wind them down?
Mr. Coffee. Again, it is a perfectly fair question and I am
not telling you that every bank should be bailed out, not even
every large bank. But if we are going to get the financial
system working again, we have to move credit through banks.
Senator Shelby. Sure.
Mr. Coffee. You can nationalize them. There still has to be
a management.
Senator Shelby. I have never advocated that.
Mr. Coffee. The government can't run a bank. It might be
that if you want to name institutions that may not deserve
further bail-outs, AIG is an example. It is not a bank and
basically the government is spending $160 billion there to pay
off the counterparties, most of whom are foreign banks.
Senator Shelby. Obviously poorly regulated.
Mr. Coffee. I am saying that the key bank institutions are
the only way we can get a corporate capital system moving
again. The money has to flow through the system, and if they
are part of the basic transmission belt, then there is a strong
argument for ensuring their survival.
Senator Shelby. Mr. Ryan, you and your organization, the
Securities Industry and Financial Markets Association, have
advocated a merger of the SEC and the CFTC. You are not by
yourself there. If a merger, should it go forward, should it
occur simultaneously with whatever broader regulatory
restructuring we undertake in this Committee? Should it be part
of the overall comprehensive structure, restructure?
Mr. Ryan. I would sort your--because I think the biggest
issue Congress has right now, you obviously have many, many
issues on your plate and sorting them by priority is an
essential component of your work right now, I am sure. So our
recommendation is that you sort them with the systemic
regulator being first and immediately come behind that with
cleaning up the many regulatory agencies that have overlapping
authority. If I was the new regulator, I may actually ask you
to do those simultaneously so I would know my job.
May I also----
Senator Shelby. It would send certainty to the market,
would it not?
Mr. Ryan. Yes, sir. I think that is possible. May I also
just provide one comment based on your last question to the
professor to my right?
Senator Shelby. Go ahead.
Mr. Ryan. Because I am probably--because I spent an awful
lot of time in this Committee when I was the OTS regulator----
Senator Shelby. You did.
Mr. Ryan. ----dealing with the RTC, and what I learned
through those 3 years brings me to a very firm view on
opposition totally to nationalization of financial
institutions. We have a process in this country through the
FDIC where we, in a sense, nationalize. We call them bridge
banks. We know that if you put very large institutions into
bridge banks and they stay there for a very long period of
time, the cost escalates enormously. You can look back to the
RTC experiences. If we ask the government to take control of a
large financial institution and run them, the cost of
resolution is going to dramatically increase. The way we are
doing it right now is much better. Thank you.
Senator Shelby. One last question, Mr. Chairman, if I can,
to Mr. Silvers. Mr. Silvers, you advocate a greater role for
long-term investors in the election, I will use your term, of
psychologically independent directors on corporate boards. What
measures would you take to ensure, sir, that these directors
whose responsibilities would flow to all shareholders of the
corporation are independent not only of management, which is
important, but also the shareholder group responsible for their
election?
Mr. Silvers. Senator Shelby, I think that there are several
specific ways of doing that. The first is that those
mechanisms, as my testimony indicated, need to be tied to a
certain amount of tenure as a shareholder. I think that is a
way of ensuring that it is not captive to individuals who are
looking for liquidity.
Senator Shelby. What do you mean by tenure of a
shareholder?
Mr. Silvers. Holding, a holding period, that anyone who
could use such a mechanism would have to be a fairly long-term
holder so that you ensure that it is not used by people who are
seeking a liquidity event as opposed to people who are seeking
the long-term health of the corporation.
Second, I think obviously that there needs to be, depending
on exactly what mechanism one uses, that all discussions in
this area have required that anyone who used such a mechanism
would be either by themselves or as part of a group a
significant set of holders. Certainly in large corporations,
more than 1 percent of the shares, and in very large public
corporations, it is a very large collection of money, or at
least it used to be until recently market events.
And then third, the most important protection here is a
very simple one, which is that we are talking about a
nomination mechanism, not an election mechanism. The majority
of stockholders would have to vote for such a person, and the
corporation, the management, at least under Delaware law, has
the right to use pretty much limitless resources to campaign
against them. I think that is the fundamental barrier.
Senator Shelby, if you would allow me, I want to say a word
or two to you about the Federal Reserve in response to your
questions to Professor Coffee.
Senator Shelby. Yes, sir.
Mr. Silvers. I believe that the concerns you raise are
profound and important ones and that they are profound in
relation to the question of whether we want--the task we are
asking a Systemic Risk Regulator to take on a fundamentally
public task. The Oversight Panel that I serve on its regulatory
reform recommendations specifically stated, if we are going to
ask the Fed to take those obligations on, the Fed must be
governed differently.
I would be comfortable, personally, with that arrangement,
with a greater degree of accountability and doing away with the
self-regulatory aspects of some of what the Fed does. But I am
convinced that that is probably not the best way to do this,
and the reason why I am convinced about that is because, A, I
am think that the Fed--there are tradeoffs with the Fed and the
other things we ask the Fed to do. And the second reason is
because, while I agree with Professor Coffee that the SEC is
not suited to be the Systemic Risk Regulator, that that job is
going to--as long as we have a twin peaks-type model where
information about our markets is flowing from different
directions within the regulatory system, that that coordination
of information and openness to information is critical. If we
ask one body to take it on, that is going to have an impact on
the flows from the other bodies.
The best answer, I think, in light of that is an agency
with staff that is governed by the heads of our twin peaks or
our three peaks. I hope we don't get to 14. How many hundred
peaks are we talking about, Lynn?
[Laughter.]
Mr. Silvers. But an agency that is governed by the
independent regulators but has its own staff and mission in
this area. And I think the Fed would play a very large role
there because they are----
Senator Shelby. And they can't be overridden by the Fed?
Mr. Silvers. Yes. That is----
Senator Shelby. That is important.
Mr. Silvers. I don't think we can give the power to
override fully public bodies charged with issues like investor
protection to the Fed. Thank you, Senator. I appreciate your
indulgence.
Senator Shelby. Thank you.
Chairman Dodd. Thank you, Senator, very much, and that is a
question that many of us have raised, given the already full
plate that the Fed has, in addition to roles they are taking
on. The obvious problem is that if you move away from the Fed
as the model, creating a whole new entity raises another whole
set of issues and that is the quandary I find myself in. I
don't disagree with Richard Shelby's point. We have all talked
about it here at various other times. And then I quickly say to
myself, so what is your alternative? And when I come to my
alternative, I find myself almost in as much of a quandary.
And so we find ourselves in this position of trying to make
a choice between an existing structure in which I can see how
this could fit--I think your point, as well--although you would
have to make some changes in this thing, or trying to create
something altogether new, which has also got its difficulties.
But it is a very critical point, obviously, and one that we are
talking about, obviously, at this point.
Anyway, with that point, Senator Bennet, I thank you for
your patience.
Senator Bennet. Thank you, Mr. Chairman.
I want to start broadly and then ask a couple of narrow
questions. Professor Coffee talked about the difference between
the culture of the banking regulator and the culture of the
securities regulator, which has been a theme that we have heard
about in this Committee, and in thinking about the new
structure, we want to make sure that that culture shifts, I
think, so that we get the kind of oversight that all of us will
feel comfortable with.
In addition to that, there is the issue of no matter what
structure you have, the constant innovation that goes on in the
market and having some assurance that the regulator is keeping
up with that innovation, as well. We want the innovation but we
also want to make sure we understand it.
And then Mr. Turner's observation that what is really
critical, as it is with all human institutions, is that you get
the right people in the job, and unfortunately, neither he nor
we have the magic wand that he called for.
But I guess the question I have is, are there thoughts from
you, Professor Coffee or others on the panel, about what we
could do in this legislation to assure that we have the kind of
attention to the changes in the market knowledge about approach
and the right people so that we can really get the job done?
Mr. Coffee. First of all, you and Senator Shelby are
definitely focused on the proper issue. The Federal Reserve may
have to change. You may have to give it a very different staff,
a very different accountability structure. You are certainly
going to want it to monitor, but I don't think you can ever
make the Fed into a strong enforcement agency. I don't think
you could ever make the SEC into a strong prudential
supervisor.
What I think you have to recognize is in terms of new
market developments, the Fed has a more universal view. The
SEC, at least as it stands today, doesn't have jurisdiction
over swaps, over-the-counter derivatives, or futures. It is not
going to know inherently what is going on in those areas. Yes,
you could merge the SEC and the CFTC, but that compounds the
political difficulties of achieving our solution by, I think,
several orders of magnitude.
And I would agree with the prioritization model that Mr.
Ryan just discussed. First create the optimal kind of Systemic
Risk Regulator, which may require changing the Fed, changing
its accountability structure, giving it a permanent staff that
would do the kind of monitoring we have in mind. But I think
that is the smaller change than designing something totally
from scratch.
Mr. Stevens. Senator, could I comment on that? I know that
many have not had a chance to absorb it, but I am very struck
by what Chairman Bernanke had said today, because everyone is
talking about his agency, of course, and he says, and I am
quoting now, ``Any new systemic risk authority should rely on
the information assessments and supervisor and regulatory
programs of existing financial supervisors and regulators
whenever possible.''
What that means to me is they don't want to take all these
functions aboard themselves. They want a very strong Capital
Markets Regulator. They want a very strong bank regulator. They
probably want a very strong Federal insurance regulator that
they can work with. The notion that they can pull all of that
inside the Fed and at the same time accomplish their
traditional missions is something I think, as I read the
speech, and this is more subtext than text, is unsettling to
the Chairman of the Fed, and with good reason, I believe.
I would say also in commenting on Chairman Dodd's quandary,
and I don't know if this is useful or not, but I spent a
considerable period of time as Chief of Staff of the National
Security Council and I have reflected a lot on that innovation
in our government. It came online after World War II and our
experience as a nation of the inability to coordinate and
integrate the efforts of our Diplomatic Service, our Armed
Forces, and the like at a time when we had burgeoning global
responsibilities as the superpower in the aftermath of World
War II. It is a Cabinet-level Council that is chaired by the
President. It has a staff whose professionalism and abilities
have been built up over time. And its function is there to
collect information, to monitor developments, to integrate and
coordinate the efforts of government.
I think it is not out of the question that you could create
a similar coordinating mechanism, and I think this is part of
what Damon is pointing toward, at a very high level with the
regulatory agencies that would pull all their expertise
together, give the chairmanship of it to someone, and maybe
that is the Chairman of the Fed, give it a permanent staff, and
allow it to be monitoring and collecting data and doing the
analysis, but in conjunction with those who are the front-line
regulators and whose expertise has got to be leveraged. At
least that is, I think, a reasonable concept on which to
reflect.
Senator Bennet. A completely unrelated question. I didn't
come here to ask it, but the Ranking Member asked about mark-
to-market and your answer is very clear. This is a place where
I have gone back and forth. If our markets were lubricated and
were doing what they were supposed to be doing, we wouldn't be
sitting here talking about investing taxpayers' money the way
we are talking about investing it to create stability in the
market.
And I wonder whether there are others here that have a
different point of view on mark-to-market in this sense. It
seems to me that there is a legitimate distinction between
assets that are held by these banks that have no collateral
behind them or very little collateral and assets that are held
by our banks that have collateral but simply have no market
right now and therefore aren't trading at all. I know there is
a pure view that says that should tell you that the assets
don't have value, but the thing I keep stumbling over is that
some have collateral and some don't have collateral and
shouldn't we be taking notice of that?
Mr. Silvers. I do not, in general, share Lynn's complete
enthusiasm for mark-to-market accounting. I think that there
are a wide range of areas in financial accounting where
historical cost accounting is actually more indicative of the
life of the business than mark-to-market. However, the
financial institutions, particularly those with demand deposits
where in theory the funds can walk out the door, are ones that
seem uniquely kind of attuned to mark-to-market principles.
In the course of the work of the Congressional Oversight
Panel, we have done two oversight--two hearings, two field
hearings in relation to our mortgage crisis, which I believe
and I think most economists believe is at the heart of what has
gone wrong here in our economy, that underlies the financial
crisis, and it is clear from those field hearings, in P.G.
County not ten miles from here and in Nevada, that even at this
late date, we do not seem to be able to get rational outcomes
out of private ordering in terms of non-performing mortgages.
We can't get the mortgage providers and the servicers to
negotiate rational outcomes with homeowners.
Now, I believe that this is related to the remnants of non-
mark-to-market accounting in banking, that effectively loans
that are never going to be worth--the banks are carrying loans
that are never going to be worth full value, even though they
are capped at high values maybe not at par, but at close to
par. And the one thing that would force them to mark them down
would be a rational settlement with the homeowner, because then
you would have to admit what you actually had.
Now, you asked about collateral. You walk through the
subdivisions and not all of them are new. In P.G. County, you
have got a lot of people who have been effectively exploited
and stripped of their homes. That is Prince George's County,
for those who are not Washingtonians, here in Maryland. You
look at those properties. There may be collateral, but it will
never support par value, never. It may return--it may recover
value. It may, 10 years from now, if the last very serious real
estate collapse is indicative, and I am afraid this is clearly
worse, in many areas, it took 10 years to recover from the bust
of the late 1980s and early 1990s. But returning to par in 10
years means you are never really worth par present value basis,
you are not going to be there.
And so I am in favor of sort of--I am kind of in the middle
of the road on these issues, but I think we need to recognize
that there could be very, very serious broad economic
consequences of indulging the fantasy that subprime loans
backed up by collapsed residential property are ever going to
be worth par. They are just not. And the consequences of that
pretense is actually throwing people out of their homes.
Senator Bennet. Mr. Chairman, could I ask one more quick
question----
Chairman Dodd. Certainly.
Senator Bennet. ----for Mr. Doe. I was very interested in
your testimony. This is a line of conversation that Senator
Warner and I have been having. I assume that your view is that
there is Federal authority now to be able to intercede, either
through the Treasury or the Fed, with the VRDO market in some
way that may give hospitals, public hospitals, schools, and
other municipal credit some relief from the lack of market that
exists for variable rate debt----
Mr. Doe. Well, I think one of the key issues associated
with that is that many of the issuers that have--that are
confronted now with challenges of restructuring their debt in
the variable rate market, as I pointed out in my testimony,
these variable securities have links to interest rate swaps
which have created all sorts of issues. And one of the things
that these--the cost of termination of the swap transaction has
become overly penal, and in some of these small towns and
counties where it is arguably there was a mismatch in terms of
appropriateness. And again, remember that the regulatory body
of the municipal industry doesn't have purview--has limited
purview, only on dealers and only on cash securities. So here
you have these cash transactions linked to swaps. It makes it a
little bit of a conundrum.
But one of the things we think where the Treasury could
step in and make a big impact is to provide subsidized loans to
municipal issuers that would help them terminate those swap
transactions, and then over time, the cost of those loans could
be recouped in future transactions and discriminate fee, and I
think that would be a really important step. So I think those
are the people and the institutions that have been most
adversely impacted.
Senator Bennet. I would like to thank the witnesses for
their testimony.
Thank you, Mr. Chairman.
Chairman Dodd. Thank you very much.
Senator Reed.
Senator Reed. Well, thank you very much, Mr. Chairman, and
thank you, gentlemen, for excellent testimony.
Just as two preliminary points, first, I want to thank
Professor Coffee for his kind words about our credit rating
legislation. Thank you very much and thanks for your help.
And then to Mr. Turner's point about the need for resources
regardless of what we do, this omnibus we are debating contains
an additional $38 million for the Securities and Exchange
Commission and the proposed budget of the President is around a
13 percent increase over the 2008 marks going forward, so I
completely concur. We can make all the structural and
legislative changes in the world, but if they don't have the
resources and the will to do the job, it won't get done.
One of the impressions I had listening to your comments is
that I think we were in this sort of false logic where
regulators of all ilk were looking at the capital of
institutions, saying we don't have to be too sensitive to their
risk evaluation assessments because they have got capital. Of
course, the capital was risk-based, so you are in this circle
around where if you don't do a good job evaluating capital
based on risk, then you don't have the capital, et cetera.
Part of this goes to Basel, the efficacy or the
effectiveness of Basel. I think that has to be looked at.
And the other issue I think has to be looked at, too, is
just the managerial capacity to run these organizations. I
think one of the issues of size is do you really have the
wherewithal, the computer systems, the structural managerial
skills to run them?
This is a long sort of preface to be saying that it appears
that this, in my view, twin peaks model we will probably adopt
in some form, that by default, perhaps the Federal Reserve will
become the regulator, and unless we make some significant
changes in the culture and the operating standards of the
Federal Reserve, we might be exactly where we were before, that
this sort of just looking at risks--not looking at risks, not
looking really well at management.
So I don't know if you have comments, Professor Coffee and
gentlemen?
Mr. Coffee. I think that is the danger. I think you are
correct and I think it does justify what I think Senator Shelby
was going for, some modification of the charter and the
requirements and responsibilities of the Federal Reserve. I
think the best way to protect transparency is to make sure that
the SEC's authority to require full disclosure is not
circumscribed. But I do think that we are now in the world
where the price of all bank stocks has fallen so low, with
Citicorp trading at $1.50, this is the time to pursue mark-to-
market because the market doesn't believe these banks have any
value. You might as well bring the accounting in accord with
the reality as the market reflects it.
I think your concern is that changes in the Federal Reserve
is a sound concern, and I can't tell you, because I am not a
Federal Reserve expert, of what the five things I would do
first to the Federal Reserve are.
Senator Reed. Does anyone else want to comment? Mr. Ryan?
Mr. Ryan. We have been talking about resources, and many of
the panelists have talked about whatever changes we make here,
let us make sure that we have the right people doing the job,
that they have adequate resources.
One specific topic I think deserves the Committee's
attention and Congress's attention is whoever is going to do
this job has to have the technology resources to get the job
done, because when we ask someone to be the systemic regulator
for our most important financial institutions, we are also, I
think, asking them to do a job that regulators have not really
done well at all, which is to look over the horizon. They are
pretty good at looking back and looking at what went wrong and
let us see if we can fix it. But we are going to ask this new
entity or the Fed to do a job that we have not really done
before and they need to have the tools to do it.
They really need to think about the technology demands,
because right now, we do not have a full understanding of the
aggregated or collective risks of all of these interconnected
entities. We have the capacity to do it from a technology,
hardware, software standpoint, but we don't really have that
done. It is going to be very expensive and it is very important
that you spend some time on that.
Senator Reed. One of the points I would note is that when
Chairman Donaldson became Chairman of the SEC, he tried to
establish a risk assessment operation. That initiative was
undone by his successor. But I think we should consider along
those lines, Mr. Ryan, requiring the systemic regulator to have
a rather independent risk assessment group that on a periodic
basis will publish to the Congress and to the people what they
consider to be the most significant pending risk and the
likelihood. That might force discussion and maybe even
sometimes action.
Professor Bullard, and then Mr. Silvers and then Mr.
Turner, and then I have one last point.
Mr. Bullard. I just wanted to add that, again, to me, the
systemic risk question is one of someone who has oversight over
a range of prudential regulatory regimes. The Fed already is
our systemic regulator. It may not have that aggregated
information as Paul was talking about, but it has the discount
window, it has the open market transactions, it has the ability
essentially to create money, although the Treasury has shown
remarkable ingenuity in creating money recently, as well.
So it already serves in that role. I think it is a separate
question as to as it sits back and decides where it needs to
take action to affect credit markets, it sees some hot spots
over here with respect to the support for some area, that it
also should not necessarily be expected to be the prudential
regulator that is in charge of monitoring what stands behind
that particular area of our financial services, because those
really are separate functions.
The systemic regulator is one who can go in and fix it with
ultimately taxpayer dollars and then tries to find situations
where it can mitigate that risk and reduce it. The prudential
regulator is the one who writes the rules that says, to back
these kinds of liabilities, these are the kinds of assets we
expect you to have, and I am not sure that those are
necessarily ones that should be or have to be housed in the
same agency.
Senator Reed. Mr. Silvers, then Mr. Turner, then I have one
final unrelated question.
Mr. Silvers. Senator Reed, first, I share your concerns
about Basel II. I think that is clearly part of the causal
fabric here for our crisis. There are three points about the
sort of managerial and task challenges associated with systemic
risk regulation.
First, the Congressional Oversight Panel in its regulatory
reform report suggested that the notion of intelligence, of
looking over the horizon in relation to financial market
systemic risk, should perhaps be delegated not to a regulatory
body but to a panel of outside experts--some of my fellow
panelists here might make good members of such a panel--whose
sole job was to look ahead and that were not intertwined in the
politics of the regulatory landscape.
Second--and this is a concern that Senator Shelby raised--
our view was it would be a very bad idea to name who is
systemically significant. In fact, it is not only a bad idea in
terms of moral hazard, but it is actually impossible to do;
that in a crisis people will--institutions will turn out to be
systemically significant that you had no idea were. And Exhibit
1 for that is Bear Stearns.
And there are other times, calm times, when very large
institutions may be allowed to fail, and probably should be;
and that rather than naming institutions, we ought to have the
capacity--and this comes to your point--the capacity for the
systemic risk regulator to work with other regulators to set
ratchets around capital requirements and around insurance costs
to discourage people getting essentially too big to fail and to
set up the financial basis to rescue them if they do.
Finally, there is, I think, some--I am not a Fed expert,
but I think there is some confusion about where the money comes
from for bailouts and rescues and so forth. The Fed does not
have the authority, as far as I know--although you all maybe
can educate me. The Fed does not have the authority to simply
expend taxpayer dollars. The Fed lends money. It is the lender
of last resort.
In a true systemic crisis, as we have just learned, we get
beyond the ability of liquidity to solve the problem, and in
that circumstance we start expending taxpayer dollars.
It is hard not to look at the TARP experience and what
preceded it and not conclude that the ad hoc quality of those
experiences did not build public confidence or political
support for what had to be done.
Given all of that, I think we need to understand that when
we ask a body to take on the role of systemic risk regulator,
that also means we are asking them to take on the role of
rescuer, and potentially to expend taxpayer dollars. And that I
think requires a set of governance mechanisms and capacities,
to your question, Senator, that we have yet really to build.
And it also requires, I think, a careful balance between
genuine public accountability and transparency, on the one
hand, and genuine independence from the all too eager desire of
everyone around to bail out their friends.
Senator Reed. Mr. Turner, do you have comments?
Mr. Turner. Just like Damon, I would say Basel II needs to
be re-examined. I expressed concern almost 8 years ago to the
Fed that it would not work, and I think if it stays the way it
is, it will contribute to further problems. I think Sheila Bair
has been very insightful on that in that regard.
As far as managing the risk, I have actually had to run a
large international semiconductor company, and in the
technology area, we had a lot of risk, and it changed very
dramatically, much faster than what it even does in the banking
industry. And what we found was, if we are going to be
successful in managing the risk, we could not do it with the
same people that we had necessarily running the operational,
the manufacturing side of the company. You needed a group of
people that were much more focused on the future and where
things were going. They needed to be looking not just ahead,
but much further ahead, and have a pulse not only what was
going on but where that turned around and took you.
After Donaldson formed the Risk Management Office, I went
and visited with him for a while. Certainly that type of
mentality plus the tools were not in that group at that point
in time. I have not seen that at the Fed in my dealings with
the Fed over the last couple decades either.
I am not sure you can get that without a major wholesale
change, and so I come back to--having gone through this and
having to manage risk myself, I come back to probably what
Damon has said, and probably the best way to put this together
across the broad spectrum would be to create the separate
organization, chaired with the board of the major chairmen of
the major agencies, but with real staff and real resources and
focused on that aspect of the business. I just do not think we
are going to get it if you put it inside one of these agencies.
And, in fact, think about it. We have had risk management--
a Risk Management Office in the Fed, in the OCC, in the SEC,
and it has not worked. And why would we turn around, given what
this devastation and travesty has cost us all, why would we go
back and say let us try it again? You know, this is not one
where I give people another swing at the bat.
Senator Reed. Mr. Chairman, you have been very kind. I have
one question, if I may.
Chairman Dodd. Sure.
Senator Reed. I will address it to Professor Coffee,
because it might be way off the beaten track. In fact, it
sounds like an extra credit question in a law exam.
[Laughter.]
Senator Reed. So here it goes. Whatever happened to Rule
10b-5? I mean, I have been listening to discussions of
potential fraud in the marketplace, securities that had no
underlying underwriting. And I grew up thinking that material
omissions as well as material commissions gives the SEC in
every capacity, as long as it is a security, to go in
vigorously to investigate, a private right of action, and yet I
have been before the Committee now for 2 and 3 years, and I do
not think anyone has brought up, you know, Rule 10b-5 actions.
Can you just sort of----
Mr. Coffee. I am glad you asked that question because it is
a good question, but there are two major limitations on Rule
10b-5. As you have heard from others on this panel, it does not
apply to aiders and abetters, even those who are conscious co-
conspirators in a fraud. That is one limitation that Congress
can address. And, two, when you try to apply Rule 10b-5 to the
gatekeepers, whether it is the accountants or the credit rating
agencies, you run up against the need to prove scienter. It is
possible to have been stupid and dumb rather than stupid and
fraudulent, and that is basically the defense of accountants
and credit rating agencies.
I think you need to look to a standard of scienter that
will at least create some threat of liability when you write an
incredibly dumb AAA credit report on securities that you have
not even investigated, because you do not do investigations as
a credit rating agency. You just assume with the facts that you
are given by management.
So I do think there is some need for updating the anti-
fraud rules for the reasons I just specified.
Senator Reed. Thank you.
Chairman Dodd. Senator Warner.
Senator Warner. Thank you, Mr. Chairman. A fascinating
panel. First of all, I commend you for asking that ``What is
the one take-away?'' question from each of these gentlemen. And
while I think there was a consensus that we need to get rid of
this shadow market, we need to make sure we get rid of this
Swiss cheese approach to regulation, I think we will be
challenged, taking some of these broad overviews and taking
them into specific legislation.
Chairman Dodd. I agree.
Senator Warner. And I appreciate your asking that question.
I want to follow up, before I get to my quick question, on
Senator Shelby's comments along the notion of the institutions
that have posed this systemic risk, the ``too big to fail''
excuse, and Damon's comments about perhaps not publishing those
that are systemic risks, but this problem we are in the middle
of the crisis now of too big to fail. And I would be curious
perhaps in a written question to the Members--I know Senator
Shelby has, I think, provocatively raised a number of times the
issue of, well, how much more on Citi and should we go ahead
and let it go through some kind of process? And the quick
response normally being, well, no, that is too big to fail.
Well, I would love to hear from the panel, perhaps in
written testimony, if you were to see the transition, dramatic
transition--and I know we are sometimes afraid of the
terminology, whether it is ``receivership'' or
``nationalization,'' some other way to get it out of the
current ditch that it is in--you know, how you would take one
of these institutions that fall into this ``too big to fail''
category that appears to have real solvency issues and get it
through a transition? And I perhaps would work with the Senator
on submitting that type of written question.
So we have seen, you know, the big take-aways on how we
regulate and where we put this prudential or systemic risk
oversight. We have seen the question of how we deal with the
current challenging institution. I want to come with my
question, and I know our time is about up, but I will start
with Mr. Pickel, but would love to hear others' comments on
this, and that is, maybe come at this from the other end.
Even if we get the risk right, with the great people that
Mr. Turner has advocated, where and how should we look at the
products? I would argue that intellectually I understand the
value of derivatives and the better pricing of risk. I candidly
would love somebody to say, How much societal value have we
gained from this additional pricing of this risk when we have
seen all of the downside that the whole system is now absorbing
because, to use your terms, you know, actions by AIG and others
of misunderstanding of the products and not taking appropriate
hedging?
I guess I have got a series of questions. How do we prevent
the current products or future products from being abused?
Should we have standards whereby if an AIG, a future AIG,
either misunderstood or went beyond protocols, that that would
set off more than an alarm bell and would require some kind of
warning? Is it simply enough to say we are going to move toward
some level of a clearinghouse? Is clearing alone enough
security? As some of the European regulators have talked about
for those products and contracts that do not go through a
clearinghouse, should there be needs of additional capital
requirements?
You know, I am all for innovation, but in some cases I
think under the guise of financial innovation and financial
engineering, we have ended up with a lot of customers,
including customers that Mr. Doe represents in terms of some of
the muni market, getting in way over their head. And I just
fear on a going-forward basis that regulation and transparency
alone may not solve the problem.
So rather than coming at it at the macro level on
regulation or on the specific issues that I think Senator
Shelby has wonderfully raised about how do we unwind one of the
``too big to fail'' institutions, I would like to look at it
from the bottom up on the products line, starting with Mr.
Pickel and then anybody else can comment.
Mr. Pickel. Yes, I think as far as the products themselves,
if you look, for instance, at the credit default swap market,
there is information that has been published by the Depository
Trust and Clearing Corporation through their trade information
warehouse, which encompasses 80 to 90 percent of credit default
swaps engaged in around the world. And the information there is
that virtually all the trades in that warehouse, essentially
all, 100 percent, are done involving at least one dealer party
who is, in fact, a regulated institution, and actually 86
percent of them are between two dealer institutions. So you
have got that structure of the institutional regulation there,
of the oversight of those individual firms looking at the
activities of those firms. And I think the Committee, again,
heard testimony from the OTS last week admitting some
shortcomings in their enforcement and in their execution of
their authority, but perhaps we should look at making sure that
they have got the ability to understand and get more detail on
the products that those individual entities are----
Senator Warner. Just a quick question, Mr. Pickel. But
those 86 percent of institutions that are involved in using
these products, are you saying the market knows all the terms
and conditions and that we have got a transparent market there?
Mr. Pickel. The parties who engage in those transactions
have access to information and have the transparency to engage
in those transactions. I think you also have regulators who
have the authority--whether they have exercised it and what
they have done with that, we should discuss further. But they
have the authority to understand what those institutions are
doing.
I think the other thing is--and we have got a very good
example of this in the credit default swap market, an effort
that goes back to September 2005, started by now----
Treasury Secretary Geithner, to pull in the regulators in a
global initiative, regulators from around the world, as well at
that time 18 major credit default swap players, dealers, and
also buy-side entities as well, to talk about issues that were
serious and needed to be addressed in the credit default swap
market at that time. And significant progress was made very
quickly, with the implicit threat--or, actually, explicit
threat, I think, from the regulators, that if you do not get
your act in order on these backlogs and assignment issues, that
they will actually stop people from trading. So the regulators
indicated that they would take that action, and the industry
responded.
The experience that we have gone through in settling credit
default swaps over the past 6 to 8 months has been
significantly facilitated by the foresight of Secretary
Geithner at that time to anticipate these problems. So that was
an important step at that time.
So I think looking at those kind of public-private
interactions where regulators and the industry work together to
identify these issues is very important going forward as well.
Senator Warner. Mr. Stevens, I would love to hear from you,
Mr. Ryan, and Mr. Silvers.
Mr. Stevens. Thank you, Senator. I think it is a really
excellent question, and I have asked myself this, and it is not
intended as a competitive observation.
If Franklin Roosevelt were to come back today and he would
find we had these enormous pooled funds that were outside,
virtually outside of any form of regulation, I think he would
say, ``I thought we solved that problem in 1940.''
We need to make sure that the evident developments--and
these are not secrets--the evident developments, major
developments in our capital markets are addressed as they
arise. Hedge fund investing is no doubt a tremendous innovation
that can be of great value. But there were trillions of dollars
in hedge funds that had no form of regulation. I think that is
something that Congress was aware of, certainly the SEC was
aware of.
You could say the same about the major pooled funds in the
money markets that will be part of the subject of our report
when it is issued. Money market mutual funds are about a $4
trillion intermediary, but we're only about a third of the
money market, which has many other pooled funds.
So I think it is a problem--and this is how I envision it--
of making sure that the capital markets regulator is staying
even with market developments, and that is going to require not
only nimbleness at a regulatory level, but, frankly, Mr.
Chairman, it requires--it puts a burden on Committees like
yours, because in many instances it is going to require the
tough work of closing regulatory gaps, providing new authority,
and even providing new resources.
I do not think, however, that the answer, Senator is
creating a new agency that only looks at products, because
those products arise and exist in the context of a larger
marketplace, and they need to be understood in that context.
Senator Warner. Mr. Ryan.
Mr. Ryan. Bob Pickel and I have basically overlapping
membership. He is very domain oriented, very specific to
derivatives, and we are basically almost all of the other
products and oversight.
We have spent a lot of time, I would say, over the last 6
months trying to figure out what should we be recommending for
a new regulatory structure. And I would say it is a uniform
view among the core members of our group and of his group that
we feel comfortable recommending a systemic risk regulator that
would have no real limits on their authority. So they would
have all markets, they would have all market participants who
are significant. It would not make any difference of their
charter, so it could be a bank, it could be an insurance
company, it could be a hedge fund. And included therein would
be their authority to deal with, for instance, derivative
products.
So we can see that there is a lack of confidence in the
system. There is a lack of confidence among Members of this
Committee, Members of Congress, members of other statutory
developing entities around the globe, and we need to address
that.
So our first attempt at this is to say let us do it through
the systemic regulator. Through the systemic regulator, we will
also expand the activity, expand the breadth and depth of what
is done from a regulatory standpoint to cover areas that have
been discussing during this panel, some of the stuff that Paul
has raised. That is the best way to do it.
We are also going to, in a very early phase, be able to
address most of the key issues and do it in a thoughtful
manner.
Senator Warner. Mr. Silvers and Mr. Doe.
Mr. Silvers. Senator, these are very acute observations you
have made about this set of questions.
First, I am pleased to see that a moment of disagreement
has emerged. My colleagues on the panel who wish to put the
burden of regulating unregulated markets, like hedge funds and
derivatives, on the systemic risk regulator are, in my opinion,
making a grave mistake. What we need is routine regulation in
those areas. That is what closing the Swiss cheese system is
about, is routine regulation, not emergency regulation, not,
you know, looking at will they kick off a systemic crisis. Just
an observation about that.
I think that the Fed's refusal to regulate mortgages was
rooted somehow in the sense that consumer protection was a kind
of--something that was not really a serious subject for serious
people. It turned out to be, of course, the thread that
unraveled the system. I think that we should learn something
from that.
When we talk about routine regulation in these areas, I
think to your question, we have got to understand that it is
more than one thing. For example, a credit default swap
contract is effectively a kind of insurance. And if someone is
writing that insurance, they should probably have some capital
behind the promise they are making. That is what we learned not
just in the New Deal but long before it about insurance itself,
which was once an exotic innovation. But we learned we had to
have capital behind it.
But that is not the extent of what we need to do. If, for
example, there are transparency issues, there are disclosure
issues associated with these kinds of contracts, for contracts
in which public securities are the underlying asset, it is
clear that we need to have those kinds of disclosures, because
if we do not, then we have essentially taken away the
transparency from our securities markets.
Now--two final points. One, derivatives and hedge funds
have something profound in common. They do not have any
substantive content as terms. They are legal vehicles for
undertaking anything imaginable. You can write a derivative
contract against anything. You can write it against the
weather, against credit risk, against currency risk, against
securities, against equity, against debt. It is just a legal
vehicle for doing things in an unregulated fashion.
A hedge fund is the same thing. The hedge fund is not an
investment strategy. It is just a legal vehicle, and it is a
legal vehicle for managing money any way you can imagine, in a
way that essentially evades the limits that have historically
been placed on bank trusts and mutual funds and so on and so
forth.
What is smart regulation here is not specific to those
terms. It is specific to those activities. It is specific to
money management. It is specific to insurance. It is specific
to securities. And that is why it is so important that when we
talk about filling these regulatory gaps, we do so in a manner
that is routine, not extraordinary.
Thank you.
Mr. Doe. Senator Warner, if I could just offer an example,
I like Mr. Silvers' comment about the subject of regulation
being routine, because I think that brings vigilance. Let me
give you just a quick example of why when I hear you ask the
question about products, why I think that is so important.
After the Lehman bankruptcy in September, on the Wednesday
following there was a liquidation, an unannounced liquidation
by a money market fund of substantial holdings of cash-
equivalent securities which had been created in the municipal
market through leverage programs and which were used--
essentially synthetics securities, so derivatives.
The liquidation, unannounced--again, a trying time in the
market in mid-September--resulted in the following day of there
being no liquidity in the municipal secondary market, where one
transaction that occurred in a distressed situation resulted in
the repricing of the entire holdings of investors that were in
mutual funds that were in individual holdings.
We estimated that, in a back-of-the-envelope kind of way,
about $5.5 billion were lost on that September 18th, solely
because an illiquid market, because of liquidation of a cash
security that was synthetic in order to fulfill the needs of
having short-term investments for these money market funds, is
that created a crisis in confidence that--and a confusion among
investors as to what was the security of the credits of the
States, of the towns that were, you know, issuing municipal
debt. And that type of concern--and that lasted through
September and October, and municipal issuers who were trying to
come to market and raise important funds for capital projects
and for operations were really inhibited by extraordinarily
penal rates.
So here we had this, you know, single event and this
cascaded, touching upon cash securities, derivative securities,
and then also tied to the supposedly the most secure cash
equivalents in these money market funds.
The other thing I think is really important and not to be
lost here, as we are talking about cash securities, we are
talking about derivatives, and a lot has been talked about
credit default swaps, the municipal market, predominantly it is
interest rate swaps. Here, again, there is not transparency.
And yet these are linked intimately with cash transactions. And
when we talk about, gee, the taxpayer is coming in and helping
to bail out the various transgressions that have occurred in
the banking system or in the financial system is that here we
have taxpayers--and I think, Senator Shelby, you had some
instances with some derivatives in your State that are getting
a lot of headlines. And taxpayers are on the hook most directly
right there. And I would argue and suggest to you the notion of
really examining this opportunity that we have in our U.S.
municipal bond market, where all these products have come to
roost, and the credit default swap market is emerging. It is in
its nascent stage in the municipal bond market. Yet it is
there, and it is creating perhaps a thinness or an illiquid
market that those derivative products is maybe creating
misconceptions about the soundness of our States and our towns
and our counties.
And so I think that when we start looking at how do we gain
transparency on these securities that are now part of the risk
management of our municipalities and how do we help so we can
understand them, so we can see them, so investors that are
putting their--are facilitating the borrowing by buying these
securities, they can see what is going on, and we can also help
to protect these issuers who, as Mr. Turner was saying--well,
as we were talking about broadly in this financial regulation
of the separation of risk management and operation, is that
here we have these--our States and our towns and counties that
are serving--wearing both hats and using complex securities
that they may not have fully understood.
So I guess when I hear you talk about products, I applaud
that, because I think that it just cannot be the people
involved. We have to look at what is being used, but also
being--the word has been used--``nimble'' so that we can adapt
regulation and be flexible so that as new innovations come in
that can be very positive but also can be seen and understood.
Senator Warner. And I think our time has expired, and my
only last comment would just be that I think we will get to
some stance where we will have some level of regulatory
oversight. My hope is that we will adhere to Mr. Turner's
suggestion that it is a nimble and well-funded regulator.
But I would say from the industry, we are going to need
your help on setting standards not just retrospectively but
prospectively. With the complexity and financial engineering
that goes on, I just do not want to be here 5 years later
looking at what the next round of new products would be and
say, ``Why didn't we see that ahead of time?'' and helping us
see what those standards--so that you do not end up with having
to pre-clear every new product at some regulator. You know, you
are going to have to really step up on this one and give us
some assistance.
Thank you, Mr. Chairman.
Chairman Dodd. Well, thank you as well, Senator Warner.
Very good questions.
Before I turn to Senator Shelby for any closing comments or
questions he has, I am struck by a couple of things. It is
exactly the point that Senator Warner was concluding with.
There is this debate about whether or not we have a principle-
based system or a rule-based system in the country. And I have
always felt I was sort of in a small, tiny minority that is
attracted to the principle-based system for the simple reason
that it seems to me almost in a way a bit more intimidating
than a rule-based system for the very reason that Senator
Warner suggested, that you end up setting standards or rules,
and within a matter of literally hours, in some cases, very
creative, imaginative people can come up and figure out some
way just to get around that rule, legally and ethically and
every other reason. And so you are back at it again because
someone has come up--now, I think the idea of a clearinghouse
makes a lot of sense, by the way, new product lines, and I know
Senator Shelby feels as strongly as I do about that.
But that in itself sort of is an indication of a problem
you have with a rule-based system, and I wonder if just quickly
any of you have any quick comments on a rule-based versus a
principle-based system that you would care to share at this
point. Professor Coffee.
Mr. Coffee. Well, I have written a long article on this
that is currently posted on SSRN. I do not think any workable
system can exist without being a combination of both. You need
the principles to backstop the rules, but you can really only
enforce rules, and particularly in our litigation-oriented
system, we want rules that let you know you are within the safe
harbor and you have done what you are supposed to do.
So I think there has to be a combination of both with
principles backstopping the rules.
Chairman Dodd. Yes. Anyone else want to comment on that?
Mr. Turner. Senator Dodd, I would agree with Jack on this
one. First of all, you know, if you look at the Ten
Commandments, half the people tell you they are principles,
half them tell you they are rules.
[Laughter.]
Mr. Turner. So I am not sure anyone knows really what a
principle or rule is. I think it does take a combination.
Principles get so broad that you just never get enforcement.
Rules get so detailed that people just skirt around them. So it
takes some common sense and a combination.
Chairman Dodd. Mr. Silvers, do you have a comment you want
to make?
Mr. Silvers. Only that one of the reasons why this
discussion has become sort of hard to follow or hard to
understand is because the concept of a principles-based system
became code, became a code word for a weak regulatory system.
Chairman Dodd. Yes.
Mr. Silvers. And, in reality, a true principles-based
system would be the strongest possible regulatory system, but
it would be one no one could live in for the reasons that my
two colleagues on the panel have outlined.
Chairman Dodd. Yes.
Mr. Pickel. I would agree in general that we prefer a
principles-based approach. There may be certain circumstances
such as with retail investors where having clearer rules for
those who engage in those markets would be appropriate. But for
the markets that I think people are engaged in derivatives, in
OTC derivatives, I think the principles approach is the best
one.
Chairman Dodd. Yes.
Mr. Bullard. I would just add, putting on a private
practice hat for a moment, that principles-based regulation is
intimidating, as you described, because what it means is that
regulators have enormous enforcement discretion, and what you
typically have, at least at the SEC, it means that they play
``gotcha'' and bring cases that are based on specific rules
that are made up under those principles as opposed to opposed
to having known ahead of time exactly how the SEC might
interpret certain positions.
Chairman Dodd. Senator Shelby.
Senator Shelby. I will be brief. Mr. Chairman, thank you
for assembling this panel. We could be here all day and
probably learn a heck of a lot.
Principles matter, but rules matter, too. I like the idea
of what Professor Coffee is saying. We might need a hybrid in
some way. If you just have principles and no rules, you know,
gosh, who is going to define them to a certain extent? But just
rules, people say, ``Well, we have got a rule. How can we get
around it? How can we evade it in some way?''
So maybe it is a combination. Who knows? But thank you for
your input, and, you know, we have an awesome task ahead of us
here. We have got to do this right. We cannot rush to it. We
have both met with the President on this and many meetings. It
has got to be comprehensive. It cannot cover every contingency.
But I think we can do better than we have been doing.
I wish my friend Senator Warner was still here, because we
agree on a lot of things, but some of the product approvals,
some of these products have got to be approved before they do
irreparable damage to, I think, the marketplace myself.
Thank you, Mr. Chairman.
Chairman Dodd. Thank you, Senator Shelby, and again, you
have been a terrific panel and offered some wonderful advice.
We will probably submit some additional questions for you over
the coming days, and we look forward to your continuing
involvement with us. We have a formal setting here, but my
intention is to also have some informal settings with
interested members here and other members who are not on the
Committee, necessarily, who would like to be a part of the
discussion as we move forward on this, because there is a
growing interest, obviously, not just on the part of this
Committee but others who care about this issue and are
interested in how we proceed.
So I am very, very grateful to all of you for your
knowledge, your background, your experience, and the
thoughtfulness with which you have prepared your testimony
today and contributing to this very, very difficult task.
Let me just say as well how much I appreciate Senator
Shelby and the other Members of the Committee. Always from time
to time we have our differences, but Senator Shelby has made
the point and I make it as well: This is one where the barriers
that we traditionally see along political lines have to really
evaporate and disappear. I personally have said I am agnostic
on the question of--I do not bring any ideological framework
whatsoever to this. I want to do something that works, that
closes gaps, that does not have that Swiss cheese look to it
where people can forum shop, in a sense, in order to avoid the
regulatory process, to make sure we have good people who are
being adequately compensated for the jobs that they are doing,
and then doing what has to be done, is engaging on a consistent
basis. These things are never done forever. There are always
new products, new ideas, new--which is the genius of this in
many ways. That is not a liability. That is an asset, in a
sense. I have often said our goal here is to, one, make sure
that we have a solid, sound system that reflects the times we
are in, but not so rigid that it in any way strangles the kind
of creativity and imagination that has drawn the world and
others to come here to make their investments, because we are
creative and imaginative. But at the same time, we do not want
to be at such creativity and imagination that we lose the kind
of protections.
Striking that balance is never perfect. It is never
perfect. It is always tilting one way or the other. And our job
is to constantly try and keep that balance, if we can, as we go
forward. And that is the challenge we have in front of us, and
so we welcome your involvement and thank you immensely for your
participation.
The Committee will stand adjourned.
[Whereupon, at 1:16 p.m., the hearing was adjourned.]
[Prepared statements and response to written questions
supplied for the record follow:]
PREPARED STATEMENT OF JOHN C. COFFEE, JR.
Adolf A. Berle Professor of Law,
Columbia University Law School
March 10, 2009
Enhancing Investor Protection and the Regulation of Securities Markets
``When the music stops, in terms of liquidity, things will get
complicated. But as long as the music is playing, you've got to get up
and dance. We're still dancing.''
----Charles Prince, CEO of Citigroup
Financial Times, July 2007
Chairman Dodd, Ranking Member Shelby, and Fellow Senators, I am
pleased and honored to be invited to testify here today.
We are rapidly approaching the first anniversary of the March 17,
2008, insolvency of Bear Stearns, the first of a series of epic
financial collapses that have ushered in, at the least, a major
recession. Let me take you back just one year ago when, on this date in
2008, the U.S. had five major investment banks that were independent of
commercial banks and were thus primarily subject to the regulation of
the Securities and Exchange Commission: Goldman Sachs, Merrill Lynch,
Morgan Stanley, Lehman Brothers, and Bear Stearns. Today, one (Lehman)
is insolvent; two (Merrill Lynch and Bear Stearns) were acquired on the
brink of insolvency by commercial banks, with the Federal Reserve
pushing the acquiring banks into hastily arranged ``shotgun''
marriages; and the remaining two (Goldman and Morgan Stanley) have
converted into bank holding companies that are primarily regulated by
the Federal Reserve. The only surviving investment banks not owned by
larger commercial banks are relatively small boutiques (e.g., Lazard
Freres). Given the total collapse of an entire class of institutions
that were once envied globally for their entrepreneurial skill and
creativity, the questions virtually ask themselves: Who failed? What
went wrong?
Although there are a host of candidates--the investment banks,
themselves, mortgage loan originators, credit-rating agencies, the
technology of asset-backed securitizations, unregulated trading in
exotic new instruments (such as credit default swaps), etc.--this
question is most pertinently asked of the SEC. Where did it err? In
overview, 2008 witnessed two closely connected debacles: (1) the
failure of a new financial technology (asset-backed securitizations),
which grew exponentially until, after 2002, annual asset-backed
securitizations exceeded the annual total volume of corporate bonds
issued in the United States, \1\ and (2) the collapse of the major
investment banks. In overview, it is clear that the collapse of the
investment banks was precipitated by laxity in the asset-backed
securitization market (for which the SEC arguably may bear some
responsibility), but that this laxity began with the reckless behavior
of many investment banks. Collectively, they raced like lemmings over
the cliff by abandoning the usual principles of sound risk management
both by (i) increasing their leverage dramatically after 2004, and (ii)
abandoning diversification in pursuit of obsessive focus on high-profit
securitizations. Although these firms were driven by intense
competition and short-term oriented systems of executive compensation,
their ability to race over the cliff depended on their ability to
obtain regulatory exemptions from the SEC. Thus, as will be discussed,
the SEC raced to deregulate. In 2005, it adopted Regulation AB (an
acronym for ``Asset-Backed''), which simplified the registration of
asset-backed securitizations without requiring significant due
diligence or responsible verification of the essential facts. Even more
importantly, in 2004, it introduced its Consolidated Supervised Entity
Program (``CSE''), which allowed the major investment banks to
determine their own capital adequacy and permissible leverage by
designing their own credit risk models (to which the SEC deferred).
Effectively, the SEC abandoned its long-standing ``net capital rule''
\2\ and deferred to a system of self-regulation for these firms, which
largely permitted them to develop their own standards for capital
adequacy.
---------------------------------------------------------------------------
\1\ See John C. Coffee, Jr., Joel Seligman & Hillary Sale,
SECURITIES REGULATION: Cases and Materials (10th ed. 2007) at 10.
\2\ See Rule 15c3-1 (``Net Capital Requirements for Brokers and
Dealers''), 17 CFR 240.15c3-1.
---------------------------------------------------------------------------
For the future, it is less important to allocate culpability and
blame than to determine what responsibilities the SEC can perform
adequately. The recent evidence suggests that the SEC cannot easily or
effectively handle the role of systemic risk regulator or even the more
modest role of a prudential financial supervisor, and it may be more
subject to capture on these issues than other agencies. This leads me
to conclude (along with others) that the U.S. needs one systemic risk
regulator who, among other tasks, would have responsibility for the
capital adequacy and safety and soundness of all institutions that are
too ``big to fail.'' \3\ The key advantage of a unified systemic risk
regulator with jurisdiction over all large financial institutions is
that it solves the critical problem of regulatory arbitrage. AIG, which
has already cost U.S. taxpayers over $150 billion, presents the
paradigm of this problem because it managed to issue billions in credit
default swaps without becoming subject to regulation by any regulator
at either the federal or state level.
---------------------------------------------------------------------------
\3\ I have made this argument in greater detail in an article with
Professor Hillary Sale, which will appear in the 75th Anniversary of
the SEC volume of the Virginia Law Review. See Coffee and Sale,
``Redesigning the SEC: Does the Treasury Have a Better Idea?''
(available on the Social Science Research Network at http://ssrn.com/
abstract=1309776).
---------------------------------------------------------------------------
But one cannot stop with this simple prescription. The next
question becomes what should be the relationship between such a
systemic risk regulator and the SEC? Should the SEC simply be merged
into it or subordinated to it? I will argue that it should not. Rather,
the U.S. should instead follow a ``twin peaks'' structure (as the
Treasury Department actually proposed in early 2008 before the current
crisis crested) that assigns prudential supervision to one agency and
consumer protection and transparency regulation to another. Around the
globe, countries are today electing between a unified financial
regulator (as typified by the Financial Services Authority (``FSA'') in
the U.K.) and a ``twin peaks'' model (which both Australia and The
Netherlands have followed). I will argue that the latter model is
preferable because it deals better with serious conflict of interest
problems and the differing cultures of securities and banking
regulators. By culture, training, and professional orientation, banking
regulators are focused on protecting bank solvency, and they
historically have often regarded increased transparency as inimical to
their interests, because full disclosure of a bank's problems might
induce investors to withdraw deposits and credit. The result can
sometimes be a conspiracy of silence between the regulator and the
regulated to hide problems. In contrast, this is one area where the
SEC's record is unblemished; it has always defended the principle that
``sunlight is the best disinfectant.'' Over the long-run, that is the
sounder rule.
If I am correct that a ``twin peaks'' model is superior, then
Congress has to make clear the responsibilities of both agencies in any
reform legislation in order to avoid predictable jurisdictional
conflicts and to identify a procedure by which to mediate those
disputes that are unavoidable.
What Went Wrong?
This section will begin with the problems in the mortgage loan
market, then turn to the failure of credit-rating agencies, and finally
examine the SEC's responsibility for the collapse of the major
investment banks.
The Great American Real Estate Bubble
The earliest origins of the 2008 financial meltdown probably lie in
deregulatory measures, taken by the U.S. Congress at the end of the
1990s, that placed some categories of derivatives and the parent
companies of investment banks beyond effective regulation. \4\ Still,
most accounts of the crisis start by describing the rapid inflation of
a bubble in the U.S. housing market. Here, one must be careful. The
term ``bubble'' can be a substitute for closer analysis and may carry a
misleading connotation of inevitability. In truth, bubbles fall into
two basic categories: those that are demand-driven and those that are
supply-driven. The majority of bubbles fall into the former category,
\5\ but the 2008 financial market meltdown was clearly a supply-driven
bubble, \6\ fueled by the fact that mortgage loan originators came to
realize that underwriters were willing to buy portfolios of mortgage
loans for asset-backed securitizations without any serious
investigation of the underlying collateral. With that recognition, loan
originators' incentive to screen borrowers for creditworthiness
dissipated, and a full blown ``moral hazard'' crisis was underway. \7\
---------------------------------------------------------------------------
\4\ Interestingly, this same diagnosis was recently given by SEC
Chairman Christopher Cox to this Committee. See Testimony of SEC
Chairman Christopher Cox before the Committee on Banking, Housing and
Urban Affairs, United States Senate, September 23, 2008. Perhaps
defensively, Chairman Cox located the origins of the crisis in the
failure of Congress to give the SEC jurisdiction over investment bank
holding companies or over-the-counter derivatives (including credit
default swaps), thereby creating a regulatory void.
\5\ For example, the high-tech Internet bubble that burst in early
2000 was a demand-driven bubble. Investors simply overestimated the
value of the Internet, and for a time initial public offerings of
``dot.com'' companies would trade at ridiculous and unsustainable
multiples. But full disclosure was provided to investors and the SEC
cannot be faulted in this bubble--unless one assigns it the very
paternalistic responsibility of protecting investors from themselves.
\6\ This is best evidenced by the work of two University of
Chicago Business School professors discussed below. See Atif Mian and
Amir Sufi, ``The Consequences of Mortgage Credit Expansion: Evidence
from the 2007 Mortgage Default Crisis'', (http://ssrn.com/
abstract=1072304) (May 2008).
\7\ Interestingly, ``moral hazard'' problems also appear to have
underlain the ``savings and loan'' crisis in the United States in the
1980s, which was the last great crisis involving financial institutions
in the United States. For a survey of recent banking crises making this
point, see Note, Anticipatory Regulation for the Management of Banking
Crises, 38 Colum. J. L. & Soc. Probs. 251 (2005).
---------------------------------------------------------------------------
The evidence is clear that, between 2001 and 2006, an extraordinary
increase occurred in the supply of mortgage funds, with much of this
increased supply being channeled into poorer communities in which
previously there had been a high denial rate on mortgage loan
applications. \8\ With an increased supply of mortgage credit, housing
prices rose rapidly, as new buyers entered the market. But at the same
time, a corresponding increase in mortgage debt relative to income
levels in these same communities made these loans precarious. A study
by University of Chicago Business School professors has found that two
years after this period of increased mortgage availability began, a
corresponding increase started in mortgage defaults--in exactly the
same zip code areas where there had been a high previous rate of
mortgage loan denials. \9\ This study determined that a one standard
deviation in the supply of mortgages from 2001 to 2004 produced a one
standard deviation increase thereafter in mortgage default rates. \10\
---------------------------------------------------------------------------
\8\ See Mian and Sufi, supra note 6, at 11 to 13.
\9\ Id. at 18-19.
\10\ Id. at 19.
---------------------------------------------------------------------------
Even more striking, however, was its finding that the rate of
mortgage defaults was highest in those neighborhoods that had the
highest rates of securitization. \11\ Not only did securitization
correlate with a higher rate of default, but that rate of default was
highest when the mortgages were sold by the loan originator to
financial firms unaffiliated with the loan originator. \12\ Other
researchers have reached a similar conclusion: conditional on its being
actually securitized, a loan portfolio that was more likely to be
securitized was found to default at a 20 percent higher rate than a
similar risk profile loan portfolio that was less likely to be
securitized. \13\ Why? The most plausible interpretation is that
securitization adversely affected the incentives of lenders to screen
their borrowers.
---------------------------------------------------------------------------
\11\ Id. at 20-21.
\12\ Id.
\13\ See Benjamin J. Keys, Tanmoy K. Mukherjee, Amit Seru, and
Vikrant Vig, ``Did Securitization Lead to Lax Screening? Evidence from
Subprime Loans,'' (http://ssrn.com/abstract=1093137) (April, 2008).
These authors conclude that securitization did result in ``lax
screening.''
---------------------------------------------------------------------------
Such a conclusion should not surprise. It simply reflects the
classic ``moral hazard'' problem that arises once loan originators did
not bear the cost of default by their borrowers. As early as March,
2008, The President's Working Group on Financial Markets issued a
``Policy Statement on Financial Market Developments'' that explained
the financial crisis as the product of five ``principal underlying
causes of the turmoil in financial markets'':
a breakdown in underwriting standards for subprime
mortgages;
a significant erosion of market discipline by those
involved in the securitization process, including originators,
underwriters, credit rating agencies, and global investors,
related in part to failures to provide or obtain adequate risk
disclosures;
flaws in credit rating agencies' assessment of subprime
residential mortgages . . . and other complex structured credit
products, . . .
risk management weaknesses at some large U.S. and European
financial institutions; and
regulatory policies, including capital and disclosure
requirements, that failed to mitigate risk management
weaknesses. \14\
---------------------------------------------------------------------------
\14\ The President's Working Group on Financial Markets, ``Policy
Statement on Financial Market Developments,'' at 1 (March 2008).
Correct as the President's Working Group was in noting the
connection between the decline of discipline in the mortgage loan
origination market and a similar laxity among underwriters in the
capital markets, it did not focus on the direction of the causality.
Did mortgage loan originators fool or defraud investment bankers? Or
did investment bankers signal to loan originators that they would buy
whatever the loan originators had to sell? The available evidence tends
to support the latter hypothesis: namely, that irresponsible lending in
the mortgage market was a direct response to the capital markets'
increasingly insatiable demand for financial assets to securitize. If
underwriters were willing to rush deeply flawed asset-backed
securitizations to the market, mortgage loan originators had no
rational reason to resist them.
The rapid deterioration in underwriting standards for subprime
mortgage loans is revealed at a glance in the following table: \15\
---------------------------------------------------------------------------
\15\ See Allen Ferrell, Jennifer Bethel and Gang Hu, Legal and
Economic Issues in Litigation Arising from the 2007-2008 Credit Crisis
(Harvard Law & Economics Discussion Paper No. 612, Harvard Law School
Program in Risk Regulation Research Paper No. 08-5) at Table 4.
Underwriting Standards in Subprime Home-Purchase Loans, 2001-2006
----------------------------------------------------------------------------------------------------------------
Debt
Year Low/No-Doc Payments/ Loan/Value ARM Share Interest-
Share Income Only Share
----------------------------------------------------------------------------------------------------------------
2001...................................... 28.5% 39.7% 84.0% 73.8% 0.0%
2002...................................... 38.6% 40.1% 84.4% 80.0% 2.3%
2003...................................... 42.8% 40.5% 86.1% 80.1% 8.6%
2004...................................... 45.2% 41.2% 84.9% 89.4% 27.3%
2005...................................... 50.7% 41.8% 83.2% 93.3% 37.8%
2006...................................... 50.8% 42.4% 83.4% 91.3% 22.8%
----------------------------------------------------------------------------------------------------------------
Source: Freddie Mac, obtained from the International Monetary Fund.
The investment banks could not have missed that low document loans
(also called ``liar loans'') rose from 28.5 percent to 50.8 percent
over the 5 year interval between 2001 and 2006 or that ``interest
only'' loans (on which there was no amortization of principal)
similarly grew from 6 percent to 22.8 percent over this same interval.
Thus, the real mystery is not why loan originators made unsound
loans, but why underwriters bought them. Here, it seems clear that both
investment and commercial banks saw high profits in securitizations and
believed they could quickly sell on a global basis any securitized
portfolio of loans that carried an investment grade rating. In
addition, investment banks may have had a special reason to focus on
securitizations: structured finance offered a level playing field where
they could compete with commercial banks, whereas, as discussed later,
commercial banks had inherent advantages at underwriting corporate debt
and were gradually squeezing the independent investment banks out of
this field. \16\ Consistent with this interpretation, anecdotal
evidence suggests that due diligence efforts within the underwriting
community slackened in asset-backed securitizations after 2000. \17\
Others have suggested that the SEC contributed to this decline by
softening its disclosure and due diligence standards for asset-backed
securitizations, \18\ in particular by adopting in 2005 Regulation AB,
which covers the issuance of asset backed securities. \19\ From this
perspective, relaxed discipline in both the private and public sectors
overlapped to produce a disaster.
---------------------------------------------------------------------------
\16\ See text and notes infra at notes 56 to 61.
\17\ Investment banks formerly had relied on ``due diligence''
firms that they employed to determine whether the loans within a loan
portfolio were within standard parameters. These firms would
investigate and inform the underwriter as to the percentage of the
loans that were ``exception'' loans (i.e., loans outside the investment
bank's normal guidelines). Subsequent to 2000, the percentage of
``exception loans'' in portfolios securitized by these banks often rose
from the former level of 25 percent to as high as 80 percent. Also, the
underwriters scaled back the intensity of the investigations that they
would authorize the ``due diligence'' firm to conduct, reducing from 30
percent to as few as 5 percent the number of loans in a portfolio that
it was to check. See Vikas Bajaj & Jenny Anderson, ``Inquiry Focuses on
Withholding of Data on Loans,'' N.Y. Times, January 12, 2008, at p. A-
1.
\18\ See Richard Mendales, ``Collateralized Explosive Devices: Why
Securities Regulation Failed to Prevent the CDO Meltdown And How To Fix
It'' (Working Paper 2008) at 36 (forthcoming in 2009, U. Ill. L. Rev.).
\19\ See Securities Act Release No. 8518 (``Asset-Backed
Securities'') (January 7, 2005, 79 FR 1506). Regulation AB codified a
series of ``no action'' letters and established disclosures standards
for all asset-backed securitizations. See 17 C.F.R. 229.1100-1123
(2005). Although it did not represent a sharp deregulatory break with
the past, Regulation AB did reduce the due diligence obligation of
underwriters by eliminating any need to assure that assets included in
a securitized pool were adequately documented. See Mendales, supra note
18.
---------------------------------------------------------------------------
Credit Rating Agencies as Gatekeepers
It has escaped almost no one's attention that the credit rating
agencies bear much responsibility for the 2008 financial crisis, with
the consensus view being that they inflated their ratings in the case
of structured finance offerings. Many reasons have been given for their
poor performance: (1) rating agencies faced no competition (because
there are really only three major rating agencies); (2) they were not
disciplined by the threat of liability (because credit rating agencies
in the U.S. appear never to have been held liable and almost never to
have settled a case with any financial payment); (3) they were granted
a ``regulatory license'' by the SEC, which has made an investment grade
rating from a rating agency that was recognized by the SEC a virtual
precondition to the purchase of debt securities by many institutional
investors; (4) they are not required to verify information (as auditors
and securities analysts are), but rather simply express views as to the
creditworthiness of the debt securities based on the assumed facts
provided to them by the issuer. \20\ These factors all imply that
credit rating agencies had less incentive than other gatekeepers to
protect their reputational capital from injury. After all, if they face
little risk that new entrants could enter their market to compete with
them or that they could be successfully sued, they had less need to
invest in developing their reputational capital or taking other
precautions. All that was necessary was that they avoid the type of
major scandal, such as that which destroyed Arthur Andersen & Co., the
accounting firm, that had made it impossible for a reputable company to
associate with them.
---------------------------------------------------------------------------
\20\ For these and other explanations, see Coffee, GATEKEEPERS:
The Professions and Corporate Governance (Oxford University Press,
2006), and Frank Partnoy, ``How and Why Credit Rating Agencies Are Not
Like Other Gatekeepers'' (http://ssrn.com/abstract=900257) (May 2006).
---------------------------------------------------------------------------
Much commentary has suggested that the credit rating agencies were
compromised by their own business model, which was an ``issuer pays''
model under which nearly 90 percent of their revenues came from the
companies they rated. \21\ Obviously, an ``issuer pays'' model creates
a conflict of interest and considerable pressure to satisfy the issuer
who paid them. Still, neither such a conflicted business model nor the
other factors listed above can explain the dramatic deterioration in
the performance of the rating agencies over the last decade. Both
Moody's and Standard & Poor were in business before World War I and
performed at least acceptably until the later 1990s. To account for
their more recent decline in performance, one must point to more recent
developments and not factors that long were present. Two such factors,
each recent and complementary with the other, do provide a persuasive
explanation for this deterioration: (1) the rise of structured finance
and the change in relationships that it produced between the rating
agencies and their clients; and (2) the appearance of serious
competition within the ratings industry that challenged the long stable
duopoly of Moody's and Standard & Poor's and that appears to have
resulted in ratings inflation.
---------------------------------------------------------------------------
\21\ See Partnoy, supra note 20.
---------------------------------------------------------------------------
First, the last decade witnessed a meteoric growth in the volume
and scale of structured finance offerings. One impact of this growth
was that it turned the rating agencies from marginal, basically break-
even enterprises into immensely profitable enterprises that rode the
crest of the breaking wave of a new financial technology.
Securitizations simply could not be sold without ``investment grade''
credit ratings from one or more of the Big Three rating agencies.
Structured finance became the rating agencies' leading source of
revenue. Indeed by 2006, structured finance accounted for 54.2 percent
of Moody's revenues from its ratings business and 43.5 percent of its
overall revenues. \22\ In addition, rating structured finance products
generated much higher fees than rating similar amounts of corporate
bonds. \23\ For example, rating a $350 million mortgage pool could
justify a fee of $200,000 to $250,000, while rating a municipal bond of
similar size justified only a fee of $50,000. \24\
---------------------------------------------------------------------------
\22\ See In re Moody's Corporation Securities Litigation, 2009
U.S. Dist. LEXIS 13894 (S.D.N.Y. February 23, 2009) at *6 (also noting
that Moody's grossed $1.635 billion from its ratings business in 2006).
\23\ See Gretchen Morgenson, ``Debt Watchdogs: Tamed or Caught
Napping?'' New York Times, December 7, 2008, at p. 1, 40.
\24\ Id.
---------------------------------------------------------------------------
Beyond simply the higher profitability of rating securitized
transactions, there was one additional difference about structured
finance that particularly compromised the rating agencies as
gatekeepers. In the case of corporate bonds, the rating agencies rated
thousands of companies, no one of which controlled any significant
volume of business. No corporate issuer, however large, accounted for
any significant share of Moody's or S&P's revenues. But with the rise
of structured finance, the market became more concentrated. As a
result, the major investment banks acquired considerable power over the
rating agencies, because each of them had ``clout,'' bringing highly
lucrative deals to the agencies on a virtually monthly basis. As the
following chart shows, the top six underwriters controlled over 50
percent of the mortgage-backed securities underwriting market in 2007,
and the top eleven underwriters each had more than 5 percent of the
market and in total controlled roughly 80 percent of this very
lucrative market on whom the rating agencies relied for a majority of
their ratings revenue: \25\
---------------------------------------------------------------------------
\25\ See Ferrell, Bethel, and Hu, supra note 15, at Table 2. For
anecdotal evidence that ratings were changed at the demand of the
investment banks, see Morgenson, supra note 23.
MBS Underwriters in 2007
--------------------------------------------------------------------------------------------------------------------------------------------------------
Proceed Amount +
Rank Book Runner Number of Market Overallotment Sold in U.S.
Offerings Share ($mill)
--------------------------------------------------------------------------------------------------------------------------------------------------------
1....................................................... Lehman Brothers 120 10.80% $100,109
2....................................................... Bear Stearns & Co., Inc. 128 9.90% 91,696
3....................................................... Morgan Stanley 92 8.20% 75,627
4....................................................... JPMorgan 95 7.90% 73,214
5....................................................... Credit Suis109 7.50% 69,503
6....................................................... Bank of America Securities LLC 101 6.80% 62,776
7....................................................... Deutsche Bank AG 85 6.20% 57,337
8....................................................... Royal Bank of Scotland Group 74 5.80% 53,352
9....................................................... Merrill Lynch 81 5.20% 48,407
10...................................................... Goldman Sachs & Co. 60 5.10% 47,696
11...................................................... Citigroup 95 5.00% 46,754
12...................................................... UBS 74 4.30% 39,832
--------------------------------------------------------------------------------------------------------------------------------------------------------
If the rise of structured finance was the first factor that
compromised the credit rating agencies, the second factor was at least
as important and had an even clearer empirical impact. Until the late
1990s, Moody's and Standard & Poor's shared a duopoly over the rating
of U.S. corporate debt. But, over the last decade, a third agency,
Fitch Ratings, grew as the result of a series of mergers and increased
its U.S. market share from 10 percent to approximately a third of the
market. \26\ The rise of Fitch challenged the established duopoly. What
was the result? A Harvard Business School study has found three
significant impacts: (1) the ratings issued by the two dominant rating
agencies shifted significantly in the direction of higher ratings; (2)
the correlation between bond yields and ratings fell, suggesting that
under competitive pressure ratings less reflected the market's own
judgment; and (3) the negative stock market reaction to bond rating
downgrades increased, suggesting that a downgrade now conveyed worse
news because the rated offering was falling to an even lower quality
threshold than before. \27\ Their conclusions are vividly illustrated
by one graph they provide that shows the correlation between grade
inflation and higher competition:
---------------------------------------------------------------------------
\26\ Bo Becker and Todd Milburn, ``Reputation and Competition:
Evidence from the Credit Rating Industry,'' Harvard Business School,
Working Paper No. 09-051 (2008) (http://ssrn.com/abstract =1278150) at
p. 4.
\27\ Id. at 17.
Under high competition, lower ratings declined and investment grade
rations soared. The authors conclude that increased competition may
impair ``the reputational mechanism that underlies the provision of
good quality ratings.'' \28\
---------------------------------------------------------------------------
\28\ Id. at 21.
---------------------------------------------------------------------------
The anecdotal evidence supports a similar conclusion: the major
rating agencies responded to the competitive threat from Fitch by
making their firms ``more client-friendly and focused on market
share.'' \29\ Put simply, the evidence implies that the rapid change
toward a more competitive environment made the competitors not more
faithful to investors, but more dependent on their immediate clients,
the issuers. From the standpoint of investors, agency costs increased.
---------------------------------------------------------------------------
\29\ See ``Ratings Game--As Housing Boomed, Moody's Opened Up,''
The Wall Street Journal, April 11, 2008, at p. A-1.
---------------------------------------------------------------------------
The Responsibility of the SEC
Each of the major investment banks that failed, merged, or
converted into bank holding companies in 2008 had survived prior
recessions, market panics, and repeated turmoil and had long histories
extending back as far as the pre-Civil War era. Yet, each either failed
or was gravely imperiled within the same basically 6 month period
following the collapse of Bear Stearns in March 2008. \30\
---------------------------------------------------------------------------
\30\ For a concise overview of these developments, see Jon
Hilsenrath, Damian Palette, and Aaron Lucchetti, ``Goldman, Morgan
Scrap Wall Street Model, Become Banks in Bid To Ride Out Crisis,'' The
Wall Street Journal, September 22, 2008, at p. A-1 (concluding that
independent investment banks could not survive under current market
conditions and needed closer regulatory supervision to establish
credibility).
---------------------------------------------------------------------------
If their uniform collapse is not alone enough to suggest the
likelihood of regulatory failure, one additional common fact unites
them: each of these five firms voluntarily entered into the SEC's
Consolidated Supervised Entity (``CSE'') Program, which was established
by the SEC in 2004 for only the largest investment banks. \31\ Indeed,
these five investment banks were the only investment banks permitted by
the SEC to enter the CSE program. A key attraction of the CSE Program
was that it permitted its members to escape the SEC's traditional net
capital rule, which placed a maximum ceiling on their debt to equity
ratios, and instead elect into a more relaxed ``alternative net capital
rule'' that contained no similar limitation. \32\ The result was
predictable: all five of these major investment banks increased their
debt-to-equity leverage ratios significantly over the brief two year
period following their entry into the CSE Program, as shown by Figure 1
below: \33\
---------------------------------------------------------------------------
\31\ See Securities Exchange Act Release No. 34-49830 (June 21,
2004), 69 FR 34428 (``Alternative Net Capital Requirements for Broker-
Dealers That Are Part of Consolidated Supervised Entities'').
\32\ The SEC's ``net capital rule,'' which dates back to 1975,
governs the capital adequacy and aggregate indebtedness permitted for
most broker-dealers. See Rule 15c3-1 (``Net Capital Requirements for
Brokers and Dealers''). 17 C.F.R. 240.15c3-1. Under subparagraph
(a)(1)(i) of this rule, aggregate indebtedness is limited to fifteen
times the broker-dealer's net capital; a broker-dealer may elect to be
governed instead by subparagraph (a)(1)(ii) of this rule, which
requires it maintain its net capital at not less than the greater of
$250,000 or two percent of ``aggregate debit items'' as computed under
a special formula that gives ``haircuts'' (i.e., reduces the valuation)
to illiquid securities. Both variants place fixed limits on leverage.
\33\ This chart comes from U.S. Securities and Exchange
Commission, Office of the Inspector General, ``SEC's Oversight of Bear
Stearns and Related Entities: The Consolidated Entity Program''
(`Report No. 446-A, September 25, 2008) (hereinafter ``SEC Inspector
General Report'') at Appendix IX at p. 120.
For example, at the time of its insolvency, Bear Stearns' gross
leverage ratio had hit 33 to 1. \34\
---------------------------------------------------------------------------
\34\ See SEC Inspector General Report at 19.
---------------------------------------------------------------------------
The above chart likely understates the true increase in leverage
because gross leverage (i.e., assets divided by equity) does not show
the increase in off-balance sheet liabilities, as the result of
conduits and liquidity puts. Thus, another measure may better show the
sudden increase in risk. One commonly used metric for banks is the
bank's value at risk (VaR) estimate, which banks report to the SEC in
their annual report on Form 10-K. This measure is intended to show the
risk inherent in their financial portfolios. The chart below shows
``Value at Risk'' for the major underwriters over the interval 2004 to
2007: \35\
---------------------------------------------------------------------------
\35\ See Ferrell, Bethel, and Hu, supra note 15, at Table 8. Value
at risk estimates have proven to be inaccurate predictors of the actual
writedowns experienced by banks. They are cited here not because they
are accurate estimates of risk, but because the percentage increases at
the investment banks was generally extreme. Even Goldman Sachs, which
survived the crisis in better shape than its rivals, saw its VaR
estimate more than double over this period.
Value at Risk, 2004-2007
------------------------------------------------------------------------
2004 2005 2006 2007
Firms ($mil) ($mil) ($mil) ($mil)
------------------------------------------------------------------------
Bank of America................. $44.1 $41.8 $41.3 -
Bear Stearns.................... 14.8 21.4 28.8 $69.3
Citigroup....................... 116.0 93.0 106.0 -
Credit Suisse................... 55.1 66.2 73.0 -
Deutsche Bank................... 89.8 82.7 101.5 -
Goldman Sachs................... 67.0 83.0 119.0 134.0
JPMorgan........................ 78.0 108.0 104.0 -
Lehman Brothers................. 29.6 38.4 54.0 124.0
Merrill Lynch................... 34.0 38.0 52.0 -
Morgan Stanley.................. 94.0 61.0 89.0 83.0
UBS............................. 103.4 124.7 132.8 -
Wachovia........................ 21.0 18.0 30.0 -
------------------------------------------------------------------------
VaR statistics are reported in the 10K or 20F (in the case of foreign
firms) of the respective firms. Note that the firms use different
assumptions in computing their Value of Risk. Some annual reports are
not yet avaialble for 2007.
Between 2004 and 2007, both Bear Stearns and Lehman more than
quadrupled their value at risk estimates, while Merrill Lynch's figure
also increased significantly. Not altogether surprisingly, they were
the banks that failed.
These facts provide some corroboration for an obvious hypothesis:
excessive deregulation by the SEC caused the liquidity crisis that
swept the global markets in 2008. \36\ Still, the problem with this
simple hypothesis is that it may be too simple. Deregulation did
contribute to the 2008 financial crisis, but the SEC's adoption of the
CSE Program in 2004 was not intended to be deregulatory. Rather, the
program was intended to compensate for earlier deregulatory efforts by
Congress that had left the SEC unable to monitor the overall financial
position and risk management practices of the nation's largest
investment banks. Still, even if the 2004 net capital rule changes were
not intended to be deregulatory, they worked out that way in practice.
The ironic bottom line is that the SEC unintentionally deregulated by
introducing an alternative net capital rule that it could not
effectively monitor.
---------------------------------------------------------------------------
\36\ For the bluntest statement of this thesis, see Stephen
Labaton, ``S.E.C. Concedes Oversight Plans Fueled Collapse,'' New York
Times, September 27, 2008, at p. 1. Nonetheless, this analysis is
oversimple. Although SEC Chairman Cox did indeed acknowledge that there
were flaws in the ``Consolidated Supervised Entity'' Program, he did
not concede that it ``fueled'' the collapse or that it represented
deregulation. As discussed below, the SEC probably legitimately
believed that it was gaining regulatory authority from the CSE Program
(but it was wrong).
---------------------------------------------------------------------------
The events leading up to the SEC's decision to relax its net
capital rule for the largest investment banks began in 2002, when the
European Union adopted its Financial Conglomerates Directive. \37\ The
main thrust of the E.U.'s new directive was to require regulatory
supervision at the parent company level of financial conglomerates that
included a regulated financial institution (e.g., a broker-dealer, bank
or insurance company). The E.U.'s entirely reasonable fear was that the
parent company might take actions that could jeopardize the solvency of
the regulated subsidiary. The E.U.'s directive potentially applied to
the major U.S. investment and commercial banks because all did
substantial business in London (and elsewhere in Europe). But the
E.U.'s directive contained an exemption for foreign financial
conglomerates that were regulated by their home countries in a way that
was deemed ``equivalent'' to that envisioned by the directive. For the
major U.S. commercial banks (several of which operated a major broker-
dealer as a subsidiary), this afforded them an easy means of avoiding
group-wide supervision by regulators in Europe, because they were
subject to group-level supervision by U.S. banking regulators.
---------------------------------------------------------------------------
\37\ See Council Directive 2002/87, Financial Conglomerates
Directive, 2002 O.J. (L 35) of the European Parliament and of the
Council of 16 December 2002 on the supplementary supervision of credit
institutions, insurance undertakings and investment firms in a
financial conglomerate and amending Council Directives. For an overview
of this directive and its rationale, see Jorge E. Vinuales, The
International Regulation of Financial Conglomerates: A Case Study of
Equivalence as an Approach to Financial Integration, 37 Cal. W. Int'l
L.J. 1, at 2 (2006).
---------------------------------------------------------------------------
U.S. investment banks had no similar escape hatch, as the SEC had
no similar oversight over their parent companies. Thus, fearful of
hostile regulation by some European regulators, \38\ U.S. investment
banks lobbied the SEC for a system of ``equivalent'' regulation that
would be sufficient to satisfy the terms of the directive and give them
immunity from European oversight. \39\ For the SEC, this offered a
serendipitous opportunity to oversee the operations of investment bank
holding companies, which authority the SEC had sought for some time.
Following the repeal of the Glass-Steagall Act, the SEC had asked
Congress to empower it to monitor investment bank holding companies,
but it had been rebuffed. Thus, the voluntary entry of the holding
companies into the Consolidated Supervised Entity program must have
struck the SEC as a welcome development, and Commission unanimously
approved the program without any partisan disagreement. \40\
---------------------------------------------------------------------------
\38\ Different European regulators appear to have been feared by
different entities. Some commercial banks saw French regulation as
potentially hostile, while U.S. broker-dealers, all largely based in
London, did not want their holding companies to be overseen by the
U.K.'s Financial Services Agency (FSA).
\39\ See Stephen Labaton, ``Agency's '04 Rule Let Banks Pile Up
Debt and Risk,'' New York Times, October 3, 2008, at A-1 (describing
major investment banks as having made an ``urgent plea'' to the SEC in
April, 2004).
\40\ See Securities Exchange Act Release No. 34-49830, supra note
31.
---------------------------------------------------------------------------
But the CSE Program came with an added (and probably unnecessary)
corollary: Firms that entered the CSE Program were permitted to adopt
an alternative and more relaxed net capital rule governing their debt
to net capital ratio. Under the traditional net capital rule, a broker-
dealer was subject to fixed ceilings on its permissible leverage.
Specifically, it either had to (a) maintain aggregate indebtedness at a
level that could not exceed fifteen times net capital, \41\ or (b)
maintain minimum net capital equal to not less than two percent of
``aggregate debit items.'' \42\ For most broker-dealers, this 15 to 1
debt to net capital ratio was the operative limit within which they
needed to remain by a comfortable margin.
---------------------------------------------------------------------------
\41\ See Rule 15c3-1(a)(1)(i)(``Alternative Indebtedness
Standard''), 17 C.F.R. 240.15c3-1(a)(1).
\42\ See Rule 15c3-1(a)(1)(ii)(``Alternative Standard''), 17
C.F.R. 240.15c3-1(a)(1)(ii). This alternative standard is framed in
terms of the greater of $250,000 or 2 percent, but for any investment
bank of any size, 2 percent will be the greater. Although this
alternative standard may sound less restrictive, it was implemented by
a system of ``haircuts'' that wrote down the value of investment assets
to reflect their illiquidity.
---------------------------------------------------------------------------
Why did the SEC allow the major investment banks to elect into an
alternative regime that placed no outer limit on leverage? Most likely,
the Commission was principally motivated by the belief that it was only
emulating the more modern ``Basel II'' standards that the Federal
Reserve Bank and European regulators were then negotiating. To be sure,
the investment banks undoubtedly knew that adoption of Basel II
standards would permit them to increase leverage (and they lobbied hard
for such a change). But, from the SEC's perspective, the goal was to
design the CSE Program to be broadly consistent with the Federal
Reserve's oversight of bank holding companies, and the program even
incorporated the same capital ratio that the Federal Reserve mandated
for bank holding companies. \43\ Still, the Federal Reserve introduced
its Basel II criteria more slowly and gradually, beginning more than a
year later, while the SEC raced in 2004 to introduce a system under
which each investment bank developed its own individualized credit risk
model. Today, some believe that Basel II represents a flawed model even
for commercial banks, while others believe that, whatever its overall
merits, it was particularly ill-suited for investment banks. \44\
---------------------------------------------------------------------------
\43\ See SEC Inspector General Report at 10-11. Under these
standards, a ``well-capitalized'' bank was expected to maintain a 10
percent capital ratio. Id. at 11. Nonetheless, others have argued that
Basel II ``was not designed to be used by investment banks'' and that
the SEC ``ought to have been more careful in moving banks on to the new
rules.'' See ``Mewling and Puking: Bank Regulation,'' The Economist,
October 25, 2008 (U.S. Edition).
\44\ For the view that Basel II excessively deferred to commercial
banks to design their own credit risk models and their increase
leverage, see Daniel K. Tarullo, BANKING ON BASEL: The Future of
International Financial Regulation (2008). Mr. Tarullo has recently
been nominated by President Obama to the Board of Governors of the
Federal Reserve Board. For the alternative view, that Basel II was
uniquely unsuited for investment banks, see ``Mewling and Puking,''
supra note 43.
---------------------------------------------------------------------------
Yet, what the evidence demonstrates most clearly is that the SEC
simply could not implement this model in a fashion that placed any real
restraint on its subject CSE firms. The SEC's Inspector General
examined the failure of Bear Stearns and the SEC's responsibility
therefor and reported that Bear Stearns had remained in compliance with
the CSE Program's rules at all relevant times. \45\ Thus, if Bear
Stearns had not cheated, this implied (as the Inspector General found)
that the CSE Program, itself, had failed. The key question is then what
caused the CSE Program to fail. Here, three largely complementary
hypotheses are plausible. First, the Basel II Accords may be flawed,
either because they rely too heavily on the banks' own self-interested
models of risk or on the highly conflicted ratings of the major credit
rating agencies. \46\ Second, even if Basel II made sense for
commercial banks, it may have been ill-suited for investment banks.
\47\ Third, whatever the merits of Basel II in theory, the SEC may have
simply been incapable of implementing it.
---------------------------------------------------------------------------
\45\ SEC Inspector General Report, 10.
\46\ The most prominent proponent of this view is Professor Daniel
Tarullo. See supra note 44.
\47\ See ``Mewling and Puking,'' supra note 43.
---------------------------------------------------------------------------
Clearly, however, the SEC moved faster and farther to defer to
self-regulation by means of Basel II than did the Federal Reserve. \48\
Clearly also, the SEC's staff was unable to monitor the participating
investment banks closely or to demand specific actions by them. Basel
II's approach to the regulation of capital adequacy at financial
institutions contemplated close monitoring and supervision. Thus, the
Federal Reserve assigns members of its staff to maintain an office
within a regulated bank holding company in order to provide constant
oversight. In the case of the SEC, a team of only three SEC staffers
were assigned to each CSE firm \49\ (and a total of only thirteen
individuals comprised the SEC's Office of Prudential Supervision and
Risk Analysis that oversaw and conducted this monitoring effort). \50\
From the start, it was a mismatch: three SEC staffers to oversee an
investment bank the size of Merrill Lynch, which could easily afford to
hire scores of highly quantitative economists and financial analysts,
implied that the SEC was simply outgunned. \51\
---------------------------------------------------------------------------
\48\ The SEC adopted its CSE program in 2004. The Federal Reserve
only agreed in principle to Basel II in late 2005. See Stavros Gadinis,
The Politics of Competition in International Financial Regulation, 49
Harv. Int'l L. J. 447, 507 n. 192 (2008).
\49\ SEC Inspector General Report at 2.
\50\ Id. Similarly, the Office of CSE Inspectors had only seven
staff. Id.
\51\ Moreover, the process effectively ceased to function well
before the 2008 crisis hit. After SEC Chairman Cox re-organized the CSE
review process in the Spring of 2007, the staff did not thereafter
complete ``a single inspection.'' See Labaton, supra note 39.
---------------------------------------------------------------------------
This mismatch was compounded by the inherently individualized
criteria upon which Basel II relies. Instead of applying a uniform
standard (such as a specific debt to equity ratio) to all financial
institutions, Basel II contemplated that each regulated financial
institution would develop a computer model that would generate risk
estimates for the specific assets held by that institution and that
these estimates would determine the level of capital necessary to
protect that institution from insolvency. Thus, using the Basel II
methodology, the investment bank generates a mathematical model that
crunches historical data to evaluate how risky its portfolio assets
were and how much capital it needed to maintain to protect them.
Necessarily, each model was ad hoc, specifically fitted to that
specific financial institution. But no team of three SEC staffers was
in a position to contest these individualized models or the historical
data used by them. Effectively, the impact of the Basel II methodology
was to shift the balance of power in favor of the management of the
investment bank and to diminish the negotiating position of the SEC's
staff. Whether or not Basel II's criteria were inherently flawed, it
was a sophisticated tool that was beyond the capacity of the SEC's
largely legal staff to administer effectively.
The SEC's Inspector General's Report bears out this critique by
describing a variety of instances surrounding the collapse of Bear
Stearns in which the SEC's staff did not respond to red flags that the
Inspector General, exercising 20/20 hindsight, considered to be
obvious. The Report finds that although the SEC's staff was aware that
Bear Stearns had a heavy and increasing concentration in mortgage
securities, it ``did not make any efforts to limit Bear Stearns
mortgage securities concentration.'' \52\ In its recommendations, the
Report proposed both that the staff become ``more skeptical of CSE
firms' risk models'' and that it ``develop additional stress scenarios
that have not already been contemplated as part of the prudential
regulation process.'' \53\
---------------------------------------------------------------------------
\52\ SEC Inspector General Report at ix.
\53\ SEC Inspector General Report at ix.
---------------------------------------------------------------------------
Unfortunately, the SEC Inspector General Report does not seem
realistic on this score. The SEC's staff cannot really hope to regulate
through gentle persuasion. Unlike a prophylactic rule (such as the
SEC's traditional net capital rule that placed a uniform ceiling on
leverage for all broker-dealers), the identification of ``additional
stress scenarios'' by the SEC's staff does not necessarily lead to
specific actions by the CSE firms; rather, such attempts at persuasion
are more likely to produce an extended dialogue, with the SEC's staff
being confronted with counter-models and interpretations by the
financial institution's managers.
The unfortunate truth is that in an area where financial
institutions have intense interests (such as over the question of their
maximum permissible leverage), a government agency in the U.S. is
unlikely to be able to obtain voluntary compliance. This conclusion is
confirmed by a similar assessment from the individual with perhaps the
most recent experience in this area. Testifying in September, 2008
testimony before the Senate Banking Committee, SEC Chairman Christopher
Cox emphasized the infeasibility of voluntary compliance , expressing
his frustration with attempts to negotiate issues such as leverage and
risk management practices with the CSE firms. In a remarkable statement
for a long-time proponent of deregulation, he testified:
Beyond highlighting the inadequacy of the . . . CSE program's
capital and liquidity requirements, the last six months--during
which the SEC and the Federal Reserve worked collaboratively
with each of the CSE firms . . . --have made abundantly clear
that voluntary regulation doesn't work. \54\
---------------------------------------------------------------------------
\54\ See Testimony of SEC Chairman Christopher Cox before the
Committee on Banking, Housing, and Urban Affairs, United States Senate,
September 23, 2008 (``Testimony Concerning Turmoil in U.S. Credit
Markets: Recent Actions Regarding Government Sponsored Entities,
Investment Banks and Other Financial Institutions''), at p. 4
(available at www.sec.gov) (emphasis added). Chairman Cox has repeated
this theme in a subsequent Op/Ed column in the Washington Post, in
which he argued that ``Reform legislation should steer clear of
voluntary regulation and grant explicit authority where it is needed.''
See Christopher Cox, ``Reinventing A Market Watchdog,'' the Washington
Post, November 4, 2008, at A-17.
His point was that the SEC had no inherent authority to order a CSE
firm to reduce its debt to equity ratio or to keep it in the CSE
Program. \55\ If it objected, a potentially endless regulatory
negotiation might only begin.
---------------------------------------------------------------------------
\55\ Chairman Cox added in the next sentence of his Senate
testimony: ``There is simply no provision in the law authorizes the CSE
Program, or requires investment bank holding companies to compute
capital measures or to maintain liquidity on a consolidated basis, or
to submit to SEC requirements regarding leverage.'' Id. This is true,
but if a CSE firm left the CSE program, it would presumably become
subject to European regulation; thus, the system was not entirely
voluntary and the SEC might have used the threat to expel a non-
compliant CSE firm. The SEC's statements about the degree of control
they had over participants in the CSE Program appear to have been
inconsistent over time and possibly defensively self-serving. But
clearly, the SEC did not achieve voluntary compliance.
---------------------------------------------------------------------------
Ultimately, even if one absolves the SEC of ``selling out'' to the
industry in adopting the CSE Program in 2004, it is still clear at a
minimum that the SEC lacked both the power and the expertise to
restrict leverage by the major investment banks, at least once the
regulatory process began with each bank generating its own risk model.
Motivated by stock market pressure and the incentives of a short-term
oriented executive compensation system, senior management at these
institutions affectively converted the process into self-regulation.
One last factor also drove the rush to increased leverage and may
best explain the apparent willingness of investment banks to relax
their due diligence standards: competitive pressure and the need to
establish a strong market share in a new and expanding market drove the
investment banks to expand recklessly. For the major players in the
asset-backed securitization market, the long-term risk was that they
might be cut off from their source of supply, if loan originators were
acquired by or entered into long-term relationships with their
competitors, particularly the commercial banks. Needing an assured
source of supply, some investment banks (most notably Lehman and
Merrill, Lynch) invested heavily in acquiring loan originators and
related real estate companies, thus in effect vertically integrating.
\56\ In so doing, they assumed even greater risk by increasing their
concentration in real estate and thus their undiversified exposure to a
downturn in that market. This need to stay at least even with one's
competitors best explains the now famous line uttered by Charles
Prince, the then CEO of Citigroup in July, 2007, just as the debt
market was beginning to collapse. Asked by the Financial Times if he
saw a liquidity crisis looming, he answered:
---------------------------------------------------------------------------
\56\ See Terry Pristin, ``Risky Real Estate Deals Helped Doom
Lehman,'' N.Y. Times, September 17, 2008, at C-6 (discussing Lehman's
expensive, multi-billion dollar acquisition of Archstone-Smith);
Gretchen Morgenson, ``How the Thundering Herd Faltered and Fell,'' N.Y.
Times, November 9, 2008, at B4-1 (analyzing Merrill Lynch's failure and
emphasizing its acquisitions of loan originators).
When the music stops, in terms of liquidity, things will get
complicated. But as long as the music is playing, you've got to
get up and dance. We're still dancing. \57\
---------------------------------------------------------------------------
\57\ See Michiyo Nakamoto & David Wighton, ``Citigroup Chief Stays
Bullish on Buy-Outs,'' Financial Times, July 9, 2007, available at
http://www.ft.com/cms/s/0/80e2987a-2e50-11dc-821c-0000779fd2ac.html
In short, competition among the major investment banks can
periodically produce a mad momentum that sometimes leads to a lemmings-
like race over the cliff. \58\ This in essence had happened in the
period just prior to the 2000 dot.com bubble, and again during the
accounting scandals of 2001-2002, and this process repeated itself
during the subprime mortgage debacle. Once the market becomes hot, the
threat of civil liability--either to the SEC or to private plaintiffs
in securities class actions--seems only weakly to constrain this
momentum. Rationalizations are always available: ``real estate prices
never fall;'' ``the credit rating agencies gave this deal a `Triple A'
rating,'' etc. Explosive growth and a decline in professional standards
often go hand in hand. Here, after 2000, due diligence standards appear
to have been relaxed, even as the threat of civil liability in private
securities litigation was growing. \59\
---------------------------------------------------------------------------
\58\ Although a commercial bank, Citigroup was no exception this
race, impelled by the high fee income it involved. From 2003 to 2005,
``Citigroup more than tripled its issuing of C.D.O.s to more than $30
billion from $6.28 billion.'' See Eric Dash and Julie Creswell,
``Citigroup Pays for a Rush to Risk'' New York Times, November 22,
2008, at 1, 34. In 2005 alone, the New York Times estimates that
Citigroup received over $500 million in fee income from these C.D.O.
transactions. From being the sixth largest issuer of C.D.O.s in 2003,
it rose to being the largest C.D.O. issuer worldwide by 2007, issuing
in that year some $49.3 billion out of a worldwide total of $442.3
billion (or slightly over 11 percent of the world volume). Id. at 35.
What motivated this extreme risk-taking? Certain of the managers
running Citigroup's securitization business received compensation as
high as $34 million per year (even though they were not among the most
senior officers of the bank). Id. at 34. This is consistent with the
earlier diagnosis that equity compensation inclines management to
accept higher and arguably excessive risk. At the highest level of
Citigroup's management, the New York Times reports that the primary
concern was ``that Citigroup was falling behind rivals like Morgan
Stanley and Goldman.'' Id. at 34 (discussing Robert Rubin and Charles
Prince's concerns). Competitive pressure is, of course, enforced by the
stock market and Wall Street's short-term system of bonus compensation.
The irony then is that a rational strategy of deleveraging cannot be
pursued by making boards and managements more sensitive to shareholder
desires.
\59\ From 1996 to 1999, the settlements in securities class
actions totaled only $1.7 billion; thereafter, aggregate settlements
rose exponentially, hitting a peak of $17.1 billion in 2006 alone. See
Laura Simmons & Ellen Ryan, ``Securities Class Action Settlements:
2006, Review and Analysis'' (Cornerstone Research 2006) at 1. This
decline of due diligence practices as liability correspondingly
increased seems paradoxical, but may suggest that at least private
civil liability does not effectively deter issuers or underwriters.
---------------------------------------------------------------------------
As an explanation for an erosion in professional standards,
competitive pressure applies with particular force to those investment
banks that saw asset-back securitizations as the core of their future
business model. In 2002, a critical milestone was reached, as in that
year the total amount of debt securities issued in asset-backed
securitizations equaled (and then exceeded in subsequent years) the
total amount of debt securities issued by public corporations. \60\
Debt securitizations were not only becoming the leading business of
Wall Street, as a global market of debt purchasers was ready to rely on
investment grade ratings from the major credit rating agencies, but
they were particularly important for the independent investment banks
in the CSE Program.
---------------------------------------------------------------------------
\60\ For a chart showing the growth of asset-backed securities in
relation to conventional corporate debt issuances over recent years,
see J. Coffee, J. Seligman, and H. Sale, SECURITIES REGULATION: Case
and Materials (10th ed. 2006) at p. 10.
---------------------------------------------------------------------------
Although all underwriters anticipated high rates of return from
securitizations, the independent underwriters had gradually been
squeezed out of their traditional line of business--underwriting
corporate securities--in the wake of the step-by-step repeal of the
Glass-Steagall Act. Beginning well before the formal repeal of that Act
in 1999, the major commercial banks had been permitted to underwrite
corporate debt securities and had increasingly exploited their larger
scale and synergistic ability to offer both bank loans and underwriting
services to gain an increasing share of this underwriting market.
Especially for the smaller investment banks (e.g., Bear Stearns and
Lehman), the future lay in new lines of business, where, as nimble and
adaptive competitors, they could steal a march on the larger and slower
commercial banks. To a degree, both did, and Merrill eagerly sought to
follow in their wake. \61\ To stake out a dominant position, the CEOs
of these firms adopted a ``Damn-the-torpedoes-full-speed-ahead''
approach that led them to make extremely risky acquisitions. Their
common goal was to assure themselves a continuing source of supply of
subprime mortgages to securitize, but in pursuit of this goal, both
Merrill Lynch and Lehman made risky acquisitions, in effect vertically
integrating into the mortgage loan origination field. These decisions,
plus their willingness to acquire mortgage portfolios well in advance
of the expected securitization transaction, left them undiversified and
exposed to large writedowns when the real estate market soured.
---------------------------------------------------------------------------
\61\ For a detailed description of Merrill, Lynch's late entry
into the asset-backed securitization field and its sometimes frenzied
attempt to catch up with Lehman by acquiring originators of mortgage
loans, see Gretchen Morgenson, ``How the Thundering Herd Faltered and
Fell,'' New York Times, November 9, 2008, at BU-1. Merrill eventually
acquired an inventory of $71 billion in risky mortgages, in part
through acquisitions of loan originators. By mid-2008, an initial
writedown of $7.9 billion forced the resignation of its CEO. As
discussed in this New York Times article, loan originators dealing with
Merrill believed it did not accurately understand the risks of their
field. For Lehman's similar approach to acquisitions of loan
originators, see text and note, supra, at note 56.
---------------------------------------------------------------------------
Regulatory Modernization: What Should Be Done?
An Overview of Recent Developments
Financial regulation in the major capital markets today follows one
of three basic organizational models:
The Functional/Institutional Model: In 2008, before the financial
crisis truly broke, the Treasury Department released a major study of
financial regulation in the United States. \62\ This document (known as
the ``Blueprint'') correctly characterized the United States as having
a ``current system of functional regulation, which maintains separate
regulatory agencies across segregated functional lines of financial
services, such as banking, insurance, securities, and futures.'' \63\
Unfortunately, even this critical assessment may understate the
dimensions of this problem of fragmented authority. In fact, the U.S.
falls considerably short of even a ``functional'' regulatory model. By
design, ``functional'' regulation seeks to subject similar activities
to regulation by the same regulator. Its premise is that no one
regulator can have, or easily develop, expertise in regulating all
aspects of financial services. Thus, the securities regulator
understands securities, while the insurance regulator has expertise
with respect to the very different world of insurance. In the Gramm-
Leach-Bliley Act of 1999 (``GLBA''), which essentially repealed the
Glass-Steagall Act, Congress endorsed such a system of functional
regulation. \64\
---------------------------------------------------------------------------
\62\ The Department of the Treasury, Blueprint for Modernized
Financial Regulatory Structure (2008) (hereinafter, ``Blueprint'').
\63\ Id. at 4 and 27.
\64\ The Conference Report to the Gramm-Leach-Bliley Act clearly
states this: Both the House and Senate bills generally adhere to the
principle of functional regulation, which holds that similar activities
should be regulated by the same regulator. Different regulators have
expertise at supervising different activities. It is inefficient and
impractical to expect a regulator to have or develop expertise in
regulating all aspects of financial services. H.R. Rep. No. 106-434, at
157 (1999), reprinted in 1999 U.S.C.C.A.N. 1252.
---------------------------------------------------------------------------
Nonetheless, the reality is that the United States actually has a
hybrid system of functional and institutional regulation. \65\ The
latter approach looks not to functional activity, but to institutional
type. Institutional regulation is seldom the product of deliberate
design, but rather of historical contingency, piecemeal reform, and
gradual evolution.
---------------------------------------------------------------------------
\65\ For this same assessment, see Heidi Mandanis Schooner &
Michael Taylor, United Kingdom and United States Responses to the
Regulatory Challenges of Modern Financial Markets, 38 Tex. Int'l L. J.
317, 328 (2003).
---------------------------------------------------------------------------
To illustrate this difference between functional and institutional
regulation, let us hypothesize that, under a truly functional system,
the securities regulator would have jurisdiction over all sales of
securities, regardless of the type of institution selling the security.
Conversely, let us assume that under an institutional system,
jurisdiction over sales would be allocated according to the type of
institution doing the selling. Against that backdrop, what do we
observe today about the allocation of jurisdiction? Revealingly, under
a key compromise in GLBA, the SEC did not receive general authority to
oversee or enforce the securities laws with respect to the sale of
government securities by a bank. \66\ Instead, banking regulators
retained that authority. Similarly, the drafters of the GLBA carefully
crafted the definitions of ``broker'' and ``dealer'' in the Securities
Exchange Act of 1934 to leave significant bank securities activities
under the oversight of bank regulators and not the SEC. \67\
Predictably, even in the relatively brief time since the passage of
GLBA in 1999, the SEC and bank regulators have engaged in a continuing
turf war over the scope of the exemptions accorded to banks from the
definition of ``broker'' and ``dealer.'' \68\
---------------------------------------------------------------------------
\66\ See 15 U.S.C. 78o-5(a)(1)(B), 15 U.S.C. 78(c)(a)(34)(G),
and 15 U.S.C. 78o-5(g)(2).
\67\ See 15 U.S.C. 78(c)(a)(4),(5).
\68\ See Kathleen Day, Regulators Battle Over Banks: 3 Agencies
Say SEC Rules Overstep Securities-Trading Law, Wash. Post, July 3,
2001, at E3. Eventually, the SEC backed down in this particular
skirmish and modified its original position. See Securities Exch. Act
Release No. 34-44570 (July 18, 2001) and Securities Exchange Age
Release No. 34-44291, 66 Fed. Reg. 27760 (2001).
---------------------------------------------------------------------------
None of this should be surprising. The status quo is hard to
change, and regulatory bodies do not surrender jurisdiction easily. As
a result, the regulatory body historically established to regulate
banks will predictably succeed in retaining much of its authority over
banks, even when banks are engaged in securities activities that from a
functional perspective should belong to the securities regulator.
``True'' functional regulation would also assign similar activities
to one regulator, rather than divide them between regulators based on
only nominal differences in the description of the product or the legal
status of the institution. Yet, in the case of banking regulation,
three different federal regulators oversee banks: the Office of the
Controller of the Currency (``OCC'') supervises national banks; the
Federal Reserve Board (``FRB'') oversees state-chartered banks that are
members of the Federal Reserve System and the Federal Deposit Insurance
Corporation (``FDIC'') supervises state-chartered banks that are not
members of the Federal Reserve System but are federally insured. \69\
Balkanization does not stop there. The line between ``banks,'' with
their three different regulators at the federal level, and ``thrifts,''
which the Office of Thrift Supervision (``OTS'') regulates, is again
more formalistic than functional and reflects a political compromise
more than a difference in activities.
---------------------------------------------------------------------------
\69\ This is all well described in the Blueprint. See Blueprint,
supra note 62, at 31-41.
---------------------------------------------------------------------------
Turning to securities regulation, one encounters an even stranger
anomaly: the United States has one agency (the SEC) to regulate
securities and another (the Commodities Future Trading Commission
(CFTC)) to regulate futures. The world of derivatives is thereby
divided between the two, with the SEC having jurisdiction over options,
while the CFTC has jurisdiction over most other derivatives. No other
nation assigns futures and securities regulation to different
regulators. For a time, the SEC and CFTC both asserted jurisdiction
over a third category of derivatives--swaps--but in 2000 Congress
resolved this dispute by placing their regulation largely beyond the
reach of both agencies. Finally, some major financial sectors (for
example, insurance and hedge funds) simply have no federal regulator.
By any standard, the United States thus falls well short of a true
system of functional regulation, because deregulation has placed much
financial activity beyond the reach of any federal regulator.
Sensibly, the Blueprint proposes to rationalize this patchwork-
quilt structure of fragmented authority through the merger and
consolidation of agencies. Specifically, it proposes both a merger of
the SEC and CFTC and a merger of the OCC and the OTS. Alas, such
mergers are rarely politically feasible, and to date, no commentator
(to our knowledge) has predicted that these proposed mergers will
actually occur.
Thus, although the Blueprint proposes that we move beyond
functional regulation, the reality is that we have not yet approached
even a system of functional regulation, as our existing financial
regulatory structure is organized at least as much by institutional
category as by functional activity. Disdaining a merely ``functional''
reorganization under which banking, insurance, and securities would
each be governed by their own federal regulator, the Blueprint instead
envisions a far more comprehensive consolidation of all these
specialized regulators. Why? In its view, the problems with functional
regulation are considerable:
A functional approach to regulation exhibits several inadequacies,
the most significant being the fact that no single regulator possesses
all the information and authority necessary to monitor systemic risk,
or the potential that events associated with financial institutions may
trigger broad dislocation or a series of defaults that affect the
financial system so significantly that the real economy is adversely
affected. \70\
---------------------------------------------------------------------------
\70\ Blueprint, supra note 62, at 4.
---------------------------------------------------------------------------
But beyond these concerns about systemic risk, the architects of
the Blueprint were motivated by a deeper anxiety: regulatory reform is
necessary to maintain the capital market competitiveness of the United
States. \71\ In short, the Blueprint is designed around two objectives:
(1) the need to better address systemic risk and the possibility of a
cascading series of defaults, and (2) the need to enhance capital
market competitiveness. As discussed later, the first concern is
legitimate, but the second involves a more dubious logic.
---------------------------------------------------------------------------
\71\ In particular, the Blueprint hypothesizes that the U.K. has
enhanced its own competitiveness by regulatory reforms, adopted in
2000, that are principles-based and rely on self regulation for their
implementation. Id. at 3.
---------------------------------------------------------------------------
The Consolidated Financial Services Regulator: A clear trend is
today evident towards the unification of supervisory responsibilities
for the regulation of banks, securities markets and insurance. \72\
Beginning in Scandinavia in the late 1980s, \73\ this trend has
recently led the United Kingdom, Japan, Korea, Germany and much of
Eastern Europe to move to a single regulator model. \74\ Although there
are now a number of precedents, the U.K. experience stands out as the
most influential. It was the first major international market center to
move to a unified regulator model, \75\ and the Financial Services and
Markets Act, adopted in 2000, went significantly beyond earlier
precedents towards a ``nearly universal regulator.'' \76\ The Blueprint
focuses on the U.K.'s experience because it believes that the U.K.'s
adoption of a consolidated regulatory structure ``enhanced the
competitiveness of the U.K. economy.'' \77\
---------------------------------------------------------------------------
\72\ For recent overviews, see Ellis Ferran, Symposium: Do
Financial Supermarkets Need Super-Regulators? Examining the United
Kingdom's Experience in Adopting the Single Financial Regulator Model,
28 Brook. J. Int'l L. 257, 257-59 (2003); Jerry W. Markham, A
Comparative Analysis of Consolidated and Functional Regulation: Super
Regulator: A Comparative Analysis of Securities and Derivative
Regulation in the United States, the United Kingdom, and Japan, 28
Brook. J. Int'l L. 319, 319-20 (2003); Giorgio Di Giorgio & Carmine D.
Noia, Financial Market Regulation and Supervision: How Many Peaks for
the Euro Area?, 28 Brook. J. Int'l L. 463, 469-78 (2003).
\73\ Norway moved to an integrated regulatory agency in 1986,
followed by Denmark in 1988, and Sweden in 1991. See D. Giorgio & D.
Noia, supra note 72, at 469-478.
\74\ See Bryan D. Stirewalt & Gary A. Gegenheimer, Consolidated
Supervision of Banking Groups in the Former Soviet Republics: A
Comparative Examination of the Emerging Trend in Emerging Markets, 23
Ann. Rev. Banking & Fin. L. 533, 548-49 (2004). As discussed later, in
some countries (most notably Japan), the change seems more one of form
than of substance, with little in fact changing. See Markham, supra
note 72, at 383-393, 396.
\75\ See Ferran, supra note 72, at 258.
\76\ See Schooner & Taylor, supra note 65, at 329. Schooner and
Taylor also observe that the precursors to the U.K.'s centralized
regulator, which were mainly in Scandinavia, had a ``predominantly
prudential focus.'' Id. at 331. That is, the unified new regulator was
more a guardian of ``safety and soundness'' and less oriented toward
consumer protection.
\77\ Blueprint, supra note 62 at 3.
---------------------------------------------------------------------------
Yet it is unclear whether the U.K.'s recent reforms provide a
legitimate prototype for the Blueprint's proposals. Here, the Blueprint
may have doctored its history. By most accounts, the U.K.'s adoption of
a single regulator model was ``driven by country-specific factors,''
\78\ including the dismal failure of a prior regulatory system that
relied heavily on self-regulatory bodies but became a political
liability because of its inability to cope with a succession of serious
scandals. Ironically, the financial history of the U.K. in the 1990s
parallels that of the United States over the last decade. On the
banking side, the U.K. experienced two major banking failures--the Bank
of Credit and Commerce International (``BCCI'') in 1991 and Barings in
1995. Each prompted an official inquiry that found lax supervision was
at least a partial cause. \79\
---------------------------------------------------------------------------
\78\ Ferran, supra note 72, at 259.
\79\ Id. at 261-262.
---------------------------------------------------------------------------
Securities regulation in the U.K. came under even sharper criticism
during the 1990s because of a series of financial scandals that were
generally attributed to an ``excessively fragmented regulatory
infrastructure.'' \80\ Under the then applicable law (the Financial
Services Act of 1986), most regulatory powers were delegated to the
Securities and Investments Board (SIB), which was a private body
financed through a levy on market participants. However, the SIB did
not itself directly regulate. Rather, it ``set the overall framework of
regulation,'' but delegated actual authority to second tier regulators,
which consisted primarily of self-regulatory organizations (SROs). \81\
Persistent criticism focused on the inability or unwillingness of these
SROs to protect consumers from fraud and misconduct. \82\ Ultimately,
the then chairman of the SIB, the most important of the SROs,
acknowledged that self-regulation had failed in the U.K. and seemed
unable to restore investor confidence. \83\ This acknowledgement set
the stage for reform, and when a new Labour Government came into power
at the end of the decade, one of its first major legislative acts (as
it had promised in its election campaign) was to dismantle the former
structure of SROs and replace it with a new and more powerful body, the
Financial Services Authority (FSA).
---------------------------------------------------------------------------
\80\ Id. at 265.
\81\ Id. at 266. The most important of these were the Securities
and Futures Authority (SFA), the Investment Managers' Regulatory
Organization (IMRO), and the Personal Investment Authority (PIA).
\82\ Two scandals in particular stood out: the Robert Maxwell
affair in which a prominent financier effectively embezzled the pension
funds of his companies and a ``pension mis-selling'' controversy in
which highly risky financial products were inappropriately sold to
pension funds without adequate supervision or disclosure. Id. at 267-
268.
\83\ Id. at 268.
---------------------------------------------------------------------------
Despite the Blueprint's enthusiasm for the U.K.'s model, the
structure that the Blueprint proposes for the U.S. more closely
resembles the former U.K. system than the current one. Under the
Blueprint's proposals, the securities regulator would be restricted to
adopting general ``principles-based'' policies, which would be
implemented and enforced by SROs. \84\ Ironically, the Blueprint relies
on the U.K. experience to endorse essentially the model that the U.K.
concluded had failed.
---------------------------------------------------------------------------
\84\ See infra notes--and accompanying text.
---------------------------------------------------------------------------
The ``Twin Peaks'' Model: As the Blueprint recognizes, not all
recent reforms have followed the U.K. model of a universal regulator.
Some nations--most notably Australia and the Netherlands--instead have
followed a ``twin peaks'' model that places responsibility for the
``prudential regulation of relevant financial institutions'' in one
agency and supervision of ``business conduct and consumer protection''
in another. \85\ The term ``twin peaks'' derives from the work of
Michael Taylor, a British academic and former Bank of England official.
In 1995, just before regulatory reform became a hot political issue in
the U.K., he argued that financial regulation had two separate basic
aims (or ``twin peaks''): (1) ``to ensure the soundness of the
financial system,'' and (2) ``to protect consumers from unscrupulous
operators.'' \86\ Taylor's work was original less in its proposal to
separate ``prudential'' regulation from ``business conduct'' regulation
than in its insistence upon the need to consolidate ``responsibility
for the financial soundness of all major financial institutions in a
single agency.'' \87\ Taylor apparently feared that if the Bank of
England remained responsible for the prudential supervision of banks,
its independence in setting interest rates might be compromised by its
fear that raising interest rates would cause bank failures for which it
would be blamed. In part for this reason, the eventual legislation
shifted responsibility for bank supervision from the Bank of England to
the FSA.
---------------------------------------------------------------------------
\85\ Blueprint, supra note 62, at 3. For a recent discussion of
the Australian reorganization, which began in 1996 (and thus preceded
the U.K.), see Schooner & Taylor, supra note65, at 340-341. The
Australian Securities and Investments Commission (ASIC) is the
``consumer protection'' agency under this ``twin peaks'' approach, and
the Australian Prudential Regulatory Authority (APRA) supervises bank
``safety and soundness.'' Still, the ``twin peaks'' model was not fully
accepted in Australia as ASIC, the securities regulator, does retain
supervisory jurisdiction over the ``financial soundness'' of investment
banks. Thus, some element of functional regulation remains.
\86\ Michael Taylor, Twin Peaks: A Regulatory Structure for the
New Century i (Centre for the Study of Financial Institutions 1995).
For a brief review of Taylor's work, see Cynthia Crawford Lichtenstein,
The Fed's New Model of Supervision for ``Large Complex Banking
Organizations'': Coordinated Risk-Based Supervision of Financial
Multinationals for International Financial Stability, 18 Transnat'l
Law. 283, 295-296 (2005).
\87\ Lichtenstein, supra note 86, at 295; Taylor, supra note 86,
at 4.
---------------------------------------------------------------------------
The Blueprint, itself, preferred a ``twin peaks'' model, and that
model is far more compatible with the U.S.'s current institutional
structure for financial regulation. But beyond these obvious points,
the best argument for a ``twin peaks'' model involves conflict of
interests and the differing culture of banks and securities regulators.
It approaches the self-evident to note that a conflict exists between
the consumer protection role of a universal regulator and its role as a
``prudential'' regulator intent on protecting the safety and soundness
of the financial institution. The goal of consumer protection is most
obviously advanced through deterrence and financial sanctions, but
these can deplete assets and ultimately threaten bank solvency. When
only modest financial penalties are used, this conflict may sound more
theoretical than real. But, the U.S. is distinctive in the severity of
the penalties it imposes on financial institutions. In recent years,
the SEC has imposed restitution and penalties exceeding $3 billion
annually, and private plaintiffs received a record $17 billion in
securities class action settlements in 2006. \88\ Over a recent ten
year period, some 2,400 securities class actions were filed and
resulted in settlements of over $27 billion, with much of this cost (as
in the Enron and WorldCom cases) being borne by investment banks. \89\
If one agency were seeking both to protect consumers and guard the
solvency of major financial institutions, it would face a difficult
balancing act to achieve deterrence without threatening bank solvency,
and it would risk a skeptical public concluding that it had been
``captured'' by its regulated firms.
---------------------------------------------------------------------------
\88\ See Coffee, Law and the Market: The Impact of Enforcement,
156 U. of Pa. L. Rev. 299 (2007) (discussing average annual SEC
penalties and class action settlements).
\89\ See Richard Booth, The End of the Securities Fraud Class
Action as We Know It, 4 Berkeley Bus. L. J. 1, at 3 (2007).
---------------------------------------------------------------------------
Even in jurisdictions adopting the universal regulator model, the
need to contemporaneously strengthen enforcement has been part of the
reform package. Although the 2000 legislation in the U.K. did not adopt
the ``twin peaks'' format, it did significantly strengthen the consumer
protection role of its centralized regulator. The U.K.'s Financial
Services and Markets Act, enacted in 2000, sets out four statutory
objectives, with the final objective being the ``reduction of financial
crime.'' \90\ According to Heidi Schooner and Michael Taylor, this
represented ``a major extension of the FSA's powers compared to the
agencies it replaced,'' \91\ and it reflected a political response to
the experience of weak enforcement by self-regulatory bodies, which had
led to the creation of the FSA. \92\ With probably unintended irony,
Schooner and Taylor described this new statutory objective of reducing
``financial crime'' as the ``one aspect of U.K. regulatory reform in
which its proponents seem to have drawn direct inspiration from U.S.
law and practice.'' \93\ Conspicuously, the Blueprint ignores that
``modernizing'' financial regulation in other countries has generally
meant strengthening enforcement.
---------------------------------------------------------------------------
\90\ See Financial Services and Markets Act, 2000, c. 8, pt. 1, 6,
http://www.opsi.gov.uk/ACTS/acts2000/pdf/ukpga_20000008_en.pdf
\91\ See Schooner & Taylor, supra note 65, at 335.
\92\ Id.
\93\ Id. at 335-36.
---------------------------------------------------------------------------
A Preliminary Evaluation: Three preliminary conclusions merit
emphasis:
First, whether the existing financial regulatory structure in the
United States is considered ``institutional'' or ``functional'' in
design, its leading deficiency seems evident: it invites regulatory
arbitrage. Financial institutions position themselves to fall within
the jurisdiction of the most accommodating regulator, and investment
banks design new financial products so as to encounter the least
regulatory oversight. Such arbitrage can be defended as desirable if
one believes that regulators inherently overregulate, but not if one
believes increased systemic risk is a valid concern (as the Blueprint
appears to believe).
Second, the Blueprint's history of recent regulatory reform
involves an element of historical fiction. The 2000 legislation in the
U.K., which created the FSA as a nearly universal regulator, was not an
attempt to introduce self-regulation by SROs, as the Blueprint seems to
assume, but a sharp reaction by a Labour Government to the failures of
self-regulation. Similarly, Japan's slow, back-and-forth movement in
the direction of a single regulator seems to have been motivated by an
unending series of scandals and a desire to give its regulator at least
the appearance of being less industry dominated. \94\
---------------------------------------------------------------------------
\94\ Japan has a history and a regulatory culture of economic
management of its financial institutions through regulatory bodies that
is entirely distinct from that of Europe or the United States. Although
it has recently created a Financial Services Agency, observers contend
that it remains committed to its traditional system of bureaucratic
regulation that supports its large banks and discourages foreign
competition. See Markham, supra note 72, at 383-92, 396. Nonetheless,
scandals have been the primary force driving institutional change there
too, and Japan's FSA was created at least in part because Japan's
Ministry of Finance (MOF) had become embarrassed by recurrent scandals.
---------------------------------------------------------------------------
Third, the debate between the ``universal'' regulator and the
``twin peaks'' alternative should not obscure the fact that both are
``superregulators'' that have moved beyond ``functional'' regulation on
the premise that, as the lines between banks, securities dealers, and
insurers blur, so regulators should similarly converge. That idea will
and should remain at the heart of the U.S. debate, even after many of
the Blueprint's proposals are forgotten.
Defining the Roles of the ``Twin Peaks'' (Systemic Risk Regulator and
Consumer Protector)--Who Should Do What?
The foregoing discussion has suggested why the SEC would not be an
effective risk regulator. It has neither the specialized competence nor
the organizational culture for the role. Its comparative advantage is
enforcement, and thus its focus should be on transparency and consumer
protection. Some also argue that ``single purpose'' agencies, such as
the SEC, are more subject to regulatory capture than are broader or
``general purpose'' agencies. \95\ To the extent that the Federal
Reserve would have responsibility for all large financial institutions
and would be expected to treat monitoring their capital adequacy and
risk management practices as among its primary responsibilities, it
does seem less subject to capture, because any failure would have high
visibility and it would bear the blame. Still, this issue is largely
academic because the SEC no longer has responsibility over any
investment banks of substantial size.
---------------------------------------------------------------------------
\95\ See Jonathan Macey, Organizational Design and Political
Control of Administrative Agencies, 8 J. Law, Economics, and
Organization 93 (1992). It can, of course, be argued which agency is
more ``single purpose'' (the SEC or the Federal Reserve), but the
latter does deal with a broader class of institutions in terms of their
capital adequacy.
---------------------------------------------------------------------------
The real issue then is defining the relationships between the two
peaks so that neither overwhelms the other.
The Systemic Risk Regulator (SRR): Systemic risk is most easily
defined as the risk of an inter-connected financial breakdown in the
financial system--much like the proverbial chain of falling dominoes.
The closely linked insolvencies of Lehman, AIG, Fannie Mae and Freddie
Mac in the Fall of 2008 present a paradigm case. Were they not bailed
out, other financial institutions were likely to have also failed. The
key idea here is not that one financial institution is too big to fail,
but rather that some institutions are too interconnected to permit any
of them to fail, because they will drag the others down.
What should a system risk regulator be authorized to do? Among the
obvious powers that it should have are the following:
a. Authority To Limit the Leverage of Financial Institutions and
Prescribe Mandatory Capital Adequacy Standards. This authority would
empower the SRR to prescribe minimum levels of capital and ceilings on
leverage for all categories of financial institutions, including banks,
insurance companies, hedge funds, money market funds, pension plans,
and quasi-financial institutions (such as, for example, G.E. Capital).
The standards would not need to be identical for all institutions and
should be risk adjusted. The SRS should be authorized to require
reductions in debt to equity ratios below existing levels, to consider
off-balance sheet liabilities (including those of partially owned
subsidiaries and also contractual agreements to repurchase or
guarantee) in computing these tests and ratios (even if generally
accepted accounting principles would not require their inclusion).
The SRR would focus its monitoring on the largest institutions in
each financial class, leaving small institutions to be regulated and
monitored by their primary regulator. For example, the SEC might
require all hedge funds to register with it under the Investment
Advisers Act of 1940, but hedge funds with a defined level of assets
(say, $25 billion in assets) would be subject to the additional and
overriding authority of the SSR.
b. Authority To Approve, Restrict and Regulate Trading in New
Financial Products. By now, it has escaped no one's attention that one
particular class of over-the-counter derivative (the credit default
swap) grew exponentially over the last decade and was outside the
jurisdiction of any regulatory agency. This was not accidental, as the
Commodities Futures Modernization Act of 2000 deliberately placed over-
the-counter derivatives beyond the general jurisdiction of both the SEC
and the CFTC. The SRR would be responsible for monitoring the growth of
new financial products and would be authorized to regulate such
practices as the collateral or margin that counter-parties were
required to post. Arguably, the SRR should be authorized to limit those
eligible to trade such instruments and could bar or restrict the
purchase of ``naked'' credit default swaps (although the possession of
this authority would not mean that the SRR would have to exercise it,
unless it saw an emergency developing).
c. Authority To Mandate Clearing Houses. Securities and options
exchanges uniformly employ clearing houses to eliminate or mitigate
credit risk. In contrast, when an investor trades in an over-the-
counter derivative, it must accept both market risk (the risk that the
investment will sour or price levels will change adversely) and credit
risk (the risk that the counterparty will be unable to perform). Credit
risk is the factor that necessitated the bailout of AIG, as its failure
could have potentially led to a cascade of failures by other financial
institutions if it defaulted on its swaps. Use of the clearing house
should eliminate the need to bail out a future AIG because its
responsibilities would fall on the clearing house to assume and the
clearing house would monitor and limit the risk that its members
assumed.
At present, several clearinghouses are in the process of
development in the United States and Europe. The SRR would be the
obvious body to oversee such clearing houses (and indeed the Federal
Reserve was already instrumental in their formation). Otherwise, some
clearing houses are likely to be formed under the SEC's supervision and
some under the CFTC's, thus again permitting regulatory arbitrage to
develop.
A final and complex question is whether competing clearing houses
are desirable or whether they should be combined into a single
centralized clearing house. This issue could also be given to the SRR.
d. Authority To Mandate Writedowns for Risky Assets. A real estate
bubble was the starting point for the 2008 crisis. When any class of
assets appreciates meteorically, the danger arises that on the eventual
collapse in that overvalued market, the equity of the financial
institution will be wiped out (or at the least so eroded as to create a
crisis in investor confidence that denies that institution necessary
financing). This tendency was palpably evident in the failure of Bear
Stearns, Lehman, Fannie Mae and Freddie Mac. If the SRR regulator
relies only on debt/equity ratios to protect capital adequacy, they
will do little good and possibly provide only illusory protections. Any
financial institution that is forced to writedown its investment in
overpriced mortgage and real estate assets by 50 percent will
necessarily breach mandated debt to equity ratios. The best answer to
this problem is to authorize the SRR to take a proactive and
countercyclical stance by requiring writedowns in risky asset classes
(at least for regulatory purposes) prior to the typically much later
point at which accountants will require such a writedown.
Candidly, it is an open question whether the SRS, the Federal
Reserve, or any banking regulator would have the courage and political
will to order such a writedown (or impose similar restraints on further
acquisitions of such assets) while the bubble was still expanding. But
Congress should at least arm its regulators with sufficient power and
direct them to use it with vigor.
e. Authority To Intervene To Prevent and Avert Liquidity Crises.
Financial institutions often face a mismatch between their assets and
liabilities. They may invest in illiquid assets or make long-term
loans, but their liabilities consist of short-term debt (such as
commercial paper). Thus, regulating leverage ratios is not alone
adequate to avoid a financial crisis, because the institution may
suddenly experience a ``run'' (as its depositors flee) or be unable to
roll over its commercial paper or other short-term debt. This problem
is not unique to banks and can be encountered by hedge funds and
private equity funds (as the Long Term Capital Management crisis
showed). The SRR thus needs the authority to monitor liquidity problems
at large financial institutions and direct institutions in specific
cases to address such imbalances (either by selling assets, raising
capital, or not relying on short-term debt).
From the foregoing description, it should be obvious that the only
existing agency in a position to take on this assignment and act as an
SRR is the Federal Reserve Board. But it is less politically
accountable than most other federal agencies, and this could give rise
to some problems discussed below.
The Consumer Protection and Transparency Agency: The creation of an
SSR would change little at the major Federal agencies having
responsibilities for investor protection. Although it might be
desirable to merge the SEC and the CFTC, this is not essential. Because
no momentum has yet developed for such a merger, I will not discuss it
further at this time.
Currently, there are over 5,000 broker-dealers registered with the
SEC. They would remain so registered, and the SRR would concern itself
only with those few whose potential insolvency could destabilize the
markets. The focus of the SEC's surveillance of broker-dealers is on
consumer protection and market efficiency, and this would not be within
the expertise of the Federal Reserve or any other potential SRR.
The SEC is also an experienced enforcement agency, while the
Federal Reserve has little, if any, experience in this area. Further,
the SEC understands disclosure issues and is a champion of
transparency, whereas banking regulators start from the unstated
premise that disclosures of risks or problems at a financial
institution is undesirable because it might provoke a ``run'' on the
bank. The SEC and the Controller of the Currency have long disagreed
about what banks should disclose in the Management Discussion and
Analysis that banks file with the SEC. Necessarily, this tension will
continue.
Resolving the Conflicts: The SEC and the PCAOB have continued to
favor ``mark to market'' accounting, while major banks have sought
relief from the write-downs that it necessitates. Suppose then that in
the future a SRR decided that ``mark to market'' accounting increased
systemic risk. Could it determine that financial institutions should be
spared from such an accounting regime on the ground that it was pro-
cyclical? This is an issue that Congress should address in any
legislation authorizing a SRR or enhancing the powers of the Federal
Reserve. I would recommend that Congress maintain authority in the SEC
to determine appropriate accounting policies, because, put simply,
transparency has been the core value underlying our system of
securities regulation.
But there are other areas where a SRR might well be entitled to
overrule the SEC. Take, for example, the problem of short selling the
stocks of financial institutions during a period of market stress.
Although the SEC did ban short selling in financial stocks briefly in
2008, one can still imagine an occasion on which the SRR and the SEC
might disagree. Here, transparency would not be an issue. Short selling
is pro-cyclical, and a SRR could determine that it had the potential to
destabilize and increase systemic risk. If it did so, its judgment
should control.
These examples are given only by way of illustration, and the
inevitability of conflicts between the two agencies is not assumed. The
President's Working Group on Financial Markets has generally been able
to work out disagreements through consultation and negotiation. Still,
in any legislation, it would be desirable to identify those core
policies (such as transparency and full disclosure) that the SRR could
not override.
The Failure of Quantitative Models: If one lesson should have been
learned from the 2008 crisis, it is that quantitative models, based on
historical data, eventually and inevitably fail. Rates of defaults on
mortgages can change (and swiftly), and housing markets do not
invariably rise. In the popular vernacular, ``black swans'' both can
occur and even become predominant. This does not mean that quantitative
models should not be used, but that they need to be subjected to
qualitative and judgmental overrides.
The weakness in quantitative models is particularly shown by the
extraordinary disparity between the value at risk estimates (VaRs)
reported by underwriters to the SEC and their eventual writedowns for
mortgage-backed securities. Ferrell, Bethel and Hu report that for a
selected group of major financial institutions the average ratio of
asset writedowns as of August 20, 2008, to VaRs reported for 2006 was
291 to 1. \96\ If financial institutions cannot accurately estimate
their exposure for derivatives and risky assets, this undermines many
of the critical assumptions underlying the Basel II Accords, and
suggests that regulators cannot defer to the institutions' own risk
models. Instead, they must reach their own judgments, and Congress
should so instruct them.
---------------------------------------------------------------------------
\96\ See Farrell, Bethel, and Hu, supra note 15, at 47.
---------------------------------------------------------------------------
The Lessons of Madoff: Implications for the SEC, FINRA, and SIPC
No time need be wasted pointing out that the SEC missed red flags
and overlooked credible evidence in the Madoff scandal. Unfortunately,
most Ponzi schemes do not get detected until it is too late. This
implies that an ounce of prevention may be worth several pounds of
penalties. More must be done to discourage and deter such schemes ex
ante, and the focus cannot be only on catching them ex post.
From this perspective focused on prevention, rather than detection,
the most obvious lesson is that the SEC's recent strong tilt towards
deregulation contributed to, and enabled, the Madoff fraud in two
important respects. First, Bernard L. Madoff Investment Securities LLC
(BMIS) was audited by a fly-by-night auditing firm with only one active
accountant who had neither registered with the Public Company
Accounting Oversight Board (``PCAOB'') nor even participated in New
York State's peer review program for auditors. Yet, the Sarbanes-Oxley
Act required broker-dealers to use a PCAOB-registered auditor. \97\
Nonetheless, until the Madoff scandal exploded, the SEC repeatedly
exempted privately held broker-dealers from the obligation to use such
a PCAOB-registered auditor and permitted any accountant to suffice.
\98\ Others also exploited this exemption. For example, in the Bayou
Hedge Fund fraud, which was the last major Ponzi scheme before Madoff,
the promoters simply invented a fictitious auditing firm and forged
certifications in its name. Had auditors been required to have been
registered with PCAOB, this would not have been feasible because
careful investors would have been able to detect that the fictitious
firm was not registered.
---------------------------------------------------------------------------
\97\ See Section 17(e)(1) of the Securities Exchange Act of 1934,
15 U.S.C. 78(q)(e)(1).
\98\ See, e.g., Securities Exch. Act Rel. No. 34-54920 (Dec. 12,
2006).
---------------------------------------------------------------------------
Presumably, the SEC's rationale for this overbroad exemption was
that privately held broker-dealers did not have public shareholders who
needed protection. True, but they did have customers who have now been
repeatedly victimized. At the end of 2008, the SEC quietly closed the
barn door by failing to renew this exemption--but only after $50
billion worth of horses had been stolen.
A second and even more culpable SEC mistake continues to date.
Under the Investment Advisers Act, investment advisers are required to
maintain client funds or securities with a ``qualified custodian.''
\99\ In principle, this requirement should protect investors from Ponzi
schemes, because an independent custodian would not permit the
investment adviser to have access to the investors' funds. Indeed, for
exactly this reason, mutual funds appear not to have experienced Ponzi-
style frauds, which have occurred only in the case of hedge funds and
investment advisers. Under Section 17(f) of the Investment Company Act,
mutual funds must use a separate custodian. But in the case of
investment advisors, the SEC permits the investment adviser to use an
affiliated broker-dealer or bank as its qualified custodian. Thus,
Madoff could and did use BMIS, his broker dealer firm, to serve as
custodian for his investment adviser activities. The net result is that
only a very tame watchdog monitors the investment adviser. Had an
independent and honest custodian held the investors' funds, Madoff
could not have recycled new investors' contributions to earlier
investors, and the custodian would have noticed that Madoff was not
actually trading. Other recent Ponzi schemes seem to have similarly
sidestepped the need for an independent custodian. At Senate Banking
Committee hearings on the Madoff debacle this January, the director of
the SEC's Office of Compliance, Inspection and Examinations estimated
that, out of the 11,300 investment advisers currently registered with
the SEC, some 1,000 to 1,500 might similarly use an affiliated broker-
dealer as their custodian. For investors, the SEC's tolerance for self-
custodians makes the ``qualified custodian'' rule an illusory
protection.
---------------------------------------------------------------------------
\99\ See Rule 206(4)-2 (``Custody of Funds or Securities of
Clients By Investment Advisers''), 17 CFR 275.206(4)-2.
---------------------------------------------------------------------------
At present, the Madoff scandal has so shaken investor confidence in
investment advisors that even the industry trade group for investment
advisers (the Investment Advisers Association) has urged the SEC to
adopt a rule requiring investment advisers to use an independent
custodian. Unfortunately, one cannot therefore assume that the SEC will
quickly produce such a rule. The SEC's staff knows that smaller
investment advisers will oppose any rule that requires them to incur
additional costs. Even if a reform rule is proposed, the staff may
still overwhelm such a rule with exceptions (such as by permitting an
independent custodian to use sub-custodians who are affiliated with the
investment adviser). Congress should therefore direct it to require an
independent custodian, across the board for mutual funds, hedge funds,
and investment advisers.
The Madoff scandal exposes shortcomings not only at the SEC but
elsewhere in related agencies. Over the last 5 years, the number of
investment advisers has grown from roughly 7,500 to 11,300--more than
one third. Given this growth, it is becoming increasingly anomalous
that there is no self-regulatory body (SRO) for investment advisers.
Although FINRA may have overstated in its claim that it had no
authority to investigate Madoff's investment adviser operations
(because it could and should have examined BMIS's performance as the
``qualified custodian'' for Madoff's investment advisory activities),
it still lacks authority to examine investment advisers. Some SRO
(either FINRA or a new body) should have direct authority to oversee
the investment adviser activities of an integrated broker-dealer firm.
Similarly, the Securities Investor Protection Corporation (SIPC)
continues to charge all broker-dealer firms the same nominal fee for
insurance without any risk-adjustment. Were it to behave like a private
insurer and charge more to riskier firms for insurance, these firms
would have a greater incentive to adopt better internal controls
against fraud. A broker-dealer that acted as a self-custodian for a
related investment adviser would, for example, pay a higher insurance
commission. Also, if higher fees were charged, more insurance (which is
currently capped at $500,000 per account) could be provided to
investors. When all broker-dealers are charged the same insurance
premium, this subsidizes the riskier firms--i.e., the future Madoffs of
the industry.
Finally, one of the most perplexing problems in the Madoff story is
why, when the SEC finally forced Madoff to register as an investment
adviser in 2006, it did not conduct an early examination of BMIS's
books and records. Red flags were flying, as Madoff (1) used an unknown
accountant, (2) served as his own self-custodian, (3) had apparently
billions of dollars in customer accounts, (4) had long resisted
registration, and (5) was the subject of plausible allegations of fraud
from credible whistle-blowers. Cost constrained as the SEC may have
been, the only conclusion that can be reached here is that the SEC has
poor criteria for evaluating the relative risk of investment advisers.
At a minimum, Congress should require a report by the SEC as to the
criteria used to determine the priority of examinations and how the SEC
proposes to change those criteria in light of the Madoff scandal.
Some have proposed eliminating the SEC's Office of Compliance,
Inspection and Examinations and combining its activities with the
Division of Investment Management. I do not see this as a panacea.
Rather, it simply reshuffles the cards. The real problem is the
criteria used to determine who should be examined. Credible allegations
of fraud need to be directed to the compliance inspectors.
Asset-Backed Securitizations: What Failed?
Asset-backed securitizations represent a financial technology that
failed. As outlined earlier, this failure seems principally
attributable to a ``moral hazard'' problem that arose under which both
loan originators and underwriters relaxed their lending standards and
packaged non-creditworthy loans into portfolios, because both found
that they could sell these portfolios at a high profit and on a global
basis--at least so long as the debt securities carried an investment
grade credit rating from an NRSRO credit rating agency.
Broad deregulatory rules contributed to this problem, and the two
most important such SEC rules are Rules 3a-7 under the Investment
Company Act \100\ and Regulation AB. \101\ Asset-backed securities
(including CDOs) are typically issued by a special purpose vehicle
(SPV) controlled by the promoter (which often may be an investment or
commercial bank). This SPV would under ordinary circumstances be deemed
an ``investment company'' and thus subjected to the demanding
requirements of the Investment Company Act--but for Rule 3a-7. That
rule exempts fixed-income securities issued by an SPV if, at the time
of sale, the securities are rated in one of the four highest categories
of investment quality by a ``nationally recognized statistical rating
organization'' (NRSRO). In essence, the SEC has delegated to the NRSROs
(essentially, at the time at least, Moody's, S&P and Fitch) the ability
exempt SPVs from the Investment Company Act. Similarly, Regulation AB
governs the disclosure requirements for ``asset-backed securities'' (as
such term is defined in Section 1101(c) of Regulation AB) in public
offerings. Some have criticized Regulation AB for being more permissive
than the federal housing agencies with respect to the need to document
and verify the loans in a portfolio. \102\ Because Regulation AB
requires that the issuer not be an investment company (see Item
101(c)(2)(i) of Regulation AB), its availability (and thus expedited
registration) also depends on an NRSRO investment grade rating.
---------------------------------------------------------------------------
\100\ 17 CFR 270.3a-7 (``Issuers of Asset-Backed Securities'').
This exemption dates back to 1992.
\101\ 17 CFR 229.1100 et seq. (``Asset-Backed Securities'').
Regulation AB was adopted in 2005, but reflects an earlier pattern of
exemptions in no-action letters.
\102\ See Mendales, supra note 18.
---------------------------------------------------------------------------
No suggestion is here intended that SPVs should be classified as
``investment companies,'' but the need for the exemption given by Rule
3a-7 shows that the SEC has considerable leverage and could condition
this exemption on alternative or additional factors beyond an NRSRO
investment grade rating. The key point is that exemptions like Rule 3a-
7 give the SEC a tool that they could use even without Congressional
legislation--if the SEC was willing to take action.
What actions should be taken to respond to the deficiencies in
asset-backed securitizations? I would suggest two basic steps: (1)
curtail the ``originate-and-distribute'' model of lending that gave
rise to the moral hazard problem, and (2) re-introduce due diligence
into the securities offering process (both for public and Rule 144A
offerings).
Restricting the ``Originate-and-Distribute'' Model of Lending. In a
bubble, everyone expects that they can pass the assets on to the next
buyer in the chain--``before the music stops.'' Thus, all tend to
economize on due diligence and ignore signs that the assets are not
creditworthy. This is because none expect to bear the costs of holding
the financial assets to maturity.
Things were not always this way. When asset-backed securitizations
began, the promoter usually issued various tranches of debt to finance
its purchase of the mortgage assets, and these tranches differed in
terms of seniority and maturity. The promoter would sell the senior
most tranche in public offerings to risk averse public investors and
retain some or all of the subordinated tranche, itself, as a signal of
its confidence in the creditworthiness of the underlying assets. Over
time, this practice of retaining the subordinated tranche withered
away. In part, this was because hedge funds would take the risk of
buying this riskier debt; in part, it was because the subordinated
tranche could be included in more complex CDOs (where
overcollateralization was the investor's principal protection), and
finally it was because in a bubbly market, investors no longer looked
for commitments or signals from the promoter.
Given this definition of the problem, the answer seems obvious:
require the promoter to retain some portion of the subordinated
tranche. This would incentivize it to buy only creditworthy financial
assets and end the ``moral hazard'' problem.
To make this proposal truly effective, however, more must be done.
The promoter would have to be denied the ability to hedge the risk on
the subordinated tranche that it retained. Otherwise it might hedge
that risk by buying a credit default swap on its own offering through
an intermediary. But this is feasible. Even in the absence of
legislation, the SEC could revise Rule 3a-7 to require, as a price of
its exemption, that the promoter (either through the SPV or an
affiliate) retain a specified percentage of the bottom, subordinated
tranche (or, if there were no subordinated tranche, of the offering as
a whole). Still, the cleaner, simpler way would be a direct legislative
requirement of a minimum retention.
2. Mandating Due Diligence. One of the less noticed but more
important developments associated with asset-backed securitization is
the rapid decline in due diligence after 2000. Once investment banks
did considerable due diligence on asset-backed securitizations, but
they outsourced the work to specialized ``due diligence'' firms. These
firms (of which Clayton Holdings, Inc. was the best known) would send
squads of ten to fifteen loan reviewers to sample the loans in a
securitized portfolio, checking credit scores and documentation. But
the intensity of this due diligence review declined over recent years.
The Los Angeles Times quotes the CEO of Clayton Holdings to the effect
that:
Early in the decade, a securities firm might have asked Clayton
to review 25 percent to 40 percent of the sub-prime loans in a
pool, compared with typically 10 percent in 2006 \103\
\103\ See E. Scott Reckard, ``Sub-Prime mortgage watchdogs kept on
leash; loan checkers say their warnings of risk were met with
indifference,'' Los Angeles Times, March 17, 2008, at C-1.
---------------------------------------------------------------------------
The President of a leading rival due diligence firm, the Bohan
Group, made an even more revealing comparison:
By contrast, loan buyers who kept the mortgages as an
investment instead of packaging them into securities would have
50 percent to 100 percent of the loans examined, Bohan
President Mark Hughes said. \104\
---------------------------------------------------------------------------
\104\ Id.
In short, lenders who retained the loans checked the borrowers
carefully, but the investment banks decreased their investment in due
diligence, making only a cursory effort by 2006. Again, this seems the
natural consequence of an originate-and-distribute model.
The actual loan reviewers employed by these firms also told the
above-quoted Los Angeles Times reporter that supervisors in these firms
would often change documentation in order to avoid ``red-flagging
mortgages.'' These employees also report regularly encountering
inflated documentation and ``liar's loans,'' but, even when they
rejected loans, ``loan buyers often bought the rejected mortgages
anyway.'' \105\
---------------------------------------------------------------------------
\105\ Id.
---------------------------------------------------------------------------
In short, even when the watchdog barked, no one at the investment
banks truly listened. Over the last several years, due diligence
practices long followed in the industry seemed to have been relaxed,
ignored, or treated as a largely optional formality. That was also the
conclusion of the President's Working Group on Financial Markets, which
in early 2008 identified ``a significant erosion of market discipline
by those involved in the securitization process, including originators,
underwriters, credit rating agencies, and global investors.'' \106\
---------------------------------------------------------------------------
\106\ See President's Working Group on Financial Markets, Policy
Statement on Financial Market Developments at 1 (March, 2008).
(emphasis added). This report expressly notes that underwriters had the
incentive to perform due diligence, but did not do so adequately.
---------------------------------------------------------------------------
Still, in the case of the investments bank, this erosion in due
diligence may seem surprising. At least over the long-term, it seems
contrary to their own self-interest. Four factors may explain their
indifference: (1) an industry-wide decline in due diligence as the
result of deregulatory reforms that have induced many underwriters to
treat legal liability as simply a cost of doing business; (2)
heightened conflicts of interest attributable to the underwriters'
position as more a principal than an agent in structured finance
offerings; (3) executive compensation formulas that reward short-term
performance (coupled with increased lateral mobility in investment
banking so that actors have less reason to consider the long-term); and
(4) competitive pressure. Each is briefly examined below, and then I
suggest some proposed reforms to address these problems.
i. The Decline of Due Diligence: A Short History: The Securities
Act of 1933 adopted a ``gatekeeper'' theory of protection, in the
belief that by imposing high potential liability on underwriters (and
others), this would activate them to search for fraud and thereby
protect investors. As the SEC wrote in 1998:
Congress recognized that underwriters occupied a unique
position that enabled them to discover and compel disclosure of
essential facts about the offering. Congress believed that
subjecting underwriters to the liability provisions would
provide the necessary incentive to ensure their careful
investigations of the offering.'' \107\
---------------------------------------------------------------------------
\107\ See SEC Release No. 7606A (``The Regulation of Securities
Offerings''), 63 Fed. Reg. 67174, 67230 (Dec. 4 1998).
Specifically, Section 11 of the Securities Act of 1933 holds the
underwriters (and certain other persons) liable for any material
misrepresentation or omission in the registration statement, without
requiring proof of scienter on the part of the underwriter or reliance
by the plaintiff. This is a cause of action uniquely tilted in favor of
the plaintiff, but then Section 11(b) creates a powerful incentive by
establishing an affirmative defense under which any defendant (other
---------------------------------------------------------------------------
than the issuer) will not be held liable if:
he had, after a reasonable investigation, reasonable ground to
believe and did believe, at the time such registration
statement became effective, that the statements made therein
were true and that there was an omission to state a material
fact required to be stated therein or necessary to make the
statements therein not misleading. 15 U.S.C. 77k (b)(3)(A).
(emphasis added)
Interpreting this provision, the case law has long held that an
underwriter must ``exercise a high degree of care in investigation and
independent verification of the company's representations.'' Feit v.
Leasco Data Processing Equip. Corp., 332 F. Supp. 554, 582 (E.D.N.Y.
1971). Overall, the Second Circuit has observed that ``no greater
reliance in our self-regulatory system is placed on any single
participant in the issuance of securities than upon the underwriter.''
Chris-Craft Indus., Inc. v. Piper Aircraft Corp., 480 F. 2d 341, 370
(2d Cir. 1973).
Each underwriter need not personally perform this investigation. It
can be delegated to the managing underwriters and to counsel, and, more
recently, the task has been outsourced to specialized experts, such as
the ``due diligence firms.'' The use of these firms was in fact strong
evidence of the powerful economic incentive that Section 11(b) of the
Securities Act created to exercise ``due diligence.''
But what then changed? Two different answers make sense and are
complementary: First, many and probably most CDO debt offerings are
sold pursuant to Rule 144A, and Section 11 does not apply to these
exempt and unregistered offerings. Second, the SEC expedited the
processing of registration statements to the point that due diligence
has become infeasible. The latter development goes back nearly thirty
years to the advent of ``shelf registration'' in the early 1980s. In
order to expedite the ability of issuers to access the market and
capitalize on advantageous market conditions, the SEC permitted issuers
to register securities ``for the shelf''--i.e., to permit the
securities to be sold from time to time in the future, originally over
a two year period (but today extended to a three year period). \108\
Under this system, ``takedowns''--i.e., actual sales under a shelf
registration statement--can occur at any time without any need to
return to the SEC for any further regulatory permission. Effectively,
this telescoped a period that was often three or four months in the
case of the traditional equity underwriting (i.e., the period between
the filing of the registration statement and its ``effectiveness,''
while the SEC reviewed the registration statement) to a period that
might be a day or two, but could be only a matter of hours.
---------------------------------------------------------------------------
\108\ See Rule 415 (17 C.F.R. 230.415)(2007).
---------------------------------------------------------------------------
Today, because there is no longer any delay for SEC review in the
case of an issuer eligible for shelf registration, an eligible issuer
could determine to make an offering of debt or equity securities and in
fact do so within a day's time. The original premise of this new
approach was that eligible issuers would be ``reporting entities'' that
filed continuous periodic disclosures (known as Form 10-Ks and Form 10-
Qs) under the Securities Exchange Act of 1934. Underwriters, the SEC
hoped, could do ``continuing due diligence'' on these issuers at the
time they filed their periodic quarterly reports in preparation for a
later, eventual public offering. This hope was probably never fully
realized, but, more importantly, this premise never truly applied to
debt offerings by issuers of asset-backed securities.
For bankruptcy and related reasons, the issuers of asset-backed
issuers (such as CDOs backed by a pool of residential mortgages) are
almost always ``special purpose vehicles'' (SPVs), created for the
single offering; they thus have no prior operating history and are not
``reporting companies'' under the Securities Exchange Act of 1934. To
enable issuers of asset-backed securities to use shelf-registration and
thus obtain immediate access to the capital markets, the SEC had to
develop an alternative rationale. And it did! To use Form S-3 (which is
a precondition for eligibility for shelf-regulation), an issuer of
asset-backed securities must receive an ``investment grade'' rating
from an ``NRSRO'' credit-rating agency. \109\ Unfortunately, this
requirement intensified the pressure that underwriters brought to bear
on credit-ratings agencies, because unless the offering received an
investment grade rating from at least one rating agency, the offering
could not qualify for Form S-3 (and so might be delayed for an
indefinite period of several months while its registration statement
received full-scale SEC review). An obvious alternative to the use of
an NRSRO investment grade rating as a condition for Form S-3
eligibility would be certification by ``gatekeepers'' to the SEC (i.e.,
attorneys and due diligence firms) of the work they performed. Form S-3
could still require an ``investment grade'' rating, but that it come
from an NRSRO rating agency should not be mandatory.
---------------------------------------------------------------------------
\109\ See Form S-3, General Instructions, IB5 (``Transaction
Requirements--Offerings of Investment Grade Asset-Backed Securities'').
---------------------------------------------------------------------------
After 2000, developments in litigation largely convinced
underwriters that it was infeasible to expect to establish their due
diligence defense. The key event was the WorldCom decision in 2004.
\110\ In WorldCom, the court effectively required the same degree of
investigation for shelf-registered offerings as for traditional
offerings, despite the compressed time frame and lack of underwriter
involvement in the drafting of the registration statement. The Court
asserted that its reading of the rule should not be onerous for
underwriters because they could still perform due diligence prior to
the offering by means of ``continuous due diligence'' (i.e., through
participation by the underwriter in the drafting of the various Form
10-Ks and Form 10-Qs that are incorporated by reference into the shelf-
registration).
---------------------------------------------------------------------------
\110\ In re WorldCom Inc. Securities Litigation, 346 F. Supp. 2d
628 (S.D.N.Y. 2004). The WorldCom decision denied the underwriters'
motion for summary judgment based on their asserted due diligence
defense, but never decided whether the defense could be successfully
asserted at trial. The case settled before trial for approximately $6.2
billion.
---------------------------------------------------------------------------
For underwriters, the WorldCom decision was largely seen as a
disaster. Their hopes--probably illusory in retrospect--were dashed
that courts would soften Securities Act 11's requirements in light of
the near impossibility of complying with due diligence responsibilities
during the shortened time frames imposed by shelf registration. Some
commentators had long (and properly) observed that the industry had
essentially played ``ostrich,'' hoping unrealistically that Rule 176
would protect them. \111\ In WorldCom's wake, the SEC did propose some
amendments to strengthen Rule 176 that would make it something closer
to a safe harbor. But the SEC ultimately withdrew and did not adopt
this proposal.
---------------------------------------------------------------------------
\111\ See Donald Langevoort, Deconstructing Section 11: Public
Offering Liability in a Continuous Disclosure Environment, 63 Law and
Contemporary Problems, U.S. 62-63 (2000).
---------------------------------------------------------------------------
As the industry now found (as of late 2004) that token or
formalistic efforts to satisfy Section 11 would not work, it faced a
bleak choice. It could accept the risk of liability on shelf offerings
or it could seek to slow them down to engage in full scale due
diligence. Of course, different law firms and different investment
banks could respond differently, but I am aware of no firms attempting
truly substantial due diligence on asset-backed securitizations.
Particularly in the case of structured finance, the business risk of
Section 11 liability seemed acceptable. After all, investment grade
bonds did not typically default or result in class action litigation,
and Section 11 has a short statute of limitations (one year from the
date that the plaintiffs are placed on ``inquiry notice''). Hence,
investment banks could rationally decide to proceed with structured
finance offerings knowing that they would be legally exposed if the
debt defaulted, in part because the period of their exposure would be
brief. In the wake of the WorldCom decision, the dichotomy widened
between the still extensive due diligence conducted in IPOs, and the
minimal due diligence in shelf offerings. As discussed below, important
business risks may have also motivated investment banks to decide not
to slow down structured finance offerings for extended due diligence.
The bottom line here then is that, at least in the case of asset-
backed shelf offerings, investment banks ceased to perform the due
diligence intended by Congress, but instead accepted the risk of
liability as a cost of doing business in this context. But that is only
the beginning of the story.
Conflicts of Interest: Traditionally, the investment bank in a
public offering played a gatekeeping role, vetting the company and
serving as an agent both for the prospective investors (who are also
its clients) and the corporate issuer. Because it had clients on both
sides of the offering, the underwriter's relationship with the issuer
was somewhat adversarial, as its counsel scrutinized and tested the
issuer's draft registration statement. But structured finance is
different. In these offerings, there is no corporate issuer, but only a
``special purpose vehicle'' (SPV) typically established by the
investment bank. The product--residential home mortgages--is purchased
by the investment bank from loan originators and may be held in
inventory by the investment bank for some period until the offering can
be effected. In part for this reason, the investment bank will
logically want to expedite the offering in order to minimize the period
that it must hold the purchased mortgages in its own inventory and at
its own risk.
Whereas in an IPO the underwriter (at least in theory) is acting as
a watchdog testing the quality of the issuer's disclosures, the
situation is obviously different in an assets-backed securities
offering that the underwriter is structuring itself. It can hardly be
its own watchdog. Thus, the quality of disclosure may suffer. Reports
have circulated that some due diligence firms advised their
underwriters that the majority of mortgages loans in some securitized
portfolio were ``exception'' loans--i.e., loans outside the bank's
normal guidelines. \112\ But the registration statement disclosed only
that the portfolio included a ``significant'' or ``substantial'' number
of such loans, not that it was predominantly composed of such loans.
This is inferior and materially deficient disclosure, and it seems
attributable to the built-in conflicts in this process.
---------------------------------------------------------------------------
\112\ See, e.g., Vikas Bajaj and Jenny Anderson, ``Inquiry Focuses
on Withholding of Data on Loans,'' New York Times, January 12, 2008, at
A-1.
---------------------------------------------------------------------------
Executive Compensation: Investment bankers are typically paid year-
end bonuses that are a multiple of their salaries. These bonuses are
based on successful completion of fee-generating deals during the year.
But a deal that generates significant income in Year One could
eventually generate significant liability in Year Two or Three. In this
light, the year-end bonus system may result in a short-term focus that
ignores or overly discounts longer-term risks.
Moreover, high lateral mobility characterizes investment banking
firms, meaning that the individual investment banker may not identify
with the firm's longer-term interests. In short, investment banks may
face serious agency costs problems, which may partly explain their
willingness to acquire risky mortgage portfolios without adequate
investigation of the collateral.
Competitive Pressure: Citigroup CEO Charles Prince's now famous
observation that ``when the music is playing, you've got to get up and
dance'' is principally a recognition of the impact of competitive
pressure. If investors are clamoring for ``investment grade'' CDOs (as
they were in 2004-2006), an investment bank understands that if it does
not offer a steady supply of transactions, its investors will go
elsewhere--and possibly not return. Thus, to hold onto a profitable
franchise, investment banks sought to maintain a steady pipeline of
transactions; this in turn lead them to seek to lock in sources of
supply. Accordingly, they made clear to loan originators their
willingness to buy all the ``product'' that the latter could supply.
Some investment banks even sought billion dollar promises from loan
originators of a minimum amount of product. Loan originators quickly
realized that due diligence was now a charade (even if it had not been
in the past) because the ``securitizing'' investment banks were
competing fiercely for supply. In a market where the demand seemed
inexhaustible, the real issue was obtaining supply, and investment
banks spent little time worrying about due diligence or rejecting a
supply that was already too scarce for their anticipated needs.
Providing Time for Due Diligence: The business model for structured
finance is today broken. Underwriters and credit rating agencies have
lost much of their credibility. Until structured finance can regain
credibility, housing finance in the United States will remain in scarce
supply.
The first lesson to be learned is that underwriters cannot be
trusted to perform serious due diligence when they are in effect
selling their own inventory and are under severe time pressure. The
second lesson is that because expedited shelf registration is
inconsistent with meaningful due diligence, the process of underwriting
structured finance offerings needs to be slowed down to permit more
serious due diligence. Shelf registration and abbreviated time
schedules may be appropriate for seasoned corporate issuers whose
periodic filings are incorporated by reference into the registration
statement, but it makes less sense in the case of a ``special purpose
vehicle'' that has been created by the underwriter solely as a vehicle
by which to sell asset-backed securities. Offerings by seasoned issuers
and by special purpose entities are very different and need not march
to the same drummer (or the same timetable).
An offering process for structured finance that was credible would
look very different than the process we have recently observed. First,
a key role would be played by the due diligence firms, but their
reports would not go only to the underwriter (who appears to have at
time ignored them). Instead, without editing or filtering, their
reports would also go directly to the credit-rating agency. Indeed, the
rating agency would specify what it would want to see covered by the
due diligence firm's report. Some dialogue between the rating agency
and the due diligence firm would be built into the process, and ideally
their exchange would be outside the presence of the underwriter (who
would still pay for the due diligence firm's services). At a minimum,
the NRSRO rating agencies should require full access to such due
diligence reports as a condition of providing a rating (this is a
principle with which these firms agree, but may find it difficult to
enforce in the absence of a binding rule).
To enable serious due diligence to take place, one approach would
be to provide that structured finance offerings should not qualify for
Form S-3 (or for any similar form of expedited SEC review). If the
process can occur in a day, the pressures on all the participants to
meet an impossible schedule will ensure that little serious
investigation of the collateral's quality will occur. An alternative
(or complementary approach) would be to direct the SEC to revise
Regulation AB to incorporate greater verification by the underwriter
(and thus its agents) of the quality of the underlying financial
assets.
Does this sound unrealistic? Interestingly, the key element in this
proposal--that that due diligence firm's report go to the credit rating
agency--is an important element in the settlement negotiated in 2008 by
New York State Attorney General Cuomo and the credit rating agencies.
\113\
---------------------------------------------------------------------------
\113\ See Aaron Lucchetti, ``Big Credit-Rating Firms Agree to
Reforms,'' The Wall Street Journal, June 6, 2008 at p. C-3.
---------------------------------------------------------------------------
The second element of this proposal--i.e., that the process be
slowed to permit some dialogue and questioning of the due diligence
firm's findings--will be more controversial. It will be argued that
delay will place American underwriters at a competitive disadvantage to
European rivals and that offerings will migrate to Europe. But today,
structured finance is moribund on both sides of the Atlantic. To revive
it, credibility must be restored to the due diligence process.
Instantaneous due diligence is in the last analysis simply a
contradiction in terms. Time and effort are necessary if the quality of
the collateral is to be verified--and if investors are to perceive that
a serious effort to protect their interests is occurring.
Rehabilitating the Gatekeepers
Credit rating agencies remain the critical gatekeeper whose
performance must be improved if structured finance through private
offerings (i.e., without government guarantees) is to become viable
again. As already noted, credit rating agencies face a concentrated
market in which they are vulnerable to pressure from underwriters and
active competition for the rating business.
At present, credit rating agencies face little liability and
perform little verification. Rather, they state explicitly that they
are assuming the accuracy of the issuer's representations. The only
force that can feasibly induce them to conduct or obtain verification
is the threat of securities law liability. Although that threat has
been historically non-existent, it can be legislatively augmented. The
credit rating agency does make a statement (i.e., its rating) on which
the purchasers of debt securities do typically rely. Thus, potential
liability does exist under Rule 10b-5 to the extent that it makes a
statement in connection with a purchase or sale of a security. The
difficult problem is that a defendant is only liable under Rule 10b-5
if it makes a material misrepresentation or omission with scienter. In
my judgment, there are few cases, if any, in which the rating agencies
actually know of the fraud. But, under Rule 10b-5, a rating agency can
be held liable if it acted ``recklessly.''
Accordingly, I would proposed that Congress expressly define the
standard of ``recklessness'' that creates liability under Rule 10b-5
for a credit rating agency to be the issuance of a rating when the
rating agency knowingly or recklessly is aware of facts indicating that
reasonable efforts have not been conducted to verify the essential
facts relied upon by its ratings methodology. A safe harbor could be
created for circumstances in which the ratings agency receives written
certification from a ``due diligence'' firm, independent of the
promoter, indicating that it has conducted sampling procedures that
lead it to believe in the accuracy of the facts or estimates asserted
by the promoter. The goal of this strategy is not to impose massive
liabilities on rating agencies, but to make it unavoidable that someone
(either the rating agency or the due diligence firm) conduct reasonable
verification. To be sure, this proposal would involve increased costs
to conduct such due diligence (which either the issuer or the
underwriter would be compelled to assume). But these costs are several
orders of magnitude below the costs that the collapse of the structured
finance market has imposed on the American taxpayer.
Conclusions
1. The current financial crisis--including the collapse of the U.S.
real estate market, the insolvency of the major U.S. investment banks,
and the record decline in the stock market--was not the product of
investor mania or the classic demand-driven bubble, but rather was the
product of the excesses of an ``originate-and-distribute'' business
model that both loan originators and investment banks followed to the
brink of disaster--and beyond. Under this business model, financial
institutions abandoned discipline and knowingly made non-creditworthy
loans because they did not expect to hold the resulting financial
assets for long enough to matter.
2. The ``moral hazard'' problem that resulted was compounded by
deregulatory policies at the SEC (and elsewhere) that permitted
investment banks to increase their leverage rapidly between 2004 and
2006, while also reducing their level of diversification. Under the
Consolidated Supervised Entity (CSE) Program, the SEC essentially
deferred to self-regulation by the five largest investment banks, who
woefully underestimated their exposure to risk.
3. This episode shows (if there ever was doubt) that in an
environment of intense competition and under the pressure of equity-
based executive compensation systems that are extraordinarily short-
term oriented, self-regulation does not work.
4. As a result, all financial institutions that are ``too big to
fail'' need to be subjected to prudential financial supervision and a
common (although risk-adjusted) standard. This can only be done by the
Federal Reserve Board, which should be given authority to regulate the
capital adequacy, safety and soundness, and risk management practices
of all large financial institutions.
5. Incident to making the Federal Reserve the systemic risk
regulator for the U.S. economy, it should receive legislative authority
to: (1) establish ceilings on debt/equity ratios and otherwise restrict
leverage at all major financial institutions (including banks, hedge
funds, money market funds, insurance companies, and pension plans, as
well as financial subsidiaries of industrial corporations); (2)
supervise and restrict the design, and trading of new financial
products (particularly including over-the-counter derivatives); (3)
mandate the use of clearinghouses, to supervise them, and in its
discretion to require their consolidation; (4) require the writedown of
risky assets by financial institutions, regardless of whether required
by accounting rule; and (5) to prevent liquidate crises by restricting
the issuance of short-term debt.
6. Under the ``twin peaks'' model, the systemic risk regulatory
agency would have broad powers, but not the power to override the
consumer protection and transparency policies of the SEC. Too often
bank regulators and banks have engaged in a conspiracy of silence to
hide problems, lest they alarm investors. For that reason, some SEC
responsibilities should not be subordinated to the authority of the
Federal Reserve.
7. As a financial technology, asset-backed securitizations have
decisively failed. To restore credibility to this marketplace, sponsors
must abandon their ``originate-and-distribute'' business model and
instead commit to retain a significant portion of the most subordinated
tranche. Only if the promoter, itself, holds a share of the weakest
class of debt that it is issuing (and on an unhedged basis) will there
be a sufficient signal of commitment to restore credibility.
8. Credit rating agencies must be compelled either to conduct
reasonable verification of the key facts that they are assuming in
their ratings methodology or to obtain such verification from
professionals independent of the issuer. For this obligation to be
meaningful, it must be backstopped by a standard of liability
specifically designed to apply to credit-rating agencies.
______
PREPARED STATEMENT OF T. TIMOTHY RYAN, JR.
President and Chief Executive Officer,
Securities Industry and Financial Markets Association
March 10, 2009
Chairman Dodd, Ranking Member Shelby, Members of the Committee; My
name is Tim Ryan and I am President and CEO of the Securities Industry
and Financial Markets Association (SIFMA). \1\ Thank you for your
invitation to testify at this important hearing. The purpose of my
testimony is to share SIFMA's views on how we might improve investor
protection as well as the regulation of our financial markets.
---------------------------------------------------------------------------
\1\ The Securities Industry and Financial Markets Association
brings together the shared interests of more than 600 securities firms,
banks and asset managers locally and globally through offices in New
York, Washington, D.C., and London. Its associated firm, the Asia
Securities Industry and Financial Markets Association, is based in Hong
Kong. SIFMA's mission is to champion policies and practices that
benefit investors and issuers, expand and perfect global capital
markets, and foster the development of new products and services.
Fundamental to achieving this mission is earning, inspiring and
upholding the public's trust in the industry and the markets. (More
information about SIFMA is available at http://www.sifma.org)
---------------------------------------------------------------------------
Overview
Our current financial crisis, which has affected nearly every
American family, underscores the imperative to modernize our financial
regulatory system. Our regulatory structure and the plethora of
regulations applicable to financial institutions are based on
historical distinctions among banks, securities firms, insurance
companies, and other financial institutions--distinctions that no
longer conform to the way business is conducted. Today, financial
services institutions perform many similar activities without regard to
their legacy charters, and often provide investors with similar
products and services, yet may be subject to different rules and to the
authority of different regulatory agencies because of the functions
performed in a bygone era.
Regulators continue to operate under authorities largely
established many decades ago. They also often operate without
sufficient coordination and cooperation and without a complete picture
of the market as a whole. For example, the Securities and Exchange
Commission (SEC) oversees brokerdealer activity. Futures firms are
regulated by the Commodity Futures Trading Commission (CFTC), while the
insurance industry is regulated by 50 State insurance regulators.
Thrifts are regulated by the Office of Thrift Supervision, and banks
may be overseen at the Federal level by the Office of the Comptroller
of the Currency, the Federal Reserve Board, or the Federal Deposit
Insurance Corporation. At the same time, some financial institutions,
such as hedge funds, largely escape regulation altogether.
As a result, our current regulatory framework is characterized by
duplicative or inconsistent regulation, and in some instances
insufficient or insufficiently coordinated oversight. The negative
consequences to the investing public of this patchwork of regulatory
oversight are real and pervasive. Investors do not have comparable
protections across the same or similar financial products. Rather, the
disclosures, standards of care and other key investor protections vary
based on the legal status of the intermediary or the product or service
being offered. For example, similar financial advisory services may be
delivered to retail clients via a broker-dealer, an investment adviser,
an insurance agent, or a trustee, thereby subjecting similar advisory
activities to widely disparate regulatory requirements. From the
perspective of financial institutions, many are subject to duplicative,
costly, and unnecessary regulatory burdens, including multiple
rulebooks, and multiple examinations and enforcement actions for the
same activity, that provide questionable benefits to investors and the
markets as a whole.
This regulatory hodgepodge unnecessarily exposes investors, market
participants, and regulators alike to the potential risk of under-
regulation, overregulation, or inconsistent regulation, both within the
U.S. and globally. A complex and overlapping regulatory structure
results in higher costs on all investors, depriving them of investment
opportunities. Simply enhancing regulatory cooperation among the many
different regulators will not be sufficient to address these issues.
In light of these concerns, SIFMA advocates simplifying and
reforming the financial regulatory structure to maximize and enhance
investor protection and market integrity and efficiency. More
specifically, we believe that a reformed--and sound--regulatory
structure should accomplish the following goals: First, it must
minimize systemic risk. Second, through a combination of structural and
substantive reforms, it must be as effective and efficient as possible,
while at the same time promoting and enhancing fair dealing and
investor protection. Finally, it should encourage consistent regulation
across the same or similar businesses and products, from country to
country, to minimize regulatory arbitrage.
Creation of a Financial Markets Stability Regulator
Systemic risk has been at the heart of the current financial
crisis. While there is no single, commonly accepted definition of
systemic risk, we think of ``systemic risk'' as the risk of a system
wide financial crisis characterized by a significant risk of the
contemporaneous failure of a substantial number of financial
institutions or of financial institutions or a financial market
controlling a significant amount of financial resources that could
result in a severe contraction of credit in the U.S. or have other
serious adverse effects on economic conditions or financial stability.
SIFMA has devoted considerable time and resources to thinking about
systemic risk, and what can be done to identify it, minimize it,
maintain financial stability and resolve a financial crisis in the
future. A regulatory reform committee of our members has met regularly
in recent months to consider these issues and to develop a workable
proposal to address them. We have sponsored roundtable discussions with
former regulators, financial services regulatory lawyers and our
members, as well as other experts, policymakers, and stakeholders to
develop solutions to the issues that have been exposed by the financial
crisis and the challenges facing our financial markets and, ultimately
and most importantly, America's investors.
Through this process, we have identified a number of questions and
tradeoffs that will confront policymakers in trying to mitigate
systemic risk. Although our members continue to consider this issue,
there seems to be consensus that we need a financial markets stability
regulator as a first step in addressing the challenges facing our
overall financial regulatory structure. The G30, in its report on
financial reform, supports a central body with the task of promoting
and maintaining financial stability, and the Treasury, in its
blueprint, also has supported a market stability regulator.
We are realistic in what we believe a financial markets stability
regulator can accomplish. It will not be able to identify the causes or
prevent the occurrence of all financial crises in the future. But at
present, no single regulator (or collection of coordinated regulators)
has the authority or the resources to collect information system-wide
or to use that information to take corrective action in a timely manner
across all financial institutions and markets regardless of charter. We
believe that a single, accountable financial markets stability
regulator will improve upon the current system.
While our position on the mission of the financial markets
stability regulator is still evolving, we currently believe that its
mission should consist of mitigating systemic risk, maintaining
financial stability and addressing any financial crisis, all of which
will benefit the investing public. It should have authority over all
markets and market participants, regardless of charter, functional
regulator or unregulated status. In carrying out its duties, the
financial markets stability regulator should coordinate with the
relevant functional regulators, as well as the President's Working
Group, as applicable, in order to avoid duplicative or conflicting
regulation and supervision. It should also coordinate with regulators
responsible for systemic risk in other countries. It should have the
authority to gather information from all financial institutions and
markets, adopt uniform regulations related to systemic risk, and act as
a lender of last resort. It should probably have a more direct role in
supervising systemically important financial organizations, including
the power to conduct examinations, take prompt corrective action and
appoint or act as the receiver or conservator of all or part of a
systemically important organization. These more direct powers would end
if a financial group were no longer systemically important.
Other Reforms That Would Enhance Investor Protection and Improve Market
Efficiency
While we believe that a financial markets stability regulator will
contribute to enhancing investor protection and improving market
efficiency, we also believe, as a second step, that we must work to
rationalize the broader financial regulatory framework to eliminate
regulatory gaps and imbalances that contribute to systemic risk.
Specifically, SIFMA believes that more effective and efficient
regulation of financial institutions--resulting in greater investor
Protection--is likely to be achieved by regulating similar activities
and firms in a similar manner and by consolidating certain financial
regulators.
Core Standards Governing Business Conduct
Currently, the regulation of the financial industry is based
predominantly on rules that were first established during the 1930s and
1940s, when the products and services offered by banks, broker-dealers,
investment advisors and insurance companies were distinctly different.
Today, however, the lines and distinctions among these companies and
the products and services they offer have become largely blurred.
Development of a single set of standards governing business conduct of
financial institutions towards individual and institutional investors,
regardless of the type of industry participant or the particular
products or services being offered, would promote and enhance investor
protection, and reduce potential regulatory arbitrage and
inefficiencies that are inherent in the existing system of multiple
regulators and multiple, overlapping rulebooks.
The core standards should be crafted so as to be flexible enough to
adapt to new products and services as well as evolving market
conditions, while providing sufficient direction for firms to establish
enhanced compliance systems. As Federal Reserve Board Chairman Ben
Bernanke once suggested, ``a consistent, principles-based, and risk-
focused approach that takes account of the benefits as well as the
risks that accompany financial innovation'' is an effective way
toprotect investors while maintaining the integrity of the marketplace.
\2\
---------------------------------------------------------------------------
\2\ See Ben S. Bernanke, Federal Reserve Board Chairman, Remarks
at the Federal Reserve Bankof Atlanta's 2007 Financial Markets
Conference, Sea Island, Georgia (May 15, 2007), at http://
www.federalreserve.gov/boarddocs/Speeches/2007/20070515/default.htm
---------------------------------------------------------------------------
This core standards approach, however, must be accompanied by
outcome-oriented rules (where rules are necessary), an open dialogue
between the regulator and regulated, and enforcement efforts focused on
addressing misconduct and fraud and protecting the investing public.
Harmonize Investment Advisor and Broker-Dealer Regulation
SIFMA has long advocated the modernization and harmonization of the
disparate regulatory regimes for investment advisory, brokerage and
other financial services in order to promote investor protection. A
2007 RAND Corporation report commissioned by the SEC found that efforts
to describe a financial service provider's duties or standard of care
in legalistic terms, such as ``fiduciary duty'' or ``suitability,''
contributes to--rather than resolves--investor confusion. \3\ Further
complicating matters, the laws that apply to many customer accounts,
such as ERISA (for employer-sponsored retirement plans) or the Internal
Revenue Code (for IRAs), have different definitions of fiduciaries, and
prohibitions on conduct and the sale of products that differ from those
under the Investment Advisers Act and state law fiduciary concepts. The
RAND report makes clear that individual investors generally do not
understand, appreciate, or care about such legal distinctions.
---------------------------------------------------------------------------
\3\ Investor and Industry Perspectives on Investment Advisers and
Broker-Dealers, RAND Institutefor Civil Justice, December 31, 2007,
available at http://www.sec.gov/news/press/2008/2008-
1_randiabdreport.pdf
---------------------------------------------------------------------------
Rather than perpetuating an obsolete regulatory regime, SIFMA
recommends the adoption of a ``universal standard of care'' that avoids
the use of labels that tend to confuse the investing public, and
expresses, in plain English, the fundamental principles of fair dealing
that individual investors can expect from all of their financial
services providers. Such a standard could provide a uniform code of
conduct applicable to all financial professionals. It would make clear
to individual investors that their financial professionals are
obligated to treat them fairly by employing the same core standards
whether the firm is a financial planner, an investment adviser, a
securities broker-dealer, a bank, an insurance agency or another type
of financial services provider. A universal standard would not limit
the ability of individual investors to contract for and receive a broad
range of services from their financial services providers, from pure
execution of customer orders to discretionary investment advice, nor
would it limit the ability of clients to define or modify relationships
with their financial services providers in ways they so choose.
As Congress contemplates regulatory reform, particularly in the
wake of the Madoff and Stanford scandals and the recent turbulence in
our financial markets, we believe that the time has come to focus on
the adoption of a universal investor standard of care.
In addition, we urge Congress to pursue a regulatory framework for
financial services providers that is understandable, practical and
provides flexibility sufficient for these intermediaries to provide
investors with both existing and future products and services. Such a
framework must also avoid artificial or vague distinctions (such as
those based on whether any investment advice is ``solely incidental''
to brokerage or whether any compensation to the financial services
provider is ``special''). Finally, the framework should support
investor choice through appropriate relief from the SEC's rigid
prohibitions against principal trading, particularly with respect to
products traded in liquid and transparent markets, which has had the
effect of foreclosing investors from obtaining more favorable pricing
on transactions based on the requirement for transaction-by-transaction
consent.
Broaden the Authority of the MSRB
The Municipal Securities Rulemaking Board (MSRB) regulates the
conduct of only broker-dealers in the municipal securities market. We
feel it is important to level the regulatory playing field by
increasing the MSRB's authority to encompass the regulation of
financial advisors, investment brokers and other intermediaries in the
municipal market to create a comprehensive regulatory framework that
prohibits fraudulent and manipulative practices; requires fair
treatment of investors, state and local government issuers of municipal
bonds and other market participants; ensures rigorous standards of
professional qualifications; and promotes market efficiencies.
Merge the SEC and CFTC
The United States is the only jurisdiction that splits the
oversight of securities and futures activities between two separate
regulatory bodies. When the CFTC was formed, financial futures
represented a very small percentage of futures activity. Now, an
overwhelming majority of futures that trade today are financial
futures. These products are nearly identical to SEC regulated
securities options from an economic standpoint, yet they are regulated
by the CFTC under a very different regulatory regime. This disparate
regulatory treatment detracts from the goal of investor protection. An
entity that combines the functions of both agencies could be better
positioned to apply consistent rules to securities and futures.
OTC Derivatives
Although OTC derivatives transactions generally are limited to
institutional participants, the use of OTC derivatives by American
businesses to manage risks and reduce funding costs provides important
benefits for our economy and, consequently, for individual investors as
well. At the same time, problems with OTC derivatives can adversely
affect the financial system and individual investors. Accordingly, we
believe that steps should be taken to further develop the
infrastructure that supports the OTC derivatives business and to
improve the regulatory oversight of that activity.
In particular, we strongly support our members' initiative to
establish a clearinghouse for credit default swaps (CDS) and we are
pleased to note that ICE US Trust LLC opened its doors for clearing CDS
transactions yesterday. We believe that development of a clearinghouse
for credit derivatives is an effective way to reduce counterparty
credit risk and, thus, promote market stability. In addition to
reducing risk, the clearinghouse will facilitate regulatory oversight
by providing a single access point for information about the CDS
transactions it processes.
We also believe that all systemically significant participants in
OTC derivatives markets should be subject to oversight by a single
systemic regulator. (It is noteworthy that the AIG affiliate that was
an active participant in the CDS market was not subject to meaningful
regulatory supervision.) The systemic regulator should be given broad
authority to promulgate rules and regulations to promote sound
practices and reduce systemic risk. We recognize that effective
regulation requires timely access to relevant information and we
believe the systemic regulator should have the necessary authority to
assure there is appropriate regulatory transparency.
Investor Protection Through International Cooperation and Coordination
Finally, the current financial crisis reminds us that markets are
global in nature and so are the risks of contagion. To promote investor
protection through effective regulation and the elimination of
disparate regulatory treatment, we believe that common regulatory
standards should be applied consistently across markets. Accordingly,
we urge that steps be taken to foster greater cooperation and
coordination among regulators in major markets in the U.S., Europe,
Asia, and elsewhere around the world. There are several international
groups in which the U.S. participates that work to further regulatory
cooperation and establish international standards, including IOSCO, the
Joint Forum, the Basel Committee on Banking Supervision, and the
Financial Stability Forum. Congress should support and encourage the
efforts of these groups.
Conclusion
Recent challenges have highlighted the necessity of reforms to
enhance investor protection. SIFMA strongly supports these efforts and
commits to be a constructive participant in the process. SIFMA stands
ready to assist the Committee as it considers regulatory reform to
minimize systemic risk, promote consistent and efficient regulation,
eliminate regulatory arbitrage, and promote capital formation--all of
which serve, directly or indirectly, the interest of investor
protection. We are confident that through our collective efforts, we
have the capacity to emerge from this crisis with stronger and more
modern regulatory oversight that will not only prepare us for the
challenges facing financial firms today and in the future, but also
help the investing public meet its financial needs and support renewed
economic growth and job creation.
______
PREPARED STATEMENT OF PAUL SCHOTT STEVENS
President and Chief Executive Officer,
Investment Company Institute
March 10, 2009
Executive Summary
Overview: Recommendations for Financial Services Regulatory Reform
The current financial crisis provides policymakers with the
public mandate needed to take bold steps to strengthen and
modernize our financial regulatory system. It is imperative to
registered investment companies (also referred to as
``funds''), as both issuers of securities to investors and
purchasers of securities in the market, that the regulatory
system ensure strong investor protection and foster competition
and efficiency in the capital markets. The ultimate outcome of
reform efforts will have a direct and lasting effect on the
fund industry and the millions of investors who choose funds to
help them save for the future.
As detailed in a recently released white paper (attached as
Appendix A), ICI recommends: (1) establishing a Systemic Risk
Regulator; (2) creating a Capital Markets Regulator
representing the combined functions of the Securities and
Exchange Commission and the Commodity Futures Trading
Commission; (3) considering consolidation of the bank
regulatory structure and authorization of an optional federal
charter for insurance companies; and (4) enhancing coordination
and information sharing among federal financial regulators.
If enacted, these reforms would improve regulators'
capability to monitor and mitigate risks across the financial
system, enhance regulatory efficiency, limit duplication, close
regulatory gaps, and emphasize the national character of the
financial services industry.
Systemic Risk Regulator
The Systemic Risk Regulator should have responsibility for:
(1) monitoring the financial markets broadly; (2) analyzing
changing conditions in domestic and overseas markets; (3)
evaluating the risks of practices as they evolve and
identifying those that are of such nature and extent that they
implicate the health of the financial system at large; and (4)
acting in coordination with other responsible regulators to
mitigate such risks.
Careful consideration should be given to how the Systemic
Risk Regulator will be authorized to perform its functions and
its relationship with other, specialized regulators.
Capital Markets Regulator
The Capital Markets Regulator should have oversight
responsibility for the capital markets, market participants,
and all financial investment products. It should be the
regulatory standard setter for funds, including money market
funds.
The agency's mission should focus on investor protection
and law enforcement, as well as maintaining the integrity of
the capital markets. Like the SEC, it should be required to
consider whether proposed regulations protect investors and
promote efficiency, competition, and capital formation.
The Capital Markets Regulator should be an independent
agency, with the resources to fulfill its mission and the
ability to attract experienced personnel who can fully grasp
the complexities of today's markets. ICI's white paper offers
recommendations for organizing and managing the new agency and
for how the agency can maximize its effectiveness.
Selected Other Areas for Reform
The Capital Markets Regulator should have express authority
to regulate in areas where there are currently gaps that have
the potential to impact the capital markets and market
participants, and to modernize regulation that has not kept
pace with changes in the marketplace. These areas include: (1)
hedge funds; (2) derivatives; (3) municipal securities; and (4)
the regulation of investment advisers and broker-dealers.
Recent Market Events and Money Market Funds
Money market funds, stringently regulated by the SEC, are
one of the most notable product innovations in American
history. These funds--which seek to offer investors stability
of principal, liquidity, and a market-based rate of return, all
at a reasonable cost--serve as an effective cash management
tool for retail and institutional investors, and are an
exceptionally important source of short-term financing in the
U.S. economy.
Until September 2008, money market funds, in some cases
with support from their sponsors, largely weathered severe
pressures in the fixed income markets that had been striking
banks and other financial services firms since 2007. In mid-
September, a series of extraordinary developments, including
the failure of Lehman Brothers, roiled financial markets around
the globe, affecting all market participants. In the midst of
this market storm, one money market fund holding a substantial
amount of Lehman commercial paper was unable to sustain its
$1.00 price per share. The news of this fund ``breaking the
buck,'' combined with broader concerns about the building
stresses in the money market and possible failures of other
financial institutions, led to heavy redemptions in prime money
market funds as investors sought safety and liquidity in
Treasury securities.
Unprecedented government initiatives--designed to provide
stability and liquidity to the markets and to support money
market funds--successfully bolstered investor confidence. To
date, the Treasury Temporary Guarantee Program for Money Market
Funds has received no claims for its guarantee, and none are
anticipated. Assuming continued progress in restoring the
health of the money market, there will be no need to extend the
Temporary Guarantee Program beyond its current one-year maximum
period.
To capture the lessons learned from recent experience, ICI
formed a Money Market Working Group of senior fund industry
leaders, led by John J. Brennan of The Vanguard Group. The
Working Group has conducted a thorough examination of how the
money market can function better, and how all funds operating
in that market, including registered money market funds, should
be regulated. The Working Group intends to report its findings,
conclusions, and recommendations later this month. We believe
that prompt implementation of its recommendations will help
assure a smooth transition away from the Temporary Guarantee
Program.
Introduction
My name is Paul Schott Stevens. I am President and CEO of the
Investment CompanyInstitute, the national association of U.S.
investment companies, including mutual funds, closed-end funds,
exchange-traded funds (ETFs), and unit investment trusts (UITs).
Members of ICI manage total assets of $9.88 trillion and serve over 93
million shareholders. ICI is pleased to testify today about investor
protection and the regulation of securities markets.
This hearing takes place at a time when the United States and a
host of other nations are grappling with the most significant financial
crisis in generations. In this country, the crisis has revealed
significant weaknesses in our current system for oversight of financial
institutions. At the same time, it offers an important opportunity for
robust dialogue about the way forward. And it provides policymakers
with the public mandate needed to take bold steps to strengthen and
modernize regulatory oversight of the financial services industry. We
strongly commend this Committee for the substantial attention it is
devoting to examining the causes of the current crisis and considering
how the regulatory system can best be improved, with particular focus
on protecting consumers and investors.
It is no exaggeration that the ultimate outcome of these reform
efforts will have a direct and lasting impact on the future of our
industry. By extension, the decisions you make will affect the millions
of American investors who choose registered investment companies (also
referred to as ``funds'') as investment vehicles to help them meet the
costs of college, their retirement needs, or other financial goals.
Funds themselves are among the largest investors in U.S. companies,
holding about one quarter of those companies' outstanding stock. Funds
also hold approximately 40 percent of U.S. commercial paper, an
important source of short-term funding for corporate America, and more
than one third of tax-exempt debt issued by U.S. municipalities. It is
thus imperative to funds, as both issuers of securities to investors
and purchasers of securities in the market, that our financial
regulatory system ensure strong protections for investors and foster
competition and efficiency within the capital markets.
Like other stakeholders, we have been thinking very hard about how
to revamp our current system so that our nation emerges from this
crisis with stronger, well-regulated institutions operating within a
fair, efficient, and transparent marketplace. Last week, ICI released a
white paper outlining detailed recommendations on how to reform the
U.S. financial regulatory system, with particular emphasis on reforms
most directly affecting the functioning of the capital markets and the
regulation of investment companies. \1\ Section II of my testimony
provides a summary of these recommendations.
---------------------------------------------------------------------------
\1\ See Investment Company Institute, Financial Services
Regulatory Reform: Discussion and Recommendations (March 3, 2009),
available at http://www.ici.org/pdf/ppr_09_reg_reform.pdf and attached
as Appendix A.
---------------------------------------------------------------------------
In addition to demonstrating the need to reform our financial
regulatory system, events of the past year have highlighted the need
for greater protections for both investors and the marketplace in
several specific areas. Section III of my testimony outlines ICI's
recommendations for legislative authority to address certain regulatory
gaps that have the potential to affect the capital markets and market
participants, and to modernize regulation that has not kept pace with
changes in the marketplace.
Finally, as discussed in Section IV of my testimony, events of the
past year have brought into sharp focus the significance of money
market funds and the critical role they play as a low-cost funding
vehicle for the American economy. While the regulatory regime for money
market funds has proven to be flexible and resilient, lessons learned
from recent events suggested the need for a thorough examination of how
the money market can function better and how all funds operating in
that market should be regulated. To that end, ICI last November formed
a working group of senior fund industry leaders with a broad mandate to
develop recommendations in these areas. The Money Market Working Group
is chaired by John J. Brennan, Chairman of The Vanguard Group, and
expects to issue a detailed report by the end of March. We would
welcome the opportunity to discuss with this Committee the
recommendations of the Money Market Working Group following the release
of its report.
Financial Services Regulatory Reform
Overview of ICI Recommendations
Broadly speaking, ICI recommends changes to our regulatory
structure that would create a framework to enhance regulatory
efficiency, limit duplication, close regulatory gaps, and emphasize the
national character of the financial services industry. To improve the
government's capability to monitor and mitigate risks across the
financial system, ICI supports the designation of a new or existing
agency or inter-agency body as a ``Systemic Risk Regulator.'' A new
``Capital Markets Regulator'' should encompass the combined functions
of the Securities and Exchange Commission and the Commodity Futures
Trading Commission, thus creating a single independent federal
regulator responsible for oversight of U.S. capital markets, market
participants, and all financial investment products. ICI further
recommends that Congress consider consolidating the regulatory
structure for the banking sector and authorizing an optional federal
charter for insurance companies. Such a regulatory framework--with one
or more dedicated regulators to oversee each major financial services
sector--would maintain specialized regulatory focus and expertise, as
well as avoid the potential for one industry sector to take precedence
over the others in terms of regulatory priorities or the allocation of
resources.
To ensure the success of this new financial regulatory structure,
there must be effectivecoordination and information sharing among the
financial regulators, including in particular the Systemic Risk
Regulator. Stronger links among these regulators should greatly assist
in developing sound policies and should facilitate U.S. cooperation
with the international regulatory community. In our white paper, we
discuss why the President's Working Group on Financial Markets, with
certain modifications, may be the most logical mechanism through which
to accomplish these purposes.
Systemic Risk Regulator
The current financial crisis has exposed the vulnerability of our
financial system to risks that have the potential to spread rapidly
throughout the system and cause significant damage. Analyses of the
causes of the current crisis suggest that systemic risks may be
occasioned by, for example, excessive leveraging, lack of transparency
regarding risky practices, and gaps in the regulatory framework.
ICI agrees with the growing consensus that our regulatory system
needs to be better equipped to anticipate and address systemic risks
affecting the financial markets. Some have called for the establishment
of a ``Systemic Risk Regulator.'' Subject to important cautions, ICI
supports designating a new or existing agency or inter-agency body to
serve in this role. We recommend that the Systemic Risk Regulator have
responsibility for: (1) monitoring the financial markets broadly; (2)
analyzing changing conditions in domestic and overseas markets; (3)
evaluating the risks of practices as they evolve and identifying those
that are of such nature and extent that they implicate the health of
the financial system at large; and (4) acting in coordination with
other responsible regulators to mitigate such risks.
The specifics of creating and empowering the Systemic Risk
Regulator will require careful attention. By way of example, to perform
its monitoring functions, this regulator likely will need information
about a range of financial institutions and market sectors. The types
of information that the regulator may require, and how the regulator
will obtain that information, are just two of the discrete issues that
will need to be fully considered.
In ICI's view, legislation establishing the Systemic Risk Regulator
should be crafted to avoid imposing undue constraints or inapposite
forms of regulation on normally functioning elements of the financial
system, or stifling innovations, competition, or efficiencies. For
example, it has been suggested that a Systemic Risk Regulator could be
given the authority to identify financial institutions that are
``systemically significant'' and to oversee those institutions
directly. Despite its seeming appeal, such an approach could have very
serious anticompetitive effects if the identified institutions were
viewed as ``too big to fail'' and thus judged by the marketplace as
safer bets than their smaller, ``less significant'' competitors.
Additionally, the Systemic Risk Regulator should be carefully
structured so as not to simply add another layer of bureaucracy or to
displace the primary regulators responsible for capital markets,
banking, or insurance. Legislation establishing the Systemic Risk
Regulator thus should define the nature of the relationship between
this new regulator and the primary regulators for these industry
sectors. The authority granted to the Systemic Risk Regulator should be
subject to explicit limitations, and the specific areas in which the
Systemic Risk Regulator and the primary regulators should work together
will need to be identified. We believe, for example, that the primary
regulators have a critical role to play as the first line of defense
for detecting potential risks within their spheres of expertise.
Capital Markets Regulator
Currently, securities and futures--and their respective markets and
market participants--are subject to separate regulatory regimes under
different federal regulators. This system reflects historical
circumstances and is out of step with the increasing convergence of
these two industries. It has resulted in jurisdictional disputes,
regulatory inefficiency, and gaps in investor protection. To bring a
consistent policy focus to U.S. capital markets, ICI recommends the
creation of a Capital Markets Regulator as a new agency that would
encompass the combined functions of the SEC and the CFTC. As the
federal regulator responsible for overseeing the capital markets and
all financial investment products, the Capital Markets Regulator--like
the SEC and the CFTC--should be established as an independent agency,
with an express statutory mission and the rulemaking and enforcement
powers necessary to carry out that mission.
It is critically important that the Capital Markets Regulator's
statutory mission focus theagency sharply on investor protection and
law enforcement, as distinct from the safety and soundness of regulated
entities. At the same time, the Capital Markets Regulator (like the SEC
today) should be required to consider, in determining whether a
proposed regulation is consistent with the public interest, both the
protection of investors and whether the regulation would promote
efficiency, competition, and capital formation. The Capital Markets
Regulator's mission also should include maintaining the integrity of
the capital markets, which will benefit both market participants and
consumers. Congress should ensure that the agency is given the
resources it needs to fulfill its mission. Most notably, the Capital
Markets Regulator must have the ability to attract personnel with the
necessary market experience to fully grasp the complexities of today's
global marketplace.
ICI envisions the Capital Markets Regulator as the regulatory
standard setter for registered investment companies, including money
market funds (as is the case now with the SEC). In so authorizing this
new agency, Congress would be continuing the important benefits that
have flowed from the shared system of federal and state oversight
established by the National Securities Markets Improvement Act of 1996.
Under this system, federal law governs all substantive regulation of
investment companies, and states have concurrent authority to protect
against fraud. We believe that this approach is consistent with the
national character of the market in which investment companies operate
and would continue to achieve the regulatory efficiencies Congress
intended, without compromising investor protection in any way.
The Capital Markets Regulator should continue to regulate
registered investment companies under the Investment Company Act of
1940. While funds are not immune to problems, the substantive
protections embodied in the Investment Company Act and related rules
have contributed significantly to the protection of investors and the
continuing integrity of funds as an investment model. Among these
protections are: (1) daily pricing and redeemability of the fund's
shares, with a requirement to use mark-to-market valuation; (2)
separate custody of fund assets (typically with a bank custodian); (3)
restrictions on complex capital structures and leveraging; (4)
prohibitions or restrictions on affiliated transactions and other forms
of self-dealing; and (5) diversification requirements. In addition,
funds are subject to more extensive disclosure and transparency
requirements than any other financial product. This regulatory
framework has proven resilient through difficult market conditions, and
has shielded fund investors from some of the problems associated with
other financial products and services. Indeed, recent experience
suggests that consideration should be given to extending the greater
discipline that has worked so well in core areas of fund regulation--
such as valuation, \2\ independent custody, affiliated transaction
prohibitions, leveraging restrictions, diversification, and
transparency--to other marketplace participants.
---------------------------------------------------------------------------
\2\ From the perspective of funds as investors in corporate and
fixed income securities, ICI believes that financial reporting that
requires the use of mark-to-market or fair value accounting to measure
the value of financial instruments serves the interests of investors
and the capital markets better than alternative cost-based measures.
For a more detailed discussion of our views, see Letter from Paul
Schott Stevens, President and CEO, Investment Company Institute, to The
Honorable Christopher Cox, Chairman, U.S. Securities and Exchange
Commission, dated November 14, 2008, available at http://www.ici.org/
statements/cmltr/08_sec_mark-to-market_com.html
---------------------------------------------------------------------------
With the establishment of a new Capital Markets Regulator, Congress
has a very valuable opportunity to ``get it right'' in terms of how the
new agency is organized and managed. Our white paper outlines several
recommendations in this regard, including the need for high-level focus
on management of the agency. We stress the importance, for example, of
the agency's having open and effective lines of internal communication,
mechanisms to facilitate internal coordination and information sharing,
and a comprehensive process for setting regulatory priorities and
assessing progress.
ICI's white paper also suggests ways in which the Capital Markets
Regulator would be able to maximize its effectiveness in performing its
responsibilities. I would like to highlight two of the most significant
suggestions for the Committee. First, the Capital Markets Regulator
should seek to facilitate close, cooperative interaction with the
entities it regulates as a means to identify and resolve problems, to
determine the impact of problems or practices on investors and the
market, and to cooperatively develop best practices that can be shared
broadly with market participants. Incorporating a more preventative
approach would likely encourage firms to step forward with self-
identified problems and proposed resolutions. Second, the Capital
Markets Regulator should establish mechanisms to stay abreast of market
and industry developments. Ways to achieve this end include hiring more
agency staff with significant prior industry experience and
establishing by statute a multidisciplinary ``Capital Markets Advisory
Committee'' comprised of private-sector representatives from all major
sectors of the capital markets.
Expected Benefits of These Reforms
If implemented, the recommended reforms outlined above and
discussed in detail in our white paper would help to establish a more
effective and efficient regulatory structure for the U.S. financial
services industry. Most significantly, these reforms would:
Improve the U.S. government's capability to monitor and
mitigate risks across our nation's financial system;
Create a regulatory framework that enhances regulatory
efficiency, limits duplication, and emphasizes the national
character of the financial services industry;
Close regulatory gaps to ensure appropriate oversight of
all market participants and investment products;
Preserve specialized regulatory focus and expertise while
avoiding the potential for uneven attention to different
industries or products;
Foster a culture of close consultation and dialogue among
U.S. financial regulators to facilitate collaboration on issues
of common concern; and
Facilitate coordinated interaction with regulators in other
jurisdictions, including with regard to risks affecting global
capital markets.
We recognize that some have criticized sector-based regulation
because it may not provide any one regulator with a full view of a
financial institution's overall business, and does not give any single
regulator authority to mandate actions designed to mitigate systemic
risks across financial markets as a whole. Our proposed approach would
address those concerns through the establishment of the Systemic Risk
Regulator to undertake this market-wide monitoring of the financial
system and through specific measures to strengthen inter-agency
coordination and information sharing.
We further believe that retaining some elements of the current
multi-agency structure would offer advantages over a single, integrated
regulator approach. Even though a single regulator could be organized
with separate units or departments focusing on different financial
services sectors, it is our understanding that, in practice, there can
be a tendency for agency leadership or staff to gravitate to certain
areas and devote insufficient attention to financial sectors perceived
to be less high profile or prone to fewer problems. Such a result has
the potential to stifle innovation valuable to consumers and produce
regulatory disparities.
Finally, we believe that a streamlining of the current regulatory
structure may be more effective and workable than an approach that
assigns regulatory responsibilities to separate agencies based on broad
regulatory objectives, such as market stability, safety and soundness,
and business conduct. These functions often are highly interrelated.
Not only could separating them prove quite challenging, but it would
force regulators to view institutions in a less integrated way and to
operate with a narrower, less informed knowledge base. For example, a
Capital Markets Regulator is likely to be more effective in protecting
investors if its responsibilities require it to maintain a thorough
understanding of capital market operations and market participants. And
while an objective-based structure could be one way to promote
consistent regulation of similar financial products and services, it is
not the only way. Under our proposed approach, minimizing regulatory
disparities for like products and services would be an express purpose
of enhanced inter-agency coordination and information sharing efforts.
Selected Other Areas for Reform
Recent experiences in the markets have underscored the need for the
Capital Markets Regulator (or, until Congress creates such a new
agency, the SEC) to have express authority to regulate in certain areas
where there are currently gaps that have the potential to impact the
capital markets and market participants, and to modernize regulation
that has not kept pace with changes in the marketplace. \3\ ICI
supports reforms for these purposes in the areas discussed below.
---------------------------------------------------------------------------
\3\ Although not necessitating legislative action, another area
for reform is regulation of credit rating agencies. ICI has long
supported increased regulatory oversight, disclosure, and transparency
requirements for credit rating agencies. We strongly support recent
regulatory initiatives that will impose additional disclosure,
reporting, and recordkeeping requirements on a nationally recognized
statistical ratings organization (NRSRO) for products that it rates.
These requirements, which are intended to increase disclosure and
transparency surrounding NRSRO policies and procedures for issuing
ratings and to increase an NRSRO's accountability for its ratings, are
a welcome step forward that should help to restore investor confidence
in the integrity of credit ratings and, ultimately, the market as a
whole. We expect to file a comment letter on the SEC's latest proposal
to enhance NRSRO regulation at the end of this month.
Hedge funds and other unregulated private pools of capital.
The Capital Markets Regulator should have the power to oversee
hedge funds and other unregulated pooled products with respect
to, at a minimum, their potential impact on the capital
markets. For example, the Capital Markets Regulator should
require nonpublic reporting of information, such as investment
positions and strategies, that could bear on systemic risk and
---------------------------------------------------------------------------
adversely impact other market participants.
Derivatives. The Capital Markets Regulator should have
clear authority to adopt measures to increase transparency and
reduce counterparty risk of certain over-the-counter
derivatives, while not unduly stifling innovation.
Municipal Securities. The Capital Markets Regulator should
be granted expanded authority over the municipal securities
market, and should use this authority to ensure that investors
have timely access to relevant and reliable information about
municipal securities offerings. Currently, the SEC and the
Municipal Securities Rulemaking Board are prohibited from
requiring issuers of municipal securities to file disclosure
documents before the securities are sold. As a result, existing
disclosures are limited, non-standardized, and often stale, and
there are numerous disparities from the corporate issuer
disclosure regime.
Investment Advisers and Broker-Dealers. The Capital Markets
Regulator should have explicit authority to harmonize the
regulatory regimes governing investment advisers and broker-
dealers. What once were real distinctions in the businesses of
advisers and brokerdealers are no longer so clear, to the point
that retail investors are largely unable to distinguish the
services of an adviser from those of a broker-dealer. These two
types of financial intermediaries, and their customers and
clients, deserve a coherent regulatory structure that provides
adequate investor protections without overlapping or
unnecessary regulation. Of particular importance is devising a
consistent standard of care in which investor protection must
be paramount. The standard thus should be a high one. We
recommend that both types of intermediaries be held to a
fiduciary duty to their clients. \4\
---------------------------------------------------------------------------
\4\ See Securities and Exchange Commission v. Capital Gains
Research Bureau, Inc., 375 U.S. 180, 84 S. Ct. 275 (1963) (holding that
Section 206 of the Investment Advisers Act of 1940 imposes a fiduciary
duty on investment advisers by operation of law).
---------------------------------------------------------------------------
Recent Market Events and Money Market Funds
Evolution and Current Significance of Money Market Funds
Money market funds are registered investment companies that seek to
maintain a stable net asset value (NAV), typically $1.00 per share.
They are comprehensively regulated under the Investment Company Act and
subject to the special requirements of Rule 2a-7 under that Act that
limit the funds' exposure to credit risk and market risk.
These strong regulatory protections, administered by the SEC for
nearly three decades, have made money market funds an effective cash
management tool for retail and institutional investors. Indeed, money
market funds represent one of the most notable product innovations in
our nation's history, with assets that have grown more than 2,000
percent (from about $180 billion to $3.9 trillion) since Rule 2a-7 was
adopted in 1983. Money market fund assets thus represent about one
third of an estimated $12 trillion U.S. ``money market,'' the term
generally used to refer to the market for debt securities with a
maturity of one year or less. \5\
---------------------------------------------------------------------------
\5\ Other participants in the money market include corporations,
state and local governments, unregistered cash pools, commercial banks,
broker-dealers, and pension funds.
---------------------------------------------------------------------------
Money market funds also are an exceptionally important source of
short-term financing in the U.S. economy. They lower the cost of
borrowing to the U.S. Treasury, businesses, and banks and finance
companies by investing in a wide array of money market instruments. By
way of example, money market funds hold roughly 40 percent of the
commercial paper issued by U.S. corporations. In addition, tax-exempt
money market funds are a significant source of funding for state and
local governments. As of December 2008, these funds had $491 billion
under management. Tax-exempt money market funds held more than 20
percent of all state and local government debt outstanding.
Money market funds seek to offer investors stability of principal,
liquidity, and a market-based rate of return, all at a reasonable cost.
Although there is no guarantee that money market funds can always
achieve these objectives (and investors are explicitly warned of this),
they have been highly successful in doing so. Since Rule 2a-7 was
adopted over 25 years ago, $325 trillion has flowed in and out of money
market funds. Yet only twice has a money market fund failed to repay
the full principal amount of its shareholders' investments. One of
these instances is directly related to recent market events and is
discussed below. The other occurred in 1994, when a small institutional
money market fund ``broke the buck'' because it had a large percentage
of its assets in adjustable-rate securities that did not return to par
at the time of an interest rate readjustment. Shareholders in that fund
ultimately received $0.96 per share (representing a 4 percent loss of
principal). In contrast, during roughly the same time period, nearly
2,400 commercial banks and savings institutions have failed in the
United States.
Impact of Recent Market Events
Until September 2008, money market funds largely had weathered
severe pressures in the fixed income market that had been striking
banks and other financial services firms since 2007. \6\ That changed
as a series of extraordinary events, in rapid succession, roiled
financial markets both in the United States and around the globe:
---------------------------------------------------------------------------
\6\ During the period from September 2007 to September 2008, many
money market fund advisers or related persons did purchase structured
investment vehicles from, or enter into credit support arrangements
with, their affiliated funds to avoid any fund shareholder losses.
On September 7, the U.S. Government placed Fannie Mae and
---------------------------------------------------------------------------
Freddie Mac intoreceivership, wiping out shareholder equity;
Long-circulated rumors about the stability of Merrill
Lynch, AIG, and Lehman Brothers gained traction;
Over the weekend of September 13-14, Merrill Lynch hastily
arranged to be sold to Bank of America;
On September 15, the federal government declined to support
Lehman Brothers, despite having arranged a buyout of Bear
Stearns, a smaller investment bank, earlier in the year. Unable
to find a buyer, Lehman declared bankruptcy; and
On September 16, the Federal Reserve Board announced a
bailout of AIG, in which the Federal Reserve Bank of New York
agreed to lend AIG up to $85 billion and to take a nearly 80
percent stake in the company.
Beginning with news of the Lehman bankruptcy on Monday, September
15, money markets in the U.S. and elsewhere began to freeze, with a
severity that was unexpected. Although Lehman's viability had been
questioned for several months, its failure--and that of Bear Stearns
several months earlier--led to mounting concerns about the health of
other financial institutions such as Wachovia, Citigroup, and many
foreign banks. There was also growing uncertainty about whether and how
the U.S. and foreign governments would support these institutions and
their creditors.
With investors running for cover, yields on Treasury securities
fell, while those on commercial paper jumped. Inter-bank rates soared
with the uncertainty about financial institutions' exposure to Lehman
and other failing financial institutions. Governments around the globe,
attempting to calm panicked markets, injected billions of dollars of
liquidity into their markets. The U.S. stock market declined nearly 5
percent on September 15 alone, reflecting broad losses to financial
companies.
Certainly the Federal Reserve seems to have been surprised by the
market's reaction to this chain of events. Appearing before this
Committee on September 23, 2008, Federal Reserve Chairman Ben Bernanke
noted:
The failure of Lehman posed risks. But the troubles at Lehman
had been well known for some time, and investors clearly
recognized--as evidenced, for example, by the high cost of
insuring Lehman's debt in the market for credit default swaps--
that the failure of the firm was a significant possibility.
Thus, we judged that investors and counterparties had had time
to take precautionary measures. While perhaps manageable in
itself, Lehman's default was combined with the unexpectedly
rapid collapse of AIG, which together contributed to the
development last week of extraordinarily turbulent conditions
in global financial markets.
Intense pressure in the money market was brought to bear, affecting
all market participants. In the midst of this market storm, a further
pressure point occurred for money market funds. The Lehman bankruptcy
meant that securities and other instruments issued by Lehman became
ineligible holdings for money market funds, in accordance with the
requirements of Rule 2a-7. One such fund that held a substantial amount
of Lehman Brothers commercial paper, the $62 billion Reserve Primary
Fund, received $25 billion in redemption requests on September 15; the
following day, September 16, its NAV dropped below $1.00 per share.
News of this development, combined with investors' broader concerns
about the building stresses in the money market and possible failures
of other financial institutions, led to heavy redemptions in prime
money market funds as investors sought safety and liquidity in Treasury
securities. To meet these unprecedented redemption requests, many money
market funds were forced to sell commercial paper and other assets. It
should be emphasized that other market participants, including
unregistered cash pools seeking to maintain a stable NAV but not
subject to Rule 2a-7, and money market funds in other jurisdictions,
experienced difficulties as least as great as those experienced by U.S.
registered money market funds.
Actions by Federal Regulators To ``Unfreeze'' the Credit Markets
The Federal Reserve and U.S. Treasury Department, seeking to cope
with completely illiquid short-term fixed income markets, on September
19 announced a series of unprecedented initiatives designed to provide
market stability and liquidity, including programs designed to support
money market funds and the commercial paper market. The Federal Reserve
established the Asset-Backed Commercial Paper Money Market Mutual Fund
Liquidity Facility (AMLF) and the Commercial Paper Funding Facility
(CPFF). \7\ The Treasury Department announced its Temporary Guarantee
Program for Money Market Funds, which guaranteed account balances as of
September 19 in money market funds that signed up for, qualified for,
and paid a premium to participate in the program. According to press
reports, virtually all money market funds signed up for the initial
term of the Treasury Temporary Guarantee Program.
---------------------------------------------------------------------------
\7\ The AMLF provided non-recourse loans at the primary credit
rate to U.S. depository institutions and bank holding companies to
finance purchases of high-quality asset-backed commercial paper (ABCP)
from money market funds. The CPFF provided a backstop to U.S. issuers
of commercial paper through a special purpose vehicle that would
purchase three-month unsecured commercial paper and ABCP directly from
eligible issuers. On February 3, 2009, the Federal Reserve extended
these and other programs for an additional six months, until October
30, 2009.
---------------------------------------------------------------------------
The government's programs successfully bolstered investor
confidence in the money market and in money market funds. Shortly after
the programs were announced, prime money market funds stabilized and,
by mid-October 2008, began to see inflows once again. By February 2009,
owing to renewed confidence in money market funds at both the retail
and institutional levels, assets of money market funds had achieved an
all-time high of just less than $3.9 trillion.
The initial three-month term of the Treasury Temporary Guarantee
Program expired on December 18, 2008, but the Treasury Department
extended the program until April 30, 2009. If extended again, the
program will expire by its own terms no later than September 18, 2009.
At the time of this hearing, an estimated $813 million has been paid in
premiums. \8\ There has been--and we are hopeful that there will be--no
occasion for the Treasury Guarantee Program to pay any claim. Assuming
continued progress in restoring the health of the money market, we
would not anticipate any need to extend the Treasury Guarantee Program
beyond the one-year maximum period.
---------------------------------------------------------------------------
\8\ See Shefali Anand, ``Treasury Pads Coffers in Bailout,'' The
Wall Street Journal (February 17, 2009), available at http://
online.wsj.com/article/SB123483112001495707.html
---------------------------------------------------------------------------
Industry-Led Reform Initiative
The market events described above have brought into sharp focus the
significance of money market funds and the critical role they play as a
low-cost funding vehicle for the American economy. To us, these events
and their impact also signaled a need to devote serious effort to
capturing the lessons learned--by conducting a thorough examination of
how the money market can function better, and how all funds operating
in that market, including registered money market funds, should be
regulated.
To that end, in November 2008 ICI formed a Money Market Working
Group, led by John J. Brennan, Chairman of The Vanguard Group. The
Working Group was given a broad mandate to develop recommendations to
improve the functioning of the money market as a whole, and the
operation and regulation of funds investing in that market. The Working
Group intends to report its findings, conclusions, and recommendations
later this month, and we look forward to sharing that information with
the Committee at that time. We believe that prompt implementation of
the Working Group's recommendations will help assure a smooth
transition away from the Treasury Guarantee Program.
Conclusion
ICI applauds the Committee for its diligent efforts on the very
important issues discussed above, and we thank you for the opportunity
to testify. We believe our recommendations for reforming financial
services regulation would have significant benefits for investors and
the capital markets. We look forward to continuing to work with the
Committee and its staff on these matters.
APPENDIX A
Investment Company Institute Financial Services Regulatory Reform:
Discussion and Recommendations--March 3, 2009
EXECUTIVE SUMMARY
Today's financial crisis has demonstrated that the current system
for oversight of U.S. financial institutions is insufficient to address
modern financial markets. Yet it also affords policymakers with the
public mandate necessary to take bold steps to strengthen and modernize
regulatory oversight of the financial services industry. In this paper,
the Investment Company Institute (ICI), the national association of
U.S. investment companies, offers its recommendations on how to achieve
meaningful reforms, with particular emphasis on those reforms that most
directly affect the functioning of the capital markets and the
regulation of investment companies (also referred to as ``funds'').
To improve the U.S. government's capability to monitor and mitigate
risks across our nation's financial system, ICI supports the
designation of a new or existing agency or inter-agency body as a
``Systemic Risk Regulator.'' As the financial crisis has shown, our
system is vulnerable to risks that have the potential to spread rapidly
throughout the system and cause significant damage. The Systemic Risk
Regulator should have responsibility for: (1) monitoring the financial
markets broadly; (2) analyzing changing conditions in domestic and
overseas markets; (3) evaluating the risks of practices as they evolve
and identifying those that are of such nature and extent that they
implicate the health of the financial system at large; and (4) acting
to mitigate such risks in coordination with other responsible
regulators. At the same time, very careful consideration should be
given to the specifics of how the Systemic Risk Regulator will be
authorized to perform its functions and how it will relate to other
financial regulators.
More broadly, ICI recommends changes to create a regulatory
framework that enhances regulatory efficiency, limits duplication,
closes regulatory gaps, and emphasizes the national character of the
financial services industry. A new ``Capital Markets Regulator'' should
encompass the combined functions of the Securities and Exchange
Commission and the Commodity Futures Trading Commission, thus creating
a single independent federal regulator responsible for oversight of
U.S. capital markets, market participants, and all financial investment
products. Also to achieve these goals, ICI recommends that Congress
consider consolidation of the regulatory structure for the banking
sector and authorization of an optional federal charter for insurance
companies. Such a regulatory framework--with one or more dedicated
regulators to oversee each major financial services sector--would
maintain specialized regulatory focus and expertise, as well as avoid
the potential for one industry sector to take precedence over the
others in terms of regulatory priorities or the allocation of
resources.
To preserve regulatory efficiencies achieved under the National
Securities Markets Improvement Act of 1996, Congress should affirm the
role of the Capital Markets Regulator as the regulatory standard setter
for all registered investment companies. The Capital Markets
Regulator's jurisdiction should include money market funds. \1\ ICI
further envisions the Capital Markets Regulator as the first line of
defense with respect to risks across the capital markets. The new
agency should be granted explicit authority to regulate in certain
areas where there are currently gaps in regulation--in particular, with
regard to hedge funds, derivatives, and municipal securities--and
explicit authority to harmonize the legal standards applicable to
investment advisers and broker-dealers. In performing its mission, the
Capital Markets Regulator should maintain a sharp focus on investor
protection and law enforcement. It also should be required to carefully
consider the impact of its rulemaking activity on efficiency,
competition and capital formation.
---------------------------------------------------------------------------
\1\ ICI has formed a Money Market Working Group that is developing
recommendations to improve the functioning of the money market and the
operation and regulation of funds investing in that market. The group
will identify needed improvements in market and industry practices;
regulatory reforms, including improvements to SEC rules governing money
market funds; and possibly legislative proposals. The Working Group
expects to report its recommendations in the first quarter of 2009.
---------------------------------------------------------------------------
Establishing the Capital Markets Regulator presents a very valuable
opportunity to ``get it right'' in terms of how the agency is organized
and managed. It is imperative, for example, that the Capital Markets
Regulator be able to keep current with market and industry developments
and understand their impact on regulatory policy. Ways to achieve this
end include hiring more agency staff with significant prior industry
experience and establishing a multidisciplinary ``Capital Markets
Advisory Committee'' comprised of private sector representatives from
all major sectors of the capital markets. There should be a high-level
focus on agency management, perhaps through the designation of a Chief
Operating Officer. To perform effectively, the agency must have open
and effective lines of internal communication, and mechanisms to
facilitate internal coordination and information sharing. We further
suggest that the agency would benefit from a comprehensive process for
setting regulatory priorities and assessing progress.
Finally, if a new U.S. financial regulatory structure is to be
successful in protecting the interests of our nation's savers and
investors, there is a critical need for effective coordination and
information sharing among the financial regulators, including in
particular the Systemic Risk Regulator. Stronger links between
regulators and an overriding sense of shared purpose would greatly
assist in sound policy development, prioritization of effort, and
cooperation with the international regulatory community. ICI observes
that the President's Working Group on Financial Markets, with certain
modifications, may be the most logical mechanism through which to
accomplish this purpose.
We strongly believe that the future of the fund industry depends
upon the existence of strong, wellregulated financial institutions
operating within a well-regulated financial marketplace that will
promote investor confidence, attract global financial business, and
enable our institutions to compete more effectively. ICI looks forward
to working with other stakeholders and policymakers to strengthen the
U.S. financial services regulatory system and to improve its ability to
meet new challenges posed by the continuing evolution of the financial
markets, market participants, and financial products.
Introduction
Well before mainstream Americans felt the widespread effects of the
current financial crisis, many policymakers and commentators were
calling for financial services regulatory reform. \2\ These efforts
reflected general agreement that our current organization for oversight
of financial institutions is insufficient to address modern financial
markets. Recent market events have served to put into much sharper
focus the many weaknesses of the current system and the many important
linkages that exist between and among the U.S. financial markets and
the markets of other developed nations.
---------------------------------------------------------------------------
\2\ See, e.g., The Department of the Treasury Blueprint for a
Modernized Financial Regulatory Structure (March 2008) (``Treasury
Blueprint''), available at http://www.treas.gov/press/releases/reports/
Blueprint.pdf; Report and Recommendations: Commission on the Regulation
of U.S. Capital Markets in the 21st Century, U.S. Chamber of Commerce
(March 2007), available at http://www.capitalmarketscommission.com/
portal/capmarkets/default.htm; Sustaining New York's and the US' Global
Financial Services Leadership (report by McKinsey & Co., Jan. 2007), at
http://www.senate.gov/schumer/SchumerWebsite/pressroom/special_reports/
2007/NY_REPORT%20_FINAL.pdf; Interim Report of the Committee on Capital
Markets Regulation (Nov. 30, 2006), available on the Committee's Web
site at http://www.capmktsreg.org/research.html
---------------------------------------------------------------------------
Yet the current financial crisis also offers an important
opportunity--the chance to have a frank and robust public dialogue
about what works and what does not. It further affords policymakers
with the public mandate necessary to take bold steps to strengthen and
modernize regulatory oversight of the financial services industry.
The debate over financial services regulatory reform will require
careful consideration of a multitude of complicated and interconnected
issues, and there are many stakeholders in the eventual outcomes of
this debate--most importantly, the nation's savers and investors. In
this paper, the Investment Company Institute (ICI), the national
association of U.S. investment companies, \3\ offers its
recommendations on how to achieve meaningful reform of financial
services regulation. We give particular emphasis to reforms that most
directly affect the functioning of the capital markets and the
regulation of investment companies (also referred to as ``funds'').
---------------------------------------------------------------------------
\3\ ICI members include mutual funds, closed-end funds, exchange-
traded funds (ETFs) and unit investment trusts (UITs).
---------------------------------------------------------------------------
Investment companies have a unique perspective on our regulatory
system, as both issuers of securities and investors in domestic and
international securities markets. It has been our experience that, in
large measure, the needs of issuers and investors are aligned--that
both will benefit from broad and efficient markets, transparency of
information, strong investor protections, and within that context, the
elimination of unnecessary regulatory impediments to innovation.
We strongly believe that the future of our industry depends upon
the existence of strong, well-regulated financial institutions
operating within a well-regulated financial marketplace that will
promote investor confidence, attract global financial business and
enable our institutions to compete more effectively. The reforms
suggested in this paper should help to build and foster such a
financial system.
Our recommendations and the benefits they are designed to achieve
are summarized in Section II below. We elaborate on our recommendations
in Section III (Establishment of a Systemic Risk Regulator), Section IV
(Formation of a New Capital Markets Regulator), Section V (Regulatory
Structure Affecting Other Financial Institutions), and Section VI
(Enhanced Inter-agency Coordination and Information Sharing). In
Section VII, we discuss in detail the expected benefits from these
reforms.
A host of different reform proposals are being advanced--by the new
Administration, members of Congress, industry groups, academics, and
others. ICI will closely follow these developments and participate in
this debate on behalf of the fund industry. We also may refine as
appropriate the views expressed in this paper.
Summary of Recommendations and Expected Benefits
Recommendations for Reform
ICI recommends that Congress:
Designate a new or existing agency or inter-agency body to
act as a Systemic Risk Regulator.
Establish a new Capital Markets Regulator encompassing the
combined functions of the Securities and Exchange Commission
and the Commodity Futures Trading Commission. The Capital
Markets Regulator should:
be the regulatory standard setter for all registered
investment companies, including money market funds;
have explicit authority to regulate in certain areas
where there are currently gaps in regulation and to harmonize
the legal standards that apply to investment advisers and
broker-dealers;
maintain a sharp focus on investor protection and law
enforcement;
carefully consider as well the impact of its rulemaking
activity on efficiency, competition and capital formation;
serve as the first line of defense with respect to
risks across the capital markets as a whole; and
take proactive steps to maximize its continuing
effectiveness, including: establishing the conditions necessary
for ongoing dialogue with the regulated industry; establishing
mechanisms to stay abreast of market/industry developments; and
developing strong capability to conduct economic analysis to
support sound rulemaking and oversight.
Consider consolidation of the regulatory structure for the
banking sector.
Authorize an optional federal charter for insurance
companies.
Enhance inter-agency coordination and information sharing
efforts, including by modernizing the Executive Order
authorizing the President's Working Group on Financial Markets.
Expected Benefits of These Reforms
ICI believes the principal benefits of these reforms would be to:
Improve the U.S. government's capability to monitor and
mitigate risks across our nation's financial system.
Create a regulatory framework that enhances regulatory
efficiency, limits duplication, and emphasizes the national
character of the financial services industry.
Close regulatory gaps to ensure appropriate oversight of
all market participants and investment products.
Preserve specialized regulatory focus and expertise and
avoid potential uneven attention to different industries or
products.
Foster a culture of close consultation and dialogue among
U.S. financial regulators to facilitate collaboration on issues
of common concern.
Facilitate coordinated interaction with regulators in other
jurisdictions, including with regard to risks affecting global
capital markets.
Establishment of a Systemic Risk Regulator
Over the past year, various policymakers and other commentators
have called for the establishment of a formal mechanism for
identifying, monitoring, and managing risks to the financial system as
a whole. For example, in a March 2008 speech, House Financial Services
Committee Chairman Barney Frank (D-MA) recommended that Congress
consider establishing a ``Financial Services Systemic Risk Regulator''
that has the capacity and power to assess risk across financial markets
and to intervene when appropriate. \4\ Around the same time, then-
Senator Barack Obama highlighted the need for a process that identifies
systemic risks to the financial system. \5\
---------------------------------------------------------------------------
\4\ See Frank Calls for Significant Changes in Financial Services
Regulation, Press Release (March 20, 2008), available at http://
financialservices.house.gov/press110/press0320082.shtml. Likewise, the
Treasury Blueprint issued shortly thereafter suggested that an optimal
regulatory structure would include the designation of a market
stability regulator responsible for overall issues of financial market
stability.
\5\ See Remarks of Senator Barack Obama: Renewing the American
Economy, New York, NY (March 27, 2008), available at http://
www.barackobama.com/2008/03/27/remarks_of_senator_barack_obam_54.php
---------------------------------------------------------------------------
The deepening financial crisis has further exposed the
vulnerability of our financial system to risks that have the potential
to spread rapidly throughout the system and cause significant damage.
It has led to a growing consensus that bold steps are needed to equip
regulators to better anticipate and address such risks. Analyses of the
causes of the current crisis suggest that systemic risks may be
occasioned by, for example: (1) excessive leveraging by financial
institutions; (2) a lack of transparency regarding risky practices; and
(3) institutions or activities that fall through gaps in the regulatory
framework. Systemic risks--whether they are attributable to excessive
risk taking by some market participants or to other causes--can
negatively impact investment companies, thereby making it more
difficult for their shareholders to achieve important financial goals.
Subject to important cautions, ICI supports the designation of a
new or existing agency or inter-agency body as a ``Systemic Risk
Regulator.'' Broadly stated, the goal in establishing a Systemic Risk
Regulator should be to provide greater overall stability to the
financial system as a whole. The Systemic Risk Regulator should have
responsibility for: (1) monitoring the financial markets broadly; (2)
analyzing changing conditions in domestic and overseas markets; (3)
evaluating the risks of practices as they evolve and identifying those
that are of such nature and extent that they implicate the health of
the financial system at large; and (4) acting to mitigate such risks in
coordination with other responsible regulators.
Very careful consideration must be given to the specifics of how
the Systemic Risk Regulator will be authorized to perform its
functions. In particular, the legislation establishing the Systemic
Risk Regulator should be crafted to avoid imposing undue constraints or
inapposite forms of regulation on normally functioning elements of the
financial system, or stifling innovations, competition or efficiencies.
By way of example, it has been suggested that a Systemic Risk Regulator
could be given the authority to identify financial institutions that
are ``systemically significant'' and to oversee those institutions
directly. Despite its seeming appeal, such an approach could have very
serious anticompetitive effects if the identified institutions were
viewed as ``too big to fail'' and thus judged by the marketplace as
safer bets than their smaller, ``less significant'' competitors. \6\
---------------------------------------------------------------------------
\6\ See, e.g., Peter J. Wallison, Regulation Without Reason: The
Group of Thirty Report, AEI Financial Services Outlook (Jan. 2009),
available at http://www.aei.org/publications/pubID.29285/pub_detail.asp
---------------------------------------------------------------------------
Additionally, the Systemic Risk Regulator should not be structured
to simply add another layer of bureaucracy or to displace the primary
regulator(s) responsible for capital markets, banking or insurance.
Legislation establishing the Systemic Risk Regulator thus should define
the nature of the relationship between this new regulator and the
primary regulator(s) for each industry sector. This should involve
placing explicit limitations on the extent of the authority granted to
the Systemic Risk Regulator, as well as identifying specific areas in
which the Systemic Risk Regulator and primary regulator(s) should work
together. We believe, for example, that the primary regulators have a
critical role to play by acting as the first line of defense with
regard to detecting potential risks within their spheres of expertise.
How these issues are resolved will have a very real impact on
registered investment companies, as both issuers and investors in the
capital markets. Money market funds, for example, are comprehensively
regulated under the Investment Company Act of 1940 and subject to
special requirements that limit the fund's exposure to credit risk and
market risk. \7\ These strong regulatory protections, administered by
the SEC for nearly three decades, have made money market funds an
effective cash management tool for retail and institutional investors
and an important source of short-term financing for American business
and municipalities. Given the size of this industry segment \8\ and its
important role in our nation's money markets, money market funds are
likely to be on the radar screen of the Systemic Risk Regulator as it
monitors the financial markets. The type of information about money
market funds that the Systemic Risk Regulator may need to perform this
function, and how the regulator will obtain that information, are just
two of the specific issues that will need to be carefully considered.
As a threshold matter, however, ICI firmly believes that regulation and
oversight of money market funds must be the province of the Capital
Markets Regulator.
---------------------------------------------------------------------------
\7\ The term ``credit risk'' refers to the exposure of securities,
through default or otherwise, to risks associated with the
creditworthiness of the issuer. The term ``market risk'' refers to the
exposure of securities to significant changes in value due to changes
in prevailing interest rates.
\8\ Money market funds had assets of approximately $3.9 trillion
under management as of February 2009.
---------------------------------------------------------------------------
ICI will closely follow the debate over the establishment of a
Systemic Risk Regulator, and will inform policymakers as to the fund
industry's views of future proposals.
Formation of a New Capital Markets Regulator
Currently, securities and futures are subject to separate
regulatory regimes under different federal regulators. This system
reflects historical circumstances that have changed significantly. As
recently as the mid-1970s, for example, agricultural products accounted
for most of the total U.S. futures exchange trading volume. By the late
1980s, a shift from the predominance of agricultural products to
financial instruments and currencies was readily apparent in the volume
of trading on U.S. futures exchanges. In addition, as new, innovative
financial instruments were developed, the lines between securities and
futures often became blurred. The existing, divided regulatory approach
has resulted in jurisdictional disputes, regulatory inefficiency, and
gaps in investor protection. With the increasing convergence of
securities and futures products, markets, and market participants, the
current system makes little sense. To bring a consistent policy focus
to U.S. capital markets, we recommend the creation of a Capital Markets
Regulator as a new agency that would encompass the combined functions
of the SEC and the CFTC.
As the federal regulator responsible for overseeing all financial
investment products, it is imperative that the Capital Markets
Regulator--like the SEC and the CFTC--be established by Congress as an
independent agency, with an express statutory mission and the
rulemaking and enforcement powers necessary to carry out that mission.
\9\ A critical part of that mission should be for the new agency to
maintain a sharp focus on investor protection and law enforcement. And
Congress should ensure that the agency is given the resources it needs
to fulfill its mission. Most notably, the Capital Markets Regulator
must have the ability to attract personnel with the necessary market
experience to fully grasp the complexities of today's global
marketplace.
---------------------------------------------------------------------------
\9\ Currently, regulatory oversight of both the securities and
futures industries involves various self-regulatory organizations. In
establishing the Capital Markets Regulator, Congress will need to
determine the appropriate role for any such organization(s).
---------------------------------------------------------------------------
Scope of Authority
ICI recommends that the Capital Markets Regulator assume on an
integrated basis the responsibilities currently handled by the SEC and
the CFTC. For the SEC, those functions include requiring public
companies to disclose financial and other information to the public;
overseeing various market participants, including securities exchanges,
broker-dealers, investment advisers, and investment companies; and
enforcing the securities laws. The SEC also oversees the setting of
accounting standards for public companies. For its part, the CFTC
regulates the commodity futures and option markets. It oversees various
entities including exchanges, clearing facilities, and market
participants such as futures commission merchants, commodity pool
operators, and commodity trading advisors. Through its oversight and
enforcement powers, it seeks to protect market users and the public
from fraud, manipulation, and abusive practices.
Of particular importance to the fund industry is to ensure that the
Capital Markets Regulator is authorized: (1) to act as the regulatory
standard setter for all registered investment companies, as is the case
now with the SEC; (2) to regulate in certain areas where there are
currently gaps that have the potential to impact the capital markets
and market participants; and (3) to regulate broker-dealers and
investment advisers in a consistent manner when they provide similar
services to investors.
1. Regulation of Registered Investment Companies: In creating the
new regulator, Congress should take note of the important benefits that
have flowed from the shared system of federal-state oversight
established by the National Securities Markets Improvement Act of 1996
(NSMIA). Under this system, federal law governs all substantive
regulation of investment companies and states have concurrent authority
to protect against fraud. NSMIA represented the judgment of Congress
that ``the system of dual federal and state securities regulation ha[d]
resulted in a degree of duplicative and unnecessary regulation . . .
that, in many instances, [was] redundant, costly, and ineffective.''
\10\ In recognition of the national character of the market in which
investment companies operate, and to secure the regulatory efficiencies
Congress intended, Congress should affirm the role of the Capital
Markets Regulator as the regulatory standard setter for registered
investment companies. The Capital Markets Regulator's regulatory
jurisdiction should include the authority to regulate money market
funds. \11\
---------------------------------------------------------------------------
\10\ Joint Explanatory Statement of the Committee of Conference,
Conference Report--National Securities Markets Improvement Act of 1996,
H.R. 3005, H.R. Conf. Rep. No. 104-864 (1996).
\11\ ICI has formed a Money Market Working Group that is
developing recommendations to improve the functioning of the money
market and the operation and regulation of funds investing in that
market. The group will identify needed improvements in market and
industry practices; regulatory reforms, including improvements to SEC
rules governing money market funds; and possibly legislative proposals.
The Working Group expects to report its recommendations in the first
quarter of 2009.
---------------------------------------------------------------------------
2. Regulatory Gaps: The Capital Markets Regulator should have
express regulatory authority in the following areas:
Hedge funds and other unregulated private pools of capital.
The Capital Markets Regulator should be authorized to provide
oversight over hedge funds and other unregulated pooled
products with respect to, at a minimum, their potential impact
on the capital markets (e.g., require nonpublic reporting of
information such as investment positions and strategies that
could bear on systemic risk and adversely impact other market
participants).
Derivatives. The Capital Markets Regulator should have
clear authority to adopt measures to increase transparency and
reduce counterparty risk of certain over-the-counter
derivatives, while not unduly stifling innovation.
Municipal Securities. The Capital Markets Regulator should
be granted expanded authority over the municipal securities
market, and use this authority to ensure that investors have
timely access to relevant and reliable information about
municipal securities offerings. Currently, the SEC and the
Municipal Securities Rulemaking Board are prohibited from
requiring issuers of municipal securities to file disclosure
documents before the securities are sold. As a result, existing
disclosures are limited, non-standardized and often stale, and
there are numerous disparities from the corporate issuer
disclosure regime.
3. Regulation of Investment Advisers and Broker-Dealers: The
Capital Markets Regulator also should have explicit authority to
harmonize the regulatory regimes governing investment advisers and
broker-dealers. What once were real distinctions in the businesses of
advisers and broker-dealers are no longer so clear, to the point that
retail investors are largely unable to distinguish the services of an
adviser from those of a broker-dealer. These two types of financial
intermediaries, and their customers and clients, deserve a coherent
regulatory structure that provides adequate investor protections--
including, in particular, a consistent standard of care--without
overlapping or unnecessary regulation.
Mission
The SEC describes its mission as ``to protect investors, maintain
fair, orderly and efficient markets, and facilitate capital
formation.'' For its part, the CFTC states that its mission is ``to
protect market users and the public from fraud, manipulation and
abusive practices related to the sale of commodity and financial
futures and options, and to foster open, competitive, and financially
sound futures and options markets.'' The differing focus expressed in
these two mission statements is reflective of historical distinctions
in the securities and futures industries, including with regard to the
purposes of their respective markets and the participants in those
markets. As growing convergence within these two industries suggests
the creation of a unified regulator for the capital markets, it is
important to consider how the mission statement for the new regulator
can best reflect this convergence.
From the perspective of the fund industry, the mission of the
Capital Markets Regulator must involve maintaining a sharp focus on
investor protection, supported by a comprehensive enforcement program.
This core feature of the SEC's mission has consistently distinguished
the agency from the banking regulators, who are principally concerned
with the safety and soundness of the financial institutions they
regulate, and it has generally served investors well over the years.
At the same time, the SEC is required by NSMIA to consider, in
determining whether a proposed regulation is consistent with the public
interest, both the protection of investors and whether the regulation
would promote efficiency, competition and capital formation. This NSMIA
requirement suggests that Congress did not view investor protection and
efficiency, competition, and capital formation as being competing
considerations, but rather determined that each is relevant to the
development of sound capital markets regulation. We strongly believe
that the Capital Markets Regulator should be subject to the same
requirements. \12\ Investors are not well served, for example, by
rulemaking actions that create significant inefficiencies or have anti-
competitive effects in the marketplace, which ultimately result in
increased costs for investors.
---------------------------------------------------------------------------
\12\ Curiously, the SEC's description of its own mission (see
http://www.sec.gov/about/whatwedo.shtml) omits any reference to
promoting competition--notwithstanding the specific requirement under
NSMIA to consider this factor.
---------------------------------------------------------------------------
Combining the market-related missions of the SEC and CFTC should be
more straightforward. Generally speaking, each agency is called upon to
maintain the integrity of the markets under its jurisdiction. The same
must be true for the new Capital Markets Regulator. As the ongoing
financial crisis demonstrates, it is imperative that the task of
maintaining market integrity be viewed broadly to include monitoring
and addressing risks across the markets as a whole. Formally assigning
some level of responsibility to the Capital Markets Regulator in this
area makes sense. Given its expertise and its position as the primary
regulator of these markets, the Capital Markets Regulator can serve as
the first line of defense with regard to detecting problems in the
capital markets. While this approach could result in some potential
overlap with the responsibilities of the Systemic Risk Regulator, we
believe that any inefficiencies may be minimized through effective
coordination and information sharing.
Agency Management and Organization
It is axiomatic in the private sector that a company's success is
directly related to the soundness of its management. The same principle
holds true for public sector entities. But management improvements take
time and serious attention, not to mention allocation of resources.
Given that they often experience frequent turnovers in leadership and
strained resources, it is not surprising that government agencies can
find it particularly difficult to undertake and sustain significant
management reforms. Establishing a new agency presents a very valuable
opportunity to ``get it right'' as part of that process.
There is also an opportunity to make sound decisions up-front about
how to organize the new agency. In so doing, it is important not to
simply use the current structure of the SEC and/or the CFTC as a
starting point. In the case of the SEC, for example, its current
organizational structure largely took shape in the early 1970s and
reflects the operation of the securities markets of that day. Rather,
the objective should be to build an organization that not only is more
reflective of today's markets, market participants and investment
products, but also will be flexible enough to regulate the markets and
products of tomorrow.
We offer the following thoughts with regard to management and
organization of the Capital Markets Regulator:
Ensure high-level focus on agency management. One approach
would be to designate a Chief Operating Officer for this
purpose.
Implement a comprehensive process for setting regulatory
priorities and assessing progress. It may be helpful to draw
upon the experience of the United Kingdom's Financial Services
Authority, which seeks to follow a methodical approach that
includes developing a detailed annual business plan
establishing agency priorities and then reporting annually the
agency's progress in meeting prescribed benchmarks.
Promote open and effective lines of communication among the
regulator's Commissioners and between its Commissioners and
staff. Such communication is critical to fostering awareness of
issues and problems as they arise, thus increasing the
likelihood that the regulator will be able to act promptly and
effectively. A range of approaches may be appropriate to
consider in meeting this goal, including whether sufficient
flexibility is provided under the Government in the Sunshine
Act, and whether the number of Commissioners should be greater
than the current number at the SEC and at the CFTC (currently,
each agency has five).
Align the inspections and examinations functions and the
policymaking divisions. This approach would have the benefit of
keeping staff in the policymaking divisions updated on current
market and industry developments, as well as precluding any de
facto rulemaking by the regulator's inspections staff.
Develop mechanisms to facilitate coordination and
information sharing among the policymaking divisions. These
mechanisms would help to ensure that the regulator speaks with
one voice.
Additional Steps To Maximize Effectiveness
ICI believes that the following proactive steps will greatly
enhance the ability of the Capital Markets Regulator to fulfill its
mission successfully when carrying out its regulatory responsibilities
and should be priorities for the new agency.
1. Establish the conditions necessary for constructive, ongoing
dialogue with the regulated industry: The Capital Markets Regulator
should seek to facilitate closer, cooperative interaction with the
entities it regulates to identify and resolve problems, to determine
the impact of problems or practices on investors and the market, and to
cooperatively develop best practices that can be shared broadly with
market participants. Incorporating a more preventative approach would
likely encourage firms to step forward with self-identified problems
and proposed resolutions. The net result is that the Capital Markets
Regulator would pursue its investor protection responsibilities through
various means not always involving enforcement measures, although
strong enforcement must remain an important weapon in the regulator's
arsenal.
2. Establish mechanisms to stay abreast of market and industry
developments: The Capital Markets Regulator would benefit from the
establishment of one or more external mechanisms designed to help the
agency stay abreast of market and industry issues and developments,
including developments and practices in non-U.S. jurisdictions as
appropriate. For example, several federal agencies--including both the
SEC and CFTC--utilize a range of advisory committees. Such committees,
which generally have significant private sector representation, may be
established to provide recommendations on a discrete set of issues
facing the agency (e.g., the SEC's Advisory Committee on Improvements
to Financial Reporting) or to provide regular information and guidance
to the agency (e.g., the CFTC's Agricultural Advisory Committee).
ICI believes that a multidisciplinary ``Capital Markets Advisory
Committee'' could be a very effective mechanism for providing the
Capital Markets Regulator with ``real world'' perspectives and insights
on an ongoing basis. We recommend that such a committee be comprised
primarily of private sector representatives from all major sectors of
the capital markets, and include one or more members representing funds
and asset managers. Additionally, the Capital Markets Advisory
Committee should be specifically established in, and required by, the
legislation creating the Capital Markets Regulator. Such a statutory
mandate would emphasize the importance of this advisory committee to
the agency's successful fulfillment of its mission.
The establishment of an advisory committee would complement other
efforts by the Capital Markets Regulator to monitor developments
affecting the capital markets and market participants. These efforts
should include, first and foremost, hiring more staff members with
significant prior industry experience. Their practical perspective
would enhance the agency's ability to keep current with market and
industry developments and better understand the impact of such
developments on regulatory policy.
3. Apply reasonably comparable regulation to like products and
services: Different investment products often are subject to different
regulatory requirements, often with good reason. At times, however,
heavier regulatory burdens have been placed on funds than on other
investment products that share similar features and are sold to the
same customer base. It does not serve investors well if the regulatory
requirements placed on funds--however well-intentioned--end up
discouraging investment advisers from entering or remaining in the fund
business, dissuading portfolio managers from managing funds as opposed
to other investment products, or creating disincentives for brokers and
other intermediaries to sell fund shares. It is critically important
for the Capital Markets Regulator to be sensitive to this dynamic in
its rulemakings. Among other things, in analyzing potential new
regulatory requirements for funds, the Capital Markets Regulator should
consider whether other investment products raise similar policy
concerns and thus should be subject to comparable requirements.
4. Develop strong capability to conduct economic analysis to
support sound rulemaking and oversight: The Capital Markets Regulator
will be best positioned to accomplish its mission if it conducts
economic analysis in various aspects of the agency's work, including
rulemaking, examinations, and enforcement. Building strong economic
research and analytical capabilities is an important way to enhance the
mix of disciplines that will inform the agency's activities. From
helping the agency look at broad trends that shed light on how markets
or individual firms are operating to enabling it to demonstrate that
specific policy initiatives are well-grounded, developing the agency's
capability to conduct economic analysis will be well worth the long-
term effort required. The agency should consider having economists
resident in each division to bring additional, important perspectives
to bear on regulatory challenges.
It is important that economic analysis play an integral role in the
rulemaking process, because many regulatory costs ultimately are borne
by investors. When new regulations are required, or existing
regulations are amended, the Capital Markets Regulator should
thoroughly examine all possible options and choose the alternative that
reflects the best trade-off between costs to, and benefits for,
investors. Effective cost-benefit analysis does not mean compromising
protections for investors or the capital markets. Rather, it challenges
the regulator to consider alternative proposals and think creatively to
achieve appropriate protections while minimizing regulatory burdens, or
to demonstrate that a proposal's costs and burdens are justified in
light of the nature and extent of the benefits that will be achieved.
\13\
---------------------------------------------------------------------------
\13\ See, e.g., Special Report on Regulatory Reform, Congressional
Oversight Panel (submitted under Section 125(b)(2) of Title I of the
Emergency Economic Stabilization Act of 2008) (Jan. 2009) (``In
tailoring regulatory responses . . . the goal should always be to
strike a reasonable balance between the costs of regulation and its
benefits. Just as speed limits are more stringent on busy city streets
than on open highways, financial regulation should be strictest where
the threats--especially the threats to other citizens--are greatest,
and it should be more moderate elsewhere.'').
---------------------------------------------------------------------------
5. Modernize regulations that no longer reflect current market
structures and practices: Financial markets and related services are
constantly evolving, frequently at a pace that can make the regulations
governing them (or the rationale behind those regulations) become less
than optimal, if not entirely obsolete. Requiring industry participants
to comply with outmoded regulations imposes unnecessary costs on both
firms and investors, may impede innovation, and, most troubling of all,
could result in inadequate protection of investors. It is thus
important that the Capital Markets Regulator engage in periodic reviews
of its existing regulations to determine whether any such regulations
should be modernized or eliminated.
6. Give heightened attention to investor education: During the
course of their lives, investors are called upon to make a variety of
investment decisions as their personal circumstances change. These
decisions may involve saving to buy a home or to finance a child's
education, building an adequate nest egg for retirement, or investing
an inheritance, to name a few. Whether they make their investment
decisions individually or with the help of a financial adviser,
investors need to be able to make informed decisions based upon their
individual needs.
The recent turmoil in the financial markets has underscored how
important it is that investors be knowledgeable and understand their
investments. Well-informed investors are more likely to develop
realistic expectations, take a long-term perspective, and understand
the trade-off between risk and reward. They are less likely to panic
and make mistakes.
To better equip investors to make good decisions about their
investments, the Capital Markets Regulator should assign a high
priority to pursuing regulatory initiatives that will help educate
investors. The SEC's new rule allowing mutual funds and exchange-traded
funds to provide a ``summary prospectus'' containing key fund
information to investors--while making additional information available
online or by mail or e-mail upon request--is an excellent example of a
forward thinking approach to better informing investors. It should
serve as a model for future disclosure improvement efforts, such as
reform of fund shareholder reports. Regulatory efforts to promote
investor education also should extend beyond funds. Investors who
purchase other types of investment products or services, such as
separately managed accounts, likewise would benefit from clear,
concise, understandable disclosure. In addition, appropriately
fashioned point of sale disclosure would help investors in all types of
retail investment products assess and evaluate broker recommendations.
The SEC has an Office of Investor Education and Advocacy and
provides some investor education resources on its Web site. These types
of efforts should be expanded, possibly in partnership with other
governmental or private entities, and better publicized. Many industry
participants, too, have developed materials and other tools to help
educate investors; additional investor outreach efforts should be
encouraged.
Process of Merging the SEC and CFTC
Legislation to merge the SEC and CFTC should outline a process by
which to harmonize the very different regulatory philosophies of the
two agencies, as well as to rationalize their governing statutes and
current regulations. We note that there is potential peril in leaving
open-ended the process of merging the two agencies. We accordingly
recommend that the legislation creating the Capital Markets Regulator
set forth a specific timetable, with periodic benchmarks and
accountability requirements, so as to ensure that the merger of the SEC
and CFTC is completed as expeditiously as possible.
The process of merging the two agencies will be lengthy, complex,
and have the potential to disrupt the functioning of the SEC, CFTC, and
their regulated industries. We suggest that, in anticipation of the
merger, the SEC and CFTC undertake detailed consultation on all
relevant issues and take all steps possible toward greater
harmonization of the agencies. This work should be facilitated by the
Memorandum of Understanding the two agencies signed last year regarding
coordination in areas of common regulatory interest. \14\ ICI believes
that its recommendations with respect to the Capital Markets Regulator
may provide a helpful framework for these efforts.
---------------------------------------------------------------------------
\14\ See SEC, CFTC Sign Agreement to Enhance Coordination,
Facilitate Review of New Derivative Products (SEC press release dated
March 11, 2008), available at http://www.sec.gov/news/press/2008/2008-
40.htm
---------------------------------------------------------------------------
Regulatory Structure Affecting Other Financial Institutions
Earlier in this paper, we have recommended the establishment of a
Systemic Risk Regulator, and we have discussed at length the need for a
new Capital Markets Regulator to oversee markets and market
participants in the securities and futures industries. In this section
and the one immediately following, we comment briefly on reforms
affecting the regulators overseeing other sectors of the U.S. financial
system (specifically, banking and insurance) and how all regulators
within the system can work together more effectively.
Regulation of the banking and insurance industries is, quite
obviously, not ICI's primary area of focus. That said, regulation of
these industries greatly affects the performance of the U.S. financial
system as a whole and the ability of investment companies to function
within that system.
ICI believes it is important, therefore, for policymakers to
carefully consider how to achieve a more rational regulatory structure
for the banking sector that consolidates duplicative regulatory
agencies and clarifies regulatory missions. Any such analysis would no
doubt need to address difficult issues concerning the future role of
state banking regulators if we are to have a more rational regulatory
system at the national level.
With regard to the insurance industry, ICI supports in concept the
idea of creating a regulator at the federal level, a reform that has
been sought by some insurance companies as a means of providing a
streamlined and efficient alternative to the current system of state
regulation. Authorizing an optional federal charter for insurers
appears to be a logical way to bridge the gap between what exists today
and the more comprehensive approach that is required for all financial
institutions operating in truly national and often international
markets. We also believe that a federal insurance regulator would
provide an important and practical enhancement to federal inter-agency
coordination and information sharing efforts, as discussed below.
Enhanced Inter-Agency Coordination and Information Sharing
A recent report examined the benefits and shortcomings of the four
primary approaches to regulatory supervision currently used in
jurisdictions around the world. \15\ The report observed that,
regardless of the type of supervisory system in place, virtually all
financial supervisors emphasized the importance of inter-agency
coordination and information sharing for successful oversight of the
financial system as a whole and for mitigation of systemic risk.
---------------------------------------------------------------------------
\15\ See Group of Thirty, The Structure of Financial Supervision:
Approaches and Challenges in a Global Marketplace (Oct. 6, 2008),
available at http://www.deloitte.com/dtt/cda/doc/content/
us_fsi_banking_G30%20Final%20Report%2010-3-08.pdf
---------------------------------------------------------------------------
Effective inter-agency coordination also plays a critical role when
there is a need to engage on financial services regulatory issues at an
international level. The variety of supervisory systems around the
world and the increasing globalization of financial markets make
coordination among U.S. regulatory agencies all the more important.
In the United States at present, a variety of mechanisms are used
to promote coordination and information sharing within our complex
regulatory system, including arrangements at both the Federal and State
levels and arrangements among federal and state agencies. These
arrangements may be specifically mandated by Congress, such as the
inter-agency Federal Financial Institutions Examination Council, or may
be initiated by the regulators themselves, such as the July 2008
Memorandum of Understanding between the Federal Reserve and the SEC to
foster greater coordination and information sharing. \16\ One
particularly important mechanism for the past two decades has been the
President's Working Group on Financial Markets, whose members are the
heads of the Treasury Department, Federal Reserve, SEC and CFTC. As
described in the Treasury Blueprint, the role of the PWG has evolved
beyond the scope of the 1988 Executive Order creating it, so that the
PWG has become a key communication and coordination mechanism for
financial policy.
---------------------------------------------------------------------------
\16\ See, e.g., SEC, FRB Sign Agreement to Enhance Collaboration,
Coordination, and Information Sharing (SEC press release dated July 7,
2008), available at http://www.sec.gov/news/press/2008/2008-134.htm
---------------------------------------------------------------------------
If efforts to streamline the U.S. financial regulatory structure
are to be successful, some of these coordination mechanisms would
almost certainly require modification or perhaps would no longer be
necessary. There would, however, still be a very critical need for
coordination and information sharing among the remaining regulatory
bodies, presumably with involvement in particular by the Systemic Risk
Regulator. The President's Working Group, with necessary modifications,
would appear to be the easiest way to achieve this end.
ICI concurs with the recommendation in the Treasury Blueprint that
the Executive Order authorizing the PWG should be modernized ``to
reinforce the group's mission and purpose . as an ongoing mechanism for
coordination and communication on financial policy matters including
systemic risk, market integrity, investor and consumer protection, and
capital markets competitiveness.'' We suggest that any new Executive
Order also discuss the following additional areas where inter-agency
coordination and information sharing are critically important: (1) the
regular exchange of information about the latest market and industry
developments, including international trends and developments; (2) the
discussion of policy initiatives that extend across jurisdictional
lines; (3) the minimization of regulatory disparities for like
financial products and services; and (4) the need to balance financial
innovation with appropriate market and investor protection safeguards.
Equally important, in ICI's view, is the role of the PWG in
fostering a culture of close consultation and dialogue among senior
officials within each regulatory sector that will carry over into each
regulator's dealings with one another. Stronger links between
regulators and an overriding sense of shared purpose would greatly
assist in sound policy development, prioritization of effort, and
cooperation with the international regulatory community.
Expected Benefits From These Reforms
If implemented, the recommended reforms outlined above would help
to establish a more effective and efficient regulatory structure for
the U.S. financial services industry. Most significantly, these reforms
would:
Improve the U.S. government's capability to monitor and
mitigate risks across our nation's financial system.
Create a regulatory framework that enhances regulatory
efficiency, limits duplication, and emphasizes the national
character of the financial services industry.
Close regulatory gaps to ensure appropriate oversight of
all market participants and investment products.
Preserve specialized regulatory focus and expertise and
avoid potential uneven attention to different industries or
products.
Foster a culture of close consultation and dialogue among
U.S. financial regulators to facilitate collaboration on issues
of common concern.
Facilitate coordinated interaction with regulators in other
jurisdictions, including with regard to risks affecting global
capital markets.
Of significant import to registered investment companies, creation
of a consolidated Capital Markets Regulator would provide a single
point of regulatory authority and consistent rulemaking and oversight
for investment products, the capital markets, and market participants.
It would create regulatory efficiencies by eliminating areas where
responsibilities overlap and by ensuring against regulatory gaps and
potential inconsistencies. A strong, integrated regulator for the
capital markets that can see ``the whole picture'' will be better
equipped to face the challenges of these rapidly evolving markets, and
thus to protect the interests of investors.
More generally, increased consolidation of financial services
regulators, combined with the establishment of a Systemic Risk
Regulator and more robust inter-agency coordination and information
sharing, should facilitate monitoring and mitigation of risks across
the financial system. It also should result in increased regulatory
efficiency, including less duplication, and help to eliminate
regulatory gaps.
Consolidation of regulatory agencies also may further the
competitive posture of the U.S. financial markets. It may make it
easier to harmonize U.S. regulations with regulations in other
jurisdictions when that is appropriate. And reducing the number of U.S.
regulatory agencies, while also strengthening the culture of
cooperation and dialogue among senior officials of the agencies, will
likely facilitate coordinated interaction with regulators around the
world.
By providing for one or more dedicated regulators to oversee each
major financial services sector, the proposed structure would maintain
the specialized focus and expertise that is a hallmark of effective
regulation. This structure also would allow appropriate tailoring of
regulation to accommodate fundamental differences in regulated
entities, products and activities. Additionally, it would avoid the
potential for one industry sector to take precedence over the others in
terms of regulatory priorities or approaches or the allocation of
regulatory resources.
ICI recognizes that some have criticized sector-based regulation
because it may not provide any one regulator with a full view of a
financial institution's overall business, and does not give any single
regulator authority to mandate actions designed to mitigate systemic
risks across financial markets as a whole. Our proposed approach would
address those concerns through the establishment of the Systemic Risk
Regulator and specific measures to strengthen inter-agency coordination
and information sharing.
We further believe that retaining some elements of the current
multi-agency structure likely would offer advantages over a single,
integrated regulator approach. Even though a single regulator could be
organized with separate units or departments focusing on different
financial services sectors, it is our understanding that, in practice,
there can be a tendency for agency staff to ``gravitate'' to certain
areas and devote insufficient attention to financial sectors perceived
to be less high profile or prone to fewer problems. Such a result has
the potential to stifle innovation valuable to consumers and produce
regulatory disparities.
Finally, we believe that a streamlining of the current regulatory
structure may be more effective and workable than an approach that
assigns regulatory responsibilities to separate agencies based on broad
regulatory objectives (e.g., market stability, safety and soundness,
and business conduct). These functions often are highly interrelated.
Not only could separating them prove quite challenging, but it would
force regulators to view institutions in a less integrated way and to
operate with a narrower, less informed knowledge base. For example, a
Capital Markets Regulator is likely to be more effective in protecting
investors if its responsibilities require it to maintain a thorough
understanding of capital market operations and market participants. And
while an objective-based structure could be one way to promote
consistent regulation of similar financial products and services, it is
not the only way. Under our proposed approach, minimizing regulatory
disparities for like products and services would be an express purpose
of enhanced inter-agency coordination and information-sharing efforts.
______
PREPARED STATEMENT OF MERCER E. BULLARD
Associate Professor,
University of Mississippi School of Law, and
President,
Fund Democracy, Inc.
March 11, 2009
Chairman Dodd, Ranking Member Shelby, Members of the Committee,
thank you for the opportunity to appear before you to discuss investor
protection issues. It is an honor and a privilege to appear before the
Committee today.
I am the Founder and President of Fund Democracy, a nonprofit
advocacy group for mutual fund shareholders, and an Associate Professor
of Law at the University of Mississippi School of Law. I founded Fund
Democracy in January 2000 to provide a voice and information source for
mutual fund shareholders on operational and regulatory issues that
affect their fund investments. Fund Democracy has attempted to achieve
this objective in a number of ways, including filing petitions for
hearings, submitting comment letters on rulemaking proposals,
testifying on legislation, publishing articles, lobbying the financial
press, and creating and maintaining an Internet Web site for the
posting of information. I also have served as a consultant and expert
witnesses for plaintiffs and defendants in a variety of securities
cases, including some that are discussed in this testimony.
This testimony focuses on investor protection issues related to
investment management and investment advisory services. Some of these
issues have arisen in connection with the current financial crisis,
such as the question of prudential regulation of money market funds.
This testimony begins with a discussion of different aspects of this
question. But many investor protection issues reflect longstanding
problems that have been left unattended by the SEC. There continue to
be significant gaps in mutual fund fee disclosure rules, reform of fund
distribution regulation is long overdue, and the SEC's fund governance
initiative seems to have been all but forgotten. The SEC continues to
allow hedge funds to offer their shares to unsophisticated investors,
and brokers continue to receive undisclosed selling compensation that
creates an incentive to sell the most remunerative funds even if they
are not the best funds for the client.
On the whole, however, the investment management industry has fared
well in the current crisis. Equity mutual funds have experienced their
largest single year loss in history, yet net redemptions have remained
small. Employee benefit plan participants generally have continued to
make regular investors in funds. The mutual fund structure has been
shown to be remarkably resilient in this time of stress. Investors seem
to have faith in mutual funds' promise to convert their accounts to
cash in short order at their next computed NAV, which is based on
actual market values as opposed to malleable accounting principles.
More money has flowed out of broker-managed accounts than mutual funds.
Only one money market fund has experienced a loss of principal
(compared with the failure of dozens of banks), and, with the playing
field with banks temporarily leveled by the Treasury's temporary
insurance program, money market funds have increased their total
assets. The investment management industry's success depends, however,
on its and its regulators' keeping pace with the needs of investors.
MONEY MARKET FUNDS
Money Market Fund Insurance \1\
---------------------------------------------------------------------------
\1\ See generally Mercer Bullard, Federally Insured Money Market
Funds and Narrow Banks: The Path of Least Insurance (Mar. 2, 2009)
available at http://ssrn.com/abstract=1351987
---------------------------------------------------------------------------
As discussed above, mutual funds have been a singular success story
in the midst of the current financial crisis. Money market funds
arguably have been the best illustration of this success. As often
happens when those who succeed are surrounded by failed competitors,
however, some have responded to the failure of a single retail money
market fund--the first in history--by demanding that money market funds
be converted to and regulated as banks. A former Fed chairman explained
this position as follows: ``If they are going to talk like a bank and
squawk like a bank, they ought to be regulated like a bank.'' The
problem with this argument is that money markets do not fail like
banks.
Since 1980, more than 3,000 U.S. banks have failed, costing
taxpayers hundreds of billions of dollars. During the same time period,
two money market funds have failed, costing taxpayers zero dollars. \2\
The lesson that the Group of 30 takes from this history is that it is
money market funds that should be regulated as banks. The lesson that
Congress should take from this history is that banks should be
regulated more like money market funds. As discussed further below,
banks routinely fail because they are permitted to invest deposits that
can be withdrawn at a moment's notice in illiquid, long-term, risky
assets. In comparison, money market funds invest in liquid, short-term,
safe assets. The Group of 30 has disparaged money market funds as
``underscor[ing] the dangers of institutions with no capital, no
supervision, and no safety net,'' yet the extraordinary stability of
money market funds relative to banks makes a mockery of their argument.
---------------------------------------------------------------------------
\2\ Ironically, money market funds to date have provided a net
positive contribution to the ongoing bailout of financial institutions.
The Treasury has collected more than $800 million in money market fund
insurance premiums only a small fraction of which, if any, are likely
ever to be paid out in claims.
---------------------------------------------------------------------------
It is banks that should be regulated like money market funds, with
the investment of insured bank deposits being limited to liquid, short-
term, safe assets. There is no longer any good policy reason to insure
bank deposits backed by longterm, risky assets. The current financial
crisis has demonstrated that banks no longer play a special role in
this market. Many types of entities now play a significant role in the
creation of liquidity through investment in long-term assets that
historically was dominated by banks. And many of these entities rely on
shortterm liabilities (i.e., funds subject to payment on demand),
including mutual funds, to fund such investments. Any regulatory regime
that seeks to mitigate the systemic risk inherent in the investment of
short-term funds in long-term ventures must consider the full spectrum
liquidity-creation mechanisms and reject a bank-centric view of finance
that distorts efficient investment and leaves unregulated large areas
of financial activities.
What should be insured is cash accounts on which the stability of
the payments system depends. The current crisis has demonstrated the
need to ensure that the cash management vehicles that form the
foundation of our payments system are absolutely secure. Deposit
insurance provides this security. Its weakness, however, is that it
also insures risks that are necessary to the provision of transactional
services. Banks are permitted to invest insured bank deposits in
longterm risky ventures, thereby destabilizing the payments system and
inflicting large losses on the insurance fund and taxpayers. Money
market funds have been a paragon of stability because they are
permitted to invest only in a diversified pool of short-term, high-
quality assets.
The answer to whether money market fund insurance should be made
permanent seems obvious. Terminating the temporary insurance program
could lead to another run on money market funds and require that the
program immediately be restored. Even if a run does not follow
termination of the program, money market funds will continue to
represent a major source of transactional services the failure of which
would threaten the viability of our payments system. Money market fund
shareholders know this. The question of whether there is an implied
federal guarantee of money market funds has been answered. The next
time that a run on money market funds seems imminent, a federal entity
will have to stop the run with a guarantee, except that without an
insurance program in place it will not have collected any premiums that
(if risk-based) might have reduced risktaking and that would have
provided non-taxpayer funds with which to cover losses. Taxpayers will
be left to back up this guarantee.
At the same time that federal insurance is extended to all
significant sources of transaction accounts, it should be used to
reduce exposure to risk from the investment of short-term deposits in
long-term, risky assets. To some extent, this would be accomplished by
making money market fund insurance permanent. The higher yields
historically offered by money market funds would siphon even more
deposits from banks reduce the attendant risk of their investment in
risky assets. Federal insurance also should be extended to a new kind
of bank that was required to invest deposits in the same kinds of
assets as money market funds. In order to enable compete such ``narrow
banks'' to compete effectively, they would be relieved of burdens
unique to banks, such as the requirements of the Community Reinvestment
Act. Narrow banks also would continued to have the advantage of access
to the discount window. Without making any changes to existing deposit
insurance coverage, extending coverage to money market funds and narrow
banks would reduce the amount of deposits subject to long-term risk and
the likelihood of failure.
Prudential Regulation
Prudential regulation, as used herein, refers to government-imposed
rules designed to ensure that adequate assets stand behind the
liabilities of financial institutions. Prudential regulation is an
inherently suboptimal approach to risk because free markets are far
more efficient at pricing risk than governments. Governments cannot
avoid injecting political considerations into the underwriting of risk,
which results in inevitably inefficient risk-minimization structures.
Nonetheless, short-term social instability can cause permanent damage
to social, political and commercial institutions. In some cases,
government intervention is necessary to mitigate potentially
destabilizing fluctuations in free markets.
Under this admittedly oversimplified framework for government
intervention in the capital markets, I would argue that our payments
systems creates the kind of risk that should not be left to free market
forces. The payments system refers to the network of providers of
transactional services that enable a non-specie-based economic system
of exchange to operate. The temporary collapse of our payments system
could leave economic activity to be conducted on a strictly barter or
specie basis until the payments system was restored. The difficulty
with leaving the payments system to the mercy of free markets is that
the social and political upheaval that might result from a temporary
collapse of our payments system could turn the collapse into a long-
term event with long-term political, social and economic consequences.
On this basis, it is advisable to support the payments system with an
unconditional government guarantee of cash accounts on which the
payments is primarily based. Notwithstanding the likely inefficiencies
of such an insurance regime, they are outweighed by the potential
benefits of protecting the payments system.
One purpose that deposit insurance serves is to guarantee bank
deposits and thereby stabilize an important foundation for the payments
system. There are two difficulties with deposit insurance, however.
First, deposit insurance covers risks that are not necessarily
attendant upon the operation of cash accounts. Cash accounts can serve
as an important linchpin of the payments system without being invested
in long-term, high-risk assets such as the types of assets in which
banks typically invest deposits. Money market funds also provide an
important linchpin on the payments system, and they do so without
taking such risks.
Second, deposit insurance is exclusive to bank deposits. It is not
available to other types of cash accounts even if those accounts pose a
similar systemic threat to the payments system. When a run on money
market funds seemed imminent in late September 2008, there was no
government guarantee to prevent the run from turning into a wholesale
transfer of assets out of money market funds. With $4 trillion in
assets, such a stampede could have shut down the payments system with
potentially devastating long-term effects. The Treasury Department
prudently installed a government guarantee and halted the run. With
temporary money market fund insurance in place, the vast majority of
assets in transaction accounts are covered by a federal guarantee.
Thus, insuring money market funds and narrow banks would promote
appropriate prudential regulation that was designed to protect the
stability of our payments system without transferring unnecessary risk
to the government and taxpayers.
Prudential Regulator
The current financial crisis has exposed a persistent flaw in our
regulatory structure. Prudential oversight should be provided through a
regulatory structure that is amenable to the regulatory philosophy that
prudential oversight entails. Prudential regulators are risk averse.
Their purpose is to prevent loss. A regulator that is tasked with
protecting investors and promoting free and efficient markets, on the
other hand, will not be risk averse. The securities laws focus on full
disclosure of material information is designed to promote and reward
risk-taking based on the efficient flow of capital to its highest value
use, even when some uses entail significant risk. Permitting such risk-
taking is inimical to the essence of prudential regulation.
In more concrete terms, the SEC's roles: (1) in protecting
investors and promoting free, efficient markets, and (2) as the
prudential regulator of brokerdealers and money market funds, are in
conflict. Similarly, banking regulators' consumer protection role has
always suffered in the shadow of its primary prudential regulator role.
The SEC's and banking regulators' contradictory positions on fair value
accounting reflect this conflict. The SEC favors accurate pricing that
reflects market values; banking regulators favor pricing that will
restore investor confidence. As a prudential regulator, the SEC's
failure to properly administer net capital rules has led to
disappearance of the five largest investment banks as independent
entities and its approach to money market funds has necessitated the
intervention of a true prudential regulator, the Treasury Department,
to stop a run on money market funds. Conversely, banking regulators'
record of consumer protection has been abysmal, with their role more
often undermining consumer protection than enhancing it.
In short, the areas of financial activity that necessitate
prudential regulation should be administered by a prudential regulator.
Investor protection and free markets should be handled by a different
regulator. Although I support the creation of a single prudential
regulator in theory, I believe it would be more realistic to shift
prudential regulation to existing banking regulators and to locate
consumer protection responsibility with respect to financial products
and services with the SEC or FTC. If prudential regulation for
insurance companies were established at the federal level, a special
prudential regulator may be needed. It is not clear that the unique
characteristics of insurance liabilities would be good fit for a
prudential regulator that was responsible for other types of financial
products. Insurance products that have predominantly investment
characteristics (e.g., equity-indexed annuities), however, should be
regulated by the SEC as to sales practices and disclosure, and by the
same federal prudential regulator that would be responsible for
overseeing money market funds and banks.
Electronic Filing of Portfolios
In January 2008, my advocacy group, Fund Democracy, and the
Consumer Federation of America, Consumer Action, AFL-CIO, Financial
Planning Association and National Association of Personal Financial
Advisors petitioned the SEC to adopt a rule requiring money market
funds to file their portfolios electronically with the SEC. \3\ The
letter was motivated by our concern that the SEC's ad hoc practice of
allowing fund sponsors to bail out their money market funds before they
broke a dollar was inadequate in a time of market turmoil. The letter
proved to be, unfortunately, prescient. Within the year, a retail money
market fund broke a dollar for the first time.
---------------------------------------------------------------------------
\3\ Petition from Fund Democracy, Consumer Federation of America,
Consumer Action, AFL-CIO, Financial Planning Association and National
Association of Personal Financial Advisors, to Nancy Morris, Secretary,
U.S. Securities and Exchange Commission (Jan. 16, 2008) available at
http://www.funddemocracy.com/MMF%20Rulemaking%20Petition.pdf
---------------------------------------------------------------------------
Money market fund regulation, whether administered by the SEC or a
true prudential regulator, should include an electronic, portfolio-
filing requirement. Electronic filing would enable the regulator to
monitor, among other things, the prices at which different money market
funds are carrying the same securities. Although small pricing
discrepancies would be inevitable and no cause for concern, large
pricing discrepancies would indicate that some fund was underpricing
or, more importantly, overpricing its shares. Moreover, filings would
show the liquidity of the market for securities and thereby provide
insight into the credibility of prevailing prices in more thinly traded
issues. As stated by the SEC when it made a similar proposal 1995,
money market fund portfolio filing would enhance regulators' ability:
``to monitor money fund compliance with the federal securities laws,
target its limited onsite examination resources, and respond in the
event of a significant market event affecting money funds and their
shareholders.'' The SEC's own justification for this proposal is far
stronger today that it was twelve years ago.
Sponsor Support
The SEC has historically dealt with the risk of a money market
fund's breaking a dollar by granting no-action relief to fund sponsors
to purchase the problem assets at par, pump cash into the fund, extend
guarantees, or take other steps to restore the fund's per share net
asset value. This continues to be an appropriate tool for addressing
the risk of money market fund failure, but it has become far too
routine. The frequent granting of no-action relief for transactions
that generally violate the affiliated transaction prohibitions of the
Investment Company Act undermines the rule of law and encourages lax
oversight by fund managers.
First, the SEC should amend the rule that exempts certain of these
transactions to cover a broader range of sponsor support mechanisms.
Sponsors should then be expected to have established written procedures
that address scenarios in which their funds may need support and the
mechanisms that the fund expects to use to provide it, if any.
Second, the sponsor's rescue policy should be disclosed in its
Statement of Additional Information (a fund filing that investors can
obtain on request or on the SEC's Web site). As indicated by Fitch's
recent announcement that it intends to revise its money market fund
rating system to reflect sponsors' rescue plans, these plans have
become material aspects of a fund's stability. Banking regulators have
previously indicated that they might not permit a bank affiliate to
bail out its money market fund. This risk also should be disclosed to
investors. As discussed in the consumer groups' January 2008 letter,
the 11th hour negotiation of the terms of sponsor support between
sponsors and SEC staff behind closed doors should not be the model by
which the SEC and the fund industry manage unexpected market events.
\41\
---------------------------------------------------------------------------
\4\ If money market fund insurance is made permanent, such sponsor
support arrangements should be formalized and made mandatory. Sponsor
support of money market funds is the functional equivalent of the
equity buffer that insured banks are required to maintain under banking
regulations. In this respect, it should be noted that claims that money
market funds have no ``capital'' are misleading. Money market funds do
have capital; it is the sponsor support that has, in dozens of
instances prevented money market funds from breaking a dollar and
resulted in a record of only two failures in almost 30 years. The
problem is that the capital support is informal and voluntary.
---------------------------------------------------------------------------
Liquidation Procedures
The haphazard liquidation of certain Reserve Funds has exposed a
significant gap in the regulatory structure for money market funds. The
complete liquidation of any mutual fund, even a highly liquid money
market fund, cannot be accomplished overnight, but there should be no
delay in the distribution of some percentage of a money market fund's
assets in short order. Money market fund shareholders use these funds
as the functional equivalent of bank accounts on which they often rely
for daily living expenses. The SEC should require that money market
fund compliance manuals include procedures that set forth the manner in
which immediate redemptions can be effected in the event that
circumstances cause the suspension of regular distributions. The FDIC
generally is able to ensure that insured depositors receive a
substantial part of their funds almost immediately following the
closure of an insured bank. While it is reasonable for some money
market fund assets to be withheld pending a final resolution by a
receiver, there is no excuse for not releasing some percentage of
shareholders' accounts in short order.
Liquidity Oversight
Many of the problems underlying the current crisis result from a
failure to incorporate liquidity risk into prudential regulation.
Although money market funds present less liquidity risk because of the
short maturity, high quality and diversification of their assets, Rule
2a-7 should require that money market fund directors specifically
consider the liquidity risk posed by the fund's portfolio. Fund
directors should be required to ensure that procedures have been
adopted and implemented that are reasonably designed to ensure that the
pricing of portfolio securities has been tested against various market
failure scenarios.
MUTUAL FUNDS
Excessive Fees
Section 36(b) of the Investment Company Act, which was passed in
1970, provides that a fund director and fund manager shall have a
fiduciary duty with respect to the fees charged by the fund, and tasks
the Commission with bringing actions against directors and fund
managers who violate this duty. The Commission has never brought a case
for excessive fees. \5\ No plaintiff has ever prevailed in litigated
claim under this provision although there have been some significant
settlements.
---------------------------------------------------------------------------
\5\ I am aware of two cases that the Commission has brought under
Section 36(b), neither of which involved an excessive fees claim. See
In the Matter of American Birthright Trust Management Company, Inc.,
Litigation Rel. No. 9266, 1980 SEC LEXIS 26 (Dec. 30, 1980); SEC v.
Fundpack, Inc., No. 79-859, 1979 WL 1238 (D.D.C., Aug. 10, 1979).
---------------------------------------------------------------------------
Recent developments have made it unlikely that a section 36(b)
claim will ever survive a motion to dismiss. Defense experts often have
argued that mutual fund fees are set in a competitive marketplace and
therefore are necessarily fair under section 36(b). In a Seventh
Circuit decision, Judge Easterbrook adopted this theory, thereby
effectively repealing the Act's private cause of action. In a split en
banc opinion, Judge Posner rejected Judge Easterbrook's analysis,
arguing that markets are not always efficient. The same Seventh Circuit
also recently ruled that an Erisa fiduciary has no duty when selecting
investments for a 401(k) plan not to choose funds that charge excessive
fees. The court granted defendants' motion to dismiss even after
accepting as true, among other things, plaintiffs' allegation that the
plan sponsor had lied to plan beneficiaries about absorbing all of the
costs of administering the plan (beneficiaries actually paid part of
the costs). The Department of Labor filed an amicus brief opposing the
defendants' position in that case.
The Supreme Court has granted certiorari in the Seventh Circuit's
36(b) case. Unfortunately, the Court has been quite hostile to private
claims under the federal securities. I support many of the statutory
limits on private claims that Congress has enacted over the last 15
years, as well as some of the interpretive restrictions imposed by the
Court. But some decisions have gone too far and/or created absurd
results. There is significant risk that the Court's decision will
result in a complete evisceration of section 36(b). This will leave
mutual fund investors at the mercy of opaque fee disclosure and no
private claim against fund managers that charge excessive fees. It is
therefore imperative that Congress strengthen the fiduciary duty
standard under section 36(b) and implement long-overdue reforms in fee
disclosure requirements.
Fiduciary Duty Standard
Section 36(b) applies a fiduciary duty to directors only with
respect to fees paid to the fund manager. When a fund's excessive fees
are attributable not to fees paid to the fund manager, but to fees paid
on account of the administrative expense of operating a small fund,
this fiduciary duty is not triggered. Thus, a fund director's decision
to offer a fund with an 8 percent or 10 percent expense ratio may be
reviewable only under the toothless state law standard that section
36(b) was designed to supplement. \6\
---------------------------------------------------------------------------
\6\ A number of years ago, my research assistant was able to
identify 18 funds in Morningstar's database with expense ratios in
excess of 5 percent, yet the average management fee for the same funds
was only 1.06 percent, and only one fund's management fee exceeded 1.29
percent.
---------------------------------------------------------------------------
Congress should enact legislation that creates a fiduciary duty for
fund directors that would require, for example, that directors
affirmatively find that the fund could be a reasonable investment in
light of its investment objective, performance history and expenses. If
a fund's fees were so high so as to render the investment irrational,
the directors would have to take action to cure the problem, such as by
merging the fund into another fund with lower fees.
Fee Disclosure
As the Commission has recognized, fund fees ``can have a dramatic
effect on an investor's return. A 1 percent annual fee, for example,
will reduce an ending account balance by 18 percent on an investment
held for 20 years.'' \7\ Notwithstanding the importance of fees, ``the
degree to which investors understand mutual fund fees and expenses
remains a significant source of concern.'' \8\ The Department of Labor
has found that employee benefit ``plan participants on average pay fees
that are higher than necessary by 11.3 basis points per year.'' \9\
---------------------------------------------------------------------------
\7\ Shareholder Reports and Quarterly Portfolio Disclosure of
Registered Management Investment Companies, Investment Company Act
Release No. 25870, Part I.B (Dec. 18, 2002).
\8\ Id. (citing a joint report of the Commission and the Office of
the Comptroller of the Currency that ``found that fewer than one in
five fund investors could give any estimate of expenses for their
largest mutual fund and fewer than one in six fund investors understood
that higher expenses can lead to lower returns'').
\9\ See Fiduciary Requirements for Disclosure in Participant-
Directed Individual Account Plans, Employee Benefits Security
Administration, U.S. Department of Labor, 73 F.R. 43014, n.13 (July 23,
2008) (``DoL Proposal'').
---------------------------------------------------------------------------
In many respects, investors' lack of understanding is directly
attributable to the way in which fees are disclosed. The current
expense ratio is misleading because it excludes what can be a fund's
single largest expense: portfolio transaction costs. 12b-1 fees are
misleading because they create the impression that funds that do not
charge 12b-1 fees therefore do not incur distribution expenses. Fund
fees are disclosed in dollars based on hypothetical amounts, rather
than a shareholder's actual costs, and the location of this disclosure
makes it unlikely that investors will pay attention to this
information. Nowhere are funds required to put their fees in context by
comparing them to fees charged by index funds and comparable managed
funds. The Commission has failed to support or actively opposed reforms
designed to address each of these problems.
Portfolio Transaction Costs: The current expense ratio, which to be
accurate should be referred to as the ``partial expense ratio,''
excludes portfolio transaction costs. Portfolio transaction costs are
the costs incurred by a fund when it trades its portfolio securities.
Some portfolio transaction costs are easy to measure. For example,
commissions paid by funds are disclosed as a dollar amount in the
Statement of Additional Information, which is provided to shareholders
only upon request. Other portfolio transaction costs must be measured
indirectly, such as spread costs, but their existence and their
substantial impact on fund expenses is no less certain.
The Commission concedes that portfolio transaction costs constitute
a significant expense for fund shareholders. ``[F]or many funds, the
amount of transaction costs incurred during a typical year is
substantial. One study estimates that commissions and spreads alone
cost the average equity fund as much as 75 basis points.'' \10\ A 2004
study commissioned by the Zero Alpha Group, a nationwide network of
fee-only investment advisory firms, found that commissions and spread
costs for large equity funds, the expenses and turnover of which are
well below average, exceeded 43 percent of the funds' expense ratios. A
2004 survey by Lipper identified at least 86 equity funds for which the
total amount paid in commissions alone exceeded the fund's total
expense ratio, in some cases by more than 500 percent. The Department
of Labor expressly cited, as a significant failing of the mutual fund
expense ratio, its omission of portfolio transaction costs, which can
equal many multiples of a fund's other expenses. \11\
---------------------------------------------------------------------------
\10\ Request for Comments on Measures to Improve Disclosure of
Mutual Fund Transaction Costs, Investment Company Act Rel. No. 26313,
at Part I (Dec. 19, 2003) (``Concept Release'') (citing John M.R.
Chalmers, Roger M. Edelen, Gregory B. Kadlec, Fund Returns and Trading
Expenses: Evidence on the Value of Active Fund Management, at 10 (Aug.
30, 2001) (available at http://finance.wharton.upenn.edu/edelen/PDFs/
MF_tradexpenses.pdf). ``These estimates omit the effect of market
impact and opportunity costs, the magnitude of which may exceed
commissions and spreads.'' Id.
\11\ See DoL Proposal, supra, at n.13.
---------------------------------------------------------------------------
Notwithstanding the significance of portfolio transaction costs,
the Commission has opposed including these costs in the mutual fund
expense ratio. In a June 9, 2003, memorandum, the Commission
demonstrated that it had already prejudged the issue of the disclosure
of portfolio transaction costs. It concluded that ``it would be
inappropriate to account for commissions as a fund expense'' and
unequivocally answered the question of ``whether it is currently
feasible to quantify and record spreads, market impacts, and
opportunity costs as a fund expense. We believe that the answer is
`no.' '' \12\ Only after reaching this decision did the Commission
proceed with the formality of issuing a concept release asking for
comment on disclosure of portfolio transaction costs, apparently for
the purpose of considering any alternative other than full inclusion in
the expense ratio. \13\ Six years later, the Commission has not taken
any action on its proposal other than to include turnover ratios (an
indirect and opaque reflection of portfolio transaction costs) with the
fee table in new the summary prospectus. The expense ratio continues to
be a partial expense ratio.
---------------------------------------------------------------------------
\12\ Memorandum from Paul F. Roye, Director, Division of
Investment Management, Securities and Exchange Commission to William H.
Donaldson, Chairman, Securities and Exchange Commission, at 28 and 30
(June 3, 2003) (available at http://financialservices.house.gov/media/
pdf/02-14-70%20memo.pdf) (``Donaldson Memorandum'').
\13\ See Concept Release, supra.
---------------------------------------------------------------------------
The Commission's position is flatly inconsistent with its
responsibility to provide the information that the marketplace needs to
promote price competition. By requiring funds to use the partial
expense ratio, the Commission is effectively forcing the public to
choose funds based on the Commission's view of the proper measure of
fund costs. The Commission's decision to second-guess the market by
deciding for investors which kinds of information they are capable of
understanding contradicts basic market principles and is inconsistent
with our capitalist system of free enterprise.
Investors logically look to the Commission to provide standardized
reporting of expenses, and it is appropriate for the Commission to
provide this service. But once the Commission has provided the
important service of providing standardized information, it should
remove itself from the market-driven determination of which information
provides the best measure of a fund's true costs.
The Commission has argued that including portfolio transaction
costs might distort fund managers' behavior. As noted above, this is
not for the Commission to judge. The marketplace should decide which
expense ratio--the partial expense ratio or a total expense that
includes portfolio transaction costs--is the best measure of a fund's
costs.
Furthermore, it is the partial expense ratio that distorts fund
managers' and investors' behavior alike. The partial expense ratio
distorts fund managers' behavior by not holding them accountable for
their decisions to spend a substantial amount of fund assets on trading
securities.
As illustrated in Exhibit A, for example, the Commission believes
that investors should only be told that the expense ratio for the PBHG
Large Cap Fund is 1.16 percent, and that they should not be told that
when commissions and spread costs are included, the Fund's expense
ratio for the period shown is 8.59 percent. \14\ The true cost of that
Fund is more than seven times the amount shown in the Commission's
expense ratio. How can it be in the best interests of investors or
consistent with free market economics to require, much less permit, the
Fund to show its total costs of 1.16 percent? The partial expense ratio
is misleading because it impliedly represents, in conjunction with
other shareholder expenses listed in the fee table, the total cost of
fund ownership.
---------------------------------------------------------------------------
\14\ Exhibit A also shows that, when commissions and spread are
included, the expenses of the Strong Discovery Fund rise from 1.50
percent to 4.50 percent, the CGM Focus Fund from 1.20 percent to 4.48
percent, and the RS Mid Cap Opportunities Fund from 1.47 percent to
7.52 percent.
---------------------------------------------------------------------------
The data in Exhibit A does not reflect outliers, but randomly
selected examples from funds with more than $100 million in assets. If
smaller funds with high turnover were considered, the differentials
would be so large as to render the Commission's partial expense ratio
fraudulent. For example, Lipper reports that the Rydex Telecom Fund's
commissions for the fiscal year ending March 31, 2003, equaled 8.04
percent of assets. By applying the Zero Alpha Group study's methodology
of estimating spread costs, we can estimate that total spread costs
during that period equaled 8.75 percent of assets. Thus, whereas the
Commission tells us that the Rydex Telecom Fund's is only 1.37 percent,
its true costs are 18.16 percent, or 13 times higher. \15\ The
Commission's partial expense ratio distorts investors' behavior because
investors obviously would make different investment decisions if they
knew the true costs of owning certain funds.
---------------------------------------------------------------------------
\15\ The Lipper data show that at least 31 funds' expense ratios
would exceed 10 percent if they include commissions and spread costs.
---------------------------------------------------------------------------
The Commission's partial expense ratio also distorts managers'
behavior because it creates an incentive for them to pay for non-
execution expenses with fund commissions. Under current law, fund
managers can payer higher commissions--that is, more than it would cost
merely to execute the fund's trades--in return for non-execution
services. By paying for these non-execution services with commissions,
or what are known as soft dollars, fund managers effectively move these
costs out of the expense ratio where they belong. This enables the fund
that uses soft dollars to show a lower partial expense ratio than a
fund that does not--even if the fund managers use identical services
and have identical operating expenses. The Commission itself has
conceded that ``[t]he limited transparency of soft dollar commissions
may provide incentives for managers to misuse soft dollar services.''
\16\
---------------------------------------------------------------------------
\16\ Concept Release at Part III.A, supra.
---------------------------------------------------------------------------
Furthermore, the nondisclosure of portfolio transaction costs
exacerbates the conflict of interest that is inherent in the payment of
soft dollars. As the Commission has recognized,
[s]oft dollar arrangements create incentives for fund advisers
to (i) direct fund brokerage based on the research provided to
the adviser rather than the quality of execution provided to
the fund, (ii) forego opportunities to recapture brokerage
costs for the benefit of the fund, and (iii) cause the fund to
overtrade its portfolio to fulfill the adviser's soft dollar
commitments to brokers. \17\
---------------------------------------------------------------------------
\17\ Donaldson Memorandum, supra, at 36. Regarding directed
brokerage, the Commission recently stated: ``We believe that the way
brokerage has been used to pay for distribution involves unmanageable
conflicts of interest that may harm funds and fund shareholders.''
Prohibition on the Use of Brokerage Commissions to Finance
Distribution, Investment Company Act Rel. No. 26356 at Part II (Feb.
24, 2004).
The continued concealment of portfolio transaction costs permits
the soft dollar conflict to operate virtually unchecked by market
forces, whereas including portfolio transaction costs in a total
expense ratio would, at least, permit the marketplace to judge the
efficacy of soft dollar arrangements. If Congress does not take steps
to eradicate soft dollars, at least it can require that these costs be
disclosed so that the market can reach its own judgments regarding
their efficacy.
Dollar Disclosure of Fees: Under current disclosure rules, funds
are not required to disclose to investors how much they pay in fees.
Many other financial services documents show investors exactly how much
they are paying the service provider, including bank statements,
insurance bills, credit card statements, mortgage loans and a host of
other documents. But mutual funds provide only an expense ratio (and a
partial one, at that, see supra) and the dollar amount of a
hypothetical account.
Congress should require that funds provide individualized dollar
disclosure of fund expenses in shareholder statements, as recommended
by the Government Accounting Office \18\ and proposed for employee
benefit plans by the Department of Labor. \19\ This requirement is
necessary for two reasons. First, although the expense ratio is
appropriate for providing comparability across different funds, it does
not pack the same import as a dollar amount. Providing investors with
the amount in dollars that they actually spent will give concrete form
to an indefinite concept and make investors consider more fully the
costs of different investment options.
---------------------------------------------------------------------------
\18\ Government Accounting Office, Mutual Funds: Information On
Trends In Fees and Their Related Disclosure (March 12, 2003).
\19\ See DoL Proposal, supra.
---------------------------------------------------------------------------
Second, placing the dollar amount of expenses in the shareholder
statement will direct shareholders' attention to the actual costs of
fund ownership. No document is more likely to be read than a
shareholder statement that shows the value of the shareholder's account
and transaction activity during the period. Whereas the prospectus and
shareholder report typically go directly from the mailbox to the trash
can, even the most uninformed investors normally open their statements
to check on the status of their accounts. There is no better way to
draw shareholders' attention to the costs of investing than to require
that the dollar amount of fees for the period be disclosed next to the
value of the investor's account.
Some members of the fund industry have opposed informing investors
about the actual costs of their fund investments on the grounds that
doing so would be too costly and might mislead investors. It appears
that MFS Investment Management, one of the largest mutual fund managers
in America, disagrees. MFS offers to include actual dollar disclosure
in investor statements, which undercuts industry arguments that
providing this information is economically infeasible. The Department
of Labor has proposed to require dollar disclosure of fees for plan
participants and the Government Accountability has recommended that the
SEC do the same.
The Commission opposes disclosure of shareholders' actual costs and
opposes including dollar disclosure in shareholder statements. The
Commission concluded its consideration of a proposal some years ago to
require funds to disclose individualized costs in shareholder
statements by expressly rejecting both concepts. Instead, the
Commission decided to require disclosure of the hypothetical fees paid
on a $1,000 account in the shareholder report, despite the facts that
the hypothetical fees paid on a $10,000 account are already disclosed
in the prospectus, and shareholders who most need to have their
attention directed to the fees that they pay are least likely to read
the shareholder report. In view of the Commission's, express opposition
to effective disclosure of actual fees paid by shareholders,
shareholders will receive disclosure of their actual fees in
shareholder statements only if Congress requires funds to provide that
information.
Fee Comparisons: Congress should take additional steps to promote
price competition in the mutual fund industry by requiring that funds
disclose fees charged by comparable funds and, for managed funds, the
fees charged by index funds. Without any context, current fee
disclosure provides no information about whether a fund's fees are
higher or lower than its peers. Current disclosure rules also do not
show the premium paid to invest in a managed funds as opposed to an
index fund. Requiring comparative information in the fee table would
enable investors to consider a fund's fees in context and evaluate how
they compare to fees across the industry.
Distribution Fees: The Commission currently requires that 12b-1
fees be disclosed on a separate line that describes those fees as
``distribution fees.'' It does not require that the fee table show the
amount spent on distribution by the fund manager out of its management
fee. This is inherently misleading, as investors often use the presence
of 12b-1 fees as a negative screen that they use to avoid paying any
distribution fees. In fact, investors in non-12b-1 fee funds may
actually pay as much or more in distribution expenses than some
investors in 12b-1 fee funds. \20\
---------------------------------------------------------------------------
\20\ In 1999, Paul Haaga, Chairman of the Investment Company
Institute and Executive Vice President of the Capital Research and
Management Company, stated at an SEC roundtable: ``the idea that
investors ought to prefer the funds that don't tell what they're
spending on distribution over the ones that do is nonsense. You know,
if you're spending money on distribution, say it. If you're not pending
money on distribution don't say it; but don't pretend that there are no
expenses there for a fund that doesn't have a 12b-1 plan.'' Conference
on the Role of Investment Company Directors, Washington, D.C. (Feb. 23
& 24, 1999) (Haaga was not ICI Chairman at this time).
---------------------------------------------------------------------------
Congress should overrule the Commission's position and require
that, if distribution fees are stated separately in the fee table, they
must reflect all distribution expenses paid by a fund, directly or
indirectly. Alternatively, Congress should require that fund expenses
be displayed in a pie chart that shows how much of a fund's fees were
spent on each type of service. The Commission's current fee table is
misleading and understates the amount of fund assets spent on
distribution.
Disclosure of Brokers' Compensation: For virtually all securities
transactions other than purchases of mutual fund shares, investors
receive a transaction confirmation that shows how much the broker was
paid in connection with the transaction. Permitting brokers to hide
their compensation on the sale of mutual funds has spawned a Byzantine
and harmful array of selling arrangements, including revenue sharing
(also known as payments for shelf space), directed brokerage, and non-
cash compensation. Mutual fund shareholders should be entitled to
receive the same information as other investors in securities in the
form of full disclosure of their brokers' compensation on fund
transaction confirmations. Such disclosure also should show how
breakpoints applied to the transaction, as well as any special
compensation received by brokers for selling particular funds.
Brokers also should be required to provide, at or before the time
the investor places the order, an estimate of compensation to be
received by the broker in connection with the transaction and the total
costs of investing in the fund. When buying a house, purchasers are
provided with an estimate of their total closing costs before making a
final decision. As discussed immediately above, however, fund
shareholders do not even receive a final statement of their actual
costs, much less an up-front estimate of such costs.
In January 2004, the Commission proposed to require brokers to
provide, both at the point-of-sale and in the transaction confirmation,
disclosure of the costs and conflicts of interest that arise from the
distribution of mutual fund shares. \21\ More than 5 years later, the
Commission has failed to take final action on its proposal. Congress
should require that the SEC take final action on disclosure
requirements that will result in brokers' customers receiving
disclosure of the broker's economic incentives in the transaction.
---------------------------------------------------------------------------
\21\ Confirmation Requirements and Point of Sale Disclosure
Requirements for Transactions in Certain Mutual Funds and Other
Securities, and Other Confirmation Requirement Amendments, and
Amendments to the Registration Form for Mutual Funds, Investment
Company Act Rel. No. 26341 (Jan. 29, 2004).
---------------------------------------------------------------------------
Distribution Arrangements
12b-1 Fees
When Congress enacted the Investment Company Act of 1940, it
expressly prohibited fund managers from using fund assets to finance
the distribution of the fund's shares. Section 12(b) of the Act
recognized the inherent conflict of interest between the manager's
desire to increase fund assets in order to increase its fees on the one
hand, and the fund's desire to hold down costs on the other hand.
Unfortunately, the policy underlying Section 12(b) has long been
abandoned, as fund assets are used for a wide range of distribution
expenses that benefit fund managers at the expense of fund
shareholders.
The policy of separating the product from its distribution was
first abandoned by the Commission when, after a prolonged review, it
adopted Rule 12b-1 in 1980. In the 1970s, mutual funds experienced
periods of net redemptions that prompted fund managers to lobby the
Commission to permit the use of fund assets to finance the distribution
of the funds' shares. Fund managers argued that net redemptions
resulted in increased costs and that the financing of distribution by
the fund would help reduce or eliminate net redemptions.
The Commission initially rejected these arguments, but ultimately
relented, provided that certain conditions were observed. For example,
the Commission required that the fund's independent directors approve
the 12b-1 plan. Among the factors that the Commission said a fund's
directors should consider when evaluating whether to adopt or renew a
12b-1 plan was the plan's effectiveness in remedying the problem that
it was designed to address, i.e., increased costs resulting from net
redemptions.
The Commission's most significant concern regarding 12b-1 fees was
the conflict of interest that they created between the fund and its
adviser. The Commission feared that 12b-1 fees would result in higher
advisory fees and the fund's adviser would not share the benefits of
asset growth. \22\ Some would argue that this is precisely what has
happened, with any growth-based economies of scale realized from 12b-1
fees being pocketed by fund managers and not shared with fund
shareholders.
---------------------------------------------------------------------------
\22\ See Donaldson Memorandum, supra, at 70-71 (``When a fund
bears its own distribution expenses, the fund's investment adviser is
spared the cost of bearing those expenses itself, and the adviser
benefits further if the fund's distribution expenditures result in an
increase in the fund's assets and a concomitant increase in the
advisory fees received by the adviser.'').
---------------------------------------------------------------------------
Of course, this analysis goes primarily to the use of 12b-1 fees
for marketing the fund, which is what Rule 12b-1 was intended to
permit. It does not address the ways in which 12b-1 are actually used
today and that were wholly unanticipated by the Commission when Rule
12b-1 was adopted. According to and Investment Company Institute
report, only 5 percent of 12b-1 fees are spent on advertising and sales
promotion, whereas 63 percent of 12b-1 fees are spent on broker
compensation. \23\
---------------------------------------------------------------------------
\23\ Use of Rule 12b-1 Fees by Mutual Funds in 1999, Investment
Company Institute, 9 Fundamentals 2 (April 2000). Funds spend the other
32 percent of 12b-1 fees on administrative services. Id.
---------------------------------------------------------------------------
The use of fund assets to compensate brokers is precisely what
Section 12(b) was intended to prohibit. This practice puts the fund
squarely in the position of underwriting its own securities. The fund's
assets are used to incentivize brokers to recommend the fund over
competing funds. The lesser the quality of the fund, the greater the
pressure on the fund and its manager to pay brokers more to sell the
fund.
This irreconcilable conflict is mirrored on the distribution side
of the business. When brokers are paid by the funds, rather than their
customers, they have an incentive to recommend the fund that offers the
biggest payout, rather than the fund that will provide the best
investment for their customers. \24\ There is another incentive for
brokers to favor arrangements whereby they are compensated by funds,
and that is the fact that the compensation from the fund is not
transparent. Whereas the payment of a front-end load is relatively
evident to the investor, the payment of a 12b-1 fee is not. It is even
less clear that the already opaque 12b-1 fee is ending up in the
broker's pocket. For this reason, brokers and investors have begun to
favor classes of fund shares where the broker is compensated by the
fund, regardless of whether that class is in the best interests of
shareholders. \25\
---------------------------------------------------------------------------
\24\ See Laura Johannes and John Hechinger, Conflicting Interests:
Why a Brokerage Giant Pushes Some Mediocre Mutual Funds, Wall St. J.
(Jan. 9, 2004); see also In the Matter of Morgan Stanley DW Inc.,
Exchange Act Rel. No. 48789 (Nov. 17, 2003).
\25\ See Complaint, Benzon v. Morgan Stanley, No. 03-03-0159 (M.D.
Tenn.). The SEC has banned the use of brokerage as compensation for
fund brokerage.
---------------------------------------------------------------------------
Thus, the Commission has created a distribution compensation
structure that is directly at odds with the interests of investors and
the Investment Company Act. Rather than tying brokers' compensation to
their relationships with their customers, where the Investment Company
Act requires that it be placed, the Commission has tied brokers'
compensation to their relationships with the funds, where the
Investment Company Act expressly forbade its placement.
Congress should reaffirm the supremacy of Section 12(b) and
prohibit funds from compensating brokers for selling fund shares.
Although this will necessarily entail the repeal of Rule 12b-1, it will
in no way limit the ways in which investors can choose to pay their
brokers. It will simply require that however brokers are compensated--
through a front-end load, back-end load, level-load, or any combination
thereof--they are compensated by their customers, not by the funds.
Thus, if a customer chooses to pay his broker on an installment basis,
at 0.50 percent each year, for example, that amount would be paid by
the customer directly or deducted from his fund account.
One might argue that, to maintain perfect legislative coherence,
Congress should also prohibit fund managers from paying for general
marketing services that are not connected to specific sales. I
disagree. The conflict is substantially reduced in this situation
because the fund manager's and the fund's interests are generally
aligned. General marketing payments do not create a direct incentive
for brokers to favor one fund group over another. General marketing
does what advertising for decades has been shown to do: promote
competition. Indeed, by locating these payments in the management fee,
the manager will be spending its own money and accordingly will have an
incentive to minimize costs. With an express requirement that
independent fund directors evaluate the efficacy of fund manager
expenditures on marketing and determine that resulting economies have
been shared with fund shareholders, expressly permitting fund managers
to use the management fee to pay for marketing would be appropriate.
Revenue Sharing
Over the last two decades, a compensation practice has evolved that
strikes at the heart of the principle of full disclosure of conflicts
of interest. Known as ``revenue sharing,'' this practice involves the
payment of a part of fees collected by a mutual fund manager to a third
party in return for administrative and/or distribution services.
Notwithstanding the somewhat pejorative term ``revenue sharing,'' there
is nothing necessarily inappropriate about the practice itself.
Broadside critiques of revenue sharing are off base. Revenue sharing
primarily reflects a compensation structure that can be a more
efficient method of compensation than direct charges by each service
provider to the client. Indeed, 12b-1 fees are functionally a kind of
revenue sharing that are subject to enhanced (but still inadequate, see
supra) disclosure requirements.
That being said, the regulation and practice of revenue sharing
disclosure has been abysmal. Revenue sharing payments are generally
included in the total fees charged by a fund. Unlike 12b-1 fees, they
are not, and are not required to be, broken out separately in the
mutual fund fee table. More importantly, they are not necessarily
disclosed by the service provider that receives them. Revenue sharing
constitutes compensation to service provider that is not part of the
fees charged directly to the client, so the client often is unaware of
the service provider's economic incentive to sell the fund. When a
broker recommends funds to clients, the broker does not disclose, and
has not been required to disclose by the SEC or FINRA, that the broker
will receive different amounts of revenue sharing payments depending on
the fund purchased. The revenue sharing payments are made under the
table; this blatant conflict of interest goes undisclosed.
This is a significant problem in the context of brokers' mutual
fund sales. The SEC and FINRA continue to defend a suitability standard
for brokers that does not require full disclosure of conflicts of
interest, even when the broker is providing individualized investment
advice to the client (as opposed to acting solely as a salesperson).
This means that brokers can recommend funds that are ``suitable''
without disclosing that they are receiving higher revenue sharing
payments from that fund's manager than they would receive from the
manager of a more suitable fund. The fees are not trivial. One SEC
settlement involving revenue sharing payments revealed that brokers
were receiving payments equal to 25 percent of the fund advisory fee in
revenue sharing payments on every sale of that fund's shares. It is
inexcusable that brokers are not required to disclose this payment
differential to their clients.
Unlike brokers subject only to a suitability standard, fiduciaries
generally have been required to disclose revenue sharing to their
clients. In SEC v. Capital Gains Research Bureau, the Supreme Court
held that that an investment adviser, as a fiduciary, was required to
disclose all material conflicts of interest to clients. Courts have
generally applied this principle to the disclosure of revenue sharing
payments on the ground that this information would be of importance to
advisory clients. The Seventh Circuit recently held, however, that an
Erisa fiduciary has no obligation to disclose revenue sharing payments
to beneficiaries as long as the total fees being paid are disclosed.
This is a truly remarkable position, especially in the wake of recent
legislation that permits conflicted persons to provide investment
advice to 401(k) plan participants on the condition that their
compensation be the same regardless of the investment option selected.
In other words, while Congress has been addressing the conflicted
advice problem by flatly prohibiting differential compensation, the
Seventh Circuit has decided not only that differential compensation can
be received by an Erisa fiduciary, it does not even need to be
disclosed.
Both the SEC and FINRA have proposed rules that, depending on their
final form, would require the disclosure of differential compensation.
These rules, like many important investment management initiatives,
have been pending for years. While the SEC has been paralyzed with
indecision, state attorneys general have sued fund managers and brokers
for their failure to disclose revenue sharing arrangements in their
prospectuses and to their clients. The SEC's failure to take a position
one way or the other has created an unpredictable patchwork of
regulation that benefits no one, especially not those who appropriately
use revenue sharing in their compensation structures. And the SEC's
failure to require the disclosure of revenue sharing payments has
allowed the practice to flourish.
Congress should not continue to wait for regulators to recognize
the obvious policy imperative of requiring full disclosure of conflicts
of interest to financial services clients. In the last six years, a
number of bills have been proposed that would, in one form or another,
require the disclosure of revenue sharing and other forms of
differential compensation. Congress should act promptly to enact some
form of this legislation. The committee reports should make it clear
that payments that create potential conflicts of interest must be
disclosed and that the legislation is intended to overrule the Seventh
Circuit's Deere decision.
Misleading Fund Share Classes
Mutual funds often offer several classes of shares that reflect
different ways of paying for distribution services. Typically, Class A
shares carry a front-end load, Class B shares a back-end load, and
Class C shares carry a level load. An investor is usually better off
buying Class A shares if he intends to hold his shares for the
longterm, and Class C shares if he may sell in the short-term. When
Class B shares are best option, it is for the shareholder who holds for
the mid-term. In some cases, however, there is virtually no shareholder
for whom Class B shares are the best option.
The Commission does not prohibit funds from offering Class B
shares, even when there is no shareholder for whom Class B shares could
be the best investment option. The Commission even rejected a rule
amendment that would have required that funds illustrate in the
prospectus the relative costs of each class of shares. Following the
Commission's lead, a federal court held in January 2004 that, even
assuming that there was no rational investor for whom Class B shares
would be the best investment, the fund had no duty to disclose this
fact in the prospectus. \26\
---------------------------------------------------------------------------
\26\ See Benzon v. Morgan Stanley, 2004 WL 62747 (M.D. Tenn.).
---------------------------------------------------------------------------
It is unconscionable that under current Commission positions a fund
can offer a class of shares that would not be the best investment for
any rational investor. Congress should require that multi-class funds
illustrate, in a graphic format, the costs of investing in different
classes over a 15-year period. In addition, Congress should require
that the fund's independent directors find, subject to a fiduciary duty
as described above, that each class of shares offered could be a
reasonable investment alternative.
Fund Advertising
Throughout the late 1990s, the Commission frequently berated the
fund industry for misleading investors by advertising short-term
performance. Funds with short life-spans routinely advertised one-year,
sometimes even 2- and 3-year annualized investment returns in excess of
100 percent. With the crash of the stock bubble in 2000, the
Commission's concerns were validated, as many of these funds
experienced huge losses, in some cases in excess of 70 percent of their
value.
The Commission's actions have not reflected its words, however. In
September 2003, the Commission adopted advertising rules that utterly
failed to address the very problems that it had identified in the late
1990s. \27\ The rules require funds to provide a telephone number or
web address where current performance information is available, as if
the problem with short-term performance was that it wasn't current
enough. The Commission also required that the text in fund ads include
the statement that ``current performance may be higher or lower than
the performance data quoted.''
---------------------------------------------------------------------------
\27\ Amendments to Investment Company Advertising Rules,
Investment Company Rel. No. 26195 (Sep. 29, 2003).
---------------------------------------------------------------------------
Fund advertisements posted following market declines in 2000-2002
demonstrate the inadequacy of the Commission's new rules. After three
years of negative returns, stock funds had a banner year in 2003. Many
of those funds are now advertising their stellar one-year performance
without any disclosure of their poor returns in 2000, 2001, and 2002.
Because they are required only to show their one-, five- and ten-year
returns, the negative returns of 2000 to 2002 are hidden from view. The
ads create a misleading impression by showing the outsized returns of
2003 without any mitigating disclosure of the down years that preceded
them and the performance volatility that those years' returns
illustrate.
For example, one ad shows SEC-mandated performance for four funds,
each of which experienced superior returns in 2003, but experienced
losses or substantially lower performance in each year from 2000 to
2002. As illustrated in the table below, the disclosure of each fund's
annual performance in the years preceding 2003 would have presented a
very different, far more accurate picture. The Commission's rulemaking
has done nothing to prevent such misleading ads, which have appeared
routinely in business and personal finance magazines in the first few
months of this year.
----------------------------------------------------------------------------------------------------------------
Disclosed* Not Disclosed**
Funds -------------------------------------------------------------------------------
2003 2002 2001 2000
----------------------------------------------------------------------------------------------------------------
Fund #1......................... 51.68% (21.27%) (7.56%) (18.10%)
Fund #2......................... 42.38% (9.37%) (12.99%) (8.96%)
Fund #3......................... 23.36% (20.44%) (3.74%) 12.25%
Fund #4......................... 29.96% (17.16%) (5.02%) 8.54%
----------------------------------------------------------------------------------------------------------------
* Source: Business 2.0 (March 2004).
** Source: Fund Prospectuses.
The Commission's rulemaking also did nothing to address the problem
of the disconnect between the advertised performance of funds and the
actual returns experienced by shareholders. As confirmed by a recent
DALBAR study, ``[i]nvestment return is far more dependent on investment
behavior than on fund performance.'' \28\ DALBAR found that the average
equity fund investor earned 2.57 percent annually over the last 19
years, in comparison with the S&P 500's 12.22 percent annual return
during the same period. This translates into a cumulative return for
the S&P 500 of 793.34 percent from 1984 to 2002, compared with equity
fund investors' actual cumulative return of 62.11 percent during the
same period.
---------------------------------------------------------------------------
\28\ DALBAR, Quantitative Analysis of Investor Behavior at 2
(2003).
---------------------------------------------------------------------------
These stunning and disheartening data illustrate, in part, a
failure of investor education and individual choice. Investors have
consistently chased the best performing funds just before they crashed,
and dumped the worst performing funds just before they recovered. This
sell-high, buy-low mentality is only encouraged by the Commission's
current approach to fund performance advertising, which permits funds
to present outsized returns with no meaningful caveats regarding their
volatility and the likelihood that performance will soon revert to the
mean. \29\
---------------------------------------------------------------------------
\29\ Notably, the Commission requires that the prospectus include
a bar chart that shows a fund's return for each of the preceding ten
years. If such a disclosure is necessary to make the prospectus not
misleading, it is unclear why the same reasoning is not applicable in
the context of a fund advertisement.
---------------------------------------------------------------------------
Not only do current rules fail to require meaningful disclosure
about the volatility of fund returns, but they also fail to place
outsized, one-year returns in the context of the market as a whole. To
illustrate, the performance of the S&P 500 for 2003 was 28.68 percent,
which puts the 51.68 percent return of the Fund cited above in a light
very different (albeit still positive) from one in which the
performance data stands alone. The Fund's advertised ten-year return of
10.58 percent would tell a different story if it were required to be
juxtaposed against the S&P 500's 11.07 percent ten-year return.
The Commission also has recognized the need for investment returns
to be considered in the context of fees, yet its rules do virtually
nothing to benefit investors in this respect. In its proposing release,
the Commission promised that its new rule would ``ensure that fund
advertisements remind fund shareholders about the availability of
information about fund charges and expenses.'' \30\ Yet the final rule
required only that fund advertisements refer investors to the
prospectus for consideration of fund expenses, among other things. \31\
In contrast, the NASD has proposed that fund advertisements include a
box that shows both the fund's maximum sales charge and its expense
ratio. \32\
---------------------------------------------------------------------------
\30\ Proposed Amendments to Investment Company Advertising Rules,
Investment Company Rel. No. 25575, Part II.C (May 17, 2002).
\31\ Amendments to Investment Company Advertising Rules, supra.
\32\ See Disclosure of Mutual Fund Expense Ratios in Performance
Advertising, National Association of Securities Dealers (Jan. 23,
2004).
---------------------------------------------------------------------------
Congress should require that fund advertisements include all
information necessary to make the information presented not misleading.
This must include, at a minimum, investment returns for each individual
year where such returns differ materially from fund's one-year
performance, disclosure of the fund's total expense ratio (i.e.,
including the fund's portfolio transaction costs) and sales charges,
and the performance and expenses of a comparable index fund.
Soft Dollars
The term ``soft dollars'' generally refers to brokerage commissions
that pay for both execution and research services. The use of soft
dollars is widespread among investment advisers. For example, total
third-party research purchased with soft dollars alone is estimated to
have exceeded $1 billion in 1998. \33\ An executive with American
Century Investment Management has testified that the research component
of soft dollar commissions costs six times the value of the execution
component. \34\
---------------------------------------------------------------------------
\33\ Inspection Report on the Soft Dollar Practices of Broker-
Dealers, Investment Advisers and Mutual Funds, Securities and Exchange
Commission, at text accompanying note 1 (Sep. 22, 1998).
\34\ Testimony of Harold Bradley, Senior Vice President, American
Century Investment Management, before the Subcommittee on Capital
Markets, Insurance and Government Sponsored Enterprises, Committee on
Financial Services, U.S. House of Representatives, at 5 (Mar. 12,
2003).
---------------------------------------------------------------------------
Soft dollar arrangements raise multiple policy concerns. The
payment of soft dollars by mutual funds creates a significant conflict
of interest for fund advisers. Soft dollars pay for research that fund
advisers would otherwise have to pay for themselves. Advisers therefore
have an incentive to cause their fund to engage in trades solely to
increase soft dollar benefits. \35\
---------------------------------------------------------------------------
\35\ Id. at 2 (the statutory safe harbor permitting soft dollars
arrangements ``encourages investment managers to use commissions paid
by investors as a source of unreported income to pay unreported
expenses of the manager'').
---------------------------------------------------------------------------
Soft dollar arrangements normally would be prohibited by the
Investment Company Act because they involve a prohibited transaction
between the fund and its adviser. \36\ Section 28(e) of the Securities
Exchange Act, however, provides a safe harbor from the Investment
Company Act for soft dollar arrangements as long as the brokerage and
research services received are reasonable in relation to the amount of
the commissions paid.
---------------------------------------------------------------------------
\36\ See Investment Company Act Section 17(e); Inspection Report
at 38, supra.
---------------------------------------------------------------------------
The conflicts of interest inherent in soft dollar arrangements are
exacerbated by current disclosure rules. The amount of fund assets
spent on soft dollars is not publicly disclosed to shareholders, so
they are unable to evaluate the extent, and potential cost, of the
adviser's conflict.
Current disclosure rules reward advisers for using soft dollars
because this practice creates the appearance that a fund is less
expensive. The expense ratio does not include commissions, which gives
advisers an incentive to pay for services with soft dollars, thereby
enabling them to lower their management fees and the fund's expense
ratio. Advisers can effectively reduce their expense ratios by spending
more on soft dollars, while the fund's actual net expenses remain
unchanged.
Finally, current disclosure rules may encourage excessive spending
on soft dollars. Advisers would tend to spend less on soft dollars if
they knew that they would be held publicly accountable for their
expenditures.
The Commission has frequently recognized but declined to address
the problem of soft dollars. As discussed above, the Commission is
opposed to including portfolio transaction costs in funds' expense
ratios, which would have the benefit of enabling the market to
determine for itself the efficacy of soft dollar arrangements. The
Commission previously proposed a rule that would require that soft
dollars costs be quantified, but decided against adopting it. \37\ When
the Commission staff last evaluated soft dollar arrangements in 1998,
it concluded that additional guidance was needed in a number of areas.
\38\ For example, the staff found that many advisers were treating
basic computer hardware--and even the electrical power needed to run
it--as research services qualifying under the Section 28(e) safe
harbor. \39\ The staff recommended that the Commission issue
interpretive guidance on these and other questionable uses of soft
dollars, but it has failed to do so.
---------------------------------------------------------------------------
\37\ Donaldson Memorandum at 13-17, supra. Fidelity recently
recommended that the Commission reconsider its decision not to require
the quantification of soft dollar costs. Ann Davis, Fidelity Wants
Trading Costs To Be Broken Down, Wall Street Journal (Mar. 15, 2004).
\38\ Inspection Report on the Soft Dollar Practices of Broker-
Dealers, Investment Advisers and Mutual Funds, Securities and Exchange
Commission, at text accompanying note 1 (Sep. 22, 1998) (``Section
28(e) Report'').
\39\ Id. at Section V.C.4.
---------------------------------------------------------------------------
In fact, the only formal action that the Commission has taken in
recent years is to expand the use of soft dollars. In December 2001,
the Commission took the position that the safe harbor should apply to
markups and markdowns in principal transactions, although Section 28(e)
expressly applies only to ``commissions.'' \40\ This position directly
contradicts not only the plain text of the statute, but also the
position taken by the Commission in 1995 that section 28(e) ``does not
encompass soft dollar arrangements under which research services are
acquired as a result of principal transactions.'' \41\ Although the
Commission has, once again, suggested that intends to narrow the scope
of soft dollars, its recent history suggests that Congressional action
is necessary. In any case, the Commission lacks the authority to ban
soft dollars.
---------------------------------------------------------------------------
\40\ Commission Guidance on the Scope of Section 28(e) of the
Exchange Act, Exchange Act Rel. No. 45194 (Dec. 27, 2001).
\41\ Investment Advisers Act Release No. 1469 (February 14, 1995).
---------------------------------------------------------------------------
There is no better evidence that the time has come to ban soft
dollars than the recognition of the insidious nature of this practice
by members of the fund industry. In addressing the fact that soft
dollars enable fund managers to use the fund's money to pay for
research used by the manager, the independent chairman of the Putnam
Funds has stated that ``[t]he best decisions get made when you buy
services with your own money.'' \42\ Similarly, MFS' chairman, Robert
Pozen,
---------------------------------------------------------------------------
\42\ Id. (quoting John Hill).
sees the soft-dollar funnel as a lucrative one for brokers, but
one that hides the true cost of such services to shareholders.
``It's all camouflaged,'' said Mr. Pozen, a former associate
general counsel of the SEC. Now, he added, ``If we want
something, if we think it's valuable, we will pay cash.'' \43\
---------------------------------------------------------------------------
\43\ Id.
A Fidelity executive has acknowledged the pro-competitive advantage
of a ban on soft dollars, stating: ``[w]e don't rule out a competitive
environment through which all research is acquired through cash rather
than commissions.'' \44\
---------------------------------------------------------------------------
\44\ Landon Thomas, Jr., Mutual Fund Tells Wall Street It Wants `
la Carte Commissions, New York Times (Mar. 16, 2004).
---------------------------------------------------------------------------
The difficulty for fund firms, however, is that without a statutory
ban on soft dollars they may suffer a competitive disadvantage MFS has
estimated that paying for its own research will reduce its advisory
fees. \45\ Fidelity has estimated that of the $1.1 billion in
commission it paid in 2003, $275 million paid for soft dollar research.
\46\ It is unrealistic to expect these fund managers to maintain the
high road at the expense of reduced advisory fees, while other fund
managers continue to pay their own research expenses through soft
dollars rather than out of their own pockets.
---------------------------------------------------------------------------
\45\ MFS Ends `Soft Dollar' System, supra.
\46\ Fidelity Wants Trading Costs To Be Broken Down, supra.
---------------------------------------------------------------------------
Fund Names and Investor Expectations
The recent collapse of the stock market has exposed a significant
gap in the regulation of mutual fund names. The average investor will
reasonably assume that funds will invest consistent with their names,
but mutual fund rules do not require that funds honor these
expectations.
To illustrate, one would expect a Target Date 2010 Fund to be
designed to fit the needs of someone who planned to retire at age 65 in
2010. Such a fund would invest in mix of stocks and bonds. The
investment of stocks carries higher risk, but this risk is necessary to
provide the growth potential needed by someone who may live 30 or more
years after retirement. The fixed income securities provide stability
to ensure that assets that will be needed for living expenses in the
near term are not exposed to risk. There is no definitive asset
allocation between stocks and fixed income securities in which a Target
Date 2010 Fund should invest, and one could not argue that under no
circumstances would it be appropriate for a 65-year-old retiree to have
an 80 percent stock / 20 percent bond mix, but such a mix would fall
well outside the generally expected asset allocation of a Target Date
2010 Fund.
Mutual fund disclosure rules would allow a Target Date 2010 Fund to
adopt such an 80 percent / 20 percent asset allocation. Notwithstanding
that the Fund's name suggests a substantially lower stock allocation,
the description of the Fund's investment objectives and style in its
prospectus could correct this misimpression and investors would be
expected to have read and understood such clarifying disclosure. Under
current prospectus liability rules, the true nature of the Fund's
aggressive asset allocation strategy could even be omitted from the
summary of its investment objectives and style in the summary
prospectus as long as corrective disclosure appeared elsewhere in the
full prospectus. (It is likely that some courts would find that even
corrective disclosure buried in the Fund's Statement of Additional
Information, which is delivered to investors only upon request, would
be a sufficient defense for prospectus liability purposes.) Thus,
investors that expect the stock allocation suggested by the name of the
Target Date 2010 Fund to be substantially lower than 80 percent and do
not carefully scrutinize other fund disclosure documents will be
subject to more risk than they expected. For example, a 45 percent
decline in the stock market would result in a 36 percent decline in the
value of their Fund shares, when they might have expected an 18 percent
or 22.5 percent based on a 40 percent or 50 percent stock allocation.
\47\
---------------------------------------------------------------------------
\47\ Tom Laurcella, For Retirement, `One Size' Isn't Always a Good
Fit, Wall. St. J. (Mar. 2, 2009) available at http://online.wsj.com/
article/SB123549381087960625.html (``A typical fund for an investor
aiming to retire 20 years from now might have at least 80 percent in
stocks. By the time the retirement date approaches, most funds
typically have less than 40 percent in stocks.'').
---------------------------------------------------------------------------
It is helpful to consider a recent example of this problem. A
particular Target Date 2010 Fund has been criticized for declining 38
percent in value, but this decline is consistent with its aggressive
asset allocation. The fourth page of the fund's prospectus (for the
relatively assiduous investor) states that each retirement fund:
is managed to the specific year of planned retirement included
in its name (the `retirement date'). The Strategies' asset
mixes will become more conservative each year until reaching
the year approximately fifteen years after the retirement date
(the `target year') at which time the asset allocation mix will
become static.
At this point, the prospectus has only reinforced the expectation
that the 2010 fund's asset allocation strategy will reflect a stock
allocation in the range of 40 percent to 50 percent. Under current law,
this disclosure by no means created an expectation on which investors
could actionably rely. Indeed the same paragraph includes a state that
15 years after retirement the static allocation would be: 27 percent
short-term bonds, 37.5 percent other fixed-income securities, 25
percent equities and 10 percent real estate investment trusts
(``REITs''). From this, a very attentive investor could assume that the
stock allocation at retirement would be fairly aggressive. The fifth
page of the prospectus includes table that shows an expected allocation
of approximately 77 percent 30 of the 2010 fund's assets to equities,
REITS and high-yield debt in the year before retirement.
The problem is that this fund's allocation is inconsistent with
what many investors will expect from a Target Date 2010 Fund. There is
no reason that this fund's sponsor should not be allowed to offer a
fund for retirees who wish to adopt an aggressive asset allocation. In
combination with other investments the retiree might hold, the
retiree's overall asset allocation might fall within the more typical
40 percent to 50 percent range. But, in the words of a Fidelity
executive, something called a ``target-date'' fund should follow a
``one size fits most'' strategy, and this fund fails that test.
This problem is not limited to target-date funds. In some 529
plans, there are asset allocations designed for children expected to
need the funds for college within one or two years that experienced
substantial losses. These losses were inconsistent with the investment
performance range one would expect from a conventionally constructed
portfolio for such a short time horizon. Some of these 529 plans
invested in bond funds that included ``short-term'' in the fund's name,
but their investment returns fall well outside of the variance one
would associate with short-term bond funds. Many ``short-term bond''
funds held outside of 529 plans have produced abnormally high losses.
To reiterate, the problem here is not that some funds have
experienced substantial losses. To the extent that investors knowingly
assumed the risk of large losses, criticizing these funds is somewhat
unfair. For example, actively managed funds that lost 60 percent of
their value while comparable markets lost only 40 percent provided
their investors with returns that were within the range of variance
from market returns that one assumes by accepting active management
risk. One could criticize such funds for poor stock-picking, but it was
the shareholder who chose to assume the active management risk that the
fund would underperform the market. Similarly, the shareholder invested
in a Target-Date 2050 Fund should expect to experience large losses
when stock markets experience significant declines. In this case, it
would be the Target-Date 2050 Fund that invested most of its assets in
money market instruments that would be contradicting the asset
allocation implied by its name.
The SEC has had the opportunity to address the potential of fund
names to mislead investors. Pursuant to a request from consumer
advocates, the SEC adopted a misleading fund names rule in 2001. The
rule fell far short of providing reasonable assurances that fund names
that strongly implied a particular investment objective or style would
stick to it. The rule allows ``stock'' funds to invest 100 percent of
their assets in cash in emergency situations, ``short-term bond'' funds
to risk substantial losses, ``value'' funds to invest primarily in
growth stocks, and ``target-date 2010'' funds to invest a more than 75
percent of their assets in equities. The SEC has taken the position
that no matter how strongly a particular fund name implies a particular
investment objective or style, the name's potential to mislead
investors can be entirely corrected through narrative disclosure that
is often buried in fund documents. The SEC staff went out of its way to
reassure fund managers that funds the included the term ``U.S.
Government'' in its name could nonetheless invest 100 percent of its
assets in securities issued by Fannie Mae and Freddie Mac. \48\ As I
testified before this Committee in 2004, the term ``U.S. Government''
implies that the fund will invest in government-guaranteed securities,
which Fannie Mae and Freddie Mac securities are not.
---------------------------------------------------------------------------
\48\ Letter from Paul F. Roye, Director, SEC Division of
Investment Management, to Craig Tyle, General Counsel, Investment
Company Institute (Oct. 17, 2003).
---------------------------------------------------------------------------
Although investors should read prospectuses carefully before
investing, I disagree that investors whose investments in a ``target-
date 2010'' fund, a ``shortterm bond'' fund or 529 plan investment
option for a 16-year-old that declined more than 40 percent in one year
are entirely to blame for their misfortune. Congress should enact
legislation that meaningfully regulates fund names. It should require
the SEC to prohibit the use of fund names that create a common
expectation among investors regarding a fund's investment objectives
and style unless the fund invests consistent with that style. The
precise scope of the rule should be left to the SEC, but there should
be no question that terms such as ``target-date,'' ``short-term bond,''
and ``value'' would be covered. Fund sponsors use these terms in fund
names precisely to communicate something about the fund to investors.
They should not be permitted to contradict the fund name's message with
qualifying disclosure in fund documents.
Some have criticized this position as requiring that the government
dictate how funds invest. This argument is a red herring designed to
divert attention from the real issue. The only restriction that would
apply would be to the names that funds are permitted to use. The new
rule would have no effect on any fund that chose a name that did not
imply a particular investment objective or style. I strongly agree that
free markets should determine what mutual funds invest in, not
regulators. Requiring that all mutual funds invest only in a portfolio
the returns of which will fall within a fairly predictable range would
be inefficient, impracticable and inconsistent with basic principles of
individual liberty. There are and should be mutual funds the variance
of the investment returns of which essentially match the scope of the
fund manager's investment discretion.
Requiring that a fund that uses a particular name produce
predictably variable returns, however, does not implicate these
concerns. When Magellan Fund manager Jeff Vinik invested a large amount
of the Fund's assets in fixed income securities prior to a run-up in
the stock market in the late 1990s, the opportunity lost by its
shareholders was a risk that they knowingly assumed. There is nothing
about the name ``Magellan Fund'' that implies that its investment
returns will reflect the variance that is characteristic of a
particular market. Indeed, the name ``Magellan'' aptly suits a fund
that may explore any and all investment opportunities around the globe.
In contrast, it is misleading that a so-named ``stock'' fund can,
consistent with its name, invest 100 percent of its assets in cash, or
that something called a ``short-term bond'' fund could lose 40 percent
of its value in a single year.
Fund Governance
As indicated by this testimony, the breadth and depth of investor
protection issues in the mutual fund industry that have been left
unattended by regulators calls for new ideas on the most efficient
structure for mutual fund regulation. The mutual fund scandal of 2003
also demonstrated the need for more independent boards. As described in
greater detail in my March 23, 2004, testimony before this Committee,
Congress should implement the following reforms to strengthen the
oversight of mutual funds:
Create a Mutual Fund Oversight Board that would have
examination and enforcement authority over funds and fund
boards.
Require that a fund's chairman be independent.
Require that a fund's board be 75 percent independent.
Prohibit former directors, officers and employees of the
fund manager from serving as independent directors.
Require that independent directors stand for election at
least once every 5 years.
The Commission does not have the authority to impose any of these
requirements on an unconditional basis. Each of these proposals
requires Congressional action.
529 PLANS
As this Committee is aware, 529 plans have become an increasingly
popular means for Americans to save for higher education. These plans
have enjoyed enormous appeal in part because they offer a unique
combination of federal and state tax benefits, high contribution
limits, matching state contributions, donor control, automatic
rebalancing and, in many cases, low costs. However, 529 plans also have
been subject to criticism on the grounds of excessive and inadequately
disclosed fees, inconsistent state tax treatment across different
plans, and questionable sales practices. The following discussion
briefly sets forth some of the issues relating to 529 plans and
proposes regulatory reforms. \49\
---------------------------------------------------------------------------
\49\ Many of these issues are discussed in greater detail at:
Mercer Bullard, The Visible Hand in Government Sponsored Financial
Services: Why States Should Not Sponsor 529 Plans, 74 U. Cin. L. Rev.
1265 (2006).
---------------------------------------------------------------------------
Regulatory Oversight
Permitting states to sell and regulate 529 plans has effectively
added 50 new regulators for tax-deferred mutual fund wrappers (e.g.,
401(k) plans, IRAs, Roth IRAs, and 403(b) plans), which are subject to
too many different regulators and sets of rules as it is. The
Commission is responsible for fee disclosure for variable annuities,
the Department of Labor is responsible for fee disclosure for employee
benefit plans, and banking regulators and the Internal Revenue Service
are responsible for fee disclosure for IRAs. Multiple disclosure
regimes confuse investors and increase the costs of offering investment
products, as each provider must tailor its program to the particular
state's requirements. The Committee should take this opportunity to
explore ways of rationalizing fee disclosure and other regulatory
aspects of various tax-deferred mutual fund wrappers.
One option would be to assign exclusive oversight of 529 plans to
the SEC. The SEC has greater experience and expertise in this area than
any other government entity, and it would bring greater independence
and objectivity to the creation and enforcement of 529 plan fee
disclosure requirements. The states, as the issuers of interests in 529
plans, lack the independence and objectivity to regulate their own
plans and to enforce any rules they might devise. Congress should
consider specifically authorizing the Commission to establish
comprehensive rules governing the 529 plan fee disclosure, and consider
expanding this responsibility to all aspects of 529 plans operations.
In addition, Congress should consider amending the municipal
securities exemption to exclude 529 plans or permitting private firms
to offer 529 plans outside of state sponsorship. The municipal
exemption under which 529 plans operate was not intended for the
offering of retail financial services, but for the conduct of bona fide
government activities. There is nothing state-specific about 529 plans
that could not be accomplished outside of the framework of a money
management structure.
Fee Disclosure
Some commentators have criticized 529 plans on the ground that the
high fees charged by many plans have reduced the potential tax benefits
of the plans. Indeed, one commentator decried a plan that consumed more
than 10 percent of participants' balances each year for two years.
Determining whether a particular fee is too high or too low, based
solely on the amount of the fee, is a difficult and uncertain exercise.
In my view, the best way to promote efficient pricing is through
standardized, transparent disclosure of fees. It is generally accepted
that standardized, transparent fee disclosure promotes competition and
reduces prices. The disclosure of 529 plan fees, however, is generally
incoherent and obscure, and 529 plans would likely be forced to reduce
their fees if adequate fee disclosure were provided.
The lack of transparent, prominent, standardized disclosure of 529
plan fees is exacerbated by factors in the 529 plan context that make
fee disclosure even more important than in other contexts. In effect,
certain governmental entities have been granted an exclusive monopoly
to sell a particular tax-deferred investment product in competition
with private providers of other tax-deferred investment products. This
intrusion of the government into the private sector may distort many
functions of the financial services markets, including the setting of
fees.
For example, investors may lower their guard when evaluating 529
plans on the assumption that a public-minded governmental entity would
sell only a high quality, low-cost investment product. In fact, states'
interests may not be aligned with plan participants' interests with
respect to negotiating fees and choosing investment options, and
investors' trust in states' motivations and interests may be misplaced.
States may have incentives to offer plans that charge high fees. States
may charge high fees as a means of increasing their general revenues,
or charge higher fees to out-of-state residents as a way to subsidize
services provided to instate participants. Political considerations
also may influence the selection of money managers and cause states to
be less diligent when negotiating fees. For example, states may favor
in-state money managers or managers that have contributed to the
election campaigns of state officials. State officials may even use 529
plan assets for self-promotion.
Further, participants in 529 plans have limited control over fees.
Mutual funds can raise advisory and 12b-1 fees only with shareholder
approval, whereas states generally can raise fees at will without
notice to participants, thereby making it more important that investors
understand the fees charged before making an investment decision. When
a mutual fund that is a 529 plan investment option seeks to raise its
fees, the state has the right to vote on the fee increase, but, as
noted above, it may not have the same interests to negotiate low fees
as plan participants have. Finally, federal law gives mutual fund
shareholders legal recourse against a fund's directors and manager with
respect to excessive fees charged by the manager, which may provide
some restraint on fees. Participants in 529 plans, however, have no
such rights absent a violation of the antifraud rules under the federal
securities laws.
Restrictions on 529 plan investment options, participants' limited
control over fees and fee increases, the costs and burdens of
transferring from one plan to another, states' monopoly on state tax
benefits, limited legal recourse against plan sponsors, and the
divergence of state and participant interests are some of the special
factors that make it especially critical that 529 plan fees be fully
disclosed in an understandable, standardized, accessible format.
These special factors militate for prompt Congressional action to
ensure that 529 plans are required to provide standardized,
transparent, prominent fee disclosure. In short, fee disclosure for 529
plans, at a minimum, should be:
Standardized, both in the way in which the fees are
calculated and the terms used to describe the fees;
Prominently disclosed relative to other information about
the plan;
Presented both as a percentage of assets and a dollar
amount, and on an illustrative and individualized basis;
Inclusive of a total expense ratio for each investment
option that includes all fees incurred in connection with an
investment in the plan, to include, among other things,
portfolio transaction costs, distribution costs, operating
costs and administrative fees, whether charged by the state,
plan manager, investment manager, or other person;
Inclusive of a pie chart that illustrates the components of
the total expense ratio according to standardized categories of
fees, such as investment management, administrative services,
and marketing and distribution;
Inclusive of information on fees charged by other 529 plans
both in a disclosure document and in an easily accessible
format on the Internet; and
Inclusive of separate disclosure of all payments received
by intermediaries for executing the transactions in plan
interests, both as a dollar amount and percentage of assets,
whether or not the payment is made directly by the participant.
As discussed above, Congress should ensure that fee disclosure
requirements for 529 plans are promulgated and enforced by an
independent, objective government entity.
Disparate State Tax Treatment
Most states that permit state deductions for 529 plans limit the
deductions to the in-state plan. This disparate state tax treatment of
529 plans distorts the marketplace for investment products. Investors
may opt for a higher-cost, in-state plan specifically in order to
receive the tax benefits of the in-state plan, or may miss out on the
in-state tax benefit offered by a low-cost in-state plan because
brokers recommend out-of-state plans that pay higher compensation to
the broker.
The disparate state tax treatment of 529 plans has the effect of
reducing price competition among 529 plans because in-state plans can
exploit their monopoly on in-state tax benefits to offset their higher
fees. This is essentially a kind of bundling, not dissimilar to a
private company that has a government-granted monopoly over one product
(state tax deductions) to help it sell another, possibly inferior
product (the 529 plan). States will inevitably exploit this monopoly to
the detriment of investors in 529 plans. The unavailability of state
tax deductions for out-of-state plans may further undermine market
efficiency and create incentives to charge higher fees, as discussed in
the next section. A small minority of states have extended their state
tax deduction to out-of-state 529 plans, but most continue to frustrate
Congress's intent in creating the plans. Congress should consider
mandating that any state tax deductions for 529 plan contributions or
distributions be reciprocal across all qualified 529 plans.
HEDGE FUNDS
Systemic Risk
There is no question that hedge funds are a potential source of
systemic risk, that is, the kind and scope of financial risk that is
systemic in the sense of posing a threat to our political, social and
economic systems. Systemic risk warrants government oversight because
our society might not be able to absorb an extreme contraction of free
financial markets without long-term damage to political, economic and
social institutions. This concern militates for appropriate prudential
oversight of hedge funds, such as requirements that they report net
positions and leverage ratios.
This does not mean that hedge funds or their advisers should be
subject to substantive regulation, however. It is important that
capital be allowed to flow to unregulated intermediaries such as hedge
funds. Investment in hedge funds is limited to sophisticated investors,
and these investors are presumed to be in the best position to protect
their interests without costly governmental oversight. Substantive
regulation of hedge funds will simply drive sophisticated capital
offshore and provide little benefit to the financial markets. As
discussed below, however, the SEC has permitted hedge funds to be sold
to unsophisticated investors in certain circumstances.
Public Offering of Hedge Funds
In 2007, the SEC effectively decided to permit hedge funds to
publicly offer their shares. These hedge funds argued that they
reflected investments in hedge fund managers, not in the funds, yet the
value of interests that they sold were predominantly depended on the
success of their funds. The financial structure of these public
companies is closer to a hedge fund than to a conventional money
manager, \50\ and the behavior of the stock prices of public hedge
funds and conventional money managers over the last two years has
reflected the significantly greater risks posed by the former. As
predicted, these publicly held hedge funds are acting like hedge funds,
not money managers. Much attention is being--and should be--paid to the
systemic risk posed by hedge funds, but too little has been paid to the
sale of hedge fund interests to unsophisticated investors. If the SEC
continues to be unwilling to ensure that hedge funds are sold only to
sophisticated investors, Congress should prohibit the public offering
of shares of these entities.
---------------------------------------------------------------------------
\50\ See After Blackstone: Should Small Investors Be Exposed to
Risks of Hedge Funds? Hearing before the Subcommitee on Domestic
Policy, Committee on Oversight and Government Reform (July 11, 2007)
(testimony of Mercer Bullard); Mercer Bullard, Regulating Hedge Fund
Managers: The Investment Company Act as a Regulatory Screen, 13
Stanford J. Law, Bus. & Fin. 286 (2008).
---------------------------------------------------------------------------
Accredited but Unsophisticated Investors
Under current law, persons with net worth of $1 million either
alone or with their spouse qualify to invest in hedge funds. The SEC
has conceded that this test, which has not been adjusted since 1982,
has made millions of new investors eligible to invest in hedge funds at
the same time that ``private pools have become increasingly complex and
involve risks not generally associated with many other issuers of
securities.'' The Commission estimated that the minimum net worth
requirement, if adjusted only for inflation and disregarding the issue
of increased complexity, would have been $1.9 million as of July 1,
2006. That year, the Commission issued a modest proposal to increase
the $1 million minimum to $2.5 million. That minimum would be
inflation-adjusted again on July 12, 2012, and every 5 years
thereafter. In 2007, the Commission requested additional comments on
the proposal, but almost three years after the initial proposal, the
Commission has yet to take final action.
The effect of the SEC's position is that a newly retired couple
with $700,000 investments and a $300,000 home--the SEC continues to
count a person's personal residence counts toward the $1 million net
worth minimum while conceding that the ``value of an individual's
primary residence may have little relevance with regard to the
individual's need for the protections of Securities Act
registration''--is sophisticated enough to invest in a hedge fund. With
$700,000 in investments, a retired couple's typical withdrawal rate
would be 4 or 5 percent annually, or about $31,000 per year, plus
Social Security income and, in some cases, a company pension. Even
assuming additional income of $20,000 per year (which would not be
needed for the couple to meet the SEC standard), it is self-evident
that this couple's net worth in no way qualifies them to risk their
retirement security in a hedge fund. Nearly three years after its
initial proposal (and 27 years after the $1 million minimum was first
established), the SEC continues to permit hedge funds to prey on
unsophisticated investors. Congress should take steps to ensure that
any individual net worth standard for private offerings bears a
reasonable relationship to the likely financial sophistication of the
purchaser.
INVESTMENT ADVISERS
Fiduciary Standard
It is hard to understand how, after years of regulatory review, the
simple question of whether those who provide individualized investment
advice should be subject to a fiduciary standard has not been answered.
It is accepted that professionals who provide individualized, technical
advice similar to investment advice--e.g., lawyers and doctors--are
fiduciaries. They are required to act solely in their clients' best
interests. They may charge higher fees than other advisers, but their
fees must be fair. The must disclose all potential conflicts of
interest to their clients. In many cases, doctors and lawyers are
prohibited from assuming a conflicted role no matter what amount of
disclosure they provide.
The Supreme Court agrees. In the Capital Gains decision, it held
that investment advisers are subject to a fiduciary duty to their
clients. Yet the SEC and FINRA have taken the position that when
brokers provide individualized advisory services to their clients, they
should not necessarily be subject to a fiduciary duty, even when they
charge a separate, asset-based fee and advertise themselves as
``financial consultants,'' ``financial planners,'' and ``wealth
managers.'' In the narrow circumstances in which the SEC would consider
a broker to be an adviser, such as when it had provided a variety of
financial planning services to a client, the SEC still would allow the
broker to revert to a non-fiduciary role in executing the financial
plan. As a practical matter, the ``financial consultant'' can provide a
generic financial plan subject to a fiduciary duty, and then take off
its fiduciary hat when selling the client mutual funds that pay the
broker higher distribution fees than other funds without disclosing the
fees. As long as the funds are suitable, which they generally will be,
the broker has acted consistent with FINRA's standards of conduct.
The SEC's approach to this issue has been consistently anti-
investor. Ten years ago, it adopted a rule that expressly eliminated
Congress's requirement that the broker exclusion apply only if the
broker receives no special compensation for investment advisory
services. The rule also read Congress's requirement that the advisory
services also be ``solely incidental'' so broadly so as to be
meaningless. The SEC took the position that advice was solely
incidental if the advisory services were provided ``in connection with
and reasonably related to'' brokerage services. As stated in an amicus
brief filed by Fund Democracy and the Consumer Federation, ``[t]he
Commission's `in connection with and reasonably related to' standard
sets no limits on the degree of advisory services provided in relation
to the brokerage services, much less in any way limit the advisory
services to those that are `minor' or otherwise `incidental.'' ''
Congress needs to take action to end this debate. For over a
decade, the SEC has been unable to muster the backbone to defend
fiduciary standards for investment advisers, and the current SEC
Chairman and one Commissioner spent years defending FINRA's self-
interested position that a suitability standard is adequate, \51\
notwithstanding that, for example, it does not require the disclosure
of conflicts of interest. Congress should enact legislation that
imposes a fiduciary duty on any persons who provide individualized
investment advice or sell products pursuant to their providing of such
individualized investment advice. Americans who naturally expect those
providing fiduciary services to act solely in their clients' best
interests are entitled to nothing less.
---------------------------------------------------------------------------
\51\ See, e.g., Letter from Mary Schapiro, Vice Chairman and
President, NASD, and Elisse Walter, Executive Vice President, NASD to
Annette Nazareth, Director, Division of Market Regulation, and Meyer
Eisenberg, Acting Director, Division of Investment Management, U.S.
Securities and Exchange Commission (Apr. 4, 2005) available at http://
www.sec.gov/rules/proposed/s72599/nasd040405.pdf
---------------------------------------------------------------------------
Madoff Scandal
It should not be necessary to include the Madoff scandal as a
separate category in this testimony, but the import of the scandal for
investment adviser regulation has been so distorted that some
clarification is necessary. We still don't know exactly how Madoff
perpetrated his fraud, except that he did so without detection for many
years. Some have argued that this reflects a failure of investment
adviser regulation despite the fact that he was exclusively regulated
as a broker-dealer during most of the period of the fraud. These
arguments may simply reflect nothing more than a short-sighted
political strategy to curry favor as the preferred choice as the SRO
for the adviser industry, but they nonetheless need to be addressed. I
agree that an SRO for advisers would be appropriate, but if the Madoff
scandal has revealed anything with respect to this issue, it is that
some regulators lack a full understanding of the nature of investment
adviser services and regulation and could not adequately protect
investors' interests in overseeing the investment adviser industry.
During most of the period during which Madoff defrauded his
clients, he was not registered as an investment adviser--he was
registered as a broker. It appears that he was not registered as an
investment adviser because the SEC had interpreted the broker exclusion
from the definition of investment adviser for ``solely incidental''
investment advice to be available for discretionary accounts. The SEC
has since abandoned this ill-advised position, but during most of
Madoff's illegal activities he was able to rely on the exclusion and
was regulated solely as a broker. Thus, while FINRA, the SRO
responsible for broker oversight, has suggested that the Madoff scandal
illustrates the risk of ``the absence of FINRA-type oversight of the
investment adviser industry,'' this position is belied by the
undisputed fact that Madoff was subject only to broker regulation
during most of the relevant period.
FINRA's position is understandable and not necessarily a negative
reflection on its capacity as the broker SRO. Its leadership lacks a
deep understanding of and experience with investment adviser
regulation, and it has a close relationship with and is naturally
protective of the brokerage industry. Its sometimes excessive
exuberance for extending its jurisdiction over functionally dissimilar
services is a common, unavoidable symptom of agency politics,
especially in the inexperienced hands of new leadership. And one would
expect that a fraud perpetrated by a man who for years served in a
variety of leadership roles with FINRA's predecessor (the NASD), and
who used the luster that the NASD gave his reputation to help entice
unknowing victims, would put FINRA on the defensive and make its
objective evaluation of the situation difficult. \52\ Indeed, FINRA's
precipitate response to the Madoff scandal is quite understandable, but
it is also, unfortunately, evidence that it is not capable of providing
effective self-regulation of the investment adviser industry.
---------------------------------------------------------------------------
\52\ See Susan Antilla, Investors happily handed Madoff Billions,
Business Times (Dec. 17, 2008) (Madoff's ``company Web site describes
him as 'a major figure in the National Association of Securities
Dealers', the regulatory agency now known as Finra. He was board
chairman of the Nasdaq Stock Market; was on the board of governors of
the NASD; sat on an advisory committee for the Securities and Exchange
Commission (SEC); and was chairman of the trading committee of Sifma,
formerly the Securities Industry Association.'').
---------------------------------------------------------------------------
This is not to say that the Madoff scandal tells us nothing about
investment adviser regulation. As noted, it demonstrates the problem of
leaving solely to broker regulation the kinds of advisory activities
that are clearly in need of investment adviser oversight. The SEC has
corrected the regulatory gap that allowed brokers who provided
discretionary advice to avoid advisory regulation. As discussed below,
the Commission should take steps to ensure that all individualized
investment advice is subject to advisory regulation.
In addition, during the last stages of the scandal Madoff was
registered as an investment adviser. His registration statement
indicated that he had custody of $17 billion in assets under
management. The Investment Advisers Act generally requires that an
investment adviser maintain custody of client assets with a broker
dealer or a bank, and in doing so relies on FINRA and banking
regulators to ensure that the custodied assets actually exist. In view
of reports that much of the Madoff related losses will be covered by
SIPC, it appears that the failed custody arrangement was with a broker.
It is unclear why, if the stolen assets were custodied by a broker,
regular broker examinations by Madoff's SRO did not uncover the fraud.
As discussed below, such prudential oversight should be assigned to a
prudential regulator, not to a regulator such as FINRA with concurrent
investor protection jurisdiction. A regulator such as FINRA should
focus solely on what it knows and does best: regulating the sales
activities of brokers.
A final word is necessary regarding the argument made by some that
the Madoff scandal demonstrates the weakness of a fiduciary standard. A
fiduciary duty is not designed to nor could it protect investors from
those who are willing to steal their money outright. The Madoff scandal
is no more a reflection on the fiduciary standard (or FINRA's lower
suitability standard) than would be a bank robbery. What would have
detected Madoff's fraud is adviser registration triggered by the
providing of individualized investment advice and competent
examinations of his custody arrangements.
Principal Trading Exemption
One of the primary reasons that brokers seek to avoid triggering
investment adviser regulation is the principal trading prohibition.
Section 206(3) of the Investment Advisers Act requires that investment
advisers obtain written notice and consent from their clients prior to
completion of the transaction in which the adviser acts in a principal
capacity. Brokers chafe under the requirement to obtain client consent
prior to every principal trade, and they hoped to be relieved of this
restriction by the SEC's proposed rule excluding virtually all brokers
managing nondiscretionary accounts from the definition of investment
adviser (known as the ``Merrill Rule''). When the Merrill Rule was
vacated by the Court of Appeals, the SEC quickly sought to accommodate
brokers' concerns by adopting an interim rule that exempted virtually
all trades not conducted in a discretionary account from section
206(3).
Before considering this interim rule, some history on the SEC's
ill-advised Merrill Rule is in order. As with the principal trading
exemption, the SEC effectively adopted the Merrill Rule without prior
notice and consent. The SEC took a no-action position with respect to
activity conducted within the proposal's purview, thereby circumventing
Administrative Procedures Act requirements. Those who opposed the rule
were left in limbo waiting for the Commission to adopt a final rule so
that it could be challenged in court. Almost five years later, the
Commission had yet take final action on the rule and the Financial
Planning Association sued to force a final resolution of the issue. The
SEC reproposed the rule in 2004, and then again in 2005. It finally
adopted the rule in 2005, after it had been in operation for almost six
years, and the Court of Appeals vacated the rule in its entirety in
2007.
Even before the Court's order went into effect, the SEC embarked on
the same path of adopting effectively final rules without prior notice
and comment. It scheduled its ``interim'' exemption from section 206(3)
to expire more than two years after its adoption. The SEC's repeated
abuse of notice and comment procedures undermines faith in the rule of
law and the administrative process, especially when it abuses its
authority by enacting broad exemptions from carefully crafted laws
enacted by Congress specifically to protect investors against abusive
transactions.
In this instance, the interim rule has created significant investor
protection gaps that continue to remain unaddressed. For example, the
rule does not expressly require firms to develop policies and
procedures that are specifically designed to detect, deter and prevent
disadvantageous principal transactions. Such procedures are necessary
to ensure that the fairness of the price at which the principal trade
is effected can be objectively verified. The market's current
difficulty in valuing certain fixed income securities that previously
were considered relatively liquid and easily valued illustrates the
potential risk. Securities that are difficult to value often are more
likely to be securities that an adviser may be attempting to dump on
its clients. The incentive to engage in the abuses that section 206(3)
is designed to prevent rises with the difficulty of determining whether
the transaction was fair. Congress should insist that the SEC take
prompt action to address this and other concerns relating to the
principal trading exemption. \53\
---------------------------------------------------------------------------
\53\ See Letter from Mercer Bullard, President and Founder, Fund
Democracy and Barbara Roper,Director of Investor Protection, Consumer
Federation of America to Nancy Morris, Secretary, U.S. Securities and
Exchange Commission (Nov. 30, 2007) available at http://
www.funddemocracy.com/principal%20trading%20ltr%2011.30.07.pdf
---------------------------------------------------------------------------
Pay-To-Play Ban
In August 1999, the SEC proposed to prohibit money managers from
engaging in pay-to-play. The Commission had thoroughly documented the
practice among public pension officials of awarding investment
management business to large political donors. \54\ The retirement
accounts of millions of our nation's schoolteachers, fire fighters,
police officers and other public servants were being invested by money
managers who qualify for the job not by earning it, but by financing
the political campaigns of public pension fund officials. The SEC's
proposal was elegantly simple. It would have required that money
managers give up any compensation they received for managing public
money for two years after the firm, its executives or agents made a
campaign contribution to an elected official or candidate who could
have influenced the selection of the money manager.
---------------------------------------------------------------------------
\54\ These findings in a series of articles published in 2001, see
Mercer Bullard, PaytoPlay in America, TheStreet.com (Apr. 26-30, 2001),
available at http://www.thestreet.com/funds/mercerbullard/1406251.html,
and a state-by-state guide to pay-to-play practices uncovered by the
SEC is posted on Fund Democracy's Web site. See http://
www.funddemocracy.com/Pay-to-Play%20Page.htm
---------------------------------------------------------------------------
The pay-to-play proposal was modeled on Rule G-37, which prohibits
municipal bond underwriters from contributing to the campaigns of
elected officials who may influence the award of bond underwriting
contracts. The rule is widely credited with cleaning up the municipal
bond industry. An unfortunate byproduct of Rule G-37 has been its
incidental effect on pay-to-play in the money manager arena. State
treasurers and other elected fiduciaries of municipal pension funds saw
campaign contributions from municipal underwriters dry up, so they
turned to money managers and lawyers doing business for the pension
funds to make up the difference.
Pay-to-play practices continue to plague the awarding of money
management business by public pension funds at the same time that
public pension underfunding has reached crisis proportions. It is
imperative that managers tasked with restoring financial stability to
public pension plans are not selected on the basis of political favor,
but on the basis of their expertise and experience. This will not
happen as long as the SEC allows investment advisers to pay-to-play in
the public money management arena. Congress should strongly encourage
the SEC to repropose the pay-to-play rule and see it through to final
adoption.
PREPARED STATEMENT OF ROBERT PICKEL
Executive Director and Chief Executive Officer,
International Swaps and Derivatives Association
March 10, 2009
Mr. Chairman and Members of the Committee, thank you very much for
allowing ISDA to testify at this hearing. We are grateful to the
Committee for the opportunity to discuss the privately negotiated
derivatives business and more specifically, the credit default swaps
market. This business is an important source of innovation for our
financial system--it is one that employs tens of thousands of
individuals in the United States and benefits thousands of American
companies across a broad range of industries.
About ISDA
ISDA, which represents participants in the privately negotiated
derivatives industry, is the largest global financial trade
association, by number of member firms. ISDA was chartered in 1985, and
today has over 850 member institutions from 56 countries on six
continents. These members include most of the world's major
institutions that deal in privately negotiated derivatives, as well as
many of the businesses, governmental entities and other end users that
rely on over-the-counter derivatives to manage efficiently the
financial market risks inherent in their core economic activities.
Since its inception, ISDA has pioneered efforts to identify and
reduce the sources of risk in the derivatives and risk management
business. Among its most notable accomplishments are: developing the
ISDA Master Agreement; publishing a wide range of related documentation
materials and instruments covering a variety of transaction types;
producing legal opinions on the enforceability of netting and
collateral arrangements; securing recognition of the risk-reducing
effects of netting in determining capital requirements; promoting sound
risk management practices; and advancing the understanding and
treatment of derivatives and risk management from public policy and
regulatory capital perspectives.
In my remarks today, I would briefly like to:
Describe how CDS contracts works and the benefits they
provide;
Clarify the level of exposure in the CDS business;
Discuss the robust infrastructure that industry
participants have developed to support the CDS business;
Review the role of CDS in today's financial crisis; and
Outline my views on the evolution of the regulatory
framework for privately negotiated derivatives.
As I cover these topics, I hope to clarify some key misperceptions
regarding the CDS business:
The first is that, even in the face of a significant
increase in defaults and the collapse of major financial
institutions, the CDS business has continued to function.
Credit derivatives have remained available as a means to manage
risk in today's financial markets;
Second, as recent events have proven out, the risks related
to the CDS business have been widely misunderstood;
Third, the CDS business operates within a robust
infrastructure that incorporates time-tested standards,
practices and principles;
Fourth, CDS are not responsible for today's financial
crisis; and
Finally, the CDS industry continues to work with
policymakers to improve and evolve how we do business.
How Credit Default Swaps Work
Credit default swaps are simple financial transactions negotiated
between two counterparties. They enable firms to transfer and more
effectively manage risk.
In the real world, CDS play an important role in the growth and
functioning of our nation's economy:
CDS facilitate the flow of credit to American businesses;
CDS lower borrowing costs for American companies; and
CDS provide vital information to the market about the
creditworthiness of borrowers.
OTC derivatives exist to serve the risk management and investment
needs of end-users such as the businesses that are the backbone of our
economy and the investors that provide funds to those businesses. The
development of OTC derivatives has followed the development of the
American economy. For centuries, foreign exchange transaction have
facilitated trade and helped American businesses expand; they were one
of the original banking powers recognized in the National Bank Act of
1863. The first OTC derivative linked to interest rates was transacted
in the early 1980s between IBM and the World Bank, helping IBM raise
funds on more favorable terms. Today, over 90 percent of the Fortune
500, 50 percent of mid-sized companies and thousands of other smaller
American companies, use OTC currency and interest rate derivatives.
Credit derivatives first appeared in the mid-1990s as a tool to help
banks diversify the credit risk in their loan portfolio, and they have
grown into a vital risk management and diversification tool. In each
case the need for these products was driven by the needs of end-users,
and their growth was a direct function of their utility to end-users.
If end-users did not want these products, they would not exist.
It might be helpful to provide an example of the needs that credit
derivatives address. Imagine a bank that wants to lend more to American
companies in a particular sector of the economy, or a particular
geographic region, but that does not have relationships with those
companies. That bank could enter into a credit derivative transaction
with a bank that does have loans to those companies, whereby the first
bank would sell protection to the second bank on those companies,
taking on some of the second bank's credit exposure to those companies
in exchange for periodic payments. This transaction benefits both
banks: the first bank diversifies its loan portfolio and earns income
and the second bank is able to lend more money to those companies and
deepen its relationship with them. Equally importantly, this
transaction also benefits the companies themselves. It expands their
funding sources and thus allows them to get better rates on their
borrowings.
CDS can also be used to hedge against other risks related to the
potential default of a borrower. For instance, an auto parts company
that is heavily reliant on one auto manufacturer as its primary
customer might seek to protect itself against the risk that
manufacturer will go out of business by purchasing protection in the
form of a CDS on that company.
These credit derivatives, so-called single-name credit default
swaps because they provide default protection on a single entity, were
the foundation of the credit derivatives market and still constitute
the vast majority of the market. These trades help American companies
raise money more cheaply, and they help American investors diversify
risk and seek out attractive investment opportunities. To that end,
Warren Buffett wrote this year in his letter to Berkshire Hathaway
shareholders that he has started to use single name CDS to sell
protection and that he would like to enter into more such transactions.
The utility of such credit derivatives to investors and to companies is
what makes them so valuable to the American economy.
Growth and Size of the CDS Business
Because of the important role they play in enabling firms to more
precisely manage risk, the CDS business has grown significantly in a
relatively short period of time. The most common measurement of the
size of the CDS business is notional amount. For CDS, this represents
the face value of the bonds and loans on which participants have
written protection.
While using notional amount as a measurement tool for the size of
the privately negotiated derivatives business has its benefits, it also
has a major drawback. Notional amount greatly overstates the actual
exposure represented by the CDS business. One reason for this is
because a seller of protection often seeks to hedge its risk by
entering into offsetting transactions. Using the example above, if the
counterparty that sold $10 million of protection wished to hedge its
risk and buy protection, it too would enter into a $10 million CDS
contract. Thus, there are now two CDS contracts outstanding with a
total notional amount of $20 million. The reality is, however, that
only $10 million is at risk.
The Depository Trust and Clearing Corporation recently began
publishing market data based on information compiled for their Trade
Information Warehouse. According to DTCC, the net notional amount
outstanding--which represents the maximum possible net funds transfer
between net sellers and net buyers of protection that could be required
upon the occurrence of a credit event--is $2.6 trillion.
This may seem like a large number, and it is. But consider what it
represents: the sum total of payouts if all reference entities were to
default. This is, to say the least, unlikely. What's more, the average
of the net notional amount across the reference entities in the DTCC
warehouse is $2.6 billion. And this actually overstates the potential
losses, because it excludes any recovery value that sellers of
protection might receive. The point here is that the net payout on an
individual reference entity basis is manageable. This was aptly
demonstrated by the Lehman default, where the amounts paid on
settlement were handled with no disruption to the system.
One additional point regarding the size and risks of the CDS
business bears mentioning. CDS do not create new risks. They enable
firms to transfer risk that already exists. This risk-shifting process
is a zero-sum arrangement; what the buyer potentially gains by buying
protection, the seller potentially loses by selling protection. The
amount that the seller of protection loses is identical to the risk
that the buyer originally held.
CDS Infrastructure
Privately negotiated derivatives are often referred to as ``OTC
derivatives,'' with the implication being that this is an unregulated
business with no structure, standards or principles governing it. As
someone who has been involved in building a robust infrastructure for
privately negotiated derivatives for virtually my entire professional
career, this misperception is perhaps the most frustrating among those
that characterize the CDS business.
The truth is, there is a robust infrastructure for CDS and other
swaps that has been developed over the past 25 years by ISDA, industry
participants and policymakers around the world. The growth, strength,
and success of the business could not have been achieved without it.
A case in point: some believe that, in the OTC derivatives
business, all kinds of firms can enter into all types of CDS contracts
with each other. This is simply not the case. The fact is, banks are
the primary market makers in the CDS business, and firms wishing to
trade CDS need to have credit lines with them. Of the trades in the
DTCC warehouse, virtually all involve at least one dealer and 86
percent are between two dealers. These dealer banks, in turn, impose a
variety of requirements on their counterparties (and vice versa) in
terms of the maximum exposure they will take, the imposition of
collateral requirements, and so on. Virtually all of the exposure in
the CDS business originates within the heavily regulated banking
system.
Another example of the industry's infrastructure at work: at the
core of every CDS transaction is a contract negotiated and entered into
between two firms. The specific terms of the contract--its amount, the
premium payment, its duration, etc.--are determined by the
counterparties and are codified in a confirmation agreement between
them.
Underlying the confirmation is the widely used ISDA Master
Agreement, which includes standardized language on definitions and
other contract terms. The ISDA Master is widely recognized as a
groundbreaking document that has enabled the growth of the risk
management industry by enhancing legal certainty and reducing credit
risk. It establishes key international contractual standards, and its
importance to the global financial community has been described as ``no
less than the creation of global law by contractual consensus.''
Reflecting its wide acceptance, the vast majority of derivatives
transactions executed annually are documented under the ISDA Master.
In addition to the standardized legal architecture governing
privately negotiated derivatives, the industry has also worked to
develop sound practices in other areas. These include risk management,
the use and management of collateral, and the incorporation of
technology into the derivatives business.
The industry's work to further strengthen and improve the
infrastructure and platform upon which it operates is never-ending. The
industry has, for example, greatly improved transparency through the
publication of information in DTCC's trade information warehouse, and
significant progress has been made to reduce operational risk in the
confirming, settling, and clearing of CDS.
The Role of CDS in Today's Financial Crisis: Bear, Lehman, AIG
Over the past year, CDS have received a significant amount of
attention because of concerns about their role in the current financial
crisis. More specifically, issues have been raised regarding whether
CDS created the financial crisis and/or played a significant part in
the Bear Stearns, Lehman Brothers, and AIG situations.
It is by now clear that the roots of the current financial crisis
lie in imprudent lending decisions, particularly with respect to
residential housing, but also extending to other areas including
consumer receivables, auto finance and commercial development. These
imprudent decisions were in part the result of an ``easy money''
environment and a mispricing of risk. They were in turn exacerbated by
distortions in ratings models that underestimated both the risk of
individual securities as well as how closely correlated the risks of
those securities were within portfolios.
If CDS did not cause the crisis, did they make it worse? Some
industry observers cite the Bear Stearns situation in answering this
question. While it may seem far longer, it was only a year ago that
Bear Stearns suffered a liquidity crisis that led to its eventual
purchase by JPMorgan Chase. As this drama unfolded, there were
widespread concerns that Bear's failure as a derivatives counterparty
would have systemic implications. The theory was the CDS and other
privately negotiated derivatives supposedly created an interlinking web
in which a shock from one participant could capsize others.
The fact is, Bear's problems were primarily related to a lack of
confidence from its lenders and its resulting inability to secure
institutional funding to run its business. It was a classic liquidity
squeeze for an institution that apparently relied too much on short-
term funding. The role of swaps in this situation was at best cursory.
As for the systemic risk fears related to Bear's role as a swaps
counterparty, subsequent events have proven this supposition to be
groundless. Lehman was larger than Bear Stearns--a bigger institution
with a bigger derivatives portfolio--and its bankruptcy created no
system fissures.
In fact, by the time of the Lehman default in September, the focus
had shifted. No longer were market observers especially worried about
the failure of a large derivatives counterparty. Concerns centered on
the implications of a failure of a reference entity upon which a
significant level of credit protection had been sold.
Here, too, however, the fears were overblown. Contrary to rumors,
the actual payout on CDS contracts in which Lehman was a reference
entity was about $5 billion--far less than some industry critics
initially thought. By all accounts, the Lehman bankruptcy and default
was processed well by the industry, testifying to its strength and
resilience.
Moving now to AIG: Last week, this Committee heard testimony on the
regulatory failures that contributed to the terrible situation at AIG.
We also heard Chairman Bernanke express his frustration with AIG,
stating that it acted like an unregulated hedge fund.
The truth, however, is far worse. First, it's clear that AIG was in
fact regulated. Its supervisors apparently knew how much mortgage risk
it was taking on in its credit protection and securities lending
business. They also knew that AIG included ratings triggers and
collateral requirements in its contracts in order to gain additional
counterparty capacity.
In addition, a hedge fund would not have been allowed to build up
such a large, uncollateralized positions with so many counterparties.
In fact AIG Financial Products operated far more recklessly than most
hedge funds or, for that matter, other businesses engaged in similar
activities. It is worth noting these practices were contrary to the
generally accepted practices advanced by ISDA for the last 20 years.
In short, the causes of the AIG situation are clear. First, AIG's
Financial Products subsidiary took on too much exposure to subprime
mortgage debt. As the ratings on that debt were downgraded, the
company's own ratings came under pressure. Under agreements with its
counterparties and customers, AIG was then forced to post ever
increasing amounts of collateral with them. In short, AIG took on too
much exposure to subprime debt, and failed to appropriately manage its
collateral and liquidity. It was a collective risk, liquidity, and
collateral management failure, facilitated by poor supervision and an
overreliance on rating agency models.
The Continued Evolution of the CDS Business
As noted previously, the CDS industry is committed to further
strengthening and improving how we do business. This includes working
with policymakers to address areas of mutual concern.
On November 14 the PWG announced a series of policy objectives for
the privately negotiated derivatives industry. The PWG broke their
recommendations into four broad categories: (1) improve the
transparency and integrity of the credit default swaps market; (2)
enhance risk management of OTC derivatives; (3) further strengthen the
OTC derivatives market infrastructure; and (4) strengthen cooperation
among regulatory authorities.
ISDA agrees with these four objectives, and believes that
continuing to pursue the improvements industry and regulators have
worked on over the last several years is key to ensuring the OTC
derivatives industry in the United States remains healthy and
competitive.
Within those four broader objectives the PWG lists a number of
specific recommendations. These can be separated into:
Recommendations for policymakers (e.g., ``Regulators should
establish consistent policy standards and risk management
expectations for CCPs or other systemically important
derivatives market infrastructures and apply those standards
consistently'');
Recommendations for industry (e.g., ``Market participants
should adopt best practices with respect to risk management for
OTC derivatives activities, including public reporting,
liquidity management, senior management oversight and
counterparty credit risk management'');
Recommendations of an operational nature (e.g., ``Details
of all credit default swaps that are not cleared through a CCP
should be retained in a central contract repository'').
These recommendations provide a helpful framework for policymakers
and industry alike to discuss while reviewing and reforming the current
regulatory structure. Of particular importance from ISDA's perspective
is the PWG's statement acknowledging the continued need for bi-lateral,
custom tailored risk management contracts. As the PWG states:
``Participants should also be able to bilaterally negotiate customized
contracts where there are benefits in doing so, subject to continued
oversight by their prudential supervisors.'' While some have posited
that all OTC derivatives contracts should be made to trade on-exchange,
as the PWG notes there will continue to be the need for customized OTC
transactions.
On the same day the PWG announced its policy objectives, it also
released a Memorandum of Understanding among the Federal Reserve, the
Commodities Futures Trading Commission and the Securities and Exchange
Commission related to regulation of central counterparties. This
Memorandum is an important step in ensuring that regulators do not work
at cross-purposes while working to facilitate the creation of a central
clearinghouse. It would be unfortunate were the creation of a CDS
clearinghouse to be unnecessarily delayed because of a lack of
agreement among federal regulators.
Conclusion
Both the role and effects of CDS in the current market turmoil have
been greatly exaggerated. CDS were not the cause, or even a large
contributor, to this turmoil. There is little dispute that ill advised
mortgage lending, coupled with improperly understood securities backed
by those loans, are the root cause of the present financial problems.
These risk management problems have in some instances been exacerbated
by a failure to appropriately manage collateral and liquidity.
CDS are valuable risk management tools. They facilitate lending and
corporate finance and provide an important price discovery function
that is useful not only within the CDS business itself but across a
much broader spectrum. The business has remained open and liquid
throughout the financial crisis, demonstrating its resiliency.
It is ISDA's hope that the facts surrounding privately negotiated
derivatives, including CDS, will highlight the benefit of these risk-
transfer tools and the robust, sound infrastructure that has developed
around them.
At the same time, recent market events clearly demonstrate that the
regulatory structure for financial services has failed. Laws and
regulations written in the 20th century, in many cases designed to
address markets which existed in the 18th century, need to be changed
to account for 21st century markets and products. An in-depth
examination of the U.S. regulatory structure is self-evidently
warranted.
In summary, privately negotiated derivatives have continued to
perform well during a greater period of stress than the world financial
system has witnessed in decades. In the wake of failures of major
market participants, both counterparties and issuers of debt, CDS
participants have settled trades in an orderly way precisely according
to the rules and procedures established by Congress and market
participants. In this respect CDS activity has been a tremendous
success. We are confident that policymakers and market participants
alike will find their prudent efforts in helping build the
infrastructure for derivatives over the last 25 years have been
rewarded.
______
PREPARED STATEMENT OF DAMON A. SILVERS
Associate General Counsel,
AFL-CIO
March 10, 2009
Good morning, Chairman Dodd and Senator Shelby. My name is Damon
Silvers, I am an Associate General Counsel of the AFL-CIO, and I am the
Deputy Chair of the Congressional Oversight Panel created under the
Emergency Economic Stabilization Act of 2008 to oversee the TARP. While
I will describe the Congressional Oversight Panel's report on
regulatory reform, my testimony reflects my views and the views of the
AFL-CIO unless otherwise noted, and is not on behalf of the Panel, its
staff or its chair, Elizabeth Warren.
The vast majority of American investors participate in the markets
as a means to secure a comfortable retirement and to send their
children to college. Most investors' goals are long term, and most
investors rely on others to manage their money. While the boom and bust
cycles of the last decade generated fees for Wall Street--in many cases
astounding fees--they have turned out to have been a disaster for most
investors. The 10-year nominal rate of return on the S&P 500 is now
negative, and returns for most other asset classes have turned out to
be more correlated with U.S. equity markets than anyone would have
imagined a decade ago.
While the spectacular frauds like the Madoff ponzi scheme have
generated a great deal of publicity, the bigger questions are (1) how
did our financial system as a whole become so weak how did our system
of corporate governance, securities regulation, and disclosure-based
market discipline fail to prevent trillions of dollars from being
invested in value-destroying activities--ranging from subprime
mortgages and credit cards, to the stocks and bonds of financial
institutions, to the credit default swaps pegged to those debt
instruments; and (2) what changes must be made to make our financial
system a more reasonable place to invest the hard earned savings of
America's working families?
My testimony today will seek to answer the second question at three
levels:
1 How should Congress strengthen the regulatory architecture to
better protect investors;
2. How should Congress think about designing regulatory jurisdiction
to better protect investors; and
3. What are some specific substantive steps Congress and the
regulators should take to shore up our system of investor
protections?
Finally, I will briefly address how to understand the challenge of
investor protection in globalized markets.
Regulatory Architecture
While there has been much discussion of the need for better
systemic risk regulation, the Congressional Oversight Panel, in its
Special Report on Regulatory Reform issued on January 29, 2009,
observed that addressing issues of systemic risk cannot be a substitute
for a robust, comprehensive system of routine financial regulation. \1\
There are broadly three types of routine regulation in the financial
markets--(1) safety and soundness regulation for insured institutions
like banks and insurance companies; (2) disclosure and fiduciary duty
regulation for issuers and money managers in the public securities
markets; and (3) substantive consumer protection regulation in areas
like mortgages, credit cards, and insurance. These are distinct
regulatory missions in significant tension with each other.
---------------------------------------------------------------------------
\1\ Congressional Oversight Panel, Special Report on Regulatory
Reform, at 3 (Jan. 29, 2009), available at http://cop.senate.gov/
documents/cop-012909-report-regulatoryreform.pdf
---------------------------------------------------------------------------
Investors, people who seek to put money at risk for the prospect of
gains, really are interested in transparency, enforcement of fiduciary
duties, and corporate governance. This is the investor protection
mission. It is often in tension with the equally legitimate regulatory
mission of protecting the safety and soundness of insured financial
institutions. A safety and soundness regulator is likely to be much
more sympathetic to regulated entities that want to sidestep telling
the investing public bad news. At the same time, investor protection is
not the same thing as consumer protection--the consumer looking for
home insurance or a mortgage is seeking to purchase a financial service
with minimal risk, not to take a risk in the hope of a profit.
Because these functions should not be combined, investor protection
should be the focus of a single agency within the broader regulatory
framework. That agency needs to have the stature and independence to
protect the principles of full disclosure by market participants and
compliance with fiduciary duties among market intermediaries. Any
solution to the problem of systemic risk prevention should involve the
agency charged with investor protection, and not supersede it.
Since the New Deal, the primary body charged with enforcing
investor protections has been the Securities and Exchange Commission.
Although the Commission has suffered in recent years from diminished
jurisdiction and leadership failure, it remains an extraordinary
government agency, whose human capital and market expertise needs to be
built upon as part of a comprehensive strategy for effective
reregulation of the capital markets.
While I have a great deal of respect for former Treasury Secretary
Paulson, there is no question that his blueprint for financial
regulatory reform was profoundly deregulatory in respect to the
Securities and Exchange Commission. \2\ He and others, like the self-
described Committee on Capital Markets Regulation led by Harvard
Professor Hal Scott, sought to dismantle the Commission's culture of
arms length, enforcement-oriented regulation and to replace it with
something frankly more captive to the businesses it regulated. \3\
While these deregulatory approaches have fortunately yet to be enacted,
they contributed to an environment that weakened the Commission
politically and demoralized its staff.
---------------------------------------------------------------------------
\2\ Department of the Treasury, Blueprint for a Modernized
Financial Regulatory Structure, at 11-13, 106-126 (Mar. 2008),
available at http://www.treas.gov/press/releases/reports/Blueprint.pdf
\3\ Committee on Capital Markets Regulation, Interim Report (Nov.
30, 2006), available at http://www.capmktsreg.org/pdfs/
11.30Committee_Interim_ReportREV2.pdf; Committee on Capital Markets
Regulation, The Competitive Position of the U.S. Public Equity Market
(Dec. 4, 2007), available at http://www.capmktsreg.org/pdfs/
The_Competitive_Position_of_the_US_Public_Equity_Market.pdf
---------------------------------------------------------------------------
While there has been a great deal of attention paid to the
Commission's failure to spot the Madoff ponzi scheme, there has been
insufficient attention to the Commission's performance in relation to
the public debt markets, where the SEC regulates more than $438.3
billion in outstanding securities related to home equity loans and
manufactured housing loans, among the riskiest types of mortgages.
Similarly, little attention has been paid to the oversight of
disclosures by the financial and homebuilding firms investing in and
trading in those securities, and perhaps most importantly, the lack of
action by the Commission once the financial crisis began. \4\
---------------------------------------------------------------------------
\4\ Securities Industry and Financial Markets Association, Market
Sector Statistics: Asset Backed Securities--Outstanding By Major Types
of Credit.
---------------------------------------------------------------------------
But elections have consequences, and one of those consequences
should be a renewed commitment by both Congress and the new
Administration to revitalizing the Commission and to rebuilding the
Commission's historic investor protection oriented culture and mission.
The President's budget reflects that type of approach in the funding it
seeks for the Commission, and the new Chair of the Commission Mary
Schapiro has appeared to be focused on just this task in her recent
statements. \5\
---------------------------------------------------------------------------
\5\ See, e.g., Speech by SEC Chairman: Address to Practising Law
Institute's ``SEC Speaks in 2009'' Program available at http://sec.gov/
news/speech/2009/spch020609mls.htm
---------------------------------------------------------------------------
A key issue the Commission faces is how to strengthen its staff.
Much of what needs to be done is in the hands of the Commission itself,
where the Chair and the Commissioners set the tone for better or for
worse. When Commissioners place procedural roadblocks in the way of
enforcing the law, good people leave the Commission and weak staff are
not held accountable. When the Chair sets a tone of vigorous
enforcement of the laws and demands a genuine dedication to investor
protection, the Commission both attracts and retains quality people.
Congress should work with the Commission to determine if changes
are needed to personnel rules to enable the Commission to attract and
retain key personnel. The Commission should look at more intensive
recruiting efforts aimed at more experienced private sector lawyers who
may be looking for public service opportunities--perhaps through a
special fellows program. On the other hand, Congress should work with
the Commission to restrict the revolving door--ideally by adopting the
rule that currently applies to senior bank examiners for senior
Commission staff--no employment with any firm whose matters the staffer
worked on within 12 months.
Regulating the Shadow Markets and the Problem of Jurisdiction
The financial crisis is directly connected to the degeneration of
the New Deal system of comprehensive financial regulation into a Swiss
cheese regulatory system, where the holes, the shadow markets, grew to
dominate the regulated markets. If we are going to lessen future
financial boom and bust cycles, Congress must give the regulators the
tools and the jurisdiction to regulate the shadow markets. In our
report of January 29, the Congressional Oversight Panel specifically
observed that we needed to regulate financial products and
institutions, in the words of President Obama, ``for what they do, not
what they are.'' \6\ We further noted in that report that shadow market
products and institutions are nothing more than new names and new legal
structures for very old activities like insurance (read credit default
swaps) and money management (read hedge funds and private equity/lbo
funds). \7\
---------------------------------------------------------------------------
\6\ Senator Barack Obama, Renewing the American Economy, Speech at
Cooper Union in New York (Mar. 27, 2008) (transcript available at
http://www.nytimes.com/2008/03/27/us/politics/27text-
obama.html?pagewanted=all); Congressional Oversight Panel, Special
Report on Regulatory Reform, at 29.
\7\ Congressional Oversight Panel, Special Report on Regulatory
Reform, at 29.
---------------------------------------------------------------------------
The Congressional Oversight Panel's report stated that shadow
institutions should be regulated by the same regulators who currently
have jurisdiction over their regulated counterparts. \8\ So, for
example, the SEC should have jurisdiction over derivatives that are
written using public debt or equity securities as their underlying
asset. The Congressional Oversight Panel stated that at a minimum,
hedge funds should also be regulated by the SEC in their roles as money
managers by being required to register as investment advisors and being
subject to clear fiduciary duties, the substantive jurisdiction of U.S.
law, and periodic SEC inspections. \9\ To the extent a hedge fund or
anyone else engages in writing insurance contracts or issuing credit,
however, it should be regulated by the bodies charged with regulating
that type of economic activity.
---------------------------------------------------------------------------
\8\ Id.
\9\ Id.
---------------------------------------------------------------------------
Some have suggested having such shadow market financial products as
derivatives and hedge funds simply regulated by a systemic regulator.
This would be a terrible mistake. Shadow market products and
institutions need to be brought under the same routine regulatory
umbrella as other financial actors. To take a specific case, while it
is a good idea to have public clearinghouses for derivatives trading,
that reform by itself is insufficient without capital requirements for
the issuers of derivatives and without disclosure and the application
of securities law principles, generally, to derivatives based on public
securities regulations. So, for example, the SEC should require the
same disclosure of short positions in public equities that it requires
of long positions in equities, whether those positions are created
through the securities themselves or synthetically through derivatives
or futures.
The historic distinctions between broker-dealers and investment
advisors have been eroding in the markets for years. In 2007, the
Federal Appeals Court for the District of Columbia issued an opinion
overturning Commission regulations seeking to better define the
boundary between the two. \10\ The Commission should look at merging
the regulation of the categories while ensuring that the new regulatory
framework preserves clear fiduciary duties to investors. As part of a
larger examination of the duties owed by both broker-dealers and
investment advisors to investors, the Commission ought to examine the
fairness and the efficacy of the use of arbitration as a form of
dispute resolution by broker-dealers. Finally, part of what must be
done in this area is to determine whether the proper regulatory
approach will require Congressional action in light of the D.C. Circuit
opinion.
---------------------------------------------------------------------------
\10\ Fin. Planning Ass'n v. SEC, 482 F.3d 481 (D.C. Cir. 2007).
---------------------------------------------------------------------------
But there is a larger point here. Financial reregulation will be
utterly ineffective if it turns into a series of rifle shots at the
particular mechanisms used to evade regulatory structures in earlier
boom and bust cycles. What is needed is a return to the jurisdictional
philosophy that was embodied in the founding statutes of federal
securities regulation--very broad, flexible jurisdiction that allowed
the SEC to follow the activities. By this principle, the SEC should
have jurisdiction over anyone over a certain size who manages public
securities, and over any contract written that references publicly
traded securities. Applying this principle would require at least
shifting the CFTC's jurisdiction over financial futures to the SEC, if
not merging the two agencies under the SEC's leadership.
Much regulatory thinking over the last couple of decades has been
shaped by the idea that sophisticated parties should be allowed to act
in financial markets without regulatory oversight. Candidly, some
investors have been able to participate in a number of relatively
lightly regulated markets based on this idea. But this idea is wrong.
Big, reckless sophisticated parties have done a lot of damage to our
financial system and to our economy. I do not mean to say that
sophisticated parties in the business of risk taking should be
regulated in the same way as auto insurers selling to the general
public. But there has to be a level of transparency, accountability,
and mandated risk management across the financial markets.
Finally, while it is not technically a shadow market, the
underregulation of the credit rating agencies has turned out to have
devastating consequences. The Congressional Oversight Panel called
particular attention to the dysfunctional nature of the issuer pays
model, and recommended a set of options for needed structural change--
from the creation of PCAOB-type oversight body to the creation of a
public or non-profit NRSRO. \11\
---------------------------------------------------------------------------
\11\ Id. at 40-44.
---------------------------------------------------------------------------
Substantive Reforms
Beyond regulating the shadow markets, the Congress and the
Securities and Exchange Commission need to act to shape a corporate
governance and investor protection regime that is favorable to long
term investors and to the channeling of capital to productive purposes.
There is no way to look at the wreckage surrounding us today in the
financial markets and not conclude we have had a regulatory regime
that, intentionally or not, facilitated grotesquely short-term thinking
and led to capital flowing in unheard of proportions to pointless or
destructive ends.
This is a large task, and I will simply point out some of the most
important steps that need to be taken in three areas--governance,
executive pay, and litigation.
First, in the area of governance, once again the weakness of
corporate boards, particularly in the financial sector, appears to be a
central theme in the financial scandal. The AFL-CIO has interviewed the
audit committees of a number of the major banks to better understand
what happened. We found in general very weak board oversight of risk--
evidenced in audit committee leadership who did not understand their
companies' risk profiles, and in boards that tolerated the weakening of
internal risk management.
Strong boards require meaningful accountability to investors.
Short-term, leveraged investors have been the most powerful voices in
corporate governance in recent years, with destructive results. The
AFL-CIO urges Congress to work with the SEC to ensure that there are
meaningful, useable ways for long-term investors to nominate and elect
psychologically independent directors to public company boards through
access to the corporate proxy. I put the stress here on long-term--
there must be meaningful holding time requirements for exercising this
right. Recent statements by SEC Chair Mary Schapiro suggest she is
focused on this area, and we urge the Congress to support her efforts.
\12\
---------------------------------------------------------------------------
\12\ Rachelle Younglai, SEC developing proxy access plans:
sources, REUTERS, Mar. 6, 2009, at http://www.reuters.com/article/
bernardMadoff/idUSTRE52609820090307
---------------------------------------------------------------------------
Second, effective investor protection requires a comprehensive
approach to reform in the area of executive pay. Proxy access is an
important first step in this area, but we should learn from the
financial crisis how destructive short-term oriented, asymmetric
executive pay can be for long-term investors and for our economy. The
focus of the Congressional Oversight Panel's recommendations in the
area of executive pay were on ending these practices in financial
institutions. \13\ Here Chairman Dodd's leadership has been very
helpful in the context of the TARP.
---------------------------------------------------------------------------
\13\ Congressional Oversight Panel, Special Report on Regulatory
Reform, at 37-40.
---------------------------------------------------------------------------
But Congress and the Administration should pursue a comprehensive
approach to executive pay reform around two concepts--equity linked pay
should be held beyond retirement, and pay packages as a whole should
reflect a rough equality of exposure to downside risk as to upside
gain. Orienting policy in this direction requires coordination between
securities regulation and tax policy. But we could begin to address
what has gone wrong in executive pay incentives by (1) developing
measurements for both the time horizon and the symmetry of risk and
reward of pay packages that could be included in pay disclosure; (2)
looking more closely at mutual fund proxy voting behavior to see if it
reflects the time horizons of the funds; (3) focusing FINRA inspections
of broker dealer pay policies on these two issues; and (4) providing
for advisory shareholder votes on pay packages. With respect to say on
pay, any procedural approaches that strengthened the hand of long term
investors in the process of setting executive compensation would be
beneficial.
Finally, Congress needs to address the glaring hole in the fabric
of investor protection created by the Central Bank of Denver and
Stoneridge cases. \14\ These cases effectively granted immunity from
civil liability to investors for parties such as investment banks and
law firms that are co-conspirators in securities frauds. It appeared
for a time after Enron that the courts were going to restore some
sanity in this area of the law on their own, by finding a private right
of action when service providers were actually not just aiders and
abetters of a fraud, but actual co-conspirators. In the Stoneridge
decision, with the Enron case looming over them, the Supreme Court made
clear Congress would have to act. The issue here of course is not
merely fairness to the investors defrauded in a particular case--it is
the incentives for financial institutions to police their own conduct.
We seem to have had a shortage of such incentives in recent years.
---------------------------------------------------------------------------
\14\ Central Bank of Denver, N.A. v. First Interstate Bank of
Denver, N.A., 511 U.S. 164 (1994); Stoneridge Investment Partners, LLC
v. Scientific-Atlanta, Inc., 128 S. Ct. 761 (2008).
---------------------------------------------------------------------------
The International Context
The Bush Administration fundamentally saw the internationalization
of financial markets as a pretext for weakening U.S. investor
protections. That approach has been discredited. It needs to be
replaced by a commitment on the part of the Obama Administration to
building a strong global regulatory floor in coordination with the
world's other major economies. This effort is vital not only for
protecting U.S. investors in global markets, but for protecting our
financial sector from the consequences of a global regulatory race to
the bottom that will inevitably end in the kind of financially driven
economic crisis that we are living through today. Congress can play a
part by seeking to strengthen its relationships with its counterpart
legislative bodies in the major world markets, and should look for
opportunities to coordinate setting regulatory standards on a global
basis. The Administration needs to make this effort a priority, and to
understand that it needs to extend beyond the narrow confines of
systemic risk and the banking system to issues of transparency and
investor protection.
However, Congress must not allow the need for global coordination
to be an impediment or a prerequisite to vigorous action to reregulate
U.S. financial markets and institutions. That task is urgent and must
be addressed if the U.S. is to recover from the blow this financial
crisis has delivered to our private capital markets' reputation as the
gold standard for transparency and accountability.
Conclusion
The task of protecting investors by reregulating our financial
system and restoring vitality to our regulators is a large one. This
testimony simply sketches the outline of an approach, and notes some
key substantive steps Congress and the Administration need to take.
This Committee has already taken a leadership role in a number of these
areas, but there is much more to be done. Even in areas where the
primary responsibility must lie with regulators, there is a much needed
role for Congress to oversee, encourage, and support the efforts of the
Administration.
While I do not speak for the Congressional Oversight Panel, I think
I am safe in saying that the Panel is honored to have been asked to
assist Congress in this effort, and is prepared to assist this
Committee in any manner the Committee finds useful. I can certainly
make that offer on behalf of the AFL-CIO. Thank you.
SUPPLEMENT--March 10, 2009
The challenge of addressing systemic risk in the future is one, but
by no means the only one, of the challenges facing Congress as Congress
considers how to reregulate U.S. financial markets following the
extraordinary events of the last 18 months.
Systemic crises in financial markets harm working people. Damaged
credit systems destroy jobs rather than create them. Pension funds with
investments in panicked markets see their assets deteriorate. And the
resulting instability undermines business' ability to plan and obtain
financing for new investments--undermining the long term growth and
competitiveness of employers and setting the stage for future job
losses. The AFL-CIO has urged Congress since 2006 to act to reregulate
shadow financial markets, and the AFL-CIO supports addressing systemic
risk, but in a manner that does not substitute for strengthening the
ongoing day to day regulatory framework, and that recognizes addressing
systemic risk both requires regulatory powers and financial resources
that can really only be wielded by a fully public body.
The concept of systemic risk is that financial market actors can
create risk not just that their institutions or portfolios will fail,
but risk that the failure of their enterprises will cause a broader
failure of other financial institutions, and that such a chain of
broader failures can jeopardize the functioning of financial markets as
a whole. The mechanisms by which this broader failure can occur involve
a loss of confidence in information, or a loss of confidence in market
actors ability to understand the meaning of information, which leads to
the withdrawal of liquidity from markets and market institutions.
Because the failure of large financial institutions can have these
consequence, systemic risk management generally is seen to both be
about how to determine what to do when a systemically significant
institution faces failure, and about how to regulated such institutions
in advance to minimize the chances of systemic crises.
Historically, the United States has had three approaches to
systemic risk. The first was prior to the founding of the Federal
Reserve system, when there was a reluctance at the Federal level to
intervene in any respect in the workings of credit markets in
particular and financial markets in general. The Federal Reserve
system, created after the financial collapse of 1907, ushered in an era
where the Federal Government's role in addressing systemic risk largely
consisted of sponsoring through the Federal Reserve system, a means of
providing liquidity to member banks, and thus hopefully preventing the
ultimate liquidity shortage that results from market participants
losing confidence in the financial system as a whole.
But then, after the Crash of 1929 and the 4 years of Depression
that followed, Congress and the Roosevelt Administration adopted a
regulatory regime whose purpose was in a variety of ways to
substantively regulate financial markets in an ongoing way. This new
approach arose out of a sense among policymakers that the systemic
financial crisis associated with the Great Depression resulted from the
interaction of weakly regulated banks with largely unregulated
securities markets, and that exposing depositors to these risks was a
systemic problem in and of itself. Such centerpieces of our regulatory
landscape as the Securities and Exchange Commission's disclosure based
system of securities regulation and the Federal Deposit Insurance
Corporation came into being not just as systems for protecting the
economic interests of depositors or investors, but as mechanisms for
ensuring systemic stability by, respectively, walling off bank
depositors from broader market risks, and ensuring investors in
securities markets had the information necessary to make it possible
for market actors to police firm risk taking and to monitor the risks
embedded in particular financial products.
In recent years, financial activity has moved away from regulated
and transparent markets and institutions and into the so-called shadow
markets. Regulatory barriers like the Glass-Steagall Act that once
walled off less risky from more risky parts of the financial system
have been weakened or dismantled. So we entered the recent period of
extreme financial instability with an approach to systemic risk that
looked a lot like that of the period following the creation of the
Federal Reserve Board but prior to the New Deal era. And so we saw the
policy response to the initial phases of the current financial crisis
primarily take the form of increasing liquidity into credit markets
through interest rate reductions and increasingly liberal provision of
credit to banks and then to non-bank financial institutions.
However, with the collapse of Lehman Brothers and the Federal
rescues of AIG, FNMA, and the FHLMC, the federal response to the
perception of systemic risk turned toward much more aggressive
interventions in an effort to ensure that after the collapse of Lehman
Brothers, there would be no more defaults by large financial
institutions. This approach was made somewhat more explicit with the
passage of the Emergency Economic Stabilization Act of 2008 and the
commencement of the TARP program. The reality was though that the TARP
program was the creature of certain very broad passages in the bill,
which generally was written with the view that the federal government
would be embarking on the purchase of troubled assets, a very different
approach than the direct infusions of equity capital that began with
the Capital Purchase Program in October of 2008.
We can now learn some lessons from this experience for the
management of systemic risk in the financial system.
First, our government and other governments around the world will
step in when major financial institutions face bankruptcy. We do not
live in a world of free market discipline when it comes to large
financial institutions, and it seems unlikely we ever will. If two
administrations as different as the Bush Administration and the Obama
Administration agree that the Federal Government must act when major
financial institutions fail, it is hard to imagine the administration
that would do differently. Since the beginning of 2008, we have used
Federal dollars in various ways to rescue either the debt or the equity
holders or both at the following companies--Bear Stearns, Indymac,
Washington Mutual, AIG, Merrill Lynch, Fannie Mae, Freddie Mac,
Citigroup, and Bank of America. But we have no clear governmental
entity charged with making the decision over which company to rescue
and which to let fail, no clear criteria for how to make such
decisions, and no clear set of tools to use in stabilizing those that
must be stabilized.
Second, we appear to be hopelessly confused as to what it means to
stabilize a troubled financial institution to avoid systemic harm. We
have a longstanding system of protecting small depositors in FDIC
insured banks, and by the way policyholders in insurance companies
through the state guarantee funds. The FDIC has a process for dealing
with banks that fail--a process that does not always result in 100
percent recoveries for uninsured creditors. Then we have the steps
taken by the Treasury Department and the Federal Reserve since Bear
Stearns collapsed. At some companies, like Fannie Mae and Freddie Mac,
those steps have guaranteed all creditors, but wiped out the equity
holders. At other companies, like Bear Stearns, AIG, and Wachovia,
while the equity holders survive, they have been massively diluted one
way or another. At others, like Citigroup and Bank of America, the
equity has been only modestly diluted when looked at on an upside
basis. It is hard to understand exactly what has happened with the
government's interaction with Morgan Stanley and Goldman Sachs, but
again there has been very little equity dilution. And then there is
poor Lehman Brothers, apparently the only non-systemic financial
institution, where everybody lost. In crafting a systematic approach to
systemically significant institutions, we should begin with the
understanding that while a given financial institution may be
systemically significant, not every layer of its capital structure
should be necessarily propped up with taxpayer funds.
Third, much regulatory thinking over the last couple of decades has
been shaped by the idea that sophisticated parties should be allowed to
act in financial markets without regulatory oversight. But this idea is
wrong. Big, reckless sophisticated parties have done a lot of damage to
our financial system and to our economy. This is not to say that
sophisticated parties in the business of risk taking should be
regulated in the same way as auto insurers selling to the general
public. But there has to be a level of transparency, accountability,
and mandated risk management across the financial markets.
Fourth, financial markets are global now. Norwegian villages invest
in U.S. mortgage backed securities. British bankruptcy laws govern the
fate of U.S. clients of Lehman Brothers, an institution that appeared
to be a U.S. institution. AIG, our largest insurance company, collapsed
because of a London office that employed 300 of AIG's 500,000
employees. Chinese industrial workers riot when U.S. real estate prices
fall. We increasingly live in a world where the least common
denominator in financial regulation rules.
So what lessons should we take away for how to manage systemic risk
in our financial system?
The Congressional Oversight Panel, in its report to Congress made
the following points about addressing systemic risk:
1. There should be a body charged with monitoring sources of
systemic risk in the financial system, but it could either be a
new body, an existing agency, or a group of existing agencies;
2. The body charged with systemic risk managements should be fully
accountable and transparent to the public in a manner that
exceeds the general accountability mechanisms present in self-
regulatory organizations;
3. We should not identify specific institutions in advance as too
big to fail, but rather have a regulatory framework in which
institutions have higher capital requirements and pay more on
insurance funds on a percentage basis than smaller institutions
which are less likely to be rescued as being too systemic to
fail.
4. Systemic risk regulation cannot be a substitute for routine
disclosure, accountability, safety and soundness, and consumer
protection regulation of financial institutions and financial
markets.
5. Ironically, effective protection against systemic risk requires
that the shadow capital markets--institutions like hedge funds
and products like credit derivatives--must not only be subject
to systemic risk oriented oversight but must also be brought
within a framework of routine capital market regulation by
agencies like the Securities and Exchange Commisson.
6. There are some specific problems in the regulation of financial
markets, such as the issue of the incentives built into
executive compensation plans and the conflict of interest
inherent in the credit rating agencies' business model of
issuer pays, that need to be addressed to have a larger market
environment where systemic risk is well managed.
7. Finally, there will not be effective reregulation of the
financial markets without a global regulatory floor.
I would like to explain some of these principles and at least the
thinking I brought to them. First, on the issue of a systemic risk
monitor, while the Panel made no recommendation, I have come to believe
that the best approach is a body with its own staff and a board made up
of the key regulators, perhaps chaired by the Chairman of the Board of
Governors of the Federal Reserve. There are several reasons for this
conclusion. First, this body must have as much access as possible to
all information extant about the condition of the financial markets--
including not just bank credit markets, but securities and commodities,
and futures markets, and consumer credit markets. As long as we have
the fragmented bank regulatory system we now have, this body would need
access to information about the state of all deposit taking
institutions. The reality of the interagency environment is that for
information to flow freely, all the agencies involved need some level
of involvement with the agency seeking the information. Connected with
the information sharing issue is expertise. It is unlikely a systemic
risk regulator would develop deep enough expertise on its own in all
the possible relevant areas of financial activity. To be effective it
would need to cooperate in the most serious way possible with all the
routine regulators where the relevant expertise would be resident.
Second, this coordinating body must be fully public. While many
have argued the need for this body to be fully public in the hope that
would make for a more effective regulatory culture, the TARP experience
highlights a much more bright line problem. An effective systemic risk
regulator must have the power to bail out institutions, and the
experience of the last year is that liquidity provision is simply not
enough in a real crisis. An organization that has the power to expend
public funds to rescue private institutions must be a public
organization--though it should be insulated from politics much as our
other financial regulatory bodies are by independent agency structures.
Here is where the question of the role of the Federal Reserve comes
in. A number of commentators and Fed officials have pointed out that
the Fed has to be involved in any body with rescue powers because any
rescue would be mounted with the Fed's money. However, the TARP
experience suggests this is a serious oversimplification. While the Fed
can offer liquidity, many actual bailouts require equity infusions,
which the Fed cannot currently make, nor should it be able to, as long
as the Fed continues to seek to exist as a not entirely public
institution. In particular, the very bank holding companies the Fed
regulates are involved in the governance of the regional Federal
Reserve Banks that are responsible for carrying out the regulatory
mission of the Fed, and would if the current structure were untouched,
be involved in deciding which member banks or bank holding companies
would receive taxpayer funds in a crisis.
These considerations also point out the tensions that exist between
the Board of Governors of the Federal Reserve System's role as central
banker, and the great importance of distance from the political
process, and the necessity of political accountability and oversight
once a body is charged with dispersing the public's money to private
companies that are in trouble. That function must be executed publicly,
and with clear oversight, or else there will be inevitable suspicions
of favoritism that will be harmful to the political underpinnings of
any stabilization effort. One benefit of a more collective approach to
systemic risk monitoring is that the Federal Reserve Board could
participate in such a body while having to do much less restructuring
that would likely be problematic in terms of its monetary policy
activity.
On the issue of whether to identify and separately regulate
systemically significant firms, another lesson of the last eighteen
months is that the decision as to whether some or all of the investors
and creditors of a financial firm must be rescued cannot be made in
advance. In markets that are weak or panicked, a firm that was
otherwise seen as not presenting a threat of systemic contagion might
be seen as doing just that. Conversely, in a calm market environment,
it maybe the better course of action to let a troubled firm go bankrupt
even if it is fairly large. Identifying firms (ITAL)ex ante as
systemically significant also makes the moral hazard problems much more
intense.
An area the Congressional Oversight Panel did not address
explicitly is whether effective systemic risk management in a world of
diversified institutions would require some type of universal systemic
risk insurance program or tax. Such a program would appear to be
necessary to the extent the federal government is accepting it may be
in a position of rescuing financial institutions in the future. Such a
program would be necessary both to cover the costs of such
interventions and to balance the moral hazard issues associated with
systemic risk management. However, there are practical problems
defining what such a program would look like, who would be covered and
how to set premiums. One approach would be to use a financial
transactions tax as an approximation. The global labor movement has
indicated its interest in such a tax on a global basis, in part to help
fund global reregulation of financial markets.
More broadly, these issues return us to the question of whether the
dismantling of the approach to systemic risk embodied in the Glass-
Steagall Act was a mistake. We would appear now to be in a position
where we cannot wall off more risky activities from less risky
liabilities like demand deposits or commercial paper that we wish to
ensure. On the other hand, it seems mistaken to try and make large
securities firms behave as if they were commercial banks. Those who
want to maintain the current dominance of integrated bank holding
companies in the securities business should have some burden of
explaining how their securities businesses plan to act now that they
have an implicit government guaranty.
Finally, the AFL-CIO believes very strongly that the regulation of
the shadow markets, and of the capital markets as a whole cannot be
shoved into the category labeled ``systemic risk regulation,'' and then
have that category be effectively a sort of night watchman effort. The
lesson of the failure of the Federal Reserve to use its consumer
protection powers to address the rampant abuses in the mortgage
industry earlier in this decade is just one of several examples going
to the point that without effective routine regulation of financial
markets, efforts to minimize the risk of further systemic breakdowns
are unlikely to succeed. We even more particularly oppose this type of
formulation that then hands responsibility in the area of systemic risk
regulation over to self-regulatory bodies.
As Congress moves forward to address systemic risk management, one
area that we believe deserves careful consideration is how much power
to give to a body charged with systemic risk management to intervene in
routine regulatory policies and practices. We strongly agree with
Professor Coffee's testimony that a systemic risk regulator should not
have the power to override investor or consumer protections. However,
there are a range of options, ranging from power so broad it would
amount to creating a single financial services superregulator, e.g.,
vesting such power in staff or a board chairman acting in an executive
capacity, to arrangements requiring votes or supermajorities, to a
system where the systemic risk regulator is more of scout than a real
regulator, limited in its power to making recommendations to the larger
regulatory community. The AFL-CIO would tend to favor a choice
somewhere more in the middle of that continuum, but we think this is an
area where further study might help policymakers formulate a well-
founded approach.
Finally, with respect to the jurisdiction and the reach of a
systemic risk regulator, we believe it must not be confined to
institutions per se, or products or markets, but must extend to all
financial activity.
In conclusion, the Congressional Oversight Panel's report lays out
some basic principles that as a Panel member I hope will be of use to
this Committee and to Congress in thinking through the challenges
involved in rebuilding a more comprehensive approach to systemic risk.
The AFL-CIO is very concerned that as Congress approaches the issue of
systemic risk it does so in a way that bolsters a broader reregulation
of our financial markets, and does not become an excuse for not
engaging in that needed broader reregulation.
AFL-CIO Executive Council Statement--Miami, Florida--March 5, 2009
Bank Bailouts
There has been a dramatic concentration of banking power since the
Gramm-Leach-Bliley Act repealed New Deal bank regulation. More than 43
percent of U.S. bank assets are held by just four institutions:
Citigroup, Bank of America, Wells Fargo and JPMorgan Chase. When these
institutions are paralyzed, our whole economy suffers. When banks
appear on the brink of collapse, as several have repeatedly since
September, government steps in. The free market rules that workers live
by do not apply to these banks.
Since Congress passed financial bailout legislation in October,
working people have seen our tax dollars spent in increasingly
secretive ways to prop up banks that we are told are healthy, until
they need an urgent bailout. In some instances, institutions that were
bailed out need another lifeline soon after. The Congressional
Oversight Panel, charged with overseeing the bailout, recently found
that the Federal Government overpaid by $78 billion in acquiring bank
stock.
The AFL-CIO believes government must intervene when systemically
significant financial institutions are on the brink of collapse.
However, government interventions must be structured to protect the
public interest, and not merely rescue executives or wealthy investors.
This is an issue of both fairness and our national interest. It makes
no sense for the public to borrow trillions of dollars to rescue
investors who can afford the losses associated with failed banks.
The most important goal of government support must be to get banks
lending again by ensuring they are properly capitalized. This requires
forcing banks to acknowledge their real losses. By feeding the banks
public money in fits and starts, and asking little or nothing in the
way of sacrifice, we are going down the path Japan took in the 1990s--a
path that leads to ``zombie banks'' and long-term economic stagnation.
The AFL-CIO calls on the Obama administration to get fair value for
any more public money put into the banks. In the case of distressed
banks, this means the government will end up with a controlling share
of common stock. The government should use that stake to force a
cleanup of the banks' balance sheets. The result should be banks that
can either be turned over to bondholders in exchange for bondholder
concessions or sold back into the public markets. We believe the debate
over nationalization is delaying the inevitable bank restructuring,
which is something our economy cannot afford.
A government conservatorship of the banks has been endorsed by
leading economists, including Nouriel Roubini, Joseph Stiglitz, and
Paul Krugman. Even Alan Greenspan has stated it will probably be
necessary.
The consequences of crippled megabanks are extraordinarily serious.
The resulting credit paralysis affects every segment of our economy and
society and destroys jobs. We urge President Obama and his team to
bring the same bold leadership to bear on this problem as they have to
the problems of economic stimulus and the mortgage crisis.
AFL-CIO Executive Council Statement--Miami, Florida--March 5, 2009
Financial Regulation
Deregulated financial markets have taken a terrible toll on
America's working families. Whether measured in lost jobs and homes,
lower earnings, eroding retirement security, or devastated communities,
workers have paid the price for Wall Street's greed. But in reality,
the cost of deregulation and financial alchemy are far higher. The
lasting damage is in missed opportunities and investments not made in
the real economy. While money poured into exotic mortgage-backed
securities and hedge funds, our pressing need for investments in clean
energy, infrastructure, education, and health care went unmet.
So the challenge of reregulating our financial markets, like the
challenge of restoring workers' rights in the workplace, is central to
securing the economic future of our country and the world. In 2006,
while the Bush administration was in the midst of plans for further
deregulation, the AFL-CIO warned of the dangers of unregulated,
leveraged finance. That call went unheeded as the financial catastrophe
gathered momentum in 2007 and 2008, and now a different day is upon us.
The costs of the deregulation illusion have become clear to all but a
handful of unrepentant ideologues, and the public cast its votes in
November for candidates who promised to end the era of rampant
financial speculation and deregulation.
In October, when Congress authorized the $700 billion financial
bailout, it also established an Oversight Panel to both monitor the
bailout and make recommendations on financial regulatory reform. The
panel's report lays the foundation for what Congress and the Obama
administration must do.
First, we must recognize that financial regulation has three
distinct purposes: (1) ensuring the safety and soundness of insured,
regulated institutions; (2) promoting transparency in financial
markets; and (3) guaranteeing fair dealing in financial markets, so
investors and consumers are not exploited. In short, no gambling with
public money, no lying and no stealing.
To achieve these goals, we need regulatory agencies with focused
missions. We must have a revitalized Securities and Exchange Commission
(SEC), with the jurisdiction to regulate hedge funds, derivatives,
private equity, and any new investment vehicles that are developed. The
Commodity Futures Trading Commission should be merged with the SEC to
end regulatory arbitrage in investor protection.
Second, we must have an agency focused on protecting consumers of
financial services, such as mortgages and credit cards. We have paid a
terrible price for treating consumer protection as an afterthought in
bank regulation.
Third, we need to reduce regulatory arbitrage in bank regulation.
At a minimum, the Office of Thrift Supervision, the regulator of choice
for bankrupt subprime lenders such as Washington Mutual and IndyMac,
should be consolidated with other federal bank regulators.
Fourth, financial stability must be a critical goal of financial
regulation. This is what is meant by creating a systemic risk
regulator. Such a regulator must be a fully public agency, and it must
be able to draw upon the information and expertise of the entire
regulatory system. While the Federal Reserve Board of Governors must be
involved in this process, it cannot undertake it on its own.
We must have routine regulation of the shadow capital markets.
Hedge funds, derivatives, and private equity are nothing new--they are
just devices for managing money, selling insurance and securities, and
engaging in the credit markets without being subject to regulation. As
President Obama said during the campaign, ``We need to regulate
institutions for what they do, not what they are.'' Shadow market
institutions and products must be subject to transparency and capital
requirements and fiduciary duties befitting what they are actually
doing.
Reform also is required in the incentives governing key market
actors around executive pay and credit rating agencies. There must be
accountability for this disaster in the form of clawbacks for pay
awarded during the bubble. According to Bloomberg, the five largest
investment banks handed out $145 billion in bonuses in the 5 years
preceding the crash, a larger amount than the GDP of Pakistan and
Egypt.
Congress and the administration must make real President Obama's
commitment to end short-termism and pay without regard to risk in
financial institutions. The AFL-CIO recently joined with the Chamber of
Commerce and the Business Roundtable in endorsing the Aspen Principles
on Long-Term Value Creation that call for executives to hold stock-
based pay until after retirement. Those principles must be embodied in
the regulation of financial institutions. We strongly support the new
SEC chair's effort to address the role played by weak boards and CEO
compensation in the financial collapse. With regard to credit rating
agencies, Congress must end the model where the issuer pays.
Financial reregulation must be global to address the continuing
fallout from deregulation. The AFL-CIO urges the Obama administration
to make a strong and enforceable global regulatory floor a diplomatic
priority, beginning with the G-20 meeting in April. The AFL-CIO has
worked closely with the European Trade Union Congress and the
International Trade Union Confederation in ensuring that workers are
represented in this process. We commend President Obama for convening
the President's Economic Recovery Advisory Board, chaired by former
Federal Reserve Chair Paul Volcker, author of the G-30 report on global
financial regulation, and we look forward to working with Chairman
Volcker in this vital area.
Reregulation requires statutory change, regulatory change,
institutional reconstruction and diplomatic efforts. The challenge is
great, but it must be addressed, even as we move forward to restore
workers' rights and revive the economy more broadly.
______
PREPARED STATEMENT OF THOMAS DOE
Chief Executive Officer,
Municipal Market Advisors
March 10, 2009
Introduction
Chairman Dodd, Senator Shelby and Committee Members: It is a
distinct pleasure that I come before you today to share my perspective
on the U.S. municipal bond industry. I am Thomas Doe, founder and CEO
of Municipal Market Advisors, that for the past 15 years has been the
leading independent research and data provider to the industry.
In addition from 2003 to 2005, I served as a public member of the
Municipal Securities Rulemaking Board (MSRB), the selfregulatory
organization (SRO) of the industry established by Congress in 1975.
The Market
There are nearly 65,000 issuers in the municipal market that are
predominantly states and local governments. Recent figures identify an
estimated $2.7 trillion in outstanding municipal debt. This is debt
that aids our communities in meeting budgets and financing society's
essential needs, whether it is building a hospital, constructing a
school, ensuring clean drinking water, or sustaining the safety of
America's infrastructure. A distinctive characteristic of the municipal
market is that many of those who borrow funds, rural counties and small
towns, are only infrequently engaged in the capital markets.
As a result, there are many issuers of debt who are inexperienced
when entering a transaction, and unable to monitor deals that may
involve the movement of interest rates or the value of derivative
products.
The Growth
According to the The Bond Buyer, the industry's trade newspaper,
annual municipal bond issuance was $29B in 1975 whereas in 2007
issuance peaked at $430B. In the past 10 years derivatives have
proliferated as a standard liability management tool for many local
governments. However, because derivatives are not regulated it is
exceptionally difficult, if not impossible, to identify the degree of
systemic, as well as specific, risk to small towns and counties that
have engaged in complex swaps and derivative transactions.
Systemic Risk Emerges
Municipal issuers themselves sought to reduce their borrowing costs
by selling bonds with a floating rate of interest, such as auction-rate
securities. Because state and local governments do not themselves have
revenues that vary greatly with interest rates, these issuers employed
interest rate swaps to hedge their risk. Issuers used the instruments
to transform their floating risk for a fixed-rate obligation.
A key factor in the growth of the leverage and derivative
structures was the prolific use of bond insurance.
The Penal Rating Scale
Municipal issuers are rated along a conservative ratings scale,
resulting in much lower ratings for school districts and states than
for private sector financial or insurance companies. Although most
state and local governments represent very little default risk to the
investor, the penal ratings scale encouraged the use of insurance for
both cash and derivatives to distribute products to investors and
facilitate issuer borrowing.
So instead of requiring more accurate ratings, the municipal
industry chose to use bond insurance to enhance an issuer's lower
credit rating to that of the higher insurance company's rating.
The last 18 months have exposed the risks of this choice when
insurance company downgrades, and auction-rate security failures,
forced numerous leveraged investors to unwind massive amounts of debt
into an illiquid secondary market. The consequence was that issuers of
new debt were forced to pay extremely high interest rates and investors
were confused by volatile evaluations of their investments.
Steps To Improve the Regulatory Context
The 34-year era of the municipal industry's self regulation must
come to an end. Today, the market would be in a much better place if:
First, the regulator were independent of the financial institutions
that create the products and facilitate issuers' borrowing.
Second, the regulator were integrated into the national regime of
regulation.
Third, the regulator's reach and authority were extended to all
financial tools and participants of the municipal transaction: ratings
agencies, insurers, evaluators, and investment and legal advisors for
both the cash and swaps transactions.
Fourth, the regulator were charged with more aggressively
monitoring market data with consumers' interests in mind, both
investors and issuers.
The good news is that this new era of regulatory oversight can be
funded by the MSRB's annual revenue $20-plus million, collected from
bond transactions, and can be staffed by the current MSRB policy and
administrative infrastructure.
Caution
I should be clear. The innovations of derivatives and swaps have a
useful application and have been beneficial for those for which they
are appropriate. However, it is also important that these instruments
become transparent and regulated with the same care as the
corresponding cash market.
Get This Done
It is critical to get this right. There is too much at stake.
Thank you for asking me to testify today, and I welcome your
questions during this session.
Municipal Market Advisors
Founded in 1995, MMA is the leading independent strategy, research
and advisory firm in the municipal bond industry. MMA's intelligent
approach to timely issues and analysis of market events has proven
invaluable to a wide range of clients. As conditions have become more
complex and difficult, MMA's recognized ability to concisely comment on
the key issues of the market is of critical importance and value. The
firm's independent research, data, market coverage and insight educate
and inform without bias or product agenda.
Our Clients: Investors, Dealers, Financial Advisors, Issuers and
Individuals: MMA's business has been predominantly portfolio managers
and dealer firms (with a focus on sales, trading and underwriting).
However, in 2007, demand for our services expanded to include issuers,
financial advisors, individuals and public finance professionals who
have recognized the increased value of accurate and insightful coverage
of current historical market conditions. MMA does not advocate on
behalf of its clients, we educate on behalf of the market.
Washington, DC--Educating and Working With Decision-Makers: MMA's
Washington DC office has enabled our firm to provide more direct
information to policy makers, regulators, trade associations and the
Federal Reserve. MMA's role is that of an educator to provide immediate
uncompromised assistance to entities that are actively engaged in
working on issues pertaining to the municipal industry.
Informing the Media: In 2008, more than 200 publications and media
outlets have sought MMA's expertise for definitive comment on the
issues confronting the industry. At no other time has accurate market
coverage been more valued, and trusted resources considered
indispensable. Unbiased information is important for correct
representation of market conditions, policy decisions and management of
portfolios.
Thomas G. Doe, Founder and CEO: Mr. Doe has been an analyst in the
municipal industry for 25 years with a consistent focus on pricing data
and information flows, investor and issuer behavior, and contextual
investing. He has addressed all of the leading groups in the municipal
industry, as Mr. Doe's insight, candor and historical context is sought
to establish a clear perspective of current conditions affecting
investors and issuers in the municipal cash and derivative markets. He
has been a featured speaker at numerous industry conferences and has
been frequently quoted in industry and national media. Mr. Doe's
leadership was recognized when he was named to a 3-year term with the
Municipal Securities Rulemaking Board, (MSRB) the regulatory entity of
the municipal securities industry in 2002. Mr. Doe received an
undergraduate degree from Colgate University in 1980 and a Masters
degree from Harvard University in 1984.
Background on How the Credit Crisis Has Affected the Municipal Market
The municipal market has suffered repeated shocks from the credit
crisis since August 2007. In a very primary sense, our sector was, and
in many ways continues to be, exposed to the same systemic risks that
collapsed the housing and securitization markets and undermined our
nation's banks. The deep interconnectedness of the municipal market
with the global financial and interest rate markets was unforeseen by
most municipal regulators, issuers, investors, advisors, lawyers, and
dealer banks; their surprise at, and misunderstanding of, the systemic
risks at work has consistently exacerbated problems over the last two
years. Further, there is little provision being made at present to
create a more resilient and stable market in the future.
The initiation of the credit crisis in municipals, as it was
elsewhere, began in 2001 and 2002, with the integration of leverage
into municipal bond buying strategies. Leveraged investment vehicles,
called Tender Option Bond (TOB) programs, borrowed low interest
(floating-rate) cash from the tax-exempt money market funds to invest
in higher yielding (fixed-rate) municipals. Not only does this strategy
capture the simple difference between the high long and low short
interest rates (the carry), but also TOB sponsors--which included hedge
funds, dealer banks, mutual funds, liquidity providers, and many
others--are placing bets on the tax-exempt market's outperformance of
carefully selected taxable bonds or swaps via interest rate hedging.
However, one of the key conditions for the safe operation of a TOB
was not implicit in the municipal market: liquidity. Because TOBs are
subject to mark-to-market accounting, margin calls, and periodic
adjustments of their leverage, they benefit from a well-traded and
accurately priced bond market. A TOB invested in securities with
unpredictable or volatile prices will itself provide unpredictable and
volatile returns. The municipal bond market, as we have detailed
elsewhere in this report, comprises 65,000 potential bond issuers and
1.5 million individual securities, most of which are rated along an
overly cautious rating scale that intentionally exaggerates the risks
and differences between individual issuers and bonds. Further,
municipal issuers have long sold bonds in serial maturities, with a
variety of interest rate coupons, call structures, security pledges,
etc. And finally, the bulk of municipal investors are households, who
either directly or indirectly though a manager, prefer to buy and hold
small pieces of multiple bond offerings: these are not active
securities trading operations.
This context was not conducive for TOBs, but, because their use of
leverage they were permitted to purchase municipal bonds at
substantially higher prices than other investors were willing to pay,
so the primary market rapidly adjusted to their needs. This entailed
the pervasive use of AAA-rated bond insurance and bank guaranties
(creating the appearance of safe homogeneity) and the facilitation of
very large, governmental-oriented bond sales carrying a standardized 5
percent coupon. For the period between 2002 and 2007, these adjustments
permitted the near doubling of annual bond issuance (from about $200Bn
to about $400Bn), and the amount of par volume municipal bonds
outstanding swelled 77 percent from $1.5T in 2001 to $2.7T today.
What's more, the rapid growth of TOB (and related strategy)
investment--along with a large increase in demand from property
casualty insurance companies riding post-9/11 waves of premiums and
profitability--allowed municipals to be priced more and more
aggressively, fulfilling the TOB investor's aim of outperformance of
the taxable bond market and encouraging ever larger allocations to this
strategy.
At the same time, the interest rates that tax-exempt money market
funds were receiving from the TOBs were better than an investor could
receive in a regular savings account, and aggressive TOB creation meant
a surfeit of product in which the money funds could invest. This
attracted more money fund deposits which, along with monetary policy,
kept short-term interest rates low. Municipal issuers themselves sought
to reduce their borrowing costs by selling long maturity, AAA-insured
bonds with a floating rate of interest, including variable-rate demand
obligations (VRDOs), which can be purchased by the money market funds,
and auction-rate securities, which were largely bought by individuals
and corporate cash managers. However, because state and local
governments do not themselves receive much revenue that floats with
short-term interest rates, these issuers also employed interest rate
swaps with these floating-rate bonds in an attempt to exchange their
floating-rate liability for a fixed one (but with the addition of
increased counterparty exposure to both a bank and the a bond insurer).
By these means, issuer interest rate swaps and derivatives became a
fundamental, but unregulated part of the municipal industry's standard
machinery, and systemic exposure to the financial sector, the bond
insurers, and, more importantly, the rating agencies' opinions of the
financial sector, and the bond insurers grew rapidly. In addition, our
market had become substantially vulnerable to fluctuations in the value
of taxable securities: remember that much if not all of the massive
investment by TOBs (estimated to have peaked near $500Bn although
little was done by the municipal regulators to even tabulate this
exposure) was hedged against the performance of Treasuries, LIBOR
swaps, or other slightly more muni-centric derivatives.
The problems with this arrangement were exposed in August 2007 with
the first surge of flight-to-safety buying of Treasury securities on
news of worsening damage to the housing sector. Stronger Treasury (and
LIBOR) prices created losses in TOB hedges, forcing margin calls that
rapidly consumed available cash. In addition, sharp increases in the
overnight lending rates pushed floating-rate product credit spreads
wider: the source of TOB leverage, loans from the money funds, grew
much more expensive, to the point where the money funds were demanding
almost as much (or more) interest than the TOBs were receiving from
their long-term, fixed-rate municipal position. Some TOBs thus began to
liquidate their positions, forcing sales of their fixed-rate bonds into
a municipal secondary market that quickly became oversupplied and
illiquid. Keep in mind that, up until that point, the TOBs had been
purchasing bonds at (and driving market clearing prices and statement
evaluations to) higher levels than traditional institutional investors
were reasonably willing to pay. Thus, when the TOBs needed to quickly
sell their bonds to these same traditional buyers, large price
concessions were required. Dealer banks helped soften the effects by
acquiring bonds into their own trading inventories, but ultimately
market pressures forced municipal bond yields sharply higher (while
Treasury yields were moving sharply lower). Higher yields attracted
enough demand to stabilize the market by the end of the month, but,
through the end of the year, nervous investors repeated this pattern of
fast selling/recovery, heightening volatility in prices, and
encouraging a steady reduction in TOB investment. For substantially
more detail on the daily and weekly evolution of our the market, please
see the complete catalog of published MMA research, available to
subscribers on our Web site and to Congressional staffs on request.
Importantly, market participants had by this time also become
increasingly concerned about the future of the bond insurers, who had
guaranteed subprime residential mortgage securitizations. Research
firms such as MMA and private investors amplified former warnings about
these companies. In particular, more cautious corporate cash managers
began selling auction-rate securities that had been marketed to them,
in part, based on the apparent safety of AAA-rated bond insurance. Once
again, dealer banks managing auction-rate programs provided liquidity
in the absence of incremental investor demand, but in December 2007,
the rating agencies sounded formal warnings about the bond insurers.
This precipitated vast selling pressure among auction-rate investors
that, in January, overwhelmed dealers' risk tolerances for buying back
additional auction paper, and auctions began to fail (please see
Auction Rate Securities, below).
Auction-rate securities paying high penalty rates attracted
investors away from other fixed- and floating-rate products, forcing
both fixed and floating rates up sharply. At the end of February 2008,
TOB programs were once again forced to sell bonds to pay margin calls,
to unwind their leverage that had grown too expensive, and to afford
investor redemptions. Extreme selling and uncertainty led to widely
divergent pricing decisions across the industry; liquidity was almost
completely interrupted, and state and local issuers were temporarily
shut out of the capital markets.
Once again, high yields galvanized demand in March, and from that
point until December 2008, the municipal market continued to face boom
and bust pricing cycles of sometimes extraordinary depth. In general,
these entailed yield-fueled, or media-driven demand bubbles that were
ultimately pricked by yet another bond insurer downgrade that renewed
fears and sometimes forced selling by leveraged bondholders. The worst
of these cycles began in September, when the collapse of Lehman
Brothers, plus concerns over other broker-dealer counterparties were
realized in investor redemptions from municipal money markets, which
put large numbers of variable rate obligations back to dealers. The
flow of bonds initially overwhelmed dealer balance sheets, forcing the
unwind of some proprietary positions, but was ultimately managed
through dramatically higher floating rates (the municipal industry's 7-
day floating rate reset from about 2 percent to 8 percent and credit
spreads to that rate widened sharply, in particular for TOBs because of
their reliance on multiple layers of bank support) and the temporary
withdrawal of a large number of floaters from active markets onto
liquidity provider balance sheets. Still, higher floating rates forced
many tender option bond programs to unwind their trades for perhaps the
final time, as investors now began demanding their money back in
earnest.
The excess supply created by forced TOB selling in September to
November, along with downgrades to the bond insurers, pushed municipal
yields sharply higher, prices lower. Institutional buyers retreated
from the public markets until the end of the year (although many large
buyers were able to buy portfolios of highly discounted bonds in the
evenings and weekends, muffling the implications of these very cheap
trades on broader market pricing), causing credit spreads to widen
dramatically. Spread widening and price declines hurt tax-exempt mutual
fund net asset values, giving the appearance of undue credit risk to
their investors and initiating perhaps the largest sequence of mutual
fund investor outflows (and thus forced selling of related holdings by
the funds) on record. And, as was well covered by the media, with
fixed-rate yields having risen to extraordinary heights, many state and
local issuers chose to table the majority of their planned primary
market loans, waiting for conditions to improve. Indeed, smaller,
lower-rated, and riskier credit issuers may have at least temporarily
been unable to access capital at all, but large states and cities were
always able to raise money; their decisions were based on price. MMA
estimates that, in 2008, more than $100Bn of planned new-money
infrastructure projects were delayed, the majority of that occurring in
the fourth quarter.
Persistent institutional demand has not yet returned to the
municipal market, but since the start of 2009, municipal fund managers
and brokerages have been highly successful attracting retail investment
on the back of both flight-to-safety allocations (out of equities) and,
more importantly, on speculation that the stimulus will ultimately
drive up municipal bond prices. In fact, yields on the kind of bonds
favored by retail investors touched two-decade lows in mid-January,
although they have since begun to retreat again. Lower-rated, risky
credit issuers (like hospitals) still face difficulty finding cost-
effective market access and even highly rated state and local
governments are commonly required to downsize new bond issues or risk
pushing market yields higher.
Summary of Regulation Issues
Introduction
The Municipal Securities Rule-making Board (MSRB) is a self-
regulatory organization (SRO) and was formulated by Congressional
statute in 1975. Please see the attached National Federation of
Municipal Analysts White Paper``Federal Securities Law Relating to
Municipal Securities,'' for background and more detail.
During Thomas Doe's tenure as a Board member from 2003 to 2005,
there was rarely a Board meeting where the subject of derivatives was
not discussed and the risks to the industry and investors were not
addressed. However, the outdated statute limited the Board's regulatory
purview to municipal cash securities and to activities of dealers and
dealer banks. Proactive action was inhibited for three reasons: (1) it
was exceptionally difficult, however well intended, for Board members
representing security firms to advocate for change that would reduce
the revenue of its firm; (2) the volunteer nature of the Board resulted
in a consistent deferral of strategy, tactics and policy to staff; (3)
the Chairman of the Board served only one year and dictated the Board's
focus, which, in our opinion, was to sustain the status quo and could
again be heavily influence by staff. Since staff, especially the
Executive Director, worked for the Board, it appeared to be
exceptionally difficult for innovation and proactive regulation to
occur.
To be fair, there is now new leadership of the MSRB's staff.
However, the negative characteristics of a: (1) short-tenured Chairman;
(2) volunteer Board; and (3) the tremendous challenge to advocate for
the investor or issuer interest, which could hurt an employer's revenue
stream, are still present. These conditions can be inhibitive toward
regulation in the best interest of the consumer--both issuers and
investors.
Opportunity
In 2009, led by the catalysts of curtailed institutional demand,
limited issuer access to the capital markets and the allegations
revolving around municipal finance practices in New Mexico, the MSRB
has suggested a review of the Congressional regulatory statute created
34 years ago. Specifically the Board has suggested an expansion of
entities to be regulated swap advisors. The willingness of the Board to
advocate change is applauded however, the action falls short.
Necessary Change
More entities should not alone be regulated, but rather legislative
language should be expanded to be inclusive of all practices and
products in which financial institutions would be involved related to
municipal finance. By regulating the products, all entities involved
with municipal finance--from creation to distribution--would be
governed by transparency and regulations, which would advance and
define a context for transactions in the municipal industry for the
protection of issuer, dealer and investor. Only in this manner can
responsibility and integrity be promoted and transparency ensured. The
byproduct of such attention to derivatives would accomplish the
disclosure required by issuers to both inform investors and those who
choose to provide capital to public entities.
Action Items
1. End the MSRB as an SRO.
2. Integrate the MSRB formally and directly into a larger entity,
possibly the Securities Exchange Commission, Treasury or
Federal Reserve.
3. Congress expand the regulatory purview of municipal regulation to
include all participants in municipal finance and all financial
tools involved in a municipal finance transaction--this would
include derivatives and swaps in addition to the cash market.
Along with dealers: advisors, ratings agencies, and evaluation
services would be included in the new regulatory scheme.
4. Ensure that the regulatory statue was adaptable and flexible to
allow regulation to be proactive and timely.
5. Include the municipal industry within an organization, where its
regulatory framework, data and action can be more easily
coordinated with larger markets. (Too often critical regulation
may not have been enacted or suggested as the industry is small
relative to equities and taxable fixed-income. One might argue
that the vulnerability of the eclectic resources of the 65,000
municipal issuers/borrowers of the industry demands more
vigilant protection because of the critical importance of the
financings to essential services and projects for town,
counties and states in the US.)
6. Mandate better regulatory coordination with its consumers--
specifically issuers and investors--not simply the dealer
community.
7. Demand greater financial forensics to mine the vast municipal
transaction data created by the Real-Time Transaction Reporting
System in order to better indentify market behavior that can
adversely impact (i.e., volatility) issuer pricing and investor
evaluations. In addition, better data analysis can better
define conditions of market liquidity to assist market
participants in risk management strategies and investors to
better use performance data measurements, specifically indices
of price performance and returns. This report highlights
significant areas where more robust data collection would have
helped manage and avert systemic risks exposed in the credit
crisis.
Conclusion
The municipal industry has evolved outside of a confined regulatory
context that is outdated and biased, and been consistently challenged
by the temptation to regulate in its self-interest. The evolution
resulted in detrimental practices and products that have proved penal
to investors, issuers and the financial institutions. The opportunity
to broaden the current regulatory framework has presented itself and in
acting to take steps to protect the public entities, which require
access to capital for infrastructure, the new broad regulation of the
municipal industry with specific attention to both derivatives and cash
financial products will provide precedence for global regulatory reform
of all derivatives.
The best news is that the MSRB's current major funding mechanism,
fees from municipal transactions (more than $20 million in 2008),
provides a revenue stream to fund expansion and transition of the
regulatory purview. In addition, the existing organizational
infrastructure of the MSRB allows for experienced personnel, technology
and data to be powerfully integrated in a revitalized context.
The municipal regulatory entity must be independent of those it
regulates and integrated within a regular Federal entity where the
industry can be included and coordinated with regulation of the larger
markets.
Disclosure and Investor Protection
For a background on municipal disclosure, MMA here quotes from the
National Federation of Municipal Analysts March 2008 ``White Paper on
Federal Securities Law Relating to Municipal Securities.'' The full
paper is attached at the end of this report.
The SEC promulgated Rule 15c2-12 (the ``Rule'' or ``Rule 15c2-
12'') in 1989 and amended the Rule in 1994 to include
continuing disclosure requirements. . . . Direct regulation of
issuers would have required repeal of the Tower Amendments, so
the Rule instead applies to municipal broker-dealers and
generally applies to financings where the principal amount
offered is $1 million or greater. The Rule applies indirectly
to issuers, effectively denying their access to the market
unless the Rule's requirements are satisfied. The Rule contains
primary disclosure requirements and continuing disclosure
requirements. With respect to continuing disclosure, the Rule
prohibits the purchase and sale of municipal securities by an
underwriter in a public offering unless the issuer or an
``obligated person'' undertakes to provide continuing
disclosure. Continuing disclosure obligations include both
periodic reporting of financial and operating information and
disclosure of the occurrence of any of a specified list of 11
events, if material. The annual information is required to
include audited financial statements when available and
material financial information and operating data of the type
included in the official statement for the securities. . . .
Independent of contractual undertakings made by issuers and
conduit borrowers and continuing disclosure obligations under
Rule 15c2-12, the SEC maintains that issuers of municipal
securities and conduit borrowers have continuing disclosure
responsibilities under Section 10(b) of the Exchange Act and
Rule 10b-5. While issuers and conduit borrowers have no
affirmative duty to disclose information (unless they are
engaged in the offering, purchase or sale of securities or
unless disclosure is required under a continuing disclosure
undertaking), if an issuer or conduit borrower chooses to
disclose information to the market it is prohibited from
disclosing information that is materially untrue or misleading,
or that contains a material omission, ``in light of the
circumstances'' in which such information is disclosed. There
are no other limits on the issuer's or the conduit borrower's
disclosure.
We also reference DPC Data's report, ``The Consequences of Poor
Disclosure Enforcement in the Municipal Securities Market'' that
provides more information on how disclosure is disseminated. Currently
disclosure occurs through a regime of several repositories (Nationally
Recognized Municipal Securities Information Repositories, or NRMSIRs),
but, with recent change in law, a single repository will exist: the
Municipal Securities Rulemaking Board. In MMA's opinion, the state of
disclosure in the municipal sector should be regarded as poor, and
recent changes in the law are unlikely to make much difference here.
Issuers, as detailed by DPC data's important (and accurate) study on
the topic, regularly fall out of compliance with stated disclosure
requirements, undermining liquidity in selected bonds and hurting
smaller investors (those without credit analysts trained to track down,
or mitigate the impact of, absent financial and operating data) who buy
bonds, in part, based on statements in the prospectus that regular
information will be disclosed.
In MMA's opinion, disclosure gaps occur because: (1) many issuer
representatives are not capital markets professionals and lose track of
their responsibilities, and (2) there is little penalty to be suffered
by the industry for not policing compliance. A specific failing of SEC
Rule 15c2-12 is its leaving the decision on whether an issuer is in
disclosure compliance to the individual participants trading the
issuer's securities. In our experience, firms have generally ignored
this requirement and continued to trade likely safe, but disclosure-
gapped bonds, albeit at a slight discount. Further, we note a pattern
of smaller issuers falling out of compliance almost immediately after a
new offering, remaining out of compliance for several years until, just
prior to another new primary market loan, the issuer will send its past
due financial information to the information repositories.
Again, MMA believes a solution to municipal disclosure problems is
available:
1. We believe Congress should require that the SEC act as arbiter to
determine whether each issuer is in compliance with their
stated disclosure requirements. This would be a very large
undertaking, potentially requiring a large staff increase by
the SEC. Should the SEC subsume the MSRB, the MSRB's funds
could offset at least a portion of the cost.
2. Bonds found to be not in compliance would be flagged, and
registered firms would be prohibited in trading in such until
either the issue's original underwriter or any other investor
can succeed in getting the issuer to remedy the gap. We are
reluctant to advise that the SEC be able to compel disclosure
directly from the issuers for fear of abridging state autonomy.
3. The SEC would keep a database to track, for every Cusip and
borrower, the number and percent of days it has been out of
compliance on all of its outstanding bond issues. This
statistic would be vitally important for potential buyers
evaluating new purchases of the borrower's securities.
4. Additionally, all firms trading municipal bonds, regardless of
their status, would need to track how many trades, and the
volume of par traded, that firm had made with disclosure-
flagged municipals Cusips. Again, this could be very important
data for investors evaluating with which firm to invest their
money.
5. MMA also believes that all tax-relevant calculations and
investigations should be included in required disclosure
topics. These include how tax-exempt bond proceeds are being
spent, on a weekly basis, the precise formula by which bond
counsel determines that a bond issue is tax exempt, and the
presence and status of any SEC investigations.
The Undisclosed Risk of Bank Bonds and Swaps
MMA's principal concern for the municipal sector in 2009 is that
variable-rate related problems will set off a wave of downgrades and
even defaults among risky sector credits (such as hospitals and private
universities), creating incremental economic loss and threatening more
investor aversion to municipal bonds generally. But the risks in
variable-rate demand obligations are not exclusive to hospitals; many
state and local governments also issued these securities and face very
similar credit challenges.
VRDOs are long maturity bonds where the interest rate is
periodically (weekly, daily, etc.) reset by a remarketing agent--
usually a dealer bank--who also attempts to make proprietary markets in
these securities among a universe of the firm's clients with a strong
focus on tax-exempt money market funds. VRDOs also entail some form of
liquidity support (structured via a letter of credit or standby
purchase agreement) from a highly rated bank. In other words, a bank is
contractually obligated to become the immediate buyer of last resort
for a VRDO, giving money market funds confidence in the liquidity of a
VRDO investment. MMA estimates that there are about $500Bn of
outstanding VRDOs at present; this number has likely increased from
$400Bn since the start of 2008 reflecting numerous post-collapse ARS
restructurings into VRDOs.
Yet today's financial markets entail substantially more investor
caution among banks and between credits generally, and large numbers of
VRDOs have been rejected by the money funds because of their reliance
on a damaged or downgraded liquidity provider (most notably DEPFA and
Dexia) or connection to a downgraded bond insurer. In the absence of
other investors or remarketing agents' inability to bring yet more
bonds onto their own balance sheets, many of these rejected bonds have
triggered their liquidity features, requiring the liquidity providers
to buy these securities directly. Provider-purchased VRDOs are referred
to as ``bank bonds,'' which the liquidity providers hold as available
for sale for a period of time (for example, 90 days), but then convert
to accelerated maturity term loans between the liquidity provider and
the issuer. It is unclear whether any municipal bank bonds have
actually yet converted to term loans, but their acceleration of
principal and penalty interest rate would reasonably require either an
immediate restructuring or a default forbearance agreement between
provider and issuer. Because there is little hope for market interest
in Dexia or DEPFA to improve, at some point, issuer defaults may become
public. MMA estimates, based on our polling of industry sources, that
there have been as many as $50Bn of rejected floaters, with perhaps
$50Bn more being kept away from liquidity providers through special--
and thus potentially temporary--intervention by securities dealers. MMA
believes that the amount of bank bonds has fallen in 2009, as issuers
are actively restructuring their bank bond obligations, although we
underscore that we are unaware of any information being collected by
any regulator or data provider on this topic.
Interestingly, the rejection of many VRDOs by the money funds has
worsened problems elsewhere in the municipal floating-rate markets.
First, it has required liquidity providers to become more cautious in
writing new policies, increasing the scarcity and cost of same for
municipal issuers. Second, by removing large swathes of floaters from
money fund ``approved'' lists, and noting: (1) the near complete
absence of TOB-related lending by the money funds (see ``Background''
section above), and (2) large, fear driven investor inflows into the
money funds, has created a severe supply/demand imbalance. Approved and
available securities are scarce and--because the funds' alternative is
not investing their funds at all--are being bid up to extremely low
yields (weekly interest rates have been close to, or well below 1
percent since November). Low benchmark floating rates, along with very
strong demand for long-maturity LIBOR swap rates, an unwinding of
arbitrageurs' interest rate hedges, and a dearth of new municipal
issuer derivative activity, has pushed the related long-maturity
municipal swap rates to very low levels. And this movement in swap
rates has greatly increased issuers' cost of terminating any
outstanding swap, complicating the restructuring of any distressed VRDO
position. Further, higher issuer swap termination costs have produced
substantial cash drains away from issuers via requirements that the
issuers collateralize their potential termination liability to their
counterparty. (Many issuers had purchased bond insurance AAAs to ward
off credit- or rate-driven collateralization requirements. But with the
insurers' losing their ratings, many issuers are no longer shielded,
sometimes removing cash to the detriment of normal operations.) As not-
for-profit hospitals have been particularly large users of this debt
structure, and as these same hospitals also face lower private pay
revenues and strained governmental reimbursements, defaults are likely
in the near term. Because swap positions are only dimly disclosed, even
sophisticated municipal investors remain largely without information on
their own portfolios' related risks.
Once again, these municipal issuer exposures to systemic risks were
accreted with little public disclosure or regulatory insight. Further,
MMA is unaware of any municipal regulator or information provider
systematically collecting information on the size and scope of this
problem. This not only inhibits better projection of potential losses,
but also prevents a more robust response from national regulators
(e.g., Treasury, the Federal Reserve, the SEC) who are struggling to
grasp the depth of the problem and coordinate their response with those
in other asset classes. Solutions are well within Federal abilities.
MMA recommends that Treasury extend subsidized loans to municipal
issuers to terminate difficult swap positions, with the cost of those
loans recouped by Treasury via a surcharge on all future issuer swap
activity. This would allow issuers to restructure their obligations
into fixed rate bonds, relieving liquidity provider balance sheets of
troubled exposures, and potentially encouraging future policy writing.
Auction-Rate Securities and Unchecked Systemic Risks
Auction-Rate Securities (ARS) are long maturity bonds where the
interest rate is periodically reset by auction among potential
investors, or failing that, set manually by a bank pursuant to an index
or (typically very high) fixed rate. Because they are valued at par,
ARS were typically purchased by individual investors as a higher-
yielding alternative to cash deposits. However, higher yields reflected
the fact that an ARS holder cannot sell their bond without an
identified buyer: a sharp distinction from other ``cash like''
instruments that required dealer banks to periodically step in as a
buyer to prevent auctions from failing. In part because of this
reliance on bank intervention, ARS programs were (and still are) set up
as proprietary trading exchanges by individual dealers, inhibiting the
easy flow of capital and information from program to program.
The implications of the ARS structure, in the context of the
municipal industry's systemic exposure to the bond insurers and the
financial counterparties were little understood by issuers, investors,
or the dealer banks themselves prior to 2007. It was the collapse of
the bond insurers in 2007 that undermined investor confidence in ARS
issuers and precipitated vast selling. (Remember that individual
investors had long been sold on the AAA virtues of bond insurance; this
myth was not so easy to dispel when bond insurer downgrades began).
Banks were initially able to use their own cash to buy back securities,
but in January 2008, bank risk tolerances prevented further purchases,
and ARS auctions began to fail. Thus, current holders were left without
a means to get out of their positions, and issuers were forced to pay
sometimes highly punitive fixed interest rates. Since that time, MMA
estimates that about two thirds of ARS issuers have restructured or
refinanced their securities, although many remain unable to do so as:
(1) refinances with liquidity-supported floating-rate debt require the
purchase of a liquidity policy from a highly rated bank--these policies
have become both scarce and expensive as U.S. banks have reduced
lending; and (2) refinances with fixed-rate debt are prevented not only
by the high fixed rates many lower-rated issuers must now pay, but also
the sometimes staggering cost of terminating the interest rate swap
most municipal issuers have connected to their bond sales (please see
bank bonds section, above).
Thus, many investors still remain stuck with highly illiquid
securities that are paying well-below-market, index-linked interest
rates. At issue is an unwillingness to allow ARS to trade at a discount
to entice potential buyers, because of the increase in potential
liability and because sub-par pricing of these holdings could result in
additional waves of mark-to-market losses for the already stressed
banks. Although private trading venues have emerged to provide
emergency assistance to distressed clients needing to liquidate their
holdings at any price, we are unaware of any broker dealer making sub-
par markets in any ARS. On the other hand, several of the large banks
have, on the intervention of state securities regulators, settled with
their individual and small institutional investors, in effect buying
ARS securities back at par. Indeed, the largest current holders of ARS
are likely the dealer firms themselves that are still carrying their
swollen inventories from 2008 and now the bonds purchased via
settlement.
ARS shows another breakdown in the municipal regulatory framework.
While there are initiatives to improve ARS price discovery, no market
participant (including investors, dealer banks, nor the regulators
themselves) knows precisely how many ARS are outstanding (MMA's
estimate was about $200Bn municipal ARS at the market's peak), how many
bonds were being placed through each dealer program, how many ARS
issuers were reliant on bond insurance for their marketing to
investors, the extent and means by which these issuers were leveraging
counterparty credit through interest rate derivatives, and how much
dealer support was being directly extended to the market. The
implications for systemic risk management, as now being discovered in
the credit crisis, are clear in these questions, which can (and should)
be extended to the still healthy, but periodically threatened,
municipal VRDO market.
Municipal Bond Ratings and Bond Insurance
Most municipal bonds are rated on a different, more conservative
rating scale than corporate bonds. Moody's and Standard & Poor's have
shown that triple-A U.S. corporate bonds have up to 10 times the
historical default rate of single-A municipals. In MMA's opinion,
neither municipal issuers, nor the individual investors who own the
large majority of outstanding paper or fund shares, understands this
point. But instead of requiring more accurate ratings, the municipal
industry (i.e., issuers, investors, and underwriters) has instead
chosen to make bonds appear safer and more similar through bond
insurance (the insurers are rated along the more generous corporate
rating scale; much of the bond insurance model distills to simple
arbitrage between the two rating scales). At its peak, the municipal
bond insurance industry entailed just nine companies whose ratings were
applied to more than 50 percent of annual municipal bond sales. And
this invited massive systemic exposure into the municipal industry as
the bond insurers carried in the risk of subprime mortgage-backed
securities, the insurers' and the financial sectors' leverage of
ratings on securitized debt, and failed rating agency models.
Attached, please find our January 2008 report, ``MMA on Corporate
Equivalent Ratings,'' and our April 2008 report, ``Second Research Note
on Moody's,'' for more detail on the problem with how municipal bonds
have been rated. In the last year, both Moody's and Fitch ratings
strongly considered reforming their muni rating processes, but both
have tabled these initiatives because of the recession. Standard and
Poor's continues to deny the existence of separate rating scales for
municipals and corporate bonds, but that agency has embarked on a plan
of sweeping upgrades to selected municipal sectors. Finally, the U.S.
House of Representatives considered the ``Municipal Bond Ratings
Fairness Act of 2008,'' which MMA believes would, for little cost to
taxpayers, successfully remediate much of the rating problem in our
sector. We strongly recommend that Congress adopt this legislation in
its current form. MMA has been a leader on the topic of ratings and the
municipal sector's use of bond insurance; we welcome any opportunities
to continue to educate Congress and its agents on these topics.
Pricing and Evaluation Issues
The events of the past 18 months have amplified the risks and
challenges associated with illiquidity and limited price discovery for
municipal bond investors and issuers.
The municipal bond industry has been challenged with a troublesome
irony. While municipal bonds have favorable low historical default
risks, the securities can be illiquid. How can a safe investment not
have liquidity? Inconsistency of primary market pricing, the eclectic
composition and multitude of issuing entities, the penal and overly
granular ratings scale, reduced number of liquidity providers, the
diminished number of AAA bond insurers and the inability to manage
interest rate and credit risk have contributed to the challenges for
the markets transactions to provide evaluation services with sufficient
price discovery. The result is that evaluations that represent the
price that investors receive on their investment firm statements or the
prices that comprise the net asset value of a mutual fund share may
bear little resemblance to an execution price should an investor choose
to buy or sell. In addition, the periodic illiquid market conditions
and limited price can result in sharp volatility that can be
misinterpreted as credit or default risk, either of which may not be
valid. In this manner the data can misinform an investor and
potentially prompt emotional and inappropriate investment decisions.
These same characteristics can also increase the difficulty for
municipal issuers to assess market conditions and accurately predict
market demand to give context for the pricing of their primary market
deal.
An aggressive, investigative and knowledgeable regulator with
access to all transactions and who conducted each transaction, can
assist consumers--both the investor and borrower--with providing a
context to ensure that the data is relied upon by consumers inspires
confidence and provides an objective context in which investors and
issuers can make decisions from the prices of their securities.
Schedule of Additional Attachments
MMA has attached the following documents, under separate Acrobat
file, in support of the arguments made herein.
NFMA White Paper ``Federal Securities Law Relating to
Municipal Securities'' March 2008
DPC Data ``The Consequences of Poor Disclosure Enforcement
in The Municipal Securities Market'' January 2009
Municipal Market Advisors ``Corporate Ratings for Munis''
January 2008
Municipal Market Advisors ``Second Research Note on
Moody's'' April 2008
PREPARED STATEMENT OF LYNN E. TURNER
Former Chief Accountant,
Securities and Exchange Commission
March 10, 2009
Thank you Chairman Dodd and Ranking Member Shelby for holding this
hearing on an issue important to not only investors in America's
capital markets, but to all who are being impacted by the current
economic devastation.
Before I start with my personal perspective on the issues
surrounding the current economic crisis and securities regulation, it
might be worthwhile to provide some background on my experience. I
serve as a trustee of a mutual fund and a public pension fund. I have
served as an executive of an international semiconductor manufacturer
as well as on the board of directors of both Fortune 500 and small cap
public companies. In the past, I served as chief accountant of the U.S.
Securities and Exchange Commission (SEC) and as a partner in one of the
major international auditing firms, where I was involved with audits
and restructurings of troubled or failed institutions. I also was the
managing director of research at a financial and proxy advisory firm.
In addition, I have also been a professor of accounting at a major U.S.
public university and an investor representative on the Public
Companies Accounting Oversight Board (PCAOB) Standards Advisory Group
and the Financial Accounting Standards Board's (FASB) Investor
Technical Advisory Committee (ITAC).
The Crisis--Bad Loans, Bad Gatekeepers, and Bad Regulation
The economic crisis of 2007-2009 has three root causes; the making
of bad loans with other peoples money, gatekeepers who sold out, and a
lack of regulation. In order to prevent a repeat of this debacle it is
of paramount importance that policy makers understand what will cure
the ``disease'' before they remedy the cause. To that end, I would urge
the committee to take the same approach it did some seven decades ago
when the Senate Banking Committee, with experienced investigators using
its subpoena powers, investigated the banking and security markets,
stock exchanges, and conduct of their participants. A similar approach
in the midst of the current crisis would give Americans and investors
hope and confidence that their interests will be served, and adequate
protections restored. Unfortunately, if the public perceives the remedy
is off target, as it has with other recent legislation, I fear the
markets will continue their downward spiral resulting in a lengthening
of the recession, or potentially worse outcome.
From my perspective, those most responsible for the current crisis
are the banks, mortgage bankers, and finance companies who took money
from depositors and investors and loaned it out to people who simply
could not, or did not repay it. In some instances predatory practices
occurred. In other instances, people borrowed more than they should
have as Americans in general ``leveraged'' their personal and corporate
balance sheets to the max. Speculators also took out loans expecting
that real estate values would continue to rise, allowing them to profit
from flipping their investments. But who can dispute that when
``liar,'' ``no doc,'' and ``Ninja loans'' are being made while banking
regulators are watching, there is something seriously wrong.
In addition to the financiers, a second problem was the
gatekeepers--the credit rating agencies and underwriters--who are
suppose to protect investors. They did anything but that. Instead they
became the facilitators of this fraud on the American public, rather
than holding up a stop sign and putting the brakes on what was
occurring. They became blinded by the dollars they were billing rather
than providing insight to the public into the perfect storm that was
forming. Recent testimony before the House of Representatives that the
rating agencies knew their models did not work, but did not fix them
was stunning. But perhaps not as stunning as the report of the SEC in
which employees of an agency stated they would rate a product even if
it had been created by a cow.
And while lenders were making bad loans in exchange for up-front
fees, and gatekeepers were falling down on the job, Federal Government
agencies were failing to supervise or regulate those under their
oversight, as well as failing to enforce laws. It is a huge public
concern that a systemic failure of financial and securities market
regulation in this country occurred. Some of this was due to the lack
of regulation of new products and institutions, such as credit default
swaps and hedge funds, but more importantly, the fundamental problem
was the lack of Federal Government regulators doing their jobs, or
lacking the resources to do so.
For example, for 13 years, as abuses of subprime lending occurred,
the Federal Reserve refused to issue regulations as mandated by the
Homeownership Equity Protection Act of 1994 (HOPEA). That legislation
specifically stated:
PROHIBITIONS--The Board, by regulation or order, shall prohibit
acts or practices in connection with-- ``(A) mortgage loans
that the Board finds to be unfair, deceptive, or designed to
evade the provisions of this section; and (B) refinancing of
mortgage loans that the Board finds to be associated with
abusive lending practices, or that are otherwise not in the
interest of the borrower.''.
Not less than once during the 3-year period beginning on the
date of enactment of this Act , and regularly thereafter, the
Board of Governors of the Federal Reserve System, in
consultation with the Consumer Advisory Council of the Board,
shall conduct a public hearing to examine the home equity loan
market and the adequacy of existing regulatory and legislative
provisions and the provisions of this subtitle in protecting
the interests of consumers, and low-income consumers in
particular . . .
Yet the Federal Reserve Board (Federal Reserve or Fed), which had
examiners in the very banks who were making mortgage loans, did
nothing. Had the Federal Reserve acted, much of the subprime disaster
might have been averted. Instead, ignoring the clarion calls of one of
its own Governors for action, the late Edward Gramlich, it was not
until 2007 that the Federal Reserve acted. But by then, much of the
damage to the American economy and capital markets had been done.
Indeed, even the Comptroller of the Currency spoke in 2006 of 3
years of lowering of lending standards. In a press release in 2006, the
Comptroller stated:
``What the Underwriting Survey says this year should give us
pause,'' Mr. Dugan said. ``Loan standards have now eased for
three consecutive years.'' The Comptroller reported
``slippage'' in commercial lending involving leverage lending
and large corporate loans as well as in retail lending with
significant easing in residential mortgage lending standards
including home equity loans. [Emphasis supplied]
Unfortunately, armed with this information and legislative
authority to fix the problem, the Comptroller of the Currency (OCC)
failed to act in earlier years. Rather than reining in these abusive
practices, the OCC permitted them to continue, with the most toxic of
the subprime loans being originated in 2006 or 2007. And today, we have
Inspector General reports that have cited the lack of action by the OCC
and Office of Thrift Supervision, leaving taxpayers and investors
exposed to losses totaling trillions of dollars.
What is equally troubling about this lack of action by the banking
regulators, is that it comes after similar problems occurred with the
crisis in the savings and loan and banking industries in the 1980s and
early 1990s. I was at the SEC at that time and watched as the Federal
Reserve who had oversight over an undercapitalized CitiBank, worked to
keep it afloat. It seems that we are seeing a repeat performance of
this situation and rather than having learned from history, we are
again repeating it. After having two swings at the bat, I wonder why
some want to make the same regulators the risk regulator for the entire
financial system in the United States. These are regulators who all too
often have been captured by the regulated.
Once again, as with Enron, a lack of transparency has also been a
contributing factor to the current crisis. Investors have time and time
again--from Bear Stearns to Lehman to Wachovia to Citigroup and Bank of
America--questioned the validity of the financial numbers they are
being provided. The prices of their stocks have reflected this lack of
credibility driven by transactions hidden off the balance sheets and
values of investments and loans that fail to reflect their real values.
Unfortunately, millions of bad loans were made that are not going
to be repaid. While financial institutions argue they will hold the
loans to maturity and be repaid, that just isn't true for loans subject
to foreclosures or short sales. And for many mortgages, they prepay and
once again are not held to maturity. At the same time, collateral
values of the underlying assets securing the loans have taken a
tremendous tumble in values. Almost 5 million Americans have lost their
jobs since this recession began impacting their ability to make their
mortgage payments. There is a years worth of inventory of unsold homes
on the market even further depressing home prices. Asset backed
securities are being sold in actual transactions at pennies on the
dollar. Yet the financial institutions continue to act like an ostrich
with their head in the sand and ignore these facts when valuing their
assets. At the same time however, the markets are looking through these
numbers and revaluing the stocks in what is an inefficient approach,
driving stocks of some of the largest financial institutions in this
country to a price that is lower than what you can buy a Happy Meal for
at McDonalds.
In 1991 the General Accounting Office (GAO) published a report
titled ``Failed Banks--Accounting and Auditing Reforms Urgently
Needed.'' In their report, the GAO noted how during the savings and
loan crisis, the failure of banks and savings and loans to promptly
reflect their loans and assets at their market values drove up the cost
to the taxpayer. I hope Congress will not allow this mistake to be
repeated by allowing banks to avoid marking their assets to market.
Managing the assets held by a financial institution and the
positions taken has also been lacking. One large institution that was
failing and required a bailout through a buyer did not even have a
chief risk officer in place as the risks that caused their demise were
entered into. This could have been avoided in if the recommendations of
the 2001 Shipley Working Group on Public Disclosure had been adopted by
the banking and securities regulators that had convened the group.
Instead, consistent with a deregulatory approach, the type of risk
disclosures the group called remained nonexistent, hiding the buildup
of risks in the financial system.
There has also been a lack of regulation of new products and
institutions. Credit rating agencies were not subject to regulation by
the SEC until after many of the subprime loans had been made. Credit
default swaps and derivatives were specifically exempted by Congress
from regulation, despite a plea for regulation from the CFTC chairman,
creating grave systemic risks for the financial system. These markets
grew to over $60 trillion, a multiple of many times the actual debt
subject to these swaps. In essence, a betting system had been
established whereby people were wagering on whether others would pay
their debt. But while we regulate betting in Las Vegas, congress chose
to specifically not regulate such weapons of mass destruction in the
capital markets. This has directly led to the more than $160 billion
bailout of the bets AIG placed, and those to whom it is indebted on
those on those bets.
Likewise, there has been a rise in a shadow banking system that
includes hedge funds and private equity firms. These funds have under
management money from many public sources, such as public pension funds
and their members and the endowments of colleges and universities. Yet
they remain largely opaque and these unregulated entities have been
allowed to co-exist alongside the regulated firms as a push was made
for less regulation. That push was advanced by an argument the markets
can regulate themselves, a perspective that has been proven to totally
lack any credibility during this decade of one scandal after another.
Others said that without regulation, these unregulated entities could
innovate and create great wealth. Unfortunately, their innovation has
not always created wealth and in other instances has been quite
destructive.
The subprime crisis, and our economic free fall, is the showcase
for what can happen without adequate regulation and enforcement. Those
who made the loans including mortgage bankers, the credit rating
agencies who put their stamp of approval on the Ninja, no doc and liar
loans, and the investment bankers who packaged them up and sold them to
an unsuspecting public were all unregulated or regulated only in a
token fashion.
Unfortunately, the deregulation of the U.S. capital markets that
many not so long ago called for, has not resulted in increased
competitiveness of the markets. Rather it has left the preeminence and
credibility of our capital markets shattered. Instead of making the
allocation of capital more efficient, it has resulted in a lack of
transparency and mispricing and misallocation of capital. Investors
have watched as over ten trillion in wealth has disappeared. And
instead of fueling a growth in our economy, we have seen it fall into a
decline the likes that haven't been seen since the great depression.
Indeed, some have now called our situation the ``Not So Great
Depression'' and one commentator, Stephen Roach of Morgan Stanley has
warned of a Japanese style economy that continues to this day to
sputter along.
Reforms--The Long Road Back
On a bipartisan basis, we have dug the hole we find ourselves in
over an extended period of time. During much of that time we have
enjoyed economic prosperity that in recent years contributed to the
``suspended disbelief'' that the good times would never end. All too
often people spoke of the ``New Economy'' and those who doubted it or
warned of dangers were treated as outcasts. But as with many a bubble
in the past, this one too has burst.
The capital markets have always been the crown jewel of our
economy--the engine that powered it. And it can once again achieve that
status, firing on all cylinders, but only if care is taken in
structuring reforms that protect the investing public.
Basic Principles
In creating regulator reform, I believe there are some critical
fundamental principles that should be established. They include:
1. Independence
2. Transparency
3. Accountability
4. Enforcement of the law
5. Adequate Resources
Independence
Those responsible for oversight, including regulators and
gatekeepers, must be independent and free of conflicts and bias when
doing their jobs. And it is not just enough that they are independent
on paper, they must be perceived by investors to be free of conflicts
avoiding arrangements that cause investors to question their
independence. They need to be free of political pressures that unduly
influence their ability to carry out their mandates to protect the
American consumer and investor. They must avoid capture by the
regulated. And their ability to get resources should not be contingent
on whether they reach a favorable decision for one special interest
group or political affiliation.
This is especially true of regulators such as the SEC and CFTC.
These agencies must avoid becoming political footballs thrown between
opposing benches. Unfortunately, that has not always been the case as
we saw recently at the SEC or with the CFTC when it asked for
regulation of credit derivatives.
Similarly, the credit rating agencies have suffered from some of
the same lack of independence the auditors did before Enron, WorldCom,
and the enactment of the Sarbanes-Oxley Act of 2002 (SOX). They became
captured by the desire to increase revenues at just about any cost,
while ignoring their gatekeeper role.
Independence also means there is a lack of conflicts that can
impact one's independent thinking. For example, when a bank originates
a subprime loan it may will ask its investment banking arm to
securitize it. But if it is a no doc, liar loan or Ninja loan, will the
investment banker perform sufficient due diligence and ensure full and
fair disclosure is made to the investors clearly delineating in plain
English what they are being sold? I doubt that has really occurred.
Unfortunately, when the Gramm-Leach-Bliley Act was passed, allowing
the creation of giant financial supermarkets, it failed to legislate
and adequately address such conflicts. In fact, it did not address them
at all leaving us with huge conflicts that have now given rise to
investments that are not suitable for the vast majority of investors.
Given this Act gave an implicit blessing to the creation of
institutions that are ``Too Big To Fail'' and knowing that after the
failure of Long Term Capital management the creation of such
institutions brings with it the backing of taxpayers money, this
serious deficiency in the laws governing regulation of conflicts of
interests in these institutions needs to be addressed in a robust
fashion.
Transparency
Transparency is the life blood of the markets. Investors allocate
their capital to those markets where they get higher returns. Investors
need the best possible financial information on which to base their
decisions as to which capital markets they will invest in, and which
companies, in order to generate the maximum possible returns.
Maximizing those returns is critical to investors, and institutions who
manage their investments, as it determines how much they will have for
retirement, or spending.
Investors will allocate their capital to those markets where
returns are maximized. While economic growth in a particular country
has a significant impact on returns for a capital market, the quality
of the information provided to those who allocate capital also
significant impacts it. In general, the better the information, the
better the decisions made, and the more efficiently capital is
allocated and returns maximized.
The U.S. capital markets have maintained their lead in
transparency, albeit our pride in that respect has been tarnished by
off balance sheeting financings, a lack of disclosures regarding the
quality of securities being sold, and credit ratings that were at best
poorly done, if not outright misleading. Nonetheless, even in today's
markets, the U.S. markets have continued to outperform foreign markets.
Accountability
Accountability clearly places the responsibility for decisions made
and actions taken. People act differently when they know they will be
held accountable. When people know there is a state trooper ahead on
the highway, they typically drive accordingly. When they know there is
no trooper, a portion of the population will hit the accelerator and
speed ahead.
There needs to be greater accountability built into the system. The
executives and boards of directors of the financial institutions that
have made the bad loans bringing our economy to its knees, causing
Americans to lose their jobs, students to have to forgo their
education, all at a great cost to the taxpayer should be held
accountable. The American public will demand nothing less.
The banking, insurance, commodities and securities regulators all
need to have greater accountability. We need to know that we have a
real cop on the beat, not just one in uniform standing on a corner.
Likewise, gatekeepers must be held accountable for the product they
provide the capital markets. Their product is critical to ensuring the
credibility of financial information needed for capital allocation.
Enforcement
We are a Nation of laws. The laws governing the capital markets and
banking in this country have been developed to provide protections for
investors and consumers alike. They provide confidence that the money
they have worked hard for, when invested, is safe from abusive,
misleading and fraudulent practices. Without such laws, people would be
much more reluctant to provide capital to banks and public companies
that can be put to work creating new plants and products and jobs.
But laws aren't worth the paper they are written on if they are not
properly enforced. An unleveled playing field in the markets brought on
by a lack of enforcement of laws providing consumer and investor
protections can have the devastating effect we are now seeing. For
example, the Financial Accounting Standards Board Chairman has written
members of this committee citing how some institutions were not
properly following the standards hiding transactions off balance sheet.
Yet to date, enforcement agencies have not brought any cases in that
regard.
And laws are not just enforced by the law enforcement agencies, but
also through private rights of actions of investors and consumers. This
is critically important as law enforcement agencies have lacked the
adequate resources to get the job done alone.
Unfortunately, in recent years we have seen an erosion of investor
and consumer rights to enforce the laws. Court cases setting up huge
hurdles to these attempts to enforce the laws have made it much more
costly taking significant time and resources to get justice. For
example, one such court decision has now made it in essence legal for
someone to knowingly aid another party in the commission of a fraud on
investors, yet be protected by the courts from legal liability. It is
akin to saying that if one drives a getaway car for a bank robber, they
can go to jail. But if one wears a white collar and provides assistance
to such a fraud in the securities market, they get a pass. Something is
just simply wrong when that is allowed to occur in our Nation. Congress
needs to remedy this promptly with legislation Senator Shelby
introduced 7 years ago in 2002.
Likewise we have seen passage of laws such as the Commodities
Modernization Act of 2000 which also put handcuffs on our enforcement
and regulatory agencies. This Act passed in the waning moments of that
Congress at the requests of special interests. Supported by government
officials, the Act specifically prevented the SEC and CFTC from
regulating the derivatives market now totaling hundreds of trillions of
dollars. These handcuffs need to be promptly removed. The securities
and commodities laws need to be clarified to give the CFTC the
authority to regulate commodities and any derivative thereof such as
carbon trading, and the SEC the authority to regulate securities and
any derivative thereof such as credit derivatives.
Adequate Resources
No one can do their job if they are not provided the proper tools,
sufficient staffing and other resources necessary for the job. This
includes being provided the necessary authority through legislation to
do the job. It means Congress has to provide a budget to these agencies
to hire sufficient number of staff. But it is not just the numbers that
count, the agencies must also be given enough money to hire staff with
sufficient experience. For example, while I was at the SEC, the budget
you provided to the agency did not give the Office of Compliance
Inspections and Examination a sufficient number of staff. And it
certainly did not provide the office with enough money to hire senior
experienced examiners who had the type of depth and breadth of
expertise in the industry that was necessary to do the job right. Whose
fault is it then when that agency fails to detects frauds through their
examinations? I would say a good part of the blame lies at the feet of
Congress.
I would urge you to take a look at how these agencies that are so
critical to the proper functioning of our markets are funded. In the
case of the SEC, it collects sufficient fees to pay for an adequate
budget. Yet each year it must go hat in hand to ask for a portion of
those fees in an amount that has not met its needs. Instead, the SEC
should be removed from the annual budget process and established as an
independently funded agency; free to keep the fees it collects to fund
its budgets.
Necessary Reforms
Once again, before legislating reforms, I would urge this committee
to undertake ``Pecora'' hearings to ensure it gets the job done right.
Some of the reforms that I believe are necessary, and which could be
examined in such hearings include the following;
Regulatory Structure: Arbitrage among banking regulators should be
eliminated, and accountability for examination and regulation of banks
centralized in one agency. To accomplish that, Congress should once
again consider the legislation offered in 1994 by the former Chairman
of this Committee, Donald Reigle. That legislation would combine the
examination function into one new agency, while having the FDIC remain
in its role as an insurer and the Federal Reserve as the central
banker. Careful consideration needs to be given to the conflicts that
arise when the central banker both sets monetary policy, such as when
it created low interest rates earlier this decade, and then regulates
the very banks such as Citigroup and Country Wide that exploit that
policy, and at the same time fails to put in place safeguards as the
Fed had been asked to do by Congress in 1994. And the mission of the
new agency, as well as the missions of the FDIC and Fed with respect to
consumer and investor protection needs to be made much more explicit.
All too often these regulators have been captured by industry, much to
the detriment of consumers and investors and in the name of safety and
soundness. Yet we have learned that what is good for consumers and
investors alike, is also good for safety and soundness, but not
necessarily the reverse.
I believe the roles of the CFTC and SEC should be clarified. I do
not support the merger of the two agencies as I don't believe the
synergies some believe exist will be achieved. I also believe
commodities and securities are fundamentally two different markets,
with significantly differing risks, and the regulator needs
significantly differing skill sets to regulate them. Accordingly, as I
have previously mentioned, I would clarify the roles of these two
agencies by giving all commodities and derivatives thereof to the CFTC
to regulate, and all securities and derivatives thereof to the SEC.
Some have argued for the creation of new agencies. To date; I have
yet to see the need for that. For example, some have argued that a
separate investor and consumer protection agency should be created.
However, when it comes to the securities markets, I believe the SEC
should continue in that role, and given the resources to do so.
Over the years, the SEC has shown it can be a strong investor
protection agency. It has only been in recent years, when quite frankly
people who did not believe in regulation were appointed to the
Commission, that it fell down on the job. By appointing investor minded
individuals to the Commission, who have a demonstrated track record of
serving and protecting the public, this problem can be fixed. Likewise
however, if a separate agency is created, but the wrong people put in
place to run it, we will see a repeat performance of what has occurred
at the SEC.
Gaps in Regulation: There are certain gaps in regulation that are
in need of fixing. Credit derivatives should become subject to
regulation by the SEC as former SEC Chairman Cox urged this committee
to do some time ago. While the establishment of a clearing house is a
positive development, in and of itself it is insufficient.
I understand the securities laws generally exclude over-the-counter
swaps from SEC regulation. This improperly limits the SEC's ability to
provide for appropriate investor protection and market quality. The OTC
derivatives market is enormous, and proper regulation is in the public
interest. The SEC would be in a better position to provide that
regulation if the following changes were made:
Repeal the exclusion of security-based swap agreements from
the definition of ``security'' under the Securities Act of 1933
and Securities Exchange Act of 1934.
Include within the definition of ``security'' financial
products that are economic derivatives for securities. It is
important to consolidate the regulatory authority at the SEC
because of its investor protection and capital markets mandate.
While the SEC has a mandate to protect investors and consumers,
other regulators may lose sight of that mission. Based on my
business and agricultural background, I have found derivatives
in agriculture and other physical commodities have a different
purpose than financial derivatives as they permit risk
management and secure supplies for users and producers of
goods.
Require all transactions in securities to be executed on a
registered securities exchange and cleared through a registered
clearing agency.
There needs to be much greater transparency for this market. The
recent reluctance of the FED to disclose the counter parties receiving
the bailout in connection with AIG is alarming but not surprising. Even
the current Fed Chairman has stated this is an agency that has been all
too opaque in the past.
There needs to be greater disclosure to the public of the trading,
pricing and positions of these arrangements. There also needs to be
disclosure identifying the counterparties when the impact of the
contracts could have a material effect on their operations, performance
or liquidity. Given the deficiencies that have existed in some
contracts, there also needs to be more transparency provided around the
nature, terms, and amounts of such contracts when they are material.
There is also a legitimate question as to whether one party should
be able to bet on whether another party will pay their debt, when the
bettor has no underlying direct interest in the debt. Certainly as we
have seen at AIG and elsewhere, these contracts can have devastating
effect. Quite frankly, they do not serve a useful purpose for investors
as a whole in the capital markets. As such, I would like to see them
prohibited.
There is also a gap in regulation of the municipal securities
market as a result of what is known as the Tower Amendment. Recent SEC
enforcement actions such as with the City of San Diego, the problems in
the auction rate securities, and the lurking problems with pension
obligation bonds, all cry out for greater regulation and transparency
in these markets. These token regulated municipal market now amount to
trillions of dollars and poses very real and significant risks.
Accordingly, as former Chairman Cox recommended, I believe Section
15B(d)--Issuance of Municipal Securities--of the Securities Act of 1934
should be deleted.
The SEC should be given authority to regulate hedge and private
equity funds that directly or indirectly take public capital including
from retail investors. They should be subject to the same type of
regulation as their counter parts in the mutual fund market. This
regulation should give the SEC the (i) authority to require the funds
to register with the SEC, (ii) give the SEC the authority to inspect
these firms, (iii) require greater transparency through public
quarterly filings of their positions and their financial statements and
(iv) give the SEC appropriate enforcement capabilities when their
conduct causes damage to investors or the financial markets and system.
As testimony before this committee in the past has demonstrated,
the SEC has insufficient authority over the credit ratings agencies
despite the roles those firms played in Enron and now the subprime
crisis. This deficiency needs to be remedied by giving the SEC the
authority to inspect credit ratings, just as Congress gave the PCAOB
the ability to inspect independent audits. In addition, the SEC should
be given the authority to fine the agencies or their employees who fail
to adequately protect investors. Greater transparency should be
provided to credit ratings themselves. And disclosure should be
required, similar to that for independent auditors of potential
conflicts of interests.
The SEC, CFTC and Banking Regulators should also be given powers to
regulate new financial products issued by those whom they regulate.
This should be accomplished through disclosure. The agencies should
have to make a determination that adequate disclosures have been made
to consumers and investors regarding the risks, terms conditions of new
products before they can be marketed. If a new product is determined by
an agency to present great risk to the financial system or investors,
the regulating agency should be empowered to prevent it from coming to
market, just as is done with new drugs.
In addition, there needs to be greater regulation of mortgage
brokers. Some States have already made progress in this regards.
However, the Federal banking regulators should be given power to
provide consumers necessary protections, if they find that state
regulators have failed to do so.
Greater Accountability Through Improved Governance and Investor
Rights: Legislation equivalent to an investor's Bill of Rights should
be adopted. Investors own the company and should have some basic
fundamental rights with respect to their ownership and investments. It
is well known that investors in the U.S. lack some of the fundamental
rights they have in foreign countries such as the United Kingdom, the
Netherlands and Australia. Yet while some argue for regulation and
regulators similar to those in foreign countries, these very same
people often oppose importing investor rights from those same countries
into our system of governance.
The excesses of executive compensation have been well documented
and need no further discussion. Some have argued investors have an
ability to directly address this by voting for or against directors on
the compensation committee of corporate boards. But that is a fallacy.
First of all, investors can only vote for, not against a director in
the system we have today. Second, some institutional investors have
direct conflicts when voting as a result of receiving fees for managing
corporate pension funds of the management they are voting on. At times
this seems to unduly and improperly influence their votes.
To remedy these shortcomings, Congress should move to adopt
legislation that would:
Require majority voting for directors and those who can't
get a majority of the votes of investors they are to represent
should be required to step down.
Require public issuers to annually submit their
compensation arrangements to a vote of their investors--
commonly referred to as ``say on pay.''
Give investors who own 3 to 4 percent of the company, the
same equal access to the proxy as management currently has.
While some argue this will give special interests an ability to
railroad corporate elections, that simply has proven not to be
the case. When special interests have tried to mobilize votes
based on their interests and not those of investors, they have
ALWAYS failed miserably.
Investors who own 5 percent or more of the stock of a
company should be permitted, as they are in other countries, to
call for a special meeting of all investors. They should also
be given the right to do so to call for a vote on
reincorporation when management and corporate boards unduly use
state laws detrimental to shareholder interests to entrench
themselves further.
Strengthen the fiduciary requirements of institutional
investors when voting on behalf of those whose money they
manage. This should extend to all such institutional investors
including mutual funds, hedge funds, public and corporate
pension funds as well as the labor pension funds.
Since voting is an integral part of and critically important to
governance, greater oversight should be put in place with respect to
those entities who advise institutions on how they should vote.
Recently a paper from the Milstein Center for Governance and
Performance at Yale has made recommendations in this regard as well. As
a former managing director of one such entity, I would support
legislation that would:
Require these entities to register with the SEC as
investment advisors, subject to inspection by the SEC. While
some have registered, others have chosen not to.
Require these entities to improve their transparency by
disclosing their voting recommendations within a reasonable
time period after the vote.
Require all institutional investors, including public,
corporate, hedge and labor pension funds to disclose their
votes, just as mutual funds are currently required to disclose
their votes.
Require that only the legal owner of a share of stock can
vote it, prohibiting those who borrow stock to unduly influence
an election by voting borrowed stock they don't even own, and
eliminating broker votes.
It should also be made explicit that the SEC has authority to set
governance standards for the mutual funds. For example, the SEC should
have the authority, and act on that authority, to require a majority of
independent directors for mutual funds, as well as an independent
chair.
Investor's rights of private actions have also been seriously
eroded in the past decade. Certainly we should not return to the abuses
of the court system that existed before the Private Securities Law
Reform Act (PSLRA) was passed. But at the same time, investors should
not have to suffer the type of conduct that contributed to Enron and
other scandals. And the SEC does not, and will not have the resources
to enforce the securities laws in all instances.
The SEC should continue to be supportive of investors' private
right of action. The SEC should also continue to support court rulings
that permit private investors to bring suits in the event of aiding and
abetting and scheme liability. In 2004, the SEC filed an amicus brief
in Simpson v. Homestore.com, Inc., upholding liability against an
individual regardless of whether or not the person made false or
misleading statements. In 2007, a request from SEC Commissioners to the
Solicitor General to submit a brief in favor of upholding scheme
liability in the case of Stoneridge v. Scientific-Atlanta was denied by
the White House, despite the urging of Senate Banking Committee
Chairman Christopher Dodd (D-CT) and House Financial Services Committee
Chairman Barney Frank (D-MA). The SEC needs to reclaim the SEC's role
of providing strong support for the right of investors to seek a
private remedy.
Investors in securities fraud cases have always had the burden of
proving that defendants' fraud caused the investors' losses. Congress
continued this policy in PSLRA. However, recent lower-court
interpretations of a 2005 Supreme Court case have improperly
transformed loss causation into an almost impossible barrier for
investors in serious cases of fraud. Congress, with the support of the
SEC, should act to fix the law in this area.
Taking advantage of the loophole in the law the courts have now
created, public companies have begun gaming the system. Specifically,
corporations may now simultaneously disclose other information--
positive and negative--in order to make their adverse disclosures
``noisy,'' so that attorneys representing shareholders will find it
more difficult, if not impossible, to satisfy loss causation
requirements. Other corporations may leak information related to the
fraud, so that the share price declines at an early date, before they
formally reveal the adverse news.
In sum, narrow lower-court standards of loss causation are allowing
dishonest conduct to avoid liability for fraudulent statements by
disclosing that the corporation's financial results have deteriorated
without specifically disclosing the truth about their prior
misrepresentations that caused the disappointing results. Insisting on
a ``fact-for-fact'' ``corrective disclosure'' allows fraudsters to
escape liability simply by not confessing.
Transparency: The lack of credible financial information has done
great damage to the capital markets. This has ranged from a lack of
information on off balance sheet transactions as was the case with
Enron, to a lack of information on the quality of assets on the balance
sheets of financial institutions, to a lack of information on risk
management at public entities, to a lack of transparency at regulators.
The lack of transparency begins with accounting standards that yet
again have failed to provide the markets and investors with timely,
comparable and relevant information. The off balance sheet transactions
that expose great risk to the markets, have once again been permitted
to be hid from view by the accounting standard setters. What is more
disturbing about this is that the standard setters were aware of these
risks and failed to act.
To remedy this serious shortcoming, and ensure the standard setters
provide a quality product to investors and the markets, I believe
Section 108 of SOX should be amended. It should require that before the
SEC recognizes an accounting standard setter for the capital markets,
either from the U.S. or internationally, that its board of trustees and
voting board members must have preferably a majority of representatives
from the investor community and certainly no less than 40 percent of
their membership should be investors with adequate skills and a
demonstrated ability to serve the public. In addition, any standard
setter should be required to have an independent funding source before
their standards are used. And finally, each standard setter should be
required to periodically reevaluate the standards they have issued, and
publicly report on the quality of their implementation. For too long
accounting standard setters have disavowed any responsibility for their
standards once they have been issued, a practice that should come to an
immediate halt.
The SEC also needs to closely monitor the current efforts of the
FASB and International Accounting Standards Board (IASB) to ensure
appropriate transactions are brought on balance sheet when a sponsoring
company controls, or effectively controls the economics of the
transaction. I fear based on developments to date, these efforts may
yet once again fail investors.
Transparency of the regulators needs to be enhanced as well so as
to establish greater accountability. For example, the regulators should
be required in their annual reports to Congress to:
Identify key risks that could affect the financial markets
and participants they regulate, and discuss the actions they
are taking to mitigate those risks. For example, the OCC and
SEC have had risk management offices for some time, yet their
reports have failed to adequately alert Congress to the
impending disaster that has now occurred. Unfortunately the SEC
risk management office was reduced to a staff of one.
They should have to provide greater detail as to their
enforcement actions including the aggregate number and nature
of the actions initiated, the number of actions in the pipeline
and average age of those cases, the number and nature of the
cases resolved and how those cases were resolved (e.g.,
litigation, settlement, case dismissed).
Banking and securities regulators should be required to
make public their examination reports. The public should be
able to see in a transparent fashion what the regulator has
found. Regulators who have found problems have all too often
failed to disclose their findings of problems to the
unsuspecting public or Congress. In some instances, the
problems identified have not been promptly addressed by the
regulator and have resulted in the need for taxpayer bailouts
amounting to hundreds of billions of dollars. That simply
should not be allowed to occur. And while some in the industry
and banking regulators have indicated such disclosure could
harm a financial institution, I believe any such harm is
questionable and certainly of much less significance than the
damage now being wrought on our economy and society.
The securities and banking regulators should also be required to
adopt greater disclosures of risks that can impact the liquidity and
capital of financial institutions. The Shipley Working Group encouraged
such disclosures. These disclosures should include greater information
regarding the internal ratings, risks and delinquencies with respect to
loans held by financial institutions. In addition, greater disclosures
should be required regarding how a company identifies and manages risk,
and changing trends in those risks, with an eye to the future.
Improve Independence and Oversight of Self Regulatory
Organizations: FINRA has been a useful participant in the capital
markets. It has provided resources that otherwise would not have been
available to regulate and police the markets. Yet serious questions
have arisen that need to be considered when improving the effectiveness
and efficiency of self regulation.
Currently the Board of FINRA includes representatives from those
who are being regulated. This is an inherent conflict and raises the
question of whose interest the Board of FINRA serves. To address this
concern, consideration should be given to establishing an independent
board, much like what Congress did when it established the PCAOB.
In addition, the arbitration system at FINRA has been shown to
favor the industry, much to the detriment of investors. While
arbitration in some instances can be a benefit, in other situations it
has been shown to be costly, time consuming, and biased towards those
who are constantly involved with it. Accordingly, FINRA's system of
arbitration should be made optional, and investors given the
opportunity to pursue their case in a court of law if they so desire to
do so.
Finally careful consideration should be given to whether or not
FINRA should be given expanded powers over investment advisors as well
as broker dealers. FINRA's drop in fines and penalties in recent years,
and lack of transparency in their annual report to the public, raises
questions about its effectiveness as an enforcement agency and
regulator. And with broker dealers involved in providing investment
advice, it is important that all who do so are governed by the same set
of regulations, ensuring adequate protection for the investing public.
Enforcement: With respect to enforcement of the securities laws,
there are a number of steps Congress should take. After all, if laws
are not adequately enforced, then in effect there is no law.
Enforcement by the SEC would be enhanced if it were granted the
power to bring civil and administrative proceedings for violations of
18 U.S.C. 1001, and seek civil money penalties therein. 18 U.S.C. 1001
is a criminal statute that provides, in pertinent part:
in any matter within the jurisdiction of the executive,
legislative, or judicial branch of the Government of the United
States, knowingly and willfully--(1) falsifies, conceals, or
covers up by any trick, scheme, or device a material fact; (2)
makes any materially false, fictitious, or fraudulent statement
or representation; or (3) makes or uses any false writing or
document knowing the same to contain any materially false,
fictitious, or fraudulent statement or entry;
shall be fined under this title, imprisoned not more than 5
years or, if the offense involves international or domestic
terrorism (as defined in section 2331), imprisoned not more
than 8 years, or both.
The SEC should be authorized to prosecute criminal violations of
the Federal securities laws where the Department of Justice declines to
bring an action. When I was at the Commission, it made a number of
criminal referrals, including such cases as the Sunbeam matter, which
DOJ declined to advance because of resource constraints. Finally the
SEC should be provided an ability to take actions for aiding and
abetting liability under the Securities Act of 1933. The Commission can
bring actions for aiding and abetting violations under the Securities
Exchange Act of 1934.
The SEC has been chronically underfunded. A dedicated, independent
financing arrangement, such as that enjoyed by the Federal Reserve,
would be useful, and is long overdue.
Finally, we have seen serious problems arise for those who have
blown the whistle on corporate fraud. Despite the provisions of SOX
designed to protect such individuals, regulatory interpretations of
that law have rendered it meaningless all too often. Congress should
fix these shortcomings, in part by giving jurisdiction over the law as
it is applicable to the securities markets, to the SEC rather than the
Department of Labor.
Conclusion
Improvements to the securities laws and regulations that will once
again ensure investors can have confidence they are playing on a level
playing field are critical to recovery of our capital markets and
economy. Such legislative changes are necessary if a recovery is to
occur, but it is equally important that when they are made, they are
changes and improvements investors perceive as being credible and
worthwhile.
Thank you and I would be happy to answer any questions.
RESPONSE TO WRITTEN QUESTIONS OF SENATOR DODD
FROM JOHN C. COFFEE, JR.
Q.1. Transparency: Are there additional types of disclosures
that Congress should require securities market participants to
make for the benefit of investors and the markets?
Also, would you recommend more transparency for investors:
1. By publicly held banks and other financial firms of off-
balance sheet liabilities or other data?
2. By credit rating agencies of their ratings methodologies
or other matters?
3. By municipal issuers of their periodic financial
statements or other data?
4. By publicly held banks, securities firms and GSEs of
their risk management policies and practices, with
specificity and timeliness?
A.1. Very simply, my answer is yes. In the case of financial
institutions, recent experience has shown that, despite the
Enron-era reforms, the major banks underwriting asset-backed
securitizations entered into ``liquidity puts'' with preferred
customers under which they agreed to repurchase those offerings
if liquidity was lost in the secondary market--and they did not
disclose these obligations on the face of their balance sheets.
This was the same use of off balance sheet financing as Enron
employed--all over again. Accounting regulators acquiesced to
pressure from banks, and once again the result endangered the
financial well being of the entire economy.
Credit rating agencies should disclose their methodologies
and assumptions (more or less as Senator Reed's bill (S. 1073)
would require).
In general, financial institutions do need to provide
better and more timely disclosure of risk management practices
on a continuing basis. Here, rather than listing specific
disclosures that should be made, I would suggest that Congress
instruct the SEC to study the recent failures and tighten its
disclosure requirements in light of such study.
Q.2. Conflicts of Interest: Concerns about the impact of
conflicts of interest that are not properly managed have been
frequently raised in many contexts--regarding accountants,
compensation consultants, credit rating agencies, and others.
For example, Mr. Turner pointed to the conflict of the board of
FINRA including representatives of firms that it regulates. The
Millstein Center for Corporate Governance and Performance at
the Yale School of Management in New Haven, Connecticut on
March 2 proposed an industry-wide code of professional conduct
for proxy services that includes a ban on a vote advisor
performing consulting work for a company about which it
provides recommendations.
In what ways do you see conflicts of interest affecting the
integrity of the markets or investor protection? Are there
conflicts affecting the securities markets and its participants
that Congress should seek to limit or prohibit?
A.2. In particular, conflicts of interest have affected the
practices of the credit rating agencies, as they both
``consult'' with issuers and rate them, and only a thin (and
possibly permeable) Chinese Wall separates the two functions
and staffs. In addition, there is the problem of forum-
shopping, as the issuer pays an initial fee to several
agencies, but only uses the higher or highest ratings (after
paying a second fee).
Proxy advisors (including Risk Metrics) are similarly
subject to the same conflicts of interest, as they also advise
both their client base of institutional investors and issuers
who specially hire them. At a minimum, such conflicts should be
disclosed to investors along with all fees received by the
proxy advisor from the issuer.
Q.3. Credit Default Swaps: There seems to be a consensus among
the financial industry, government officials, and industry
observers that bringing derivative instruments such as credit
default swaps under increased regulatory oversight would be
beneficial to the nation's economy. Please summarize your
recommendations on the best way to oversee these instruments.
A.3. The best response is to mandate the use of clearinghouses
in the trading of over-the-counter derivatives. Such
clearinghouses would in turn specify margin and mark-to-market
procedures for such instruments, subject to the general
oversight of the Fed or the SEC/CFTC (depending on the
instrument). The industry will respond to this proposal by
saying such a rule should only apply to ``standardized''
derivatives. But there is no clear line between
``standardized'' and ``customized'' derivatives, and good
lawyers can make any derivative customized in about 10 minutes
if it will enable the issuer to escape additional regulatory
costs. Thus, Congress or the SEC must draw a careful line so as
not to permit the clearinghouse requirement to be trivialized.
Q.4. Corporate Governance--Majority Vote for Directors, Proxy
Access, Say on Pay: The Council of Institutional Investors,
which represents public, union and corporate pension funds with
combined assets that exceed $3 trillion, has called for
``meaningful investor oversight of management and boards'' and
in a letter dated December 2, 2008, identified several
corporate governance provisions that ``any financial markets
regulatory reform legislation [should] include.'' Please
explain your views on the following corporate governance
issues:
1. Requiring a majority shareholder vote for directors to be
elected in uncontested elections;
2. Allowing shareowners the right to submit amendment to
proxy statements;
3. Allowing advisory shareowner votes on executive cash
compensation plans.
A.4. I support the SEC's proposals on access to the proxy
statement and would require ``say on pay'' (i.e., an advisory
shareholder vote on compensation) by legislation. I would urge
Congress to expressly authorize the SEC to adopt its proposals
on shareholder access to the proxy litigation, as otherwise
there is certain to be litigation about the SEC's authority.
Nor is the outcome of this litigation free from doubt. With
respect to majority voting, I do not think it is necessary to
overrule state law by mandating majority voting on directors,
as the majority of the Fortune 1000 already follow this
practice.
Q.5. Credit Rating Agencies: Please identify any legislative or
regulatory changes you believe are warranted to improve the
oversight of credit rating agencies.
In addition, I would like to ask your views on two specific
proposals:
1. The Peterson Institute report on ``Reforming Financial
Regulation, Supervision, and Oversight'' recommended
reducing conflicts of interest in the major rating
agencies by not permitting them to perform consulting
activities for the firms they rate.
2. The G30 Report ``Financial Reform; A Framework for
Financial Stability'' recommended that regulators
should permit users of ratings to hold NRSROs
accountable for the quality of their work product.
Similarly, Professor Coffee recommended creating
potential legal liability for recklessness when
``reasonable efforts'' have not been made to verify
``essential facts relied upon by its ratings
methodology.''
A.5. I favor the provisions set forth in the Reed Bill (S.
1073) and in the more recent proposals made by the Bipartisan
Policy Counsel's Credit Rating Agency Task Force. I agree that
the rating agencies face a conflict when they perform
consulting services for companies that they rate. Liability for
``recklessness'' makes sense, but should be accompanied by a
safe harbor that establishes clear standards that will enable
the rating agency to avoid liability (as the Reed Bill does).
Although Congress cannot resolve the First Amendment issues
that the rating agencies raise in their defense, Congress can
make legislative findings of fact (to which most courts do give
deference) that find that credit ratings (particularly those on
structured finance products) do not relate to matters of public
concern and so do not merit constitutional protection beyond
that normally accorded ``commercial speech.'' Finally, I would
urge a statutory ceiling on the liability of a credit rating
agency for any one rating or transaction, which ceiling would
apply in both Federal and State court actions.
Q.6. Hedge Funds: On March 5, 2009, the Managed Funds
Association testified before the House Subcommittee on Capital
Markets and said: ``MFA and its members acknowledge that at a
minimum the hedge fund industry as a whole is of systemic
relevance and, therefore, should be considered within the
systemic risk regulatory framework.'' MFA supported the
creation or designation of a ``single central systemic risk
regulator'' that (1) has ``the authority to request and
receive, on a confidential basis, from those entities that it
determines . . . to be of systemic relevance, any information
that the regulator determines is necessary or advisable to
enable it to adequately assess potential risks to the financial
system,'' (2) has a mandate of protection of the financial
system, but not investor protection or market integrity and (3)
has the authority to ensure that a failing market participant
does not pose a risk to the entire financial system.
Do you agree with MFA's position? Do you feel there should
be regulation of hedge funds along these lines or otherwise?
A.6. I agree with the MFA's position. Systemic risk should be
delegated to a different agency than the agency charged with
consumer protection, as there are potential conflicts between
these two roles. In short, a ``twin peaks'' model should be
followed. Hedge funds are not inherently different than AIG in
that any large financial institution could potentially fail in
a manner that endangered counterparties and could therefore
pose a systemic risk to the financial system.
Q.7. Self-Regulatory Organizations: How do you feel the self-
regulatory securities organizations have performed during the
current financial crisis? Are there changes that should be made
to the self-regulatory organizations to improve their
performance? Do you feel there is still validity in maintaining
the self-regulatory structure or that some powers should be
moved to the SEC or elsewhere?
A.7. Principally, I believe that pre-dispute arbitration
agreements should be limited, as the process is often unfair to
investors. Beyond that, the position of investment advisers,
who have no SRO, is anomalous and should be re-examined. I
express no view on whether they should form their own SRO or be
brought under FINRA.
In overview, the SROs did not cause (but did little to
prevent) the 2008 financial crisis. Conceivably, they could
have discovered Mr. Madoff's fraud (but the SEC bears the
greater responsibility). The SRO structure has some value,
particularly because SROs are self-funding and can tax the
industry. Also, they enforce ``fair and equitable'' rules that
are far broader than the SEC's typically narrower anti-fraud
rules.
Q.8. Structure of the SEC: Please share your views as to
whether you feel that the current responsibilities and
structure of the SEC should be changed.
Please comment on the following specific proposals:
1. Giving some of the SEC's duties to a systemic risk
regulator or to a financial services consumer
protection agency;
2. Combining the SEC into a larger ``prudential'' financial
services regulator;
3. Adding another Federal regulators' or self-regulatory
organizations' powers or duties to the SEC.
A.8. I do not believe that merging the SEC with another
regulator is sensible or politically feasible in the short run.
Nor do I think that the SEC should (at least over the short-
run) assume any of the duties of other regulators. However, the
SEC's ``Consolidated Supervised Entity'' Program, which was
begun in 2004, clearly failed in 2008 and should not in any
form be re-created. Large financial institutions (such as
Goldman, Sachs or Morgan Stanley) are better monitored by the
Federal Reserve as Tier One Bank Holding Companies, and the
capital markets have greater confidence in the Fed's monitoring
ability.
Q.9. SEC Staffing, Funding, and Management: The SEC has a staff
of about 3,500 full-time employees and a budget of $900
million. It has regulatory responsibilities with respect to
approximately: 12,000 public companies whose securities are
registered with it; 11,300 investment advisers; 950 mutual fund
complexes; 5,500 broker-dealers (including 173,000 branch
offices and 665,000 registered representatives); 600 transfer
agents, 11 exchanges; 5 clearing agencies; 10 nationally
recognized statistical rating organizations; SROs such as the
Financial Industry Regulatory Authority, the Municipal
Securities Rulemaking Board and the Public Company Accounting
Oversight Board.
To perform its mission effectively, do you feel that the
SEC is appropriately staffed? funded? managed? How would you
suggest that the Congress could improve the effectiveness of
the SEC?
A.9. The SEC needs more funds and more staff. This is best
accomplished by making the SEC at least partially ``self-
funding,'' specifically by allowing the SEC to keep the fees
and other charges that it levies on issuers, brokers and other
regulated entities. I would not, however, recommend that the
SEC keep civil penalties and fines, as this would raise both
due process and ``appearance of justice'' issues that are best
avoided.
------
RESPONSE TO WRITTEN QUESTIONS OF SENATOR CRAPO
FROM JOHN C. COFFEE, JR.
Q.1. Do you all agree with Federal Reserve Board Chairman
Bernanke's remarks today about the four key elements that
should guide regulatory reform?
First, we must address the problem of financial institutions
that are deemed too big--or perhaps too interconnected--to
fail. Second, we must strengthen what I will call the financial
infrastructure--the systems, rules, and conventions that govern
trading, payment, clearing, and settlement in financial
markets--to ensure that it will perform well under stress.
Third, we should review regulatory policies and accounting
rules to ensure that they do not induce excessive
procyclicality--that is, do not overly magnify the ups and
downs in the financial system and the economy. Finally, we
should consider whether the creation of an authority
specifically charged with monitoring and addressing systemic
risks would help protect the system from financial crises like
the one we are currently experiencing.
Would a merger or rationalization of the roles of the SEC
and CFTC be a valuable reform, and how should that be
accomplished?
How is it that AIG was able to take such large positions
that it became a threat to the entire financial system? Was it
a failure of regulation, a failure of a product, a failure of
risk management, or some combination?
How should we update our rules and guidelines to address
the potential failure of a systematically critical firm?
A.1. Bernanke's Comments: I would strongly agree with Chairman
Bernanke's above quoted remarks, and I believe that his final
question about the desirability of a systemic risk regulator
must be answered in the affirmative (although the identity of
that regulators can be reasonably debated). The term ``too big
to fail'' is a misnomer. In reality, a systemic risk regulator
must have the authority to identify financial institutions that
are ``too interconnected to fail'' and to regulate their
capital structure and leverage so that they do not fail and
thereby set off a chain reaction.
SEC/CFTC Merger: Although a merger of the SEC and the CFTC
would be desirable, it is not an essential reform that must be
accomplished to respond effectively to the current financial
crisis (and it would be a divisive issue that might stall
broader reform legislation). At most, I would suggest that
jurisdiction over financial futures be transferred from the
CFTC to the SEC. An even narrower transfer would be to give the
SEC jurisdiction over single stock futures and narrow-based
stock indexes. Over the counter derivatives might be divided
between the two in terms of whether the derivative related to a
security or a stock index (in which case the SEC would receive
jurisdiction) or to something else (in which case the CFTC
should have jurisdiction).
The AIG Failure: AIG's failure perfectly illustrates the
systemic risk problem (because its failure could have caused a
parade of falling financial dominoes). It also illustrates the
multiple causes of such a failure. AIG Financial Products,
Inc., the key subsidiary, was principally based in London and
was the subsidiary of the parent of the insurance company. As a
non-insurance subsidiary of an insurance holding company, it
was beyond the effective oversight of the New York State
Insurance Commissioner, and there is no Federal insurance
regulator. Although AIG also owned a small thrift, the Office
of Thrift Supervision (OTS) could not really supervise an
unrelated subsidiary operating in London. Thus, this was a case
of a financial institution that fell between the regulatory
cracks.
But it was also a case of a private governance failure
caused by excessive and short-term executive compensation. The
CEO of AIG Financial Products (Mr. Cassano) received well over
a $100 million in compensation during a several year period
between 2002 and 2006. This gave him a strong bias toward
short-term profit maximization and incentivized him to continue
to write credit default swaps for their short term income,
while ignoring the long term risk to AIG of a default (for
which no reserves were established). Thus, there were both
private and public failures underlying the AIG collapse.
Procedures for Failure of a ``Systematically Critical
Firm'': The Lehman bankruptcy will remain in the courts for a
decade or more, with considerable uncertainty overhanging the
various outcomes. In contrast, the FDIC can resolve a bank
failure over a weekend. This suggests the superiority of a
resolution-like procedure following the FDIC model, given the
uncertainty and resulting potential for panic in the case of a
failure of any major financial institution. Both the Bush and
Obama Administrations have endorsed such a FDIC-like model to
reduce the prospect of a financial panic. I note, however, that
one need not bail out all counterparties at the level of 100
percent, as a lesser level of protection would avert any panic,
while also leaving the counterparties with a strong incentive
to monitor the solvency of their counterparty.
------
RESPONSE TO WRITTEN QUESTIONS OF SENATOR VITTER
FROM JOHN C. COFFEE, JR.
Q.1. Are you concerned that too much reliance on investor
protection through private right of action against the credit
ratings agencies will dramatically increase both the number of
law suits the companies will have to deal with as well as their
cost of doing business? Have you thought about alternative ways
to ensure adequate investor protections that will not result in
driving capital from the U.S. in the same way that the fear of
litigation and costs created by Sarbanes-Oxley has resulted in
a decline in new listings in American capital markets?
A.1. I have two independent responses: First, authorizing a
cause of action along the lines that Senator Reed's bill (S.
1073--``The Rating Accountability and Transparency Enhancement
Act of 2009'') does should not increase the number or aggregate
recoveries in securities litigation to any significant degree.
This is because the Reed bill's proposed cause of action
against credit rating agencies contains an important safe
harbor under which a credit rating agency that conducts due
diligence or hires an independent due diligence firm will be
protected against suit. In this light, I believe the real
impact of this provision will be ex ante, rather than ex post,
meaning that it will change the rating agency's behavior so as
to avert litigation, rather than affecting the overall
incidence or outcome of suits against it.
Secondly, I have elsewhere proposed that all securities
litigation against secondary defendants (i.e., persons other
than the issuer or underwriter) be subject to a ceiling on
damages to protect against such litigation causing their
insolvency. So limited, securities litigation against secondary
participants could deter, but not destroy.
------
RESPONSE TO WRITTEN QUESTIONS OF SENATOR DODD
FROM T. TIMOTHY RYAN, JR.
Q.1. Transparency: Are there additional types of disclosures
that Congress should require securities market participants to
make for the benefit of investors and the markets?
Also, would you recommend more transparency for investors:
1. LBy publicly held banks and other financial firms of off-
balance sheet liabilities or other data?
2. LBy credit rating agencies of their ratings methodologies
or other matters?
3. LBy municipal issuers of their periodic financial
statements or other data?
4. LBy publicly held banks, securities firms and GSEs of
their risk management policies and practices, with
specificity and timeliness?
A.1. We do not believe that there are ``additional types of
disclosures'' that should be required regarding participants in
the capital markets. The disclosures currently required in
financial statements have increased substantially over the last
decade and in particular over the last year, and have led some
commentators to observe that re-organizing current mandated
disclosures might result in a more concise but more
intelligible set of disclosures for readers of financial
statements. We think that such an approach would be more
beneficial than simply increasing the amount of data required.
Both FASB and the SEC have recognized the need for
additional disclosure regarding off-balance sheet exposures.
Those requirements will be phasing in over the course of this
year and we believe will meaningfully increase the amount of
information in the marketplace regarding off-balance sheet
assets and liabilities.
Last year, SIFMA formed a global, investor-led task force
to identify and examine key issues related to the credit
ratings paradigm. SIFMA's Credit Rating Agency Task Force
issued its recommendations last July. These recommendations
included the following related to disclosure:
LCRAs should provide enhanced, clear, concise, and
standardized disclosure of CRA rating methodologies;
LCRAs should disclose results of due diligence and
examination of underlying asset data examinations, and
limitations on available data, as well as certain other
information relied upon by the CRAs in the ratings
process;
LCRAs should provide disclosure of CRA surveillance
procedures; this will foster transparency, and allow
market users of ratings to understand their bases and
limitations;
LCRAs should provide access to data regarding CRA
performance; this will allow investors to assess how
CRAs differ both in the performance of their initial
ratings, and in their ongoing surveillance of existing
ratings; and
LCRA fee structures, and identities of top payors,
should be disclosed by CRAs to their regulators.
The report discusses these recommendations in detail. A
copy of the report is attached and can also be found at http://
www.sifma.org/capital_markets/docs/SIFMA-CRARecommendations.pdf
Municipal issuers are already required to provide annual
financial statements and material event notice disclosures to
investors pursuant to SEC Rule 15(c)2-12, and plans are
underway for this information to be more easily accessed under
the Municipal Securities Rulemaking Board's Electronic
Municipal Market Access system (EMMA). As of July 1, 2009, EMMA
will be the new central filing repository for municipal issuer
annual financial statements and material events notices as well
as a free Internet-based transparency vehicle for retail
investors seeking this information.
The current disclosure requirements already encompass a
great deal of information on the risk management policies/
practices of banks and securities firms, particularly in the
MD&A of financial statements. While improvements in the
presentation and intelligibility of such disclosures may occur
as a result of preparer interaction with investors, analysts,
and other users of financial statements, we do not believe that
at this point additional requirements per se are warranted.
Q.2. Conflicts of Interest: Concerns about the impact of
conflicts of interest that are not properly managed have been
frequently raised in many contexts--regarding accountants,
compensation consultants, credit rating agencies, and others.
For example, Mr. Turner pointed to the conflict of the board of
FINRA including representatives of firms that it regulates. The
Millstein Center for Corporate Governance and Performance at
the Yale School of Management in New Haven, Connecticut on
March 2 proposed an industry-wide code of professional conduct
for proxy services that includes a ban on a vote advisor
performing consulting work for a company about which it
provides recommendations.
In what ways do you see conflicts of interest affecting the
integrity of the markets or investor protection? Are there
conflicts affecting the securities markets and its participants
that Congress should seek to limit or prohibit?
A.2. SIFMA agrees that effective management of conflicts of
interest by market intermediaries builds confidence in the
integrity of financial markets and promotes investor
protection. We generally support initiatives to identify and
manage conflicts and we believe the industry has made
significant steps over the past several years to develop best
practices in this area. For example, our members have developed
an ``Investor's Bill of Rights'' which states that each
investor has the right ``to be apprised of significant
conflicts of interest identified in a financial relationship
between an investor and his or her broker-dealer or account
representative.'' This resource is available at http://
www.sifma.org/private_client/pdf/SIFMA_InvestorRights.pdf. In
another example, our members have also developed ``Principles
for Managing the Distributor-Individual Relationship'' for
structured products that includes guidance on managing
potential conflicts. This resource is available at http://
www.sifma.org/private_client/pdf/GlobalRSPDistributor-
PrinciplesFinal.pdf. Similarly, our proposed universal standard
of care for retail investors, would also require financial
professionals to provide full and fair disclosure of all
material facts, including material conflicts of interest.
Q.3. Credit Default Swaps: There seems to be a consensus among
the financial industry, government officials, and industry
observers that bringing derivative instruments such as credit
default swaps under increased regulatory oversight would be
beneficial to the nation's economy. Please summarize your
recommendations on the best way to oversee these instruments.
A.3. Credit default swaps are important financial tools that
allow companies across America access to capital at lower cost
by allowing banks to efficiently hedge exposure to debt of
these companies. We believe there is wide agreement that steps
should be taken to address issues that have arisen in
connection with CDS and other derivatives, but care should be
taken not to impair the usefulness of these products. In
particular, as recognized by the President's Working Group,
market participants should be allowed to enter into customized
bilateral contracts in order to accomplish their risk
management objectives. We believe that Congress should consider
subjecting all systemically significant participants in
derivatives markets, whether they are investors or dealers, to
oversight by a single Federal regulator with broad authority to
identify who is systemically significant, to consult with
industry and develop principles for prudent management of risk,
to promulgate appropriate rules based on those principles, and
to access information necessary to carry out its oversight
responsibilities. Among other things, we believe the principles
adopted by the systemic regulator should encourage submission
of standardized credit default swaps to clearing houses that
are subject to Federal regulatory oversight. This will help
assure adequate collateral posting and decrease aggregate
leverage in the financial system, both of which will reduce
overall levels of risk. Because financial markets and the
activities of major market participants are global, it is
important that the Federal systemic risk regulator consult and
coordinate with regulators in major markets outside of the
United States.
Q.4. Corporate Governance--Majority Vote for Directors, Proxy
Access, Say on Pay: The Council of Institutional Investors,
which represents public, union and corporate pension funds with
combined assets that exceed $3 trillion, has called for
``meaningful investor oversight of management and boards'' and
in a letter dated December 2, 2008, identified several
corporate governance provisions that ``any financial markets
regulatory reform legislation [should] include.'' Please
explain your views on the following corporate governance
issues:
1. LRequiring a majority shareholder vote for directors to be
elected in uncontested elections;
2. LAllowing shareowners the right to submit amendment to
proxy statements;
3. LAllowing advisory shareowner votes on executive cash
compensation plans.
A.4. While requiring a majority shareholder vote for directors
to be elected in uncontested elections may promote selection of
well-qualified directors, SIFMA would note that for some
issuers, this could be a difficult requirement to meet. The SEC
is likely to prohibit brokerdealers from voting the
uninstructed shares of clients in director elections in a
current rulemaking project. SIFMA is concerned about protecting
the privacy of those clients who object to direct contact with
issuers. This will also increase costs for issuers who will
need to spend significant sums to get out the vote. Finally,
small issuers would likely be disadvantaged because their
shares are more often held by retail investors rather than
large, institutional investors.
SIFMA does not have a position on this proposal but notes
that the thresholds for when a shareholder or group of
shareholders would be granted access to an issuer's proxy
should be high enough so that the process isn't abused. The
process for allowing shareholders to submit amendments to proxy
statements would need to be clear and to minimize costs,
uniform. It is very important that shareholders have the right
to vote; the submission of amendments should not inhibit
today's efficient and timely process for the arrival of proxy
statements. Building in time for amendments could further delay
the transmission of proxy material, this and other practical
issues presented by the proposal warrant careful consideration.
As the Committee knows, a number of SIFMA members are
subject to the TARP requirement to conduct an advisory
shareowner vote on compensation. Several firms have either
implemented this requirement or are working toward meeting it.
We caution that the results of this year's advisory votes may
not be emblematic of potential future advisory votes. This year
many of the TARP companies did not pay senior management any
bonuses and it is quite possible that this factored heavily in
the votes in favor. Also, many TARP companies that implemented
the advisory vote were required to do so in a hasty manner and
as a result included the ``boilerplate'' proposal language
without thinking through what language made sense for their
particular company. SIFMA has yet to take a position on whether
annual advisory shareowner votes should be mandated for all
public issuers as suggested by the Council of Institutional
Investors or whether other mechanisms, such as issuer-specific
surveys, would be more helpful to enhance communications
between Boards and shareowners.
Q.5. Credit Rating Agencies: Please identify any legislative or
regulatory changes you believe are warranted to improve the
oversight of credit rating agencies.
In addition, I would like to ask your views on two specific
proposals:
1. LThe Peterson Institute report on ``Reforming Financial
Regulation, Supervision, and Oversight'' recommended
reducing conflicts of interest in the major rating
agencies by not permitting them to perform consulting
activities for the firms they rate.
2. LThe G30 Report ``Financial Reform; A Framework for
Financial Stability'' recommended that regulators
should permit users of ratings to hold NRSROs
accountable for the quality of their work product.
Similarly, Professor Coffee recommended creating
potential legal liability for recklessness when
``reasonable efforts'' have not been made to verify
``essential facts relied upon by its ratings
methodology.''
A.5. Last year, SIFMA formed a global, investor-led task force
to identify and examine key issues related to the credit
ratings paradigm. SIFMA's Credit Rating Agency Task Force
issued its recommendations last July. A copy of the report is
attached and can also be found at http://www.sifma.org/
capital_markets/docs/SIFMA-CRA-Recommendations.pdf.
SIFMA's CRA Task Force found that there is a perception by
some that the degree and nature of interaction between CRAs and
issuers during the ratings process may result in conflicts of
interest. This perception undermines investor confidence in the
accuracy and reliability of ratings. These perceived conflicts
can arise both from the interaction between CRAs and issuers in
the course of a CRA assigning a rating to a particular
security, and from the CRAs' provision of consulting or
advisory services.
The Task Force noted that each of five major CRAs (A.M.
Best, DBRS, Fitch, Moody's, and Standard & Poor's) committed in
their Joint Response to the IOSCO Consultation Report on the
Role of Credit Rating Agencies in Structured Finance Markets to
``plainly indicate'' that it does ``not and will not provide
consulting or advisory services to the issuers the [CRA]
rates.''
In order to provide clarity to market participants, the
Task Force recommended that ``core'' rating services be clearly
defined by the CRAs and distinguished from such ``consulting or
advisory'' services. The Task Force further recommended that
CRAs clarify that ``consulting or advisory'' services exclude
other ``ancillary'' services provided to issuers and
intermediaries in the ordinary course of business.
The Task Force viewed the CRAs' permissible ``core''
services as including:
1. Lthe assignment and monitoring of public, private, and
private placement ratings;
2. Lissuance of credit estimates and hypothetical ratings,
including requested Rating Evaluation Service and
Rating Advisory Service (RES/RAS) services regarding
issuer-proposed structures of hypothetical securities,
indicative, or preliminary ratings, and impact
assessments;
3. Lhybrid securities assessment services;
4. Linternal assessments;
5. Lratings coverage of project and infrastructure finance
transactions and hybrid securities;
6. Ldissemination of press releases and rating reports (that
include the rating opinion);
7. Lresearch reports and other publications, including
methodologies, models, newsletters, commentaries, and
industry studies;
8. Lregular oral and written dialogue with issuers,
intermediaries, investors, sponsors,regulators,
legislators, trade organizations, and the media; and
9. Lconducting and participating in conferences, speaking
engagements, and educational seminars.
In particular, the Task Force believes that these ``core''
services include the iterative process that occurs between an
issuer, arranger, underwriter, and CRA during the rating of
structured finance, project and infrastructure finance, and
hybrid securities.
The Task Force believes there is a misperception by some
that this type of ``core'' interaction is essentially a
consultation service by CRAs that gives rise to an insuperable
conflict of interest, and which undermines the integrity and
reliability of the resulting rating. As described above,
however, the process of rating structured finance, project and
infrastructure finance, and hybrid securities necessarily
involves an iterative give-and-take between the issuer,
arranger, underwriter, and CRA as part of the ``core'' services
performed by the CRA.
In light of this, the Task Force did not recommend placing
limitations on this iterative process. Rather, the Task Force
recommended that CRAs maintain an adequate governance structure
that includes policies, procedures, mechanisms, and firewalls
designed to minimize the likelihood that conflicts of interest
will arise, and to manage the conflicts of interest that do
arise.
Similarly, ``ancillary'' services, in the view of the Task
Force, are permissible rating-related services that are
generally segregated by the CRA into separate business groups.
The Task Force views examples of ``ancillary'' services as
including, among others, market implied ratings (MIRS), KMV
credit risk management, data services, credit risk solutions,
and indices.
SIFMA has not taken a position on legal liability for
NRSROs.
Q.6. Hedge Funds: On March 5, 2009, the Managed Funds
Association testified before the House Subcommittee on Capital
Markets and said: ``MFA and its members acknowledge that at a
minimum the hedge fund industry as a whole is of systemic
relevance and, therefore, should be considered within the
systemic risk regulatory framework.'' MFA supported the
creation or designation of a ``single central systemic risk
regulator'' that (1) has ``the authority to request and
receive, on a confidential basis, from those entities that it
determines . . . to be of systemic relevance, any information
that the regulator determines is necessary or advisable to
enable it to adequately assess potential risks to the financial
system,'' (2) has a mandate of protection of the financial
system, but not investor protection or market integrity and (3)
has the authority to ensure that a failing market participant
does not pose a risk to the entire financial system.
Do you agree with MFA's position? Do you feel there should
be regulation of hedge funds along these lines or otherwise?
A.6. We support giving the financial markets stability
regulator the authority to gather information from all U.S.
financial institutions and markets in order to identify
systemic risk and maintain financial stability. We believe this
authority should apply to all financial institutions,
regardless of charter, and regardless of whether they are
currently functionally regulated or not, including hedge funds
or private equity funds. One of the lessons learned from recent
experience is that sectors of the market can be systemically
important, even though no single institution in that sector is
a significant player. The financial markets stability regulator
will need information necessary to form and maintain a picture
of the overall systemic risks in the U.S. financial system. The
financial markets stability regulator should also have the
authority, in consultation with any relevant Federal regulator,
to make uniform rules to the extent necessary to reduce
systemic risk and promote financial stability.
As noted above, we have proposed that the financial markets
stability regulator should probably have a more direct role in
supervising systemically important financial institutions or
groups. Such systemically important financial institutions or
groups could include currently unregulated institutions, such
as hedge funds or private equity funds, although we do not
believe the financial markets stability regulator should become
the functional regulator for such unregulated institutions. We
agree with others that hedge funds should be regulated by a
merged SEC and CFTC.
Because we believe the financial markets stability
regulator should have the authority to address a financial
crisis, we believe such a regulator should have certain
resolution powers, including the authority to appoint itself or
another Federal regulatory agency as the conservator or
receiver of any systemically important financial institution or
group.
Q.7. Self-Regulatory Organizations: How do you feel the self-
regulatory securities organizations have performed during the
current financial crisis? Are there changes that should be made
to the self-regulatory organizations to improve their
performance? Do you feel there is still validity in maintaining
the self-regulatory structure or that some powers should be
moved to the SEC or elsewhere?
A.7. The SRO structure remains a viable regulatory framework.
Supplemented by government oversight, this tiered regulatory
system can provide a greater level of investor protection than
the government alone might be able to achieve. Self-policing by
professionals who have the requisite working knowledge and
expertise about both marketplace intricacies and the technical
aspects of regulation creates a self-regulatory system with
valuable checks and balances. SRO performance may be improved
by eliminating duplicative regulation, filling regulatory gaps,
and harmonizing standards that are appropriately applicable to
all investment services providers. In harmonizing standards,
however, we note that just as one size does not fit all broker-
dealers, it also does not fit all market users. There is a
world of difference between an individual investor seeking to
invest his/her retirement savings and a multi-billion dollar
hedge fund implementing a sophisticated trading strategy.
Indeed, there is a similar difference between a high net worth
individual managing substantial assets and retail market
participants seeking to save for college. While all
participants must be protected from fraud, we need a flexible
regulatory structure that can differentiate between the various
types of market participants when it comes to mandatory
prophylactic rules and requirements.
Q.8. Structure of the SEC: Please share your views as to
whether you feel that the current responsibilities and
structure of the SEC should be changed.
Please comment on the following specific proposals:
1. LGiving some of the SEC's duties to a systemic risk
regulator or to a financial services consumer
protection agency;
2. LCombining the SEC into a larger ``prudential'' financial
services regulator;
3. LAdding another Federal regulators' or self-regulatory
organizations' powers or duties to the SEC.
A.8. We have testified that we are in support of a merger of
the SEC and the CFTC. The longstanding focus of the SEC has
been investor protection, and we believe that this should
continue to be so with any regulatory reform. We do not see a
systemic risk regulator taking over any of the SEC's duties;
rather, such a regulator would work with existing Federal
functional regulators such as the SEC. For example, we agree
with others that hedge funds should be regulated by a merged
SEC and CFTC.
The SEC's investor protection mandate could be expanded to
other areas or products, and so we do not see the need for a
separate financial services consumer protection agency for that
purpose. On the other end of the spectrum, we do not see
support for moving to an FSA-type model of a single prudential
financial services regulator. Certain inefficiencies that
result from the regulation of activities by the states and the
SEC could be eliminated by vesting the regulatory authority for
those activities in the SEC.
Q.9. SEC Staffing, Funding, and Management: The SEC has a staff
of about 3,500 full-time employees and a budget of $900
million. It has regulatory responsibilities with respect to
approximately: 12,000 public companies whose securities are
registered with it; 11,300 investment advisers; 950 mutual fund
complexes; 5,500 broker-dealers (including 173,000 branch
offices and 665,000 registered representatives); 600 transfer
agents, 11 exchanges; 5 clearing agencies; 10 nationally
recognized statistical rating organizations; SROs such as the
Financial Industry Regulatory Authority, the Municipal
Securities Rulemaking Board and the Public Company Accounting
Oversight Board.
To perform its mission effectively, do you feel that the
SEC is appropriately staffed? funded? managed? How would you
suggest that the Congress could improve the effectiveness of
the SEC?
A.9. If the SEC is to take on a greater responsibility with
respect to the regulation of hedge funds and other private
equity vehicles, we believe that current staffing levels are
inadequate. The SEC is currently unable to examine investment
advisers in a timely manner, and with enhanced responsibilities
their resources will be even more stretched. As a result, we
believe additional funding of the SEC will be necessary. We
would also support an internal reorganization of the SEC such
that the examination functions, such as broker-dealer and
investment adviser examinations, are combined with, and
reporting into, the policy making units, so that the SEC speaks
and acts consistently on policy issues.
RESPONSE TO WRITTEN QUESTIONS OF SENATOR CRAPO
FROM T. TIMOTHY RYAN, JR.
Q.1. Do you all agree with Federal Reserve Board Chairman
Bernanke's remarks today about the four key elements that
should guide regulatory reform?
First, we must address the problem of financial institutions
that are deemed too big--or perhaps too interconnected--to
fail. Second, we must strengthen what I will call the financial
infrastructure--the systems, rules, and conventions that govern
trading, payment, clearing, and settlement in financial
markets--to ensure that it will perform well under stress.
Third, we should review regulatory policies and accounting
rules to ensure that they do not induce excessive
procyclicality--that is, do not overly magnify the ups and
downs in the financial system and the economy. Finally, we
should consider whether the creation of an authority
specifically charged with monitoring and addressing systemic
risks would help protect the system from financial crises like
the one we are currently experiencing.
A.1. We agree with Chairman Bernanke's remarks and support the
proposal to establish a financial markets stability regulator.
At present, no single regulator (or collection of coordinated
regulators) has the authority or the resources to collect
information system-wide or to use that information to take
corrective action across all financial institutions and markets
regardless of charter. The financial markets stability
regulator will help fill these gaps.
We have proposed that the financial markets stability
regulator should have authority over all financial institutions
and markets, regardless of charter, functional regulator or
unregulated status, including the authority to gather
information from all financial institutions and markets, and to
make uniform regulations related to systemic risk. This could
include review of regulatory policies and rules to ensure that
they do not induce excessive procyclicality.
We have proposed that the financial markets stability
regulator should probably have a more direct role in
supervising systemically important financial institutions or
groups. This would address the risks associated with financial
institutions that may be deemed ``too big to fail.'' Such
systemically important financial institutions or groups could
also include primary dealers, securities clearing agencies,
derivatives clearing organizations and payment system
operators, which would help strengthen the financial
infrastructure, another key element of Chairman Bernanke's
proposal for regulatory reform.
Q.2. Would a merger or rationalization of the roles of the SEC
and CFTC be a valuable reform, and how should that be
accomplished?
A.2. We have testified that we are in support of a merger of
the SEC and the CFTC. The U.S. is the only jurisdiction that
splits the oversight of securities and futures activities
between two separate regulatory bodies. When the CFTC was
formed, financial futures represented a very small percentage
of futures activity. Now, an overwhelming majority of futures
that trade today are financial futures. These products are
nearly identical to SEC regulated securities options from an
economic standpoint, yet they are regulated by the CFTC under a
very different regulatory regime. This disparate regulatory
treatment detracts from the goal of investor protection. An
entity that combines the functions of both agencies could be
better positioned to apply consistent rules to securities and
futures. We would support legislation to accomplish such a
merger.
Q.3. How is it that AIG was able to take such large positions
that it became a threat to the entire financial system? Was it
a failure of regulation, a failure of a product, a failure of
risk management, or some combination?
A.3. We believe the problems at AIG resulted from a combination
of several factors. Its affiliate, AIG Financial Products, sold
large amounts of credit protection in the form of credit
default swaps on collateralized debt obligations with exposure
to subprime mortgages, without hedging the risk it was taking
on. At the same time, AIG's top credit rating gave many of its
counterparties a false sense of security. Accordingly, many of
the CDS agreements it negotiated provided that AIG would not be
required to post collateral so long as it maintained a
specified credit rating. AIG apparently believed its credit
rating would never be downgraded, which enabled it to ignore
the risk it would ever have to post collateral. Moreover, AIG
appears to have under-estimated the default risk of the CDOs on
which it sold credit protection, thus underestimating the size
of its obligation to post large amounts of collateral in the
event of its credit rating downgrade. While others might have
made similar errors, it seems AIG in particular did not
adequately account for the correlation of default risk among
the different geographic areas where the mortgage assets
underlying the CDOs originated. The market value of those CDOs
fell by much more than AIG anticipated, leading to much greater
collateral demands than it could possibly meet. It also appears
that AIG Financial Products was not subject to adequate,
effective regulatory oversight. All these factors are specific
to AIG; its problems did not result from an inherent defect in
CDS as a product.
Q.4. How should we update our rules and guidelines to address
the potential failure of a systematically critical firm?
A.4. One of the most important gaps exposed during the current
financial crisis was the lack of Federal resolution powers for
systemically important financial groups. We believe that the
proposed financial stability regulator should have the
authority to appoint itself or another Federal regulatory
agency as the conservator or receiver of any systemically
important financial institution and all of its affiliates. Such
conservator or receiver should have resolution powers similar
to those contained in Sections 11 and 13 of the Federal Deposit
Insurance Act. But because the avoidance powers, priorities and
distribution schemes of the FDIA are very different from those
in the Bankruptcy Code or other specialized insolvency laws
that would otherwise apply to various companies in a
systemically important financial group, the proposed resolution
authority needs to be harmonized with the Bankruptcy Code and
such other laws to avoid disrupting the reasonable expectations
of creditors, counterparties and other stakeholders. Otherwise,
the new resolution authority itself could create legal
uncertainty and systemic risk.
The Treasury's proposed resolution authority for
systemically significant financial companies is a good first
start, but its scope needs to be expanded to apply to all of
the companies that comprise a systemically important financial
group while the gap between its substantive provisions and
those in the Bankruptcy Code and other specialized insolvency
codes that would otherwise apply needs to be reduced in order
to protect the reasonable expectations of creditors,
counterparties and other stakeholders.
------
RESPONSE TO WRITTEN QUESTIONS OF SENATOR DODD
FROM PAUL SCHOTT STEVENS
Q.1. Transparency: Are there additional types of disclosures
that Congress should require securities market participants to
make for the benefit of investors and the markets? Also, would
you recommend more transparency for investors:
LBy publicly held banks and other financial firms of
off-balance sheet liabilities or other data?
LBy credit rating agencies of their ratings
methodologies or other matters?
LBy municipal issuers of their periodic financial
statements or other data?
LBy publicly held banks, securities firms and GSEs
of their risk management policies and practices, with
specificity and timeliness?
A.1. Investment companies provide extensive disclosures and are
highly transparent, especially as compared to many other
investment products. As investors, investment companies
generally favor efforts to increase transparency in the
securities markets, unless countervailing policy objectives
dictate otherwise or the information would not be meaningful to
investors.
Two specific areas in which ICI believes additional
disclosure should be required are credit rating agencies and
municipal securities. We strongly supported the Securities and
Exchange Commission's recent credit rating agency proposals--
which would impose additional disclosure, reporting, and
recordkeeping requirements on rating agencies for rating
structured finance products--as an important first step to
restoring investor confidence in the integrity of credit
ratings and, ultimately, the market as a whole. We believe,
however, that more must be done to increase disclosure and
transparency not only in the area of structured finance
products but also with respect to other debt securities,
particularly municipal securities. We have urged the SEC to
expand many of its proposed requirements for credit rating
agencies to include these additional categories of securities,
and to support legislation that would extend increased
disclosure requirements to the issuers of these instruments. We
also have recommended a number of additional disclosures to be
made by rating agencies and issuers that should enhance
disclosure for investors in a meaningful way. \1\ We believe
the SEC currently has authority to implement many of our
recommendations. Others (such as the repeal of the Tower
Amendment and certain changes to improve municipal securities
disclosure, discussed below) would require Congressional
action.
---------------------------------------------------------------------------
\1\ See Letter from Karrie McMillan, General Counsel, Investment
Company Institute, to Florence Harmon, Acting Secretary, U.S.
Securities and Exchange Commission, dated July 25, 2008; Letter from
Karrie McMillan, General Counsel, Investment Company Institute, to
Elizabeth M. Murphy, Secretary, U.S. Securities and Exchange
Commission, dated March 26, 2009.
---------------------------------------------------------------------------
Rating Agency Disclosure: ICI recommends the following
additional disclosures, which go beyond the SEC's recent
proposals, to improve the transparency of ratings and the
rating process:
LRating agencies should be required to provide
public disclosure of any material deviations between
the credit rating implied by a rating model and the
final credit rating issued.
LRating agencies should make more timely disclosure
of their rating actions.
LRating agencies should disclose additional
information regarding staffing issues, including
personnel turnover and resource levels.
LRating agencies should disclose certain information
about the ongoing review of their ratings.
LRating agencies should disclose additional
information regarding rating stability, including when
and how downgrades are conducted and the severity of
potential downgrades.
LRating agencies should disclose additional
information regarding conflicts of interest.
LRating agencies should be required to use
standardized performance measurement statistics to
facilitate comparability of these statistics.
LRating agencies should be required to conduct due
diligence on the information they review to issue
ratings and to provide related disclosure.
We also recommend that the SEC apply these suggested
disclosure requirements in a consistent manner to all types of
rating agencies. In addition, to realize the full potential of
a meaningful and effective disclosure regime, we recommend that
the SEC require the standardized presentation of disclosure
information in a presale report issued by the rating agencies.
Municipal Securities Issuer Disclosure: ICI strongly urges
Congress and the SEC to improve the content and timing of
required disclosures regarding municipal securities. The Tower
Amendment, adopted by Congress in 1975, prohibits the SEC and
the Municipal Securities Rulemaking Board from directly or
indirectly requiring issuers of municipal securities to file
documents with them before the securities are sold. As we have
stated numerous times, because of these restrictions, the
disclosure regime for municipal securities is woefully
inadequate, and the regulatory framework is insufficient for
investors in today's complex marketplace. \2\
---------------------------------------------------------------------------
\2\ See, e.g., Letter from Karrie McMillan, General Counsel,
Investment Company Institute, to Florence Harmon, Acting Secretary,
U.S. Securities and Exchange Commission, dated September 22, 2008.
---------------------------------------------------------------------------
Legislative action regarding the Tower Amendment will be
necessary to fully develop an adequate disclosure regime for
municipal securities, including imposing certain disclosure
requirements directly on municipal issuers. We would strongly
support such action. We also recommend that Congress clarify
the legal responsibilities of officials of municipal issuers
for the disclosure documents that they authorize. In
particular, Congress should spell out the responsibilities of
underwriters with respect to municipal securities offering
statements and the responsibilities of bond counsel and other
participants in municipal offerings.
In the meantime, important steps can be taken to improve
municipal securities disclosure without legislative action. In
particular, ICI recommends that the SEC expand the list of
information that is required to be disclosed under current SEC
rules. \3\ For example, the rule provision concerning notice of
material events should be modified to more fully reflect the
types of events that are material to today's investors. These
events should include, among others, material litigation or
regulatory action, pending or threatened, or failure to meet
any financial covenants contained in the bond documents
(especially the failure to make any monthly/quarterly payments
due under the bond documents).
---------------------------------------------------------------------------
\3\ ICI is not advocating a wholesale replication of the corporate
disclosure framework for municipal securities. Instead, we are
recommending a regulatory regime designed expressly for the needs of
the municipal securities market.
---------------------------------------------------------------------------
We also recommend changes to ensure that issuer financial
information is provided to the public on a timely basis. \4\
Specifically, the SEC should establish meaningful timeframes
for the delivery of information required pursuant to the
undertakings in an issuer's continuing disclosure agreement.
For example, issuers should be required to file financial
reports within 180 days of the end of the fiscal year, instead
of the more common practice of 270 days after fiscal year end.
Also, if audited financial statements are not available within
the recommended timeframe, issuers should be required to issue
unaudited financials in the interim, as appropriate, in
accordance with guidelines established by the National
Federation of Municipal Analysts. Timely reporting would
enhance the usefulness of the information reported, including
by alerting investors to those issuers that may be experiencing
problems that could affect the credit quality or other
characteristics of their securities.
---------------------------------------------------------------------------
\4\ Rule 15c2-12 under the Securities Exchange Act of 1934
currently requires information about municipal securities issuers to be
provided only annually. In contrast, corporate issuers are subject to
quarterly reporting requirements. Moreover, the rule does not provide
any outside deadline for the disclosure of financial information, thus
leaving the timing completely to the discretion of the issuer. As a
result, investors often receive financial information anywhere from
three months to twelve months, or even longer, following the end of a
fiscal year.
---------------------------------------------------------------------------
Other Matters: In response to the question posed, ICI has
no specific recommendations to offer regarding disclosure by
publicly held banks or other financial institutions of off-
balance sheet liabilities or other data. As a general matter,
however, we would support additional transparency of off-
balance sheet entities and activities. Such transparency should
provide investment companies and other investors with important
information about potential risk exposures faced by the
companies in which they invest and should help avoid the market
inefficiencies and other adverse consequences that the current
lack of transparency has engendered. \5\ We understand that the
Financial Accounting Standards Board is working on revisions to
two of its standards (FAS 140 and FIN 46) that are intended to
address deficiencies in the accounting and disclosure of risks
associated with off-balance sheet entities (e.g., structured
investment vehicles) that were revealed during the current
financial crisis. We look forward to the implementation of
these improvements.
---------------------------------------------------------------------------
\5\ See Written Testimony of Elizabeth F. Mooney, CFA, CPA,
Accounting Analyst, Capital Group Companies, before the U.S. Senate
Committee on Banking, Housing, and Urban Affairs, Subcommittee on
Securities, Insurance, and Investment, Hearing on ``Transparency in
Accounting: Proposed Changes in Accounting for Off-Balance Sheet
Entities'' (September 18, 2008).
---------------------------------------------------------------------------
ICI likewise has no formal position on whether publicly
held banks, securities firms, and GSEs should be required to
disclose their risk management policies and practices. We would
caution, however, that ``risk management'' means different
things to different people and can also have varying
connotations depending on the context. General disclosure would
be of little value, and specific disclosure could create
opportunities for exploitation. Disclosure describing policies
and practices also would not convey how effective (or
ineffective) any particular set of policies and practices are
likely to be. Such disclosure therefore might create a false
sense of security about an entity's ability to cope with
various risks. We do not intend to suggest that sound risk
management policies and practices are not important; we merely
question the usefulness of required public disclosure
concerning such policies and practices.
Q.2. Conflicts of Interest: Concerns about the impact of
conflicts of interest that are not properly managed have been
frequently raised in many contexts--regarding accountants,
compensation consultants, credit rating agencies, and others.
For example, Mr. Turner pointed to the conflict of the board of
FINRA including representatives of firms that it regulates. The
Millstein Center for Corporate Governance and Performance at
the Yale School of Management in New Haven, Connecticut on
March 2 proposed an industry-wide code of professional conduct
for proxy services that includes a ban on a vote advisor
performing consulting work for a company about which it
provides recommendations.
In what ways do you see conflicts of interest affecting the
integrity of the markets or investor protection? Are there
conflicts affecting the securities markets and its participants
that Congress should seek to limit or prohibit?
A.2. Conflicts of interest that are unknown or not properly
managed can have a negative impact on financial markets and
market participants. ICI agrees, therefore, that it is
important to identify conflicts or potential conflicts and
determine how they can best be addressed, including through
regulation. The appropriate solution may vary depending on the
nature and extent of the conflict as well as the context in
which it arises. For example, sometimes disclosure can be an
effective tool for addressing conflicts, by putting investors
and the marketplace on notice and allowing them to evaluate the
significance and impact of the conflict. In other
circumstances, different measures may be called for, such as
restricting or prohibiting conduct or transactions that present
conflicts.
The laws governing investment advisers and investment
companies have employed both of these approaches. Under the
Investment Advisers Act of 1940, an investment adviser must
disclose conflicts to clients, and often must seek their
consent to proceed with a transaction notwithstanding a
conflict. By contrast, the Investment Company Act of 1940
addresses potential conflicts of interest in the context of
investment company (fund) operations by prohibiting certain
transactions between a fund and fund insiders or affiliated
organizations (such as the corporate parent of the fund's
adviser). The Investment Company Act authorizes the Securities
and Exchange Commission (SEC) to grant exemptions by rule or
order to the extent such exemptions are consistent with the
underlying objectives of the statute. Pursuant to this
authority, the SEC has issued exemptive rules and orders
containing conditions designed to ensure that the interests of
fund investors are amply protected.
The restrictions on affiliated transactions under the
Investment Company Act are widely viewed as a core investor
protection and one that has served funds and their investors
very well over nearly 70 years. Congress may wish to consider
whether it would be appropriate and beneficial to apply similar
restrictions to other financial market participants or
activities, coupled with exemptive authority similar to that
granted to the SEC. At the same time, as noted above, ICI
recognizes that this approach does not fit every situation that
involves conflicts of interest. \6\
---------------------------------------------------------------------------
\6\ Likewise, a disclosure and consent model would be
impracticable in the context of a pooled investment vehicle if each
investor in the pooled vehicle were required to provide consent.
---------------------------------------------------------------------------
While ICI does not have any specific legislative
recommendations at this time regarding conflicts of interest
that may affect the integrity of the markets or investor
protection, we have commented extensively in the debate over
possible regulatory actions to address conflicts of interest
involving credit rating agencies--one of the issues mentioned
in the question above. \7\ We provide our comments on that
topic below.
---------------------------------------------------------------------------
\7\ ICI believes that the SEC currently has authority under the
Credit Rating Agency Reform Act of 2006 to implement the necessary
regulatory reforms to address rating agency conflicts of interest.
---------------------------------------------------------------------------
The SEC has recently increased the list of conflicts of
interest that must be disclosed and managed by rating agencies
or, alternatively, that are prohibited. ICI supported these
amendments but we believe that more should be done in this
area. \8\ We recommend that the SEC require additional
disclosures by rating agencies regarding their conflicts of
interest including, for example, the number of other products
rated by a rating agency for a particular issuer. In addition,
the SEC recently adopted a requirement that rating agencies
disclose information regarding the conflict of being paid by
certain parties to rate structured finance products. The
targeted conflict of interest, however, is not confined to the
ratings of these instruments. The disclosure requirement
therefore should be extended to ratings of municipal
securities.
---------------------------------------------------------------------------
\8\ See Letter from Karrie McMillan, General Counsel, Investment
Company Institute, to Florence Harmon, Acting Secretary, U.S.
Securities and Exchange Commission, dated July 25, 2008; Letter from
Karrie McMillan, General Counsel, Investment Company Institute, to
Elizabeth M. Murphy, Secretary, U.S. Securities and Exchange
Commission, dated March 26, 2009.
---------------------------------------------------------------------------
Public disclosure of conflict of interest information
should improve transparency surrounding the information and
processes used by rating agencies for rating products. It will
provide users of ratings with a more complete picture of a
rating agency's rating process and expose that process to
greater scrutiny. This exposure, in turn, should promote the
issuance of more accurate, high-quality ratings, and could
prevent rating agencies from being unduly influenced to produce
higher than warranted ratings. It should also assist investors
and other market professionals in performing an independent
assessment of these products. To achieve these goals, it is
critical that the SEC's rules governing conflicts of interest
be actively enforced and that rating agencies be held
accountable for any failures to comply with the rules and their
policies and procedures adopted under the rules.
Moreover, to fully and properly address concerns about
conflicts of interest, ICI believes the government should
ensure that regulatory reforms for rating agencies are applied
in a uniform and consistent manner equally to all types of
credit rating agencies. Each type of rating agency business
model--be it issuer-paid, subscriber-paid, or other \9\--poses
concerns and harbors conflicts of interest. Indeed, it is not
clear that one model poses fewer risks of conflicts or
invariably produces higher quality ratings.
---------------------------------------------------------------------------
\9\ While the focus of the current debate has been on issuer-paid
versus subscriber-paid models, we recognize that there may be other
compensation models worthy of consideration that may better incentivize
rating agencies to produce high quality ratings. For example, payment
for public ratings could be linked to ``quality provided'' as
determined by the end user--the investors. We believe these or other
models should be subject to the same regulatory oversight as the more
common issuer-paid and subscriber-paid models.
Q.3. Credit Default Swaps: There seems to be a consensus among
the financial industry, government officials, and industry
observers that bringing derivative instruments such as credit
default swaps under increased regulatory oversight would be
beneficial to the nation's economy. Please summarize your
---------------------------------------------------------------------------
recommendations on the best way to oversee these instruments.
A.3. As we stated in our testimony, we believe that a single
independent Federal regulator should be responsible for
oversight of U.S. capital markets, market participants, and all
financial investment products. We envision this ``Capital
Markets Regulator'' as a new regulator that would encompass the
combined functions of the Securities and Exchange Commission
(SEC) and those of the Commodity Futures Trading Commission
that are not agriculture-related. The Capital Markets Regulator
should have express authority to regulate derivatives,
including credit default swaps (CDS), including clear authority
to adopt measures to increase transparency and reduce
counterparty risk, while not unduly stifling innovation. \10\
---------------------------------------------------------------------------
\10\ To the extent that no Capital Markets Regulator is formed, we
believe that the SEC is the regulator best suited to provide effective
oversight of financial derivatives, including CDS.
---------------------------------------------------------------------------
The current initiatives toward centralized clearing for CDS
are a positive step in this regard. Central clearing of CDS
should help reduce counterparty risk and bring transparency to
trading in the types of CDS that can be standardized. We
support these initiatives.
Not all CDS are sufficiently standardized to be centrally
cleared, however, and institutional investors will continue to
need to conduct over-the-counter transactions in CDS.
Accordingly, we do not support efforts to require the mandatory
clearing of all CDS. We do support, however, reasonable
reporting requirements for these CDS transactions in order to
allow the Capital Markets Regulator to have enough data on the
CDS market to provide effective oversight.
Institutional market participants should also be required
to make periodic public disclosure of their CDS positions. SEC
registered investment companies currently make these types of
periodic public disclosures. To the extent that registered
funds buy or sell CDS, they provide extensive quarterly
financial statement disclosures that typically include both
textual note disclosure on the nature and operation of CDS and
tabular disclosure describing the terms of outstanding CDS at
the report date. Textual note disclosures typically include:
objectives, strategies, risks, cash flows, and credit events
requiring performance. Tabular disclosures typically include:
the reference entity, the counterparty, the pay/receive fixed
rate, the expiration date, the notional amount, and the
unrealized appreciation/depreciation (i.e., the fair value of
the position). The Financial Accounting Standards Board (FASB)
has recently taken steps to improve disclosures by the sellers
of credit derivatives. \11\ We fully supported that effort, and
will continue to support similar initiatives that we believe
will improve marketplace transparency in derivatives.
---------------------------------------------------------------------------
\11\ See FASB Staff Position No. 133-1 and FIN 45-4, Disclosures
about Credit Derivatives and Certain Guarantees: An Amendment of FASB
Statement No. 133 and FASB Interpretation No. 45; and Clarification of
the Effective Date of FASB Statement No. 161 (Sept. 12, 2008),
available at http://fasb.org/pdf/fsp_fas133-1&fin45-4.pdf
Q.4. Corporate Governance--Majority Vote for Directors, Proxy
Access, Say on Pay: The Council of Institutional Investors,
which represents public, union and corporate pension funds with
combined assets that exceed $3 trillion, has called for
``meaningful investor oversight of management and boards'' and
in a letter dated December 2, 2008, identified several
corporate governance provisions that ``any financial markets
regulatory reform legislation [should] include'' Please explain
---------------------------------------------------------------------------
your views on the following corporate governance issues:
1. LRequiring a majority shareholder vote for directors to be
elected in uncontested elections;
2. LAllowing shareowners the right to submit amendment to
proxy statements;
3. LAllowing advisory shareowner votes on executive cash
compensation plans.
A.4. Investment companies (funds) are major shareholders in
public companies and support strong governance and effective
management of all companies whose shares they own. Funds
typically are charged with seeking to maximize returns on
behalf of fund investors, and they use a variety of methods to
influence corporate conduct to this end. These methods include
deciding whether to invest in a company, or to continue to hold
shares; engaging directly with company management; and voting
proxies for the shares they own.
Since 2004, funds--alone among all institutional
investors--have been required to publicly disclose their proxy
votes. As a result of this unique disclosure requirement, the
manner in which fund firms vote proxies has been intensely
scrutinized, and critics have sought to politicize fund
portfolio management.
While critics have mischaracterized the data, the
availability of fund voting records demonstrates how funds use
the corporate franchise to promote the interests of their
shareholders. ICI published a report last year on a study we
conducted of more than 3.5 million votes cast by funds in the
12-month period ended June 30, 2007. Our report, Proxy Voting
by Registered Investment Companies: Promoting the Interests of
Fund Shareholders, made numerous important findings including,
among others, that: (1) funds devote substantial resources to
proxy voting; (2) funds vote proxies in accordance with their
board-approved guidelines; and (3) funds do not reflexively
vote ``with management,'' as some critics claim, but rather
make nuanced judgments in determining how to vote on both
management and shareholder proposals. \12\
---------------------------------------------------------------------------
\12\ The report is available at http://www.ici.org/stats/res/
per14-01.pdf
---------------------------------------------------------------------------
ICI has recommended that Congress extend proxy vote
disclosure requirements to other institutional investors, and
we reiterate that recommendation here. Greater transparency
around proxy voting by institutional investors should enhance
the quality of the debate concerning how the corporate
franchise is used.
We are not the only proponents for increased transparency
about the proxy votes of other institutional investors. Senator
Edward M. Kennedy (D-MA) commissioned a 2004 GAO study entitled
``Pension Plans: Additional Transparency and Other Actions
Needed in Connection with Proxy Voting,'' which concluded,
among other things, that workers and retirees would benefit
from increased transparency in proxy voting by pension plans.
Similarly, in his testimony for the March 10, 2009, Senate
Banking Committee hearing, former Securities and Exchange
Commission Chief Accountant Lynn Turner expressed support for
legislation to ``require all institutional investors, including
public, corporate and labor pension funds to disclose their
votes, just as mutual funds currently are required to disclose
their votes.'' \13\ House Financial Services Committee Chairman
Barney Frank (D-MA) also has expressed interest in considering
this issue. \14\ If disclosure of proxy votes promotes
important public policy objectives, then similar requirements
should apply to all institutional investors.
---------------------------------------------------------------------------
\13\ See Statement of Lynn E. Turner Before the Senate Committee
on Banking, Housing, and Urban Affairs on Enhancing Investor Protection
and the Regulation of the Securities Markets (March 10, 2009) at 13.
\14\ See Siobhan Hughes, Rep. Frank Plans Hearing on Disclosure of
Proxy Votes, Dow Jones News Service, March 22, 2007.
---------------------------------------------------------------------------
Below we provide our views on shareholder access to company
proxy materials for director-related bylaw amendments and
shareholder advisory votes on executive pay.
Proxy Access: In their dual role as major, long-term
investors in securities of public companies and as issuers with
their own shareholders and boards of directors, funds have a
valuable perspective to offer on the topic of shareholder
access to company proxy materials and the need to appropriately
balance the interests of shareholders with those of company
management. ICI generally supports affording certain
shareholders direct access to a company's proxy materials for
director-related bylaw amendments. \15\ We agree that long-term
shareholders with a significant stake in a company have a
legitimate interest in having a voice in the company's
corporate governance. We believe that the ability to submit
bylaw amendments concerning director nomination procedures
could be an effective additional tool for use by funds and
others to enhance shareholder value.
---------------------------------------------------------------------------
\15\ ICI has presented its views on this matter in Congressional
testimony and in a comment letter to the SEC. See Statement of Paul
Schott Stevens, President and CEO, Investment Company Institute, Before
the Committee on Financial Services, United States House of
Representatives on ``SEC Proxy Access Proposals: Implications for
Investors'' (September 27, 2007); Letter from Karrie McMillan, General
Counsel, Investment Company Institute, to Ms. Nancy M. Morris,
Secretary, U.S. Securities and Exchange Commission, dated October 2,
2007, available at http://www.ici.org/statements/cmltr/
07_sec_proxy_access_com.html#TopOfPage
---------------------------------------------------------------------------
At the same time, the privilege of proxy access should not
be granted lightly. The Federal securities laws should not
facilitate efforts to use a company's proxy machinery--at
company expense--to advance parochial or short-term interests
not shared by the company's other shareholders. Instead, the
regulatory scheme should be crafted to afford access to a
company's proxy only when the interests of shareholder
proponents are demonstrably aligned with those of long-term
shareholders.
To achieve this objective, appropriate limits on the
ability to use company resources to propose changes to a
company's governing documents are critically important. In our
view, these limits should include:
LRestricting the privilege of proxy access to
shareholders who do not acquire or hold the securities
for the purpose of changing or influencing control of
the company. Shareholders seeking to change or
influence control of the company should be required to
follow the regulatory framework for proxy contests and
bear the related costs.
LRequiring shareholder proponents to demonstrate
that they are long-term stakeholders with a significant
ownership interest. We recommend that there be a
meaningful required holding period, such as two years,
to provide assurance that shareholder proponents are
committed to the long-term mission of the company,
rather than seeking the opportunity for personal gain
and quick profits or advancement of parochial interests
at the expense of the company and other shareholders.
Similarly, we support establishing a relative high
minimum ownership threshold that would encourage
shareholders to come together to effect change. We
believe a five percent ownership threshold may not be
sufficiently high to assure that the company's proxy
machinery would be used to advance the common interests
of many shareholders in addressing legitimate concerns
about the management and operation of the company.
Consideration should be given to varying the required
ownership threshold based on factors such as the
company's market capitalization. The Securities and
Exchange Commission (SEC) should study share ownership
and holding period information to arrive at well-
reasoned criteria that will encourage would-be
shareholder proponents to work together to achieve
goals that benefit all shareholders.
LExcluding borrowed shares from the determination of
ownership level and holding period. Beneficial
ownership of shares should be required to assure that
the proponents' interests truly are aligned with those
of long-term shareholders.
Another important element of proxy access is disclosure.
Shareholder proponents should be required to provide disclosure
for inclusion in proxy materials that would allow a company's
other shareholders to make informed voting decisions (e.g.,
information about their background, intentions, and course of
dealing with the company). SEC rules also should hold
shareholder proponents--and not companies--responsible for the
disclosure those shareholders provide.
SEC Chairman Mary Schapiro recently indicated that the SEC
will soon consider a proposal ``to ensure that a company's
owners have a meaningful opportunity to nominate directors.''
\16\ We look forward to reviewing and commenting on the SEC's
proposal.
---------------------------------------------------------------------------
\16\ See SEC Speech: Address to the Council of Institutional
Investors, by Chairman Mary L. Schapiro, U.S. Securities and Exchange
Commission (May 6, 2009), available at http://www.sec.gov/news/speech/
2009/spch040609mls.htm
---------------------------------------------------------------------------
Say on Pay: As noted above, funds are significant holders
of public companies. When deciding whether to invest in a
company or to continue to hold its stock, funds consider many
factors, including how the company compensates its top
executives. This information is important because it allows
funds to decide whether (1) there is an alignment of interests
between the executives running the company and the shareholders
who own the company and (2) executives have incentives to
maximize value for shareholders. ICI has supported SEC efforts
to ensure that investors receive clear and complete disclosure
regarding executive pay packages.
The financial crisis has fanned the flames of public
outrage over executive compensation, particularly where such
compensation appears to be grossly excessive in light of a
company's performance or where the compensation seems to
promote the short-term interests of managers over the longer-
term interests of shareholders. Funds are deeply mindful of
these issues. ICI would not oppose requiring public companies
to put the compensation packages of their key executive
officers to a non-binding advisory vote of shareholders as an
additional way to encourage sound decision-making by companies
regarding the composition of executive pay packages.
We strongly urge, however, that any such requirement be
coupled with requiring other institutional investors to
disclose their proxy votes, as we recommend above. Otherwise,
the votes of funds on executive compensation, but not those of
any other institutional investor, would be subject to scrutiny
and, often we feel, unfair second-guessing. Moreover, the
potential benefits of greater transparency of the proxy voting
process would seem to be particularly evident here, where the
public disclosure of executive compensation votes would
maximize their influence over management.
Q.5. Credit Rating Agencies: A. Please identify any legislative
or regulatory changes you believe are warranted to improve the
oversight of credit rating agencies. In addition, I would like
to ask your views on two specific proposals:
1. LThe Peterson Institute report on ``Reforming Financial
Regulation, Supervision, and Oversight'' recommended
reducing conflicts of interest in the major rating
agencies by not permitting them to perform consulting
activities for the firms they rate.
2. LThe G30 Report ``Financial Reform; A Framework for
Financial Stability'' recommended that regulators
should permit users of ratings to hold NRSROs
accountable for the quality of their work product.
Similarly, Professor Coffee recommended creating
potential legal liability for recklessness when
``reasonable efforts'' have not been made to verify
``essential facts relied upon by its ratings
methodology.''
A.5. Measures To Improve the Oversight of Rating Agencies: ICI
is committed to the objective of improving the rating process
to make ratings more accurate and useful to investors and to
promote the sound functioning of our capital markets. \17\ We
recommend several regulatory measures to enhance the oversight
of credit rating agencies and thereby improve the quality,
accuracy, and integrity of ratings and the rating process. \18\
Generally speaking, our recommendations would enhance
disclosure, address conflicts of interest, and hold rating
agencies accountable for their ratings. \19\
---------------------------------------------------------------------------
\17\ ICI recently participated in the SEC's Roundtable on the
oversight of credit rating agencies in an effort to further the
discussion on ways in which to improve ratings and the ratings process.
See Statement of Paul Schott Stevens, President and CEO, Investment
Company Institute, SEC Roundtable on Oversight of Credit Rating
Agencies, dated April 15, 2009, available at http://www.ici.org/home/
09_oversight_stevens_stmt.html#TopOfPage. See also Statements of Paul
Schott Stevens, President, Investment Company Institute, on the
``Credit Rating Agency Duopoly Relief Act of 2005,'' before the
Committee on Financial Services, U.S. House of Representatives
(November 29, 2005) and on ``Assessing the Current Oversight and
Operation of Credit Rating Agencies,'' before the Committee on Banking,
Housing, and Urban Affairs, U.S. Senate (March 7, 2006).
\18\ See Letter from Karrie McMillan, General Counsel, Investment
Company Institute, to Florence Harmon, Acting Secretary, U.S.
Securities and Exchange Commission, dated July 25, 2008; Letter from
Karrie McMillan, General Counsel, Investment Company Institute, to
Elizabeth M. Murphy, Secretary, U.S. Securities and Exchange
Commission, dated March 26, 2009.
\19\ We believe the SEC currently has authority to implement many
of our recommendations. Others (such as certain changes to improve
municipal securities disclosure, discussed below) would require
Congressional action.
---------------------------------------------------------------------------
Specifically, we recommend that the Securities and Exchange
Commission (SEC) improve disclosure about credit ratings and
the rating process for structured finance securities and other
debt securities. Public disclosure of information about a
credit rating agency's policies, procedures, and other
practices relating to rating decisions will allow investors to
evaluate more effectively a rating agency's independence,
objectivity, capability, and operations. Disclosure will serve
as a powerful additional mechanism for ensuring the integrity
and quality of the credit ratings themselves. To realize the
full potential of such a disclosure regime, the SEC should
require the standardized presentation of this information in a
presale report issued by the rating agencies.
The SEC also should take steps to strengthen the incentives
to produce quality ratings, because such incentives are clearly
insufficient in the current system. To this end, the SEC should
require rating agencies to conduct ``due diligence''
assessments of the information they review to issue ratings.
This should help build investor confidence in ratings and the
rating process over time, by enabling users of ratings to gauge
both the accuracy of the information being analyzed by the
rating agency and the rating agency's ability to assess the
creditworthiness of the underlying security. We also recommend
that rating agencies have greater legal accountability to
investors for their ratings. Both of these recommendations
should encourage rating agencies to improve the quality of
their ratings.
Today's rating system is hampered by deep concerns about
conflicts of interest, poor disclosure, and lack of
accountability. To address these concerns effectively, the SEC
should apply necessary regulatory reforms in a consistent
manner to all types of credit rating agencies. A consistent
approach is not only critical to improving ratings quality and
allowing investors to identify and assess potential conflicts
of interest, but also to increasing competition among rating
agencies. The SEC must also employ a consistent and active
approach to enforcement of the oversight regime, holding rating
agencies accountable for any failures to comply with the SEC's
rules and the rating agency's own policies and procedures
adopted under the rules. Finally, we recommend that the SEC
address the need for better disclosure by certain issuers
(e.g., expand issuer disclosure for structured finance
products, expand and standardize issuer disclosure for asset-
backed securities, and require that disclosure for asset-backed
securities be ongoing). In addition, we recommend that the SEC
improve issuer disclosure for municipal securities. \20\ Better
disclosure will assist investors in making their own risk
assessments and should foster better quality ratings.
---------------------------------------------------------------------------
\20\ See Letter from Karrie McMillan, General Counsel, Investment
Company Institute, to Florence Harmon, Acting Secretary, U.S.
Securities and Exchange Commission, dated September 22, 2008.
---------------------------------------------------------------------------
Controlling Conflicts of Interest--Limiting or Prohibiting
Consulting Activities: Addressing conflicts of interest at
rating agencies is particularly important given the role that
ratings play in today's capital markets. For this reason, ICI
has recommended that the SEC require rating agencies to
disclose information, including: (1) any material ancillary
business relationships between a rating agency and an issuer
and (2) information regarding the separation of a rating
agency's consulting and rating activities. \21\ If such
information is available, we believe that it is unnecessary to
prohibit rating agencies from performing any consulting
activities for the firms they rate. The SEC already has
prohibited rating agencies from rating a product in which the
rating agency has been consulted on the structure of the
product. We believe that this measure, in combination with the
disclosure we have recommended, should curtail opportunities
for questionable conduct. In addition, it should put investors
on notice regarding potential conflicts of interest arising
from a rating agency's consulting business and provide
investors with the information needed to assess the ability and
effectiveness of a rating agency to manage those conflicts of
interest.
---------------------------------------------------------------------------
\21\ See Letter from Karrie McMillan, General Counsel, Investment
Company Institute, to Florence Harmon, Acting Secretary, U.S.
Securities and Exchange Commission, dated July 25, 2008.
---------------------------------------------------------------------------
Enhancing Accountability, Due Diligence, and Legal
Liability of Rating Agencies: Given the role of ratings in the
investment process and the use of ratings by investors, ICI
agrees with the recommendation in the G30 Report and by
Professor Coffee: credit rating agencies should have greater
legal accountability for their ratings. Currently, investors do
not have sufficient legal recourse against rating agencies if,
for example, a rating agency issues an erroneous rating.
We believe that the exemption for nationally recognized
statistical rating organizations (NRSROs) from Section 11 of
the Securities Act of 1933 should be reconsidered. \22\ Under
current regulations, the SEC exempts NRSROs, but not other
rating agencies, from treatment as experts subject to liability
under Section 11 and, thus, allows NRSRO ratings in
prospectuses and financial reports. Although the SEC has stated
that NRSROs remain subject to antifraud rules, the NRSROs have
steadfastly maintained that, under the First Amendment, they
cannot be held liable for erroneous ratings absent a finding of
malice.
---------------------------------------------------------------------------
\22\ Section 11 under the Securities Act creates liability for
issuers and certain professionals who prepared or certified any part of
a registration statement for any materially false statements or
omissions in the registration statement.
---------------------------------------------------------------------------
While it may be argued that rating agencies should not be
liable for an erroneous rating as such, they should, at a
minimum, have some accountability for ratings issued in
contravention of their own disclosed procedures and standards.
As we have stated in the past, even if the First Amendment
applies to credit ratings, it should not immunize rating
agencies for false or misleading disclosures to the SEC and to
the investing public. Quite simply, if a rating agency obtains
an NRSRO designation based on, for example, a specific ratings
process, it should be held accountable to the SEC and to
investors if it fails to follow that process.
A rating agency's ability to continue to claim First
Amendment rights also has been questioned based on the business
decisions and the roles undertaken by rating agencies over the
last decade. Rating agencies have abandoned their former
practice of rating most or all securities whether or not hired
to do so, and rating agencies have become deeply involved in
the structuring of complex securities, which are normally not
sold to retail investors. These changes warrant serious
attention when considering whether rating agencies still merit
the protection that the First Amendment may have provided to
them in their more traditional role. \23\
---------------------------------------------------------------------------
\23\ Rating agencies have cited the First Amendment in statements
to Congress, the courts, and the investing public, stating that their
ratings are opinions only--not ``recommendations or commentary on the
suitability of a particular investment.'' See, e.g., Statement of Deven
Sharma, President, Standard & Poor's, on ``Credit Rating Agencies and
the Financial Crisis,'' before the Committee on Oversight and
Government Reform, U.S. House of Representatives (October 22, 2008).
See also Not ``The World's Shortest Editorial'': Why the First
Amendment Does Not Shield Rating Agencies From Liability for Over-
Rating CDOs, David J. Grais and Kostas D. Katsiris, Grais & Ellsworth,
Bloomberg Law Reports (November 2007).
---------------------------------------------------------------------------
In addition to increasing legal accountability for rating
agencies, we believe that rating agencies would have greater
ability to produce high quality and more reliable ratings if
they were required to conduct better due diligence and
verification. Under current SEC rules, it is difficult for a
user of a rating to gauge the accuracy of the information being
analyzed by the rating agency and, thus, evaluate the rating
agency's ability to assess the creditworthiness of a structured
finance product. \24\ Rating agencies are required neither to
verify the information underlying a structured finance product
received from an issuer nor to compel issuers to perform due
diligence or to obtain reports concerning the level of due
diligence performed by issuers of structured finance products.
---------------------------------------------------------------------------
\24\ Current rules only require that rating agencies provide a
description of: (1) the public and nonpublic sources of information
used in determining credit ratings, including information and analysis
provided by third-party vendors; (2) whether and how information about
verification performed on assets underlying structured finance
securities is relied upon in determining credit ratings; and (3)
whether and how assessments of the quality of originators of structured
finance securities factor into the determination of credit ratings.
---------------------------------------------------------------------------
To address these concerns, we recommend that credit rating
agencies be required to conduct due diligence on the
information they review to issue ratings. In addition, to raise
investor confidence in the quality of ratings and the rating
process as a whole, the due diligence requirements should apply
(as appropriate) to all rated debt securities, not only
structured finance products. Specifically, we recommend that:
LRating agencies be required to have policies and
procedures in place reasonably sufficient to assess the
credibility of the information they receive from
issuers and underwriters.
LRating agencies disclose these policies and
procedures, the specific steps taken to verify
information about the assets underlying a security, and
the results of the verification process.
LRating agencies disclose the limitations of the
available information or data, any actions they take to
compensate for any missing information or data, and any
risks involved with the assumptions and methodologies
they use in providing a rating.
LRating agencies be required to certify that the
rating agency has satisfied its stated policies and
procedures for performing due diligence on the security
being rated.
Q.6. Hedge Funds: On March 5, 2009, the Managed Funds
Association testified before the House Subcommittee on Capital
Markets and said: ``MFA and its members acknowledge that at a
minimum the hedge fund industry as a whole is of systemic
relevance and, therefore, should be considered within the
systemic risk regulatory framework.'' MFA supported the
creation or designation of a ``single central systemic risk
regulator'' that (1) has ``the authority to request and
receive, on a confidential basis, from those entities that it
determines . . . to be of systemic relevance, any information
that the regulator determines is necessary or advisable to
enable it to adequately assess potential risks to the financial
system,'' (2) has a mandate of protection of the financial
system, but not investor protection or market integrity and (3)
has the authority to ensure that a failing market participant
does not pose a risk to the entire financial system.
Do you agree with MFA's position? Do you feel there should
be regulation of hedge funds along these lines or otherwise?
A.6. Systemic Risk Regulation: Over the past year, various
policymakers and other commentators have called for the
establishment of a formal mechanism for identifying,
monitoring, and managing risks to the financial system as a
whole. ICI concurs with those commentators and with the Managed
Funds Association (MFA) that creation of such a mechanism is
necessary. The ongoing financial crisis has highlighted the
vulnerability of our financial system to risks that have the
potential to spread rapidly throughout the system and cause
significant damage. A mechanism that will allow Federal
regulators to look across the system should equip them to
better anticipate and address such risks.
Generally speaking, MFA's statement about a ``single
central systemic risk regulator'' touches on some of the same
themes that ICI addressed in its March 3, 2009, white paper,
Financial Services Regulatory Reform: Discussion and
Recommendations. \25\ In our white paper, we endorsed the
designation of a new or existing agency or inter-agency body as
a ``Systemic Risk Regulator.'' Broadly stated, the goal in
establishing a Systemic Risk Regulator should be to provide
greater overall stability to the financial system as a whole.
The Systemic Risk Regulator should have responsibility for: (1)
monitoring the financial markets broadly; (2) analyzing
changing conditions in domestic and overseas markets; (3)
evaluating the risks of practices as they evolve and
identifying those that are of such nature and extent that they
implicate the health of the financial system at large; and (4)
acting to mitigate such risks in coordination with other
responsible regulators.
---------------------------------------------------------------------------
\25\ See Financial Services Regulatory Reform: Discussion and
Recommendations, which is available at http://www.ici.org/pdf/
ppr_09_reg_reform.pdf. We note that the white paper was included as an
attachment to ICI's written testimony.
---------------------------------------------------------------------------
In ICI's view, Congress should determine the composition
and authority of the Systemic Risk Regulator with two important
cautions in mind. First, the legislation establishing the
Systemic Risk Regulator should be crafted to avoid imposing
undue constraints or inapposite forms of regulation on normally
functioning elements of the financial system, or stifling
innovations, competition or efficiencies. By way of example, it
has been suggested that a Systemic Risk Regulator could be
given the authority to identify financial institutions that are
``systemically significant'' and to oversee those institutions
directly. Such an approach could have very serious
anticompetitive effects if the identified institutions were
viewed as ``too big to fail'' and thus judged by the
marketplace as safer bets than their smaller, ``less
significant'' competitors. \26\
---------------------------------------------------------------------------
\26\ See, e.g., Peter J. Wallison, Regulation Without Reason: The
Group of Thirty Report, AEI Financial Services Outlook (Jan. 2009),
available at http://www.aei.org/publications/pubID.29285/pub_detail.asp
---------------------------------------------------------------------------
Second, the Systemic Risk Regulator should not be
structured to simply add another layer of bureaucracy or to
displace the primary regulator(s) responsible for capital
markets, banking or insurance. We strongly concur with MFA that
the Systemic Risk Regulator should focus principally on
protecting the financial system--as discussed in detail in our
white paper, we believe that a strong and independent Capital
Markets Regulator (or, until such agency is established by
Congress, the Securities and Exchange Commission) should focus
principally on the equally important mandates of protecting
investors and maintaining market integrity. Legislation
establishing the Systemic Risk Regulator should define the
nature of the relationship between this new regulator and the
primary regulator(s) for each industry sector. This should
involve carefully defining the extent of the authority granted
to the Systemic Risk Regulator, as well as identifying
circumstances under which the Systemic Risk Regulator and
primary regulator(s) should coordinate their efforts and work
together. We believe, for example, that the primary regulators
have a critical role to play by acting as the first line of
defense with regard to detecting potential risks within their
spheres of expertise.
In view of the two cautions outlined above, ICI believes
that the Systemic Risk Regulator would be best structured as a
statutory council comprised of senior Federal regulators.
Membership should include, at a minimum, the Secretary of the
Treasury, Chairman of the Federal Reserve Board of Governors,
and the heads of the Federal bank and capital markets
regulators (and insurance regulator, if one emerges at the
Federal level).
Regulation of the Hedge Fund Industry--Appropriate Focus of
Regulatory Oversight: In 2004, the Securities and Exchange
Commission (SEC) adopted a rule to require hedge fund advisers
to register with the SEC as investment advisers. ICI supported
this registration requirement as a way to provide the SEC with
reliable, current, and meaningful information about this
significant segment of the capital markets without adversely
impacting the legitimate operations of hedge fund advisers.
Many ICI member firms--all of whom are registered with the
SEC--currently operate hedge funds and have found that
registration is not overly burdensome and does not interfere
with their investment activities.
In June 2006, the SEC's hedge fund adviser registration
rule was struck down by the U.S. Court of Appeals for the D.C.
Circuit. The following month, in testimony before this
Committee, then SEC Chairman Christopher Cox commented that the
rule's invalidation had forced the SEC ``back to the drawing
board to devise a workable means of acquiring even basic census
data that would be necessary to monitor hedge fund activity in
a way that could mitigate systemic risk.'' \27\
---------------------------------------------------------------------------
\27\ See Written Testimony of SEC Chairman Before the U.S. Senate
Committee on Banking, Housing and Urban Affairs (July 25, 2006)
(concerning the regulation of hedge funds), available at http://
www.sec.gov/news/testimony/2006/ts072506cc.htm
---------------------------------------------------------------------------
In our white paper, we call for this regulatory gap to be
closed. Specifically, ICI recommends that the Capital Markets
Regulator (or SEC) have express regulatory authority to provide
oversight over hedge funds (through their advisers) with
respect to, at a minimum, their potential impact on the capital
markets. \28\ For example, similar to MFA's recommendation, we
state that the regulator could require nonpublic reporting of
information such as investment positions and strategies that
could bear on systemic risk and adversely impact other market
participants.
---------------------------------------------------------------------------
\28\ It is imperative, of course, that the Capital Markets
Regulator (or SEC) be organized and staffed, and have sufficient
resources, to effectively perform this oversight function.
---------------------------------------------------------------------------
We continue to believe that hedge fund adviser registration
is an appropriate response to address the risks that hedge
funds can pose to the capital markets and other market
participants. In this regard, the Capital Markets Regulator (or
SEC) may wish to consider the adoption of specific rules under
the Investment Advisers Act of 1940 that are tailored to the
specific business practices of, and market risks posed by,
hedge funds. Areas of focus for such rulemaking should include,
for example, disclosure regarding valuation practices and the
calculation of investment performance; both of these areas have
been criticized as lacking transparency and presenting the
potential for abuse.
ICI does not support, however, requiring the registration
of individual hedge funds with the SEC. Rather, as discussed in
detail below, ICI believes there must continue to be a strict
dividing line between registered, highly regulated investment
companies and unregistered, lightly regulated hedge funds. A
registration requirement for hedge funds would blur this line,
invariably causing confusion for both investors and the
marketplace. This confusion would likely exacerbate already
imprecise uses of the term ``fund'' to refer to investment
pools, whether registered or not. Further, we believe it is
imperative to keep any problems in the hedge fund area from
bleeding over in the public's mind to include mutual funds,
which are owned by almost half of all U.S. households.
Maintaining the distinctions between investment companies
and hedge funds--Compared to registered investment companies,
which are subject to the comprehensive and rigorous regulatory
regime set forth in the Investment Company Act of 1940 and
related rules, hedge funds are lightly regulated investment
products. Hedge funds are effectively outside the purview of
the Investment Company Act by reason of Sections 3(c)(1) and
3(c)(7), which require that the hedge fund is not making or
proposing to make a public offer of its securities and that
those securities be sold only to certain specific groups of
investors. These provisions thus place express statutory limits
on both the offer and the sale of securities issued by a hedge
fund. ICI firmly believes that these limits must be preserved
and should be reconfirmed in any legislation enacted to
regulate hedge funds or their advisers.
No general solicitation or public advertising by hedge
funds--Despite clear statutory language precluding a hedge fund
from ``making or proposing to make a public offer of its
securities,'' there have been several occasions in the recent
past where the hedge fund industry has argued that it should be
able to advertise through the public media, while remaining
free from the regulatory restrictions and shareholder
protections imposed by the Investment Company Act.
Additionally, in 2003, the SEC staff recommended that the SEC
consider eliminating the prohibition on general solicitation in
offerings by certain hedge funds. ICI emphatically opposes any
such efforts, because allowing hedge funds organized pursuant
to Sections 3(c)(1) and 3(c)(7) to engage in any form of
general solicitation or public advertising is fundamentally
inconsistent with hedge funds' exclusion from regulation under
the Investment Company Act.
Section 3(c)(7) was added to the Investment Company Act in
1996, in apparent recognition that the full panoply of
investment company regulation is not necessary for hedge funds
(and other private investment pools) offered and sold only to
financially sophisticated investors able to bear the risk of
loss associated with their investment. The ``no public
offering'' language used by Congress in Section 3(c)(7)
generally tracks the language in Section 4(2) of the Securities
Act of 1933. For almost five decades, the SEC has taken the
position that public advertising is inconsistent with a
nonpublic offering of securities under Section 4(2). \29\
---------------------------------------------------------------------------
\29\ See Non-Public Offering Exemption, SEC Rel. No. 33-4552 (Nov.
6, 1962) at text preceding n.2, text preceding n.3 (``Consideration
must be given not only to the identity of the actual purchasers but
also to the offerees. Negotiations or conversations with or general
solicitations of an unrestricted and unrelated group of prospective
purchasers for the purpose of ascertaining who would be willing to
accept an offer of securities is inconsistent with a claim that the
transaction does not involve a public offering even though ultimately
there may only be a few knowledgeable purchasers . . . . Public
advertising of the offerings would, of course, be incompatible with a
claim of a private offering.'').
---------------------------------------------------------------------------
In its rulemaking to implement Section 3(c)(7) and related
provisions, the SEC observed that ``while the legislative
history . . . does not explicitly discuss Section 3(c)(7)'s
limitation on public offerings by Section 3(c)(7) funds, the
limitation appears to reflect Congress's concerns that
unsophisticated individuals not be inadvertently drawn into
[such] funds.'' \30\ A member of Congress intimately involved
in this debate later concurred with the SEC's interpretation in
a letter to then SEC Chairman Arthur Levitt. His letter further
explained:
---------------------------------------------------------------------------
\30\ See Privately Offered Investment Companies, SEC Rel. No. IC-
22597 (April 3, 1997), at n.5.
In 1996, as part of the National Securities Markets Improvement
Act, Congress reaffirmed that hedge funds should not be
publicly marketed, specifically adding this restriction to a
modernized hedge fund exemption that was included in the final
bill. As you will recall, I was one of the authors of this
provision . . . I believe that the Congress has appropriately
drawn the lines regarding hedge fund marketing, and intend to
strongly oppose any effort to liberalize them. \31\
---------------------------------------------------------------------------
\31\ See Letter from Rep. Edward J. Markey (D-Mass.) to SEC
Chairman Arthur Levitt, dated Dec. 18, 2000.
Any form of general solicitation or public advertising of
unregistered hedge funds would surely cause investors to
confuse such funds with registered, highly regulated investment
companies. It also would present greater opportunities for
perpetrators of securities fraud to identify and target
unsophisticated investors. This potential for investor
confusion and fraudulent activity would be compounded by the
fact that the SEC simply would not have the resources to
monitor advertisements by hedge funds--whether legitimate or
fraudulent--in any meaningful way.
For all of these reasons, ICI firmly believes that there
must continue to be a strict prohibition on any form of general
solicitation or public advertising in connection with hedge
fund offerings.
Limitations on who may invest in hedge funds--No less
critical is the need to ensure that interests in hedge funds
are sold only to financially sophisticated investors able to
bear the economic risk of their investment. To this end, ICI
believes that the accredited investor standards in Regulation D
under the Securities Act of 1933 (which determine investor
eligibility to participate in unregistered securities offerings
by hedge funds and other issuers) should be immediately
adjusted to correct for the substantial erosion in those
standards since their adoption in 1982. This one-time
adjustment should be coupled with periodic future adjustments
to keep pace with inflation. Specifically, ICI has recommended
that the SEC's Office of Economic Analysis be required to reset
the accredited investor thresholds every 5 years, so that the
percentage of the population qualifying as accredited investors
would remain stable over time. This would entail a
straightforward economic analysis that could be performed using
widely available government databases. \32\
---------------------------------------------------------------------------
\32\ For a detailed discussion of the Institute's views on these
issues, see Letter from Paul Schott Stevens, President and CEO,
Investment Company Institute, to Nancy M. Morris, Secretary, U.S.
Securities and Exchange Commission, dated Oct. 9, 2007, available at
http://www.sec.gov/comments/s7-18-07/s71807-37.pdf
---------------------------------------------------------------------------
Also in this regard, ICI continues to support the SEC's
2006 proposal to raise the eligibility threshold for
individuals wishing to invest in hedge funds (and other private
investment pools) organized under Section 3(c)(1) of the
Investment Company Act. Specifically, an individual would need
to be an ``accredited investor'' based upon specified net worth
or income levels, as is now required, and own at least $2.5
million in investments. According to the SEC, this new two-step
approach would mirror the existing eligibility requirements
that Congress determined were appropriate for investors in
hedge funds organized under Section 3(c)(7).
Q.7. Self-Regulatory Organizations: How do you feel the self-
regulatory securities organizations have performed during the
current financial crisis? Are there changes that should be made
to the self-regulatory organizations to improve their
performance? Do you feel there is still validity in maintaining
the self-regulatory structure or that some powers should be
moved to the SEC or elsewhere?
A.7. Self-regulatory organizations (SROs) form an integral part
of the current system of securities markets oversight. ICI has
had a longstanding interest in the effective and efficient
operation of SROs, and we support an examination of their role
and operations. We believe there may be several ways to improve
SROs' performance and operations, particularly through
enhancements to their rules and rulemaking processes, and their
governance structure.
SRO rules should be crafted both to protect investors and
to promote efficiency, competition and capital formation. To
achieve these objectives, it is critically important that SROs
consider the relative costs and benefits of their rules. We
have recommended on several occasions that Congress by law, or
the SEC by rule, require that all SROs evaluate the costs and
benefits of their rule proposals prior to submission to the SEC
and establish a process for reexamining certain existing rules.
\33\ This process should be designed to determine whether the
rules are working as intended, whether there are satisfactory
alternatives of a less burdensome nature, and whether changes
should be made.
---------------------------------------------------------------------------
\33\ See Statement of the Investment Company Institute on the
Review of the U.S. Financial Markets and Global Markets
Competitiveness, Submitted to the Senate Republican Capital Markets
Task Force, U.S. Senate (February 25, 2008) and Submission of the
Investment Company Institute to the Department of the Treasury, Review
of the Regulatory Structure Associated with Financial Institutions
(December 7, 2007).
---------------------------------------------------------------------------
The SRO rulemaking process itself serves important policy
goals, including, among other things, assuring that interested
persons have an opportunity to provide input regarding SRO
actions that could have a significant effect on the market and
market participants. ICI has supported amendments that would
improve the ability of interested persons to submit comments on
SRO actions. In particular, we have recommended extending the
length of the comment period for any significant SRO proposal.
\34\
---------------------------------------------------------------------------
\34\ See Letters from Craig S. Tyle, General Counsel, Investment
Company Institute, to Jonathan G. Katz, Secretary, U.S. Securities and
Exchange Commission, dated April 6, 2001, and Dorothy M. Donohue,
Associate Counsel, Investment Company Institute, to Jonathan G. Katz,
Secretary, U.S. Securities and Exchange Commission, dated June 4, 2004.
---------------------------------------------------------------------------
Finally, ICI supports efforts to strengthen SRO governance
processes. \35\ For example, to ensure that the views of
investors are adequately represented, we have recommended that
SROs be required to have sufficient representation from funds
and other institutional investors in their governance
structures. In addition, to address concerns that SROs are
inherently subject to conflicts of interest, consideration
should be given to requiring SRO boards to have an appropriate
balance between public members and members with industry
expertise. \36\
---------------------------------------------------------------------------
\35\ See, e.g., Letter from Ari Burstein, Associate Counsel,
Investment Company Institute, to Jonathan G. Katz, Secretary, U.S.
Securities and Exchange Commission, dated March 8, 2005.
\36\ See Regulating Broker-Dealers and Investment Advisers:
Demarcation or Harmonization?, Speech by SEC Commissioner Elisse B.
Walter Before the Mutual Fund Directors Forum Ninth Annual Policy
Conference (May 5, 2009), available at http://www.sec.gov/news/speech/
2009/spch050509ebw.htm
Q.8. Structure of the SEC: Please share your views as to
whether you feel that the current responsibilities and
structure of the SEC should be changed.
Please comment on the following specific proposals:
1. LGiving some of the SEC's duties to a systemic risk
regulator or to a financial services consumer
protection agency;
2. LCombining the SEC into a larger ``prudential'' financial
services regulator;
3. LAdding another Federal regulators' or self-regulatory
organizations' powers or duties to the SEC.
A.8. Investment companies (funds) are both major holders of
securities issued by public companies and issuers of securities
(fund shares) held by almost half of all U.S. households. As
such, they have a vested interest in the effective regulation
of the capital markets by a strong and independent regulator.
Funds and their shareholders stand to benefit if that regulator
has the tools it needs to fulfill important policy objectives,
such as: preserving the integrity of the capital markets;
ensuring the adequacy and accuracy of periodic disclosures by
public issuers; and promoting fund regulation that protects
investors, encourages innovation, and does not hinder market
competition.
As discussed in its March 3, 2009, white paper, Financial
Services Regulatory Reform: Discussion and Recommendations,
\37\ ICI supports the creation of a new Capital Markets
Regulator that would encompass the combined functions of the
Securities and Exchange Commission (SEC) and those of the
Commodity Futures Trading Commission (CFTC) that are not
agriculture-related. In our response below to part A of the
question, we briefly discuss this recommendation and our
suggestions relating to the Capital Markets Regulator's
responsibilities and structure. Pending, or in the absence of,
Congressional action to create a Capital Markets Regulator,
most of our recommendations just as appropriately could be
applied to the SEC. Where appropriate for ease of discussion,
we use the term ``agency'' to refer equally to the SEC or a new
Capital Markets Regulator.
---------------------------------------------------------------------------
\37\ See Financial Services Regulatory Reform: Discussion and
Recommendations, which is available at http://www.ici.org/pdf/
ppr_09_reg_reform.pdf. We note that the white paper was included as an
attachment to ICI's written testimony.
---------------------------------------------------------------------------
We then address the issues outlined in part B of the
question in the context of a discussion about how the SEC or a
new Capital Markets Regulator should fit within the broader
financial services regulatory framework.
Reforming the Responsibilities and Structure of the SEC: To
bring a consistent policy focus to U.S. capital markets, ICI
strongly recommends the creation of a new Capital Markets
Regulator. Currently, securities and futures are subject to
separate regulatory regimes under different Federal regulators.
This system reflects historical circumstances that have changed
significantly. As recently as the mid-1970s, for example,
agricultural products accounted for most of the total U.S.
futures exchange trading volume. By the late 1980s, a shift
from the predominance of agricultural products to financial
instruments and currencies was readily apparent in the volume
of trading on U.S. futures exchanges. In addition, as new,
innovative financial instruments were developed, the lines
between securities and futures often became blurred. The
existing, divided regulatory approach has resulted in
jurisdictional disputes between the SEC and the CFTC,
regulatory inefficiency, and gaps in investor protection and
market oversight. With the increasing convergence of securities
and futures products, markets, and market participants, the
current system simply makes no sense.
As envisioned by ICI, the Capital Markets Regulator would
be a single, independent Federal regulator responsible for
oversight of U.S. capital markets, market participants, and all
financial investment products. It would have an express
statutory mission and the rulemaking and enforcement powers
necessary to carry out that mission. \38\ From the perspective
of the fund industry, the mission of the Capital Markets
Regulator must involve maintaining a sharp focus on investor
protection, supported by a comprehensive enforcement program.
This core feature of the SEC's mission has consistently
distinguished the SEC from the banking regulators, who are
principally concerned with the safety and soundness of the
financial institutions they regulate, and it has generally
served investors well over the years.
---------------------------------------------------------------------------
\38\ Currently, regulatory oversight of both the securities and
futures industries involves various self-regulatory organizations. In
establishing a Capital Markets Regulator, Congress would need to
determine the appropriate role for any such organization(s).
---------------------------------------------------------------------------
Examination of the recent financial crisis has prompted
calls for Congress to close regulatory gaps to ensure
appropriate oversight of all market participants and investment
products. In our white paper, we recommend that the Capital
Markets Regulator (or SEC) have express regulatory authority to
provide oversight with regard to hedge funds, derivatives, and
municipal securities. We further recommend that the agency be
given explicit authority to harmonize the legal standards
applicable to investment advisers and broker-dealers.
How a regulatory agency is managed, and the details of its
organizational structure, can have significant implications for
the agency's effectiveness. In our white paper, we offer the
following suggestions with regard to management and
organization of the Capital Markets Regulator (or SEC).
LEnsure high-level focus on agency management. One
approach would be to designate a Chief Operating
Officer for this purpose.
LImplement a comprehensive process for setting
regulatory priorities and assessing progress. It may be
helpful to draw upon the experience of the United
Kingdom's Financial Services Authority, which seeks to
follow a methodical approach that includes developing a
detailed annual business plan establishing agency
priorities and then reporting annually the agency's
progress in meeting prescribed benchmarks.
LPromote open and effective lines of communication
among the Commissioners and between the Commissioners
and staff. Such communication is critical to fostering
awareness of issues and problems as they arise, thus
increasing the likelihood that the agency will be able
to act promptly and effectively. A range of approaches
may be appropriate to consider in meeting this goal,
including whether sufficient flexibility is provided
under the Government in the Sunshine Act, and whether
the number of Commissioners should be greater than the
current number at the SEC (five).
LAlign the inspections and examinations functions
and the policymaking divisions. This approach would
have the benefit of keeping staff in the policymaking
divisions updated on current market and industry
developments, as well as precluding any de facto
rulemaking by the agency's inspections staff.
LDevelop mechanisms to facilitate coordination and
information sharing among the policymaking divisions.
These mechanisms would help to ensure that the agency
speaks with one voice.
How the SEC (or a New Capital Markets Regulator) Fits
Within the Broader Financial Services Regulatory Framework:
Today's financial crisis has demonstrated that the current
system for oversight of U.S. financial institutions is
insufficient to address modern financial markets. In its white
paper, ICI recommends changes to create a regulatory framework
that enhances regulatory efficiency, limits duplication, closes
regulatory gaps, and emphasizes the national character of the
financial services industry. In brief, ICI supports:
LCreating a consolidated Capital Markets Regulator,
as discussed above;
LEstablishing a ``Systemic Risk Regulator'' that
would identify, monitor and manage risks to the
financial system as a whole;
LConsidering consolidation of the regulatory
structure for the banking sector;
LAuthorizing an optional Federal charter for
insurance companies; and
LPromoting effective coordination and information
sharing among the various financial regulators,
including in particular the new Systemic Risk
Regulator.
Increased consolidation of financial services regulators,
combined with the establishment of a Systemic Risk Regulator
and more robust inter-agency coordination and information
sharing, should facilitate monitoring and mitigation of risks
across the financial system. We believe that consolidation of
regulatory agencies also may further the competitive posture of
the U.S. financial markets and could make it easier, when
appropriate, to harmonize U.S. regulations with regulations in
other jurisdictions. Reducing the number of U.S. regulatory
agencies, while also strengthening the culture of cooperation
and dialogue among senior officials of the agencies, will
likely facilitate coordinated interaction with regulators
around the world.
By providing for one or more dedicated regulators to
oversee each major financial services sector, this proposed
structure would maintain the specialized focus and expertise
that is a hallmark of effective regulation. This structure also
would allow appropriate tailoring of regulation to accommodate
fundamental differences in regulated entities, products and
activities. Additionally, it would avoid the potential for one
industry sector to take precedence over the others in terms of
regulatory priorities or approaches or the allocation of
regulatory resources.
In particular, the regulatory structure favored by ICI
would preserve the important distinctions between the mission
of the Capital Markets Regulator (or SEC), which is sharply
focused on investor protection, and that of the banking
regulators, which is principally concerned with the safety and
soundness of the banking system. Both regulatory approaches
have a critical role to play in ensuring a successful and
vibrant financial system, but neither should be allowed to
trump the other. For this reason, we believe it would be
inappropriate to combine the SEC into a larger ``prudential''
financial services regulator, a move that could result in
diminished investor protections.
Preserving regulatory balance, and bringing to bear
different perspectives, is a theme that has influenced ICI's
thinking on how to structure a Systemic Risk Regulator. In our
white paper, we suggested that very careful consideration must
be given to the specifics of how a Systemic Risk Regulator
would be authorized to perform its functions. We offered two
important cautions in that regard. First, we recommended that
the legislation establishing the Systemic Risk Regulator should
be crafted to avoid imposing undue constraints or inapposite
forms of regulation on normally functioning elements of the
financial system, or stifling innovations, competition or
efficiencies. Second, we recommended that the Systemic Risk
Regulator should not be structured to simply add another layer
of bureaucracy or to displace the primary regulator(s)
responsible for capital markets, banking, or insurance.
Legislation establishing the Systemic Risk Regulator thus
should define the nature of the relationship between this new
regulator and the primary regulator(s) for each industry
sector. This should involve carefully defining the extent of
the authority granted to the Systemic Risk Regulator, as well
as identifying circumstances under which the Systemic Risk
Regulator and primary regulator(s) should coordinate their
efforts and work together. We believe, for example, that the
primary regulators have a critical role to play by acting as
the first line of defense with regard to detecting potential
risks within their spheres of expertise.
In view of the two cautions outlined above, ICI believes
that the Systemic Risk Regulator would be best structured as a
statutory council comprised of senior Federal regulators.
Membership should include, at a minimum, the Secretary of the
Treasury, Chairman of the Federal Reserve Board of Governors,
and the heads of the Federal bank and capital markets
regulators (and insurance regulator, if one emerges at the
Federal level).
Finally, we note that the question requests comment on
whether some of the SEC's duties should be given to a financial
services consumer protection agency. As a general matter, we
observe that Federal regulators must improve their ability to
keep up with new market developments. This will require both
nimbleness at the regulatory level and Congressional
willingness to close regulatory gaps, provide new authority
where appropriate, and even provide additional resources. ICI
does not believe, however, that it would be helpful to create a
new ``financial products safety commission'' or ``financial
services consumer protection agency.'' Financial products and
services arise and exist in the context of a larger
marketplace, and they need to be understood in that context.
The primary regulator is best positioned to perform this
function.
Q.9. SEC Staffing, Funding, and Management: The SEC has a staff
of about 3,500 full-time employees and a budget of $900
million. It has regulatory responsibilities with respect to
approximately: 12,000 public companies whose securities are
registered with it; 11,300 investment advisers; 950 mutual fund
complexes; 5,500 broker-dealers (including 173,000 branch
offices and 665,000 registered representatives); 600 transfer
agents, 11 exchanges; 5 clearing agencies; 10 nationally
recognized statistical rating organizations; SROs such as the
Financial Industry Regulatory Authority, the Municipal
Securities Rulemaking Board and the Public Company Accounting
Oversight Board.
To perform its mission effectively, do you feel that the
SEC is appropriately staffed? funded? managed? How would you
suggest that the Congress could improve the effectiveness of
the SEC?
A.9. Investment companies (funds) are both major holders of
securities issued by public companies and issuers of securities
(fund shares) that are held by almost half of all U.S.
households. As such, they have a vested interest in effective
regulation of the capital markets by a strong and independent
regulator. Funds, and therefore their shareholders, stand to
benefit if that regulator has the tools it needs to fulfill
important policy objectives, such as: preserving the integrity
of the capital markets; ensuring the adequacy and accuracy of
periodic disclosures by public issuers; and promoting fund
regulation that protects our investors, encourages innovation,
and does not hinder market competition.
As discussed in our March 3, 2009, white paper, Financial
Services Regulatory Reform: Discussion and Recommendations,
\39\ ICI supports the creation of a new Capital Markets
Regulator that would encompass the combined functions of the
Securities and Exchange Commission (SEC) and the Commodity
Futures Trading Commission (CFTC). The white paper makes a
series of recommendations--including several concerning the
staffing, funding, and management of the Capital Markets
Regulator--aimed at maximizing this regulator's ability to
perform its mission effectively. Pending, or in the absence of,
Congressional action to create a Capital Markets Regulator,
most of our recommendations just as appropriately could be
applied to the SEC. An outline of those recommendations is
included in the response below. Where appropriate for ease of
discussion, we use the term ``agency'' to refer equally to the
SEC or a new Capital Markets Regulator.
---------------------------------------------------------------------------
\39\ See Financial Services Regulatory Reform: Discussion and
Recommendations, which is available at http://www.ici.org/pdf/
ppr_09_reg_reform.pdf. We note that the white paper was included as an
attachment to ICI's written testimony.
---------------------------------------------------------------------------
Agency Funding, and Staffing: ICI consistently has called
for adequate funding for the SEC in order to support its
critical regulatory functions. We note that, in testimony
before the House Financial Services Committee in March of this
year, SEC Commissioner Elisse Walter stated that the SEC's
examination and enforcement resources are inadequate to keep
pace with the growth and innovation in the securities markets.
\40\ We believe that Congress must seriously consider any
suggestion from senior SEC officials that additional resources
are required. We were pleased, therefore, to hear about the
recent bipartisan effort, led by Senators Charles Schumer (D-
NY) and Richard Shelby (R-AL) and endorsed by SEC Chairman Mary
Schapiro, to increase the SEC's budget by $20 million for
fiscal years 2010 and 2011 in order to add enforcement staff
and fund needed technology upgrades.
---------------------------------------------------------------------------
\40\ See Testimony of Elisse B. Walter, Commissioner, U.S.
Securities and Exchange Commission, Before the House Committee on
Financial Services, Concerning Securities Law Enforcement in the
Current Financial Crisis (March 20, 2009) at 30-31.
---------------------------------------------------------------------------
ICI believes that the agency also must have greater ability
(and resources) to attract and retain professional staff having
significant prior industry experience. Their practical
perspectives would enhance the agency's ability to keep current
with market and industry developments and better understand the
impact of such developments on regulatory policy. The SEC's
announcement of a new Industry and Market Fellows Program is an
encouraging step in the right direction. \41\ As discussed
further below, the agency also should build strong economic
research and analytical capabilities and should consider having
economists resident in each division.
---------------------------------------------------------------------------
\41\ See SEC Announces New Initiative to Identify and Assess Risks
in Financial Markets (April 30, 2009), available at http://www.sec.gov/
news/press/2009/2009-98.htm
---------------------------------------------------------------------------
Examination of the recent financial crisis has prompted
calls for Congress to close regulatory gaps to ensure
appropriate oversight of all market participants and investment
products. In our white paper, we recommend that the Capital
Markets Regulator (or SEC) have express regulatory authority to
provide oversight with regard to hedge funds, derivatives, and
municipal securities. To the extent that the scope of the
agency's responsibilities is expanded, it will be imperative
that it have sufficient staffing and resources to effectively
perform all of its oversight functions.
Agency Management and Organization: How a regulatory agency
is managed, and the details of its organizational structure,
can have significant implications for the agency's
effectiveness. In our white paper, we offer the following
suggestions with regard to agency management and organization.
LEnsure high-level focus on agency management. One
approach would be to designate a Chief Operating
Officer for this purpose.
LImplement a comprehensive process for setting
regulatory priorities and assessing progress. It may be
helpful to draw upon the experience of the United
Kingdom's Financial Services Authority, which seeks to
follow a methodical approach that includes developing a
detailed annual business plan establishing agency
priorities and then reporting annually the agency's
progress in meeting prescribed benchmarks.
LPromote open and effective lines of communication
among the Commissioners and between the Commissioners
and staff. Such communication is critical to fostering
awareness of issues and problems as they arise, thus
increasing the likelihood that the agency will be able
to act promptly and effectively. A range of approaches
may be appropriate to consider in meeting this goal,
including whether sufficient flexibility is provided
under the Government in the Sunshine Act, and whether
the number of Commissioners should be greater than the
current number at the SEC (five).
LAlign the inspections and examinations functions
and the policymaking divisions. This approach would
have the benefit of keeping staff in the policymaking
divisions updated on current market and industry
developments, as well as precluding any de facto
rulemaking by the agency's inspections staff.
LDevelop mechanisms to facilitate coordination and
information sharing among the policymaking divisions.
These mechanisms would help to ensure that the agency
speaks with one voice.
Improving Agency Effectiveness: Our white paper recommends
the following additional ways to enhance the agency's ability
to fulfill its mission successfully when carrying out its
regulatory responsibilities:
1. LEstablish the conditions necessary for constructive,
ongoing dialogue with the regulated industry: The
agency should seek to facilitate closer, cooperative
interaction with the entities it regulates to identify
and resolve problems, to determine the impact of
problems or practices on investors and the market, and
to cooperatively develop best practices that can be
shared broadly with market participants. Incorporating
a more preventative approach would likely encourage
firms to step forward with self-identified problems and
proposed resolutions. The net result is that the agency
would pursue its investor protection responsibilities
through various means not always involving enforcement
measures, although strong enforcement must remain an
important weapon in the agency's arsenal.
2. LEstablish mechanisms to stay abreast of market and
industry developments: The agency would benefit from
the establishment of one or more external mechanisms
designed to help it stay abreast of market and industry
issues and developments, including developments and
practices in non-U.S. jurisdictions as appropriate. For
example, several Federal agencies--including both the
SEC and CFTC--utilize a range of advisory committees.
Such committees, which generally have significant
private sector representation, may be established to
provide recommendations on a discrete set of issues
facing the agency (e.g., the SEC's Advisory Committee
on Improvements to Financial Reporting) or to provide
regular information and guidance to the agency (e.g.,
the CFTC's Agricultural Advisory Committee).
LICI believes that a multidisciplinary ``Capital Markets
Advisory Committee'' could be a very effective
mechanism for providing the agency with ``real world''
perspectives and insights on an ongoing basis. We
recommend that such a committee be comprised primarily
of private sector representatives from all major
sectors of the capital markets, and include one or more
members representing funds and asset managers.
Additionally, the Capital Markets Advisory Committee
should be specifically established in, and required by,
legislation. Such a statutory mandate would emphasize
the importance of this advisory committee to the
agency's successful fulfillment of its mission.
LThe establishment of an advisory committee would
complement other efforts by the agency to monitor
developments affecting the capital markets and market
participants. These efforts should include, first and
foremost, hiring more staff members with significant
prior industry experience. As indicated above, their
practical perspectives would enhance the agency's
ability to keep current with market and industry
developments and better understand the impact of such
developments on regulatory policy.
3. LApply reasonably comparable regulation to like products
and services: Different investment products often are
subject to different regulatory requirements, often
with good reason. At times, however, heavier regulatory
burdens have been placed on some investment products or
services than on others, even where they share similar
features and are sold to the same customer base. It
does not serve investors well if the regulatory
requirements placed on funds--which serve over 93
million investors--end up discouraging investment
advisers from entering or remaining in the fund
business, dissuading portfolio managers from managing
funds as opposed to other investment products, or
creating disincentives for brokers and other
intermediaries to sell fund shares. It is critically
important for the agency to be sensitive to this
dynamic in its rulemakings.
LAmong other things, in analyzing potential new regulatory
requirements for funds or in other situations as
appropriate, the agency should consider whether other
investment products raise similar policy concerns and
thus should be subject to comparable requirements. In
this regard, we note that separately managed accounts
sometimes are operated much like mutual funds and other
investment companies and yet do not offer the same
level of investor protection. For example, as the
fallout from the Ponzi scheme perpetrated by Bernard
Madoff has highlighted, separately managed accounts are
not subject to all of the restrictions on custody
arrangements that serve to protect fund assets, and
existing rules leave room for abuse. \42\
---------------------------------------------------------------------------
\42\ The SEC has scheduled an open meeting on May 14, 2009, at
which it will consider proposed rule amendments designed to enhance the
protections provided to advisory clients when they entrust their funds
and securities to an investment adviser. The SEC's meeting announcement
indicates that if adopted, the amendments would require investment
advisers having custody of client funds and securities to obtain a
surprise examination by an independent public accountant, and, unless
the client assets are maintained with an independent custodian, obtain
a review of custodial controls from an independent public accountant.
See SEC News Digest (May 7, 2009), available at http://www.sec.gov/
news/digest/2009/dig050709.htm
4. LDevelop strong capability to conduct economic analysis to
support sound rulemaking and oversight: The agency will
be best positioned to accomplish its mission if it
conducts economic analysis in various aspects of the
agency's work, including rulemaking, examinations, and
enforcement. Building strong economic research and
analytical capabilities is an important way to enhance
the mix of disciplines that will inform the agency's
activities. From helping the agency look at broad
trends that shed light on how markets or individual
firms are operating to enabling it to demonstrate that
specific policy initiatives are well-grounded,
developing the agency's capability to conduct economic
analysis will be well worth the long-term effort
required. The agency should consider having economists
resident in each division to bring additional,
important perspectives to bear on regulatory
---------------------------------------------------------------------------
challenges.
LIt is important that economic analysis play an integral
role in the rulemaking process, because many regulatory
costs ultimately are borne by investors. When new
regulations are required, or existing regulations are
amended, the agency should thoroughly examine all
possible options and choose the alternative that
reflects the best trade-off between costs to, and
benefits for, investors. Effective cost benefit
analysis does not mean compromising protections for
investors or the capital markets. Rather, it challenges
the regulator to consider alternative proposals and
think creatively to achieve appropriate protections
while minimizing regulatory burdens, or to demonstrate
that a proposal's costs and burdens are justified in
light of the nature and extent of the benefits that
will be achieved. \43\
---------------------------------------------------------------------------
\43\ See, e.g., Special Report on Regulatory Reform, Congressional
Oversight Panel (submitted under Section 125(b)(2) of Title I of the
Emergency Economic Stabilization Act of 2008) (Jan. 2009) (``In
tailoring regulatory responses . . . the goal should always be to
strike a reasonable balance between the costs of regulation and its
benefits. Just as speed limits are more stringent on busy city streets
than on open highways, financial regulation should be strictest where
the threats-especially the threats to other citizens--are greatest, and
it should be more moderate elsewhere.'').
5. LModernize regulations that no longer reflect current
market structures and practices: Financial markets and
related services are constantly evolving, frequently at
a pace that can make the regulations governing them (or
the rationale behind those regulations) become less
than optimal, if not entirely obsolete. Requiring
industry participants to comply with outmoded
regulations imposes unnecessary costs on both firms and
investors, may impede innovation, and, most troubling
of all, could result in inadequate protection of
investors. It is thus important that the agency engage
in periodic reviews of its existing regulations to
determine whether any such regulations should be
---------------------------------------------------------------------------
modernized or eliminated.
6. LGive heightened attention to investor education: The
recent turmoil in the financial markets has underscored
how important it is that investors be knowledgeable and
understand their investments. Well-informed investors
are more likely to develop realistic expectations, take
a long-term perspective, and understand the trade-off
between risk and reward. They are less likely to panic
and make mistakes.
LTo better equip investors to make good decisions about
their investments, the agency should assign a high
priority to pursuing regulatory initiatives that will
help educate investors. The SEC has an Office of
Investor Education and Advocacy and provides some
investor education resources on its Web site. These
types of efforts should be expanded, possibly in
partnership with other governmental or private
entities, and better publicized. Many industry
participants, too, have developed materials and other
tools to help educate investors; additional investor
outreach efforts should be encouraged.
Q.10. Systemic Risk Regulatory Structure: You have put forth
the idea of a systemic risk regulator that is organized as a
committee of financial regulatory heads. Could you please
elaborate on the structure and organization of such a systemic
risk regulatory you are suggesting? Also, please describe the
positives and negatives of such an arrangement and the reasons
why it would be superior to other possibilities.
A.10. In light of the financial crisis, it is imperative that
Congress establish in statute responsibility to address risks
to the financial system at large. For certain specific and
identifiable purposes, such as assuring effective consolidated
global supervision of the largest bank holding companies and
overseeing the robust functioning of the payment and settlement
system as appropriate, this systemic risk management
responsibility might be lodged with the Federal Reserve Board.
Beyond this context, however, I recommend that systemic risk
management responsibility should be assigned to a statutory
council comprised of senior Federal regulators.
In concept, such a council would be similar to the National
Security Council (NSC), which was established by the National
Security Act of 1947. In the aftermath of World War II,
Congress recognized the need to assure better coordination and
integration of ``domestic, foreign, and military policies
relating to the national security'' and the ongoing assessment
of ``policies, objectives, and risks.'' The 1947 Act
established the NSC under the President as a Cabinet-level
council with a dedicated staff. In succeeding years, the NSC
has proved to be a key mechanism utilized by Presidents to
address the increasingly complex and multi-faceted challenges
of national security policy. It was my honor from 1987-89 to
serve as statutory head (i.e., Executive Secretary) of the NSC.
As with national security, addressing risks to the
financial system at large requires, in my view, diverse inputs
and perspectives. Membership of such a council accordingly
should draw upon a broad base of expertise, and should include
at a minimum the Secretary of the Treasury, Chairman of the
Board of Governors of the Federal Reserve System, and the heads
of the Federal bank and capital markets regulators (and
insurance regulator, if one emerges at the Federal level). As
with the NSC, flexibility should exist to enlist other
regulators into the work of the council on specific issues as
required--including, for example, State insurance regulators
and self-regulatory organizations.
By statute, the council should have a mandate to monitor
conditions and developments in the domestic and international
financial markets, to assess their implications for the health
of the financial system at large, to identify regulatory
actions to be taken to address systemic risks as they emerge,
to assess the effectiveness of these actions, and to advise the
President and the Congress on emerging risks and necessary
legislative or regulatory responses. The council would be
responsible for coordinating and integrating the national
response to systemic financial risks, but it would not have a
direct operating role (much as the NSC coordinates and
integrates military and foreign policy that is implemented by
the Defense or State Department and not by the NSC itself).
Rather, responsibility for addressing identified risks would
lie with the existing functional regulators, who would act
pursuant to their normal statutory authorities but under the
council's direction.
The Secretary of the Treasury, as the senior-most member of
the council, should be designated chairman. An executive
director, appointed by the President, should run the day-to-day
operations of the council and serve as head of the council's
staff. The council should meet on a regular basis, with an
interagency process coordinated through the council's staff to
support and follow through on its ongoing deliberations.
To accomplish its mission, the council should have the
support of a dedicated, highly-experienced staff. The staff
should represent a mix of disciplines (e.g., economics, finance
and law) and should consist of individuals seconded from
government departments and agencies (Federal and state), as
well as recruited from the private sector with a business,
professional or academic background. As with the NSC, the
staff's focus would be to support the work of the council as
such, and thus the staff would operate independently from the
functional regulators. Nonetheless, the background and
experience of the staff would help assure the kind of strong
working relationships with the functional regulators necessary
for the council's success. Such a staff could be recruited and
at work in a relatively short period of time. The focus in
recruiting such a staff should be on quality, not quantity, and
the council's staff accordingly should not and need not be
large.
Such a council structure has many advantages to recommend
it:
LSystemic risks may arise in different ways and
affect different parts of the domestic and global
financial system. No existing agency or department has
a comprehensive frame of reference or the necessary
expertise to assess and respond to any and all such
risks. Creating such an all-purpose systemic risk
manager would be a long and complex undertaking, and
would involve developing expertise that duplicates that
which exists in today's functional regulators. The
council structure by contrast would enlist the
expertise of the entire regulatory community in
identifying and devising strategies to mitigate
systemic risks. It also could be established and begin
operation much more quickly.
LThe council structure would avoid risks inherent in
the leading alternative that has been proposed--i.e.,
designating an existing agency like the Federal Reserve
Board to serve as an all-purpose systemic risk
regulator. In this role, the Federal Reserve
understandably may tend to view risks and risk
mitigation through its lens as a bank regulator focused
on prudential regulation and ``safety and soundness''
concerns, potentially to the detriment of consumer and
investor protection concerns and of non-bank financial
institutions. In my view, a council such as I have
outlined would bring all these competing perspectives
to bear and, as a result, would seem far more likely to
strike the proper balance.
LSuch a council would provide a high degree of
flexibility in convening those Federal and State
regulators whose input and participation is necessary
to addressing a specific issue, without creating an
unwieldy or bureaucratic structure. As is the case with
the NSC, the council should have a core membership of
senior Federal officials and the ability to expand its
participants on an ad hoc basis when a given issue so
requires.
LWith an independent staff dedicated solely to
pursuing the council's agenda, the council would be
well positioned to test or challenge the policy
judgments or priorities of various functional
regulators. Moreover, by virtue of their participation
on the council, the various functional regulators are
themselves likely to be more attentive to emerging
risks or regulatory gaps. This would help assure a far
more coordinated and integrated approach. Over time,
the council also would assist in identifying and
promoting political consensus about significant
regulatory gaps and necessary policy responses.
LThe council model anticipates that functional
regulators, as distinct from the council itself, would
be charged with implementing regulations to mitigate
systemic risks as they emerge. This operational role is
appropriate because the functional regulators will have
the greatest in depth knowledge of their respective
regulated industries. Nonetheless, the council and its
staff will have an important independent role in
evaluating the effectiveness of the measures taken by
functional regulators to mitigate systemic risk and,
where necessary, in prompting further actions.
LA potential criticism of the council structure is
that it may diffuse responsibility and pose
difficulties in assuring proper follow-through by the
functional regulators. I agree it is important that the
council have ``teeth,'' and this can be accomplished,
in crafting the legislation, through appropriate
amendments to the organic statutes governing the
functional regulators.
------
RESPONSE TO WRITTEN QUESTIONS OF SENATOR CRAPO
FROM PAUL SCHOTT STEVENS
Q.1. Do you all agree with Federal Reserve Board Chairman
Bernanke's remarks today about the four key elements that
should guide regulatory reform?
First, we must address the problem of financial institutions
that are deemed too big--or perhaps too interconnected--to
fail. Second, we must strengthen what I will call the financial
infrastructure--the systems, rules, and conventions that govern
trading, payment, clearing, and settlement in financial
markets--to ensure that it will perform well under stress.
Third, we should review regulatory policies and accounting
rules to ensure that they do not induce excessive
procyclicality--that is, do not overly magnify the ups and
downs in the financial system and the economy. Finally, we
should consider whether the creation of an authority
specifically charged with monitoring and addressing systemic
risks would help protect the system from financial crises like
the one we are currently experiencing.
A.1. In his March 10 speech to the Council on Foreign
Relations, Chairman Bernanke suggested that policymakers should
begin to think about ``reforms to the financial architecture,
broadly conceived, that could help prevent a similar
[financial] crisis from developing in the future.'' He further
highlighted the need for ``a strategy that regulates the
financial system as a whole, in a holistic way.'' ICI concurs
with Chairman Bernanke that the four areas outlined in the
question, and discussed in turn below, are key elements of such
a strategy. It bears emphasizing that this list is not
exclusive (and that Chairman Bernanke himself did not suggest
otherwise). In ICI's view, other key elements of a reform
strategy include consolidating and strengthening the primary
regulators for each financial sector, and ensuring more
effective coordination and information sharing among those
regulators. These issues are addressed in detail in ICI's March
3, 2009, white paper, Financial Services Regulatory Reform:
Discussion and Recommendations. \1\
---------------------------------------------------------------------------
\1\ See Financial Services Regulatory Reform: Discussion and
Recommendations, which is available at http://www.ici.org/pdf/
ppr_09_reg_reform.pdf. We note that the white paper was included as an
attachment to ICI's written testimony.
---------------------------------------------------------------------------
``Too big to fail'': ICI agrees that the notion of
financial institutions that are too big or too interconnected
to fail deserves careful attention. The financial crisis has
highlighted how the activities of large financial institutions
can have wide-ranging effects on the economy. It is incumbent
upon policymakers and other interested parties to consider how
best to mitigate the risks that the activities of large
financial institutions can pose to the financial system as a
whole.
As part of this analysis, one issue is how to define what
is meant by ``too big to fail.'' If it means that certain large
financial institutions will receive either explicit or implicit
Federal guarantees of their debt, such institutions will gain
an unfair competitive advantage. Allowing these institutions to
borrow at risk-free (or near risk-free) interest rates could
encourage them to take excessive risks, and may cause them to
grow faster than their competitors, both of which potentially
would magnify systemic risks. Ultimately, U.S. taxpayers would
bear the costs of such actions.
Chairman Bernanke echoed these concerns in his March 10
remarks. He described the undesirable effects if market
participants believe that a firm is considered too big to fail,
indicating that this belief:
reduces market discipline and encourages excessive risk-taking
by the firm. It also provides an artificial incentive for firms
to grow, in order to be perceived as too big to fail. And it
creates an unlevel playing field with smaller firms, which may
not be regarded as having implicit government support.
Moreover, government rescues of too-big-to-fail firms can be
costly to taxpayers, as we have seen recently.
Legislative or regulatory reforms aimed at addressing risks
to the financial system posed by the activities of large and
complex financial firms must be designed to avoid these
results.
Strengthening the Financial Infrastructure: ICI strongly
concurs with Chairman Bernanke's comments about the need to
strengthen the financial infrastructure, in order to improve
the ability of the financial system to withstand future shocks
and ``reduc[e] the range of circumstances in which systemic
stability concerns might prompt government intervention.'' For
example, we support current initiatives toward centralized
clearing for credit default swaps (CDS). Central clearing
should help reduce counterparty risk and bring transparency to
trading in the types of CDS that can be standardized. Not all
CDS are sufficiently standardized to be centrally cleared,
however, and institutional investors will continue to need to
conduct over-the-counter transactions in CDS. For those
transactions, we support reasonable reporting requirements, in
order to ensure that regulators have enough data on the CDS
market to provide effective oversight. In addition, we would be
generally supportive of efforts to improve the market for
repurchase agreements. Steps such as those we have outlined may
serve to deepen the relevant markets, encourage buyers and
sellers to continue to transact during times of market turmoil
and, in particular, help foster greater price transparency.
We further concur with Chairman Bernanke's assessment of
the importance of money market funds--particularly their
``crucial role'' in the commercial paper market and as a
funding source for businesses--and his call for policymakers to
consider ``how to increase the resiliency of those funds that
are susceptible to runs.'' Similarly, Treasury Secretary
Geithner has outlined the Administration's position on systemic
risk and called for action in six areas, including the adoption
of new requirements for money market funds to reduce the risk
of rapid withdrawals. In this regard, ICI and its members,
working through our Money Market Working Group, recently issued
a comprehensive report outlining a range of measures to
strengthen money market funds and help them withstand difficult
market conditions in the future. \2\ More specifically, the
Working Group's recommendations are designed to strengthen and
preserve the unique attributes of a money market fund as a low-
cost, efficient cash management tool that provides a high
degree of liquidity, stability in principal value, and a
market-based yield. The proposed standards and regulations
would ensure that money market funds are better positioned to
sustain prolonged and extreme redemption pressures and that
mechanisms are in place to ensure that all shareholders are
treated fairly if a fund sees its net asset value fall below
$1.00.
---------------------------------------------------------------------------
\2\ See Report of the Money Market Working Group, Investment
Company Institute (March 17, 2009), available at http://www.ici.org/
pdf/ppr_09_mmwg.pdf
---------------------------------------------------------------------------
Secretary Geithner specifically identified the SEC as the
agency to implement any new requirements for money market
funds. ICI wholeheartedly concurs that the SEC, as the primary
regulator for money market funds, is uniquely qualified to
evaluate and implement potential changes to the existing scheme
of money market fund regulation. SEC Chairman Shapiro and
members of her staff have indicated on several occasions that
her agency is currently conducting such a review on an
expedited basis, and we are pleased that the review will
include consideration of the Working Group's recommendations.
Preventing Excessive Procyclicality: Some financial
institutions have criticized the use of mark-to-market
accounting in the current environment as overstating losses,
diminishing bank capital, and exacerbating the crisis. Others
have applauded its use as essential in promptly revealing the
extent of problem assets and the deteriorating financial
condition of institutions. Investment companies, as investors
in securities, rely upon financial reporting that accurately
portrays the results and financial position of companies
competing for investment capital. ICI supports the work of the
Financial Accounting Standards Board and its mission to develop
financial reporting standards that provide investors with
relevant, reliable and transparent information about corporate
financial performance. Certainly, regulatory policies and
accounting rules should not induce excessive procyclicality. At
the same time, accounting standards should not be modified to
achieve any objective other than fair and accurate reporting to
investors and the capital markets. Any concerns regarding the
procyclical effects of mark-to-market accounting on lending
institutions' capital may be better addressed through changes
to capital standards themselves. Consideration should be given
to, for example, developing countercyclical capital standards
and requiring depositaries and other institutions to build up
capital more amply in favorable market conditions and thus
position themselves to weather unfavorable conditions more
easily.
Monitoring and Addressing Systemic Risk: Over the past
year, various policymakers and other commentators have called
for the establishment of a formal mechanism for identifying,
monitoring, and managing risks to the financial system as a
whole. ICI concurs with those commentators that creation of
such a mechanism is necessary. The ongoing financial crisis has
highlighted the vulnerability of our financial system to risks
that have the potential to spread rapidly throughout the system
and cause significant damage. A mechanism that will allow
Federal regulators to look across the system should equip them
to better anticipate and address such risks.
In its recent white paper on regulatory reform, ICI
endorsed the designation of a new or existing agency or inter-
agency body as a ``Systemic Risk Regulator.'' Broadly stated,
the goal in establishing a Systemic Risk Regulator should be to
provide greater overall stability to the financial system as a
whole. The Systemic Risk Regulator should have responsibility
for: (1) monitoring the financial markets broadly; (2)
analyzing changing conditions in domestic and overseas markets;
(3) evaluating the risks of practices as they evolve and
identifying those that are of such nature and extent that they
implicate the health of the financial system at large; and (4)
acting to mitigate such risks in coordination with other
responsible regulators.
In ICI's view, Congress should determine the composition
and authority of the Systemic Risk Regulator with two important
cautions in mind. First, the legislation establishing the
Systemic Risk Regulator should be crafted to avoid imposing
undue constraints or inapposite forms of regulation on normally
functioning elements of the financial system, or stifling
innovations, competition or efficiencies. Second, the Systemic
Risk Regulator should not be structured to simply add another
layer of bureaucracy or to displace the primary regulator(s)
responsible for capital markets, banking or insurance. Rather,
the Systemic Risk Regulator should focus principally on
protecting the financial system--as discussed in detail in our
white paper, we believe that a strong and independent Capital
Markets Regulator (or, until such agency is established by
Congress, the SEC) should focus principally on the equally
important mandates of protecting investors and maintaining
market integrity.
Legislation establishing the Systemic Risk Regulator should
define the nature of the relationship between this new
regulator and the primary regulator(s) for each industry
sector. This should involve carefully defining the extent of
the authority granted to the Systemic Risk Regulator, as well
as identifying circumstances under which the Systemic Risk
Regulator and primary regulator(s) should coordinate their
efforts and work together. We believe, for example, that the
primary regulators have a critical role to play by acting as
the first line of defense with regard to detecting potential
risks within their spheres of expertise.
We recognize that it may be appropriate, for example, to
lodge responsibility for ensuring effective consolidated global
supervision of the largest bank holding companies with a
designated regulator such as the Federal Reserve Board. Beyond
this context, however, and in view of the two cautions outlined
above, ICI believes that responsibility for systemic risk
management more broadly should be assigned to a Systemic Risk
Regulator structured as a statutory council comprised of senior
Federal regulators. Membership should include, at a minimum,
the Secretary of the Treasury, Chairman of the Federal Reserve
Board of Governors, and the heads of the Federal bank and
capital markets regulators (and insurance regulator, if one
emerges at the Federal level).
Q.2. Would a merger or rationalization of the roles of the SEC
and CFTC be a valuable reform, and how should that be
accomplished?
A.2. Establishment of a New Capital Markets Regulator: ICI
strongly believes that a merger or rationalization of the roles
of the Securities and Exchange Commission (SEC) and Commodity
Futures Trading Commission (CFTC) would be a valuable reform.
Currently, securities and futures are subject to separate
regulatory regimes under different Federal regulators. This
system reflects historical circumstances that have changed
significantly. As recently as the mid-1970s, for example,
agricultural products accounted for most of the total U.S.
futures exchange trading volume. By the late 1980s, a shift
from the predominance of agricultural products to financial
instruments and currencies was readily apparent in the volume
of trading on U.S. futures exchanges. In addition, as new,
innovative financial instruments were developed, the lines
between securities and futures often became blurred. The
existing, divided regulatory approach has resulted in
jurisdictional disputes, regulatory inefficiency, and gaps in
investor protection and market oversight. With the increasing
convergence of securities and futures products, markets, and
market participants, the current system simply makes no sense.
To bring a consistent policy focus to U.S. capital markets, ICI
strongly recommends the creation of a Capital Markets Regulator
as a new agency that would encompass the combined functions of
the SEC and those of the CFTC that are not agriculture-related.
As the Federal regulator responsible for overseeing all
financial investment products, it is imperative that the
Capital Markets Regulator--like the SEC and the CFTC--be
established by Congress as an independent agency, with an
express statutory mission and the rulemaking and enforcement
powers necessary to carry out that mission. A critical part of
that mission should be for the new agency to maintain a sharp
focus on investor protection and law enforcement. And Congress
should ensure that the agency is given the resources it needs
to fulfill its mission. Most notably, the Capital Markets
Regulator must have the ability to attract personnel with the
necessary market experience to fully grasp the complexities of
today's global marketplace.
To preserve regulatory efficiencies achieved under the
National Securities Markets Improvement Act of 1996, Congress
should affirm the role of the Capital Markets Regulator as the
regulatory standard setter for all registered investment
companies. ICI further envisions the Capital Markets Regulator
as the first line of defense with respect to identifying and
addressing risks across the capital markets. The new agency
should be granted explicit authority to regulate in certain
areas where there are currently gaps in regulation--in
particular, with regard to hedge funds, derivatives, and
municipal securities--and explicit authority to harmonize the
legal standards applicable to investment advisers and
brokerdealers. These areas are discussed in greater detail in
ICI's March 3, 2009, white paper, Financial Services Regulatory
Reform: Discussion and Recommendations. \3\
---------------------------------------------------------------------------
\3\ See Financial Services Regulatory Reform: Discussion and
Recommendations, which is available at http://www.ici.org/pdf/
ppr_09_reg_reform.pdf. We note that the white paper was included as an
attachment to ICI's written testimony.
---------------------------------------------------------------------------
Organization and Management of the Capital Markets
Regulator: In the private sector, a company's success is
directly related to the soundness of its management. The same
principle holds true for public sector entities. Establishing a
new agency presents a very valuable opportunity to ``get it
right'' as part of that process. There is also an opportunity
to make sound decisions up-front about how to organize the new
agency. In so doing, it is important not to simply use the
current structure of the SEC and/or the CFTC as a starting
point. The SEC's current organizational structure, for example,
largely took shape in the early 1970s and reflects the
operation of the securities markets of that day. Rather, the
objective should be to build an organization that not only is
more reflective of today's markets, market participants and
investment products, but also will be flexible enough to
regulate the markets and products of tomorrow.
ICI offers the following thoughts with regard to
organization and management of the Capital Markets Regulator:
LEnsure high-level focus on agency management. One
approach would be to designate a Chief Operating
Officer for this purpose.
LImplement a comprehensive process for setting
regulatory priorities and assessing progress. It may be
helpful to draw upon the experience of the United
Kingdom's Financial Services Authority, which seeks to
follow a methodical approach that includes developing a
detailed annual business plan establishing agency
priorities and then reporting annually the agency's
progress in meeting prescribed benchmarks.
LPromote open and effective lines of communication
among the regulator's Commissioners and between its
Commissioners and staff. Such communication is critical
to fostering awareness of issues and problems as they
arise, thus increasing the likelihood that the
regulator will be able to act promptly and effectively.
A range of approaches may be appropriate to consider in
meeting this goal, including whether sufficient
flexibility is provided under the Government in the
Sunshine Act, and whether the number of Commissioners
should be greater than the current number at the SEC
and at the CFTC (currently, each agency has five).
LAlign the inspections and examinations functions
and the policymaking divisions. This approach would
have the benefit of keeping staff in the policymaking
divisions updated on current market and industry
developments, as well as precluding any de facto
rulemaking by the regulator's inspections staff.
LDevelop mechanisms to facilitate coordination and
information sharing among the policymaking divisions.
These mechanisms would help to ensure that the
regulator speaks with one voice.
Process of Merging the SEC and CFTC: Legislation to merge
the SEC and CFTC should outline a process by which to harmonize
the very different regulatory philosophies of the two agencies,
as well as to rationalize their governing statutes and current
regulations. There is potential peril in leaving open-ended the
process of merging the two agencies. ICI accordingly recommends
that the legislation creating the Capital Markets Regulator set
forth a specific timetable, with periodic benchmarks and
accountability requirements, to ensure that the merger of the
SEC and CFTC is completed as expeditiously as possible.
The process of merging the two agencies will be lengthy,
complex, and have the potential to disrupt the functioning of
the SEC, CFTC, and their regulated industries. ICI suggests
that, in anticipation of the merger, the SEC and CFTC undertake
detailed consultation on all relevant issues and take all steps
possible toward greater harmonization of the agencies. This
work should be facilitated by the Memorandum of Understanding
the two agencies signed last year regarding coordination in
areas of common regulatory interest. \4\ ICI believes that its
recommendations with respect to the Capital Markets Regulator,
outlined in detail in its white paper, may provide a helpful
framework for these efforts.
---------------------------------------------------------------------------
\4\ See SEC, CFTC Sign Agreement to Enhance Coordination,
Facilitate Review of New Derivative Products (SEC press release dated
March 11, 2008), available at http://www.sec.gov/news/press/2008/2008-
40.htm
Q.3. How is it that AIG was able to take such large positions
that it became a threat to the entire Financial system? Was it
a failure of regulation, a failure of a product, a failure of
---------------------------------------------------------------------------
risk management, or some combination?
A.3. ICI does not have particular insight to offer with regard
to AIG, the size of its positions in credit default swaps
(CDS), and the effect that those positions ultimately had on
the broader financial markets. Nevertheless, our sense is that
the answers lie in a combination of all the factors outlined
above. We note that Congress seems poised to establish a
bipartisan commission to investigate the causes of the current
financial crisis. A thorough examination of what happened with
AIG would no doubt be a very useful part of the commission's
inquiry.
With regard to CDS generally, ICI believes that a single
independent Federal regulator for capital markets should have
clear authority to adopt measures to increase transparency and
reduce counterparty risk, while not unduly stifling innovation.
\5\ We support current initiatives toward centralized clearing
for CDS, which should help to reduce counterparty risk and
bring transparency to trading in the types of CDS that can be
standardized. Not all CDS are sufficiently standardized to be
centrally cleared, however, and institutional investors will
continue to need to conduct over-the counter transactions in
CDS. For those transactions, we support reasonable reporting
requirements, in order to ensure that regulators have enough
data on the CDS market to provide effective oversight. Finally,
we believe that all institutional market participants should be
required to periodically disclose their CDS positions publicly,
as funds are currently required to do.
---------------------------------------------------------------------------
\5\ In our March 3, 2009 white paper, Financial Services
Regulatory Reform: Discussion and Recommendations (which is available
at http://www.ici.org/pdf/ppr_09_reg_reform.pdf), ICI recommended the
creation of a Capital Markets Regulator as a new agency that would
encompass the combined functions of the SEC and those of the CFTC that
are not agriculture-related. To the extent that no Capital Markets
Regulator is formed, we believe that the SEC is the regulator best
suited to provide effective oversight of financial derivatives,
including CDS.
Q.4. How should we update our rules and guidelines to address
---------------------------------------------------------------------------
the potential failure of a systematically critical firm?
A.4. Experience during the financial crisis has prompted calls
to establish a better process for dealing with large,
diversified financial institutions whose solvency problems
could have significant adverse effects on the financial system
or the broader economy. Depository institutions already have in
place a resolution framework administered by the Federal
Deposit Insurance Corporation. In contrast, other
``systemically important'' financial institutions facing
insolvency either have to rely on financial assistance from the
government (as was the case with AIG) or file for bankruptcy
(as was the case with Lehman Brothers).
The Treasury Department has expressed concern that these
``options do not provide the government with the necessary
tools to manage the resolution of [a financial institution]
efficiently and effectively in a manner that limits systemic
risk with the least cost to the taxpayer.'' \6\ Treasury has
sent draft legislation to Congress that is designed to address
this concern. The legislation would authorize the FDIC to take
a variety of actions (including appointing itself as
conservator or receiver) with respect to a ``financial
company'' if the Treasury Secretary, in consultation with the
President and based on the written recommendation of the
Federal Reserve Board and the ``appropriate Federal regulatory
agency,'' makes a systemic risk determination concerning that
company.
---------------------------------------------------------------------------
\6\ See Treasury Proposes Legislation for Resolution Authority
(March 25, 2009), available at http://www.treas.gov/press/releases/
tg70.htm
---------------------------------------------------------------------------
ICI agrees that it would be helpful to establish rules
governing the resolution of certain large, diversified
financial institutions in order to minimize the impact of the
potential failure of such an institution on the financial
system and consumers as a whole. Such a resolution process
could benefit investors, including investment companies (and
their shareholders). The rules for a federally-facilitated wind
down should be clearly established so that creditors and other
market participants understand the process that will be
followed and the likely ramifications. Uncertainty associated
with ad hoc approaches that differ from one resolution to the
next will be very destabilizing to the financial markets. Clear
rules and a transparent process are critical to bolster
confidence and avoid potentially creating reluctance on the
part of market participants to transact with an institution
that is perceived to be ``systemically important.''
In determining which institutions might be subject to this
resolution process, we recommend taking into consideration not
simply ``size'' or the specific type of institution but
critical factors such as the nature and extent of an
institution's leverage and trading positions, the nature of its
borrowing relationships, the amount of difficult-to-value
assets on its books, its off-balance sheet liabilities, and the
degree to which it engages in activities that are opaque or
unregulated.
More broadly, the reforms recommended in ICI's recent white
paper, \7\ if enacted, would lead to better supervision of
systemically critical financial institutions and would help
avoid in the future the types of situations that have arisen in
the financial crisis, such as the failure or near failure of
systemically important firms. Our recommendations include:
---------------------------------------------------------------------------
\7\ See Financial Services Regulatory Reform: Discussion and
Recommendations, which is available at http://www.ici.org/pdf/
ppr_09_reg_reform.pdf. We note that the white paper was included as an
attachment to ICI's written testimony.
LEstablishing a ``Systemic Risk Regulator'' that
would identify, monitor and manage risks to the
---------------------------------------------------------------------------
financial system as a whole;