[Senate Hearing 110-982]
[From the U.S. Government Publishing Office]
S. Hrg. 110-982
TURMOIL IN U.S. CREDIT MARKETS: THE ROLE OF CREDIT RATING AGENCIES
=======================================================================
HEARING
before the
COMMITTEE ON
BANKING,HOUSING,AND URBAN AFFAIRS
UNITED STATES SENATE
ONE HUNDRED TENTH CONGRESS
SECOND SESSION
ON
ISSUES INVOLVING THE RATING OF STRUCTURED FINANCE INSTRUMENTS BY THE
NATIONALLY RECOGNIZED STATISTICAL RATING ORGANIZATIONS (NRSROS), AS
WELL AS RECENT INITIATIVES THAT THE NRSROS HAVE ADOPTED AND
RECOMMENDATIONS FOR LEGISLATIVE, REGULATORY AND VOLUNTARY CHANGES TO
IMPROVE THE CREDIT RATING PROCESS
__________
TUESDAY, APRIL 22, 2008
__________
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
50-399 WASHINGTON : 2010
-----------------------------------------------------------------------
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 WAYNE ALLARD, Colorado
EVAN BAYH, Indiana MICHAEL B. ENZI, Wyoming
THOMAS R. CARPER, Delaware CHUCK HAGEL, Nebraska
ROBERT MENENDEZ, New Jersey JIM BUNNING, Kentucky
DANIEL K. AKAKA, Hawaii MIKE CRAPO, Idaho
SHERROD BROWN, Ohio ELIZABETH DOLE, North Carolina
ROBERT P. CASEY, Pennsylvania MEL MARTINEZ, Florida
JON TESTER, Montana BOB CORKER, Tennessee
Shawn Maher, Staff Director
William D. Duhnke, Republican Staff Director and Counsel
Dean V. Shahinian, Counsel
Roger M. Hollingsworth, Professional Staff Member
Didem Nisanci, Professional Staff Member
David Stoopler, Professional Staff Member
Jayme Roth, Professional Staff Member
Brian Filipowich, Legislative Assistant
Megan Bartley, Legislative Assistant
Jason Rosenberg, Legislative Assistant
Mark Osterle, Republican Counsel
Andrew Olmem, Republican Counsel
Tewana Wilkerson, Republican Professional Staff Member
Courtney Geduldig, Republican Legislative Assistant
Dawn Ratliff, Chief Clerk
Shelvin Simmons, IT Director
Jim Crowell, Editor
C O N T E N T S
----------
TUESDAY, APRIL 22, 2008
Page
Opening statement of Chairman Dodd............................... 1
Opening statements, comments, or prepared statements of:
Senator Shelby............................................... 3
Senator Menendez............................................. 4
Senator Reed................................................. 6
Senator Allard............................................... 6
Senator Schumer.............................................. 7
Senator Tester............................................... 8
WITNESSES
Christopher Cox, Chairman, Securities and Exchange Commission.... 10
Prepared statement........................................... 61
Response to written questions of:
Chairman Dodd............................................ 141
Senator Shelby........................................... 142
Senator Menendez......................................... 146
Senator Bunning.......................................... 149
John C. Coffee, Jr., Adolf A. Berle Professor of Law, Columbia
University Law School.......................................... 34
Prepared statement........................................... 69
Response to written questions of:
Senator Shelby........................................... 150
Vickie A. Tillman, Executive Vice President for Credit Market
Services, Standard & Poor's.................................... 37
Prepared statement........................................... 89
Response to written questions of:
Chairman Dodd............................................ 152
Senator Shelby........................................... 155
Senator Menendez......................................... 161
Claire Robinson, Senior Managing Director, Moody's Investors
Service........................................................ 39
Prepared statement........................................... 106
Response to written questions of:
Chairman Dodd............................................ 186
Senator Shelby........................................... 196
Senator Menendez......................................... 210
Stephen W. Joynt, President and Chief Executive Officer, Fitch
Ratings........................................................ 41
Prepared statement........................................... 119
Response to written questions of:
Chairman Dodd............................................ 216
Senator Shelby........................................... 220
Senator Menendez......................................... 225
Arturo Cifuentes, Ph.D., Managing Director, R.W. Pressprich & Co. 43
Prepared statement........................................... 125
Response to written questions of:
Senator Shelby........................................... 240
Additional Material Supplied for the Record
Aaron Lucchetti, The Wall Street Journal, ``As housing boomed,
Moody's opened up,'' article dated April 11, 2008.............. 242
TURMOIL IN U.S. CREDIT MARKETS: THE ROLE OF CREDIT RATING AGENCIES
----------
TUESDAY, APRIL 22, 2008
U.S. Senate,
Committee on Banking, Housing, and Urban Affairs,
Washington, DC.
The Committee met at 10:12 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, and my
apologies to my colleagues and others for being a few minutes
late this morning. I was taking the shuttle down, which is
always a bit of roll of the dice. So I apologize for being a
few minutes late, but I want to thank all in attendance for
being here this morning. Let me share some opening thoughts. I
will turn to Senator Shelby and then to other Members of the
Committee who would like to make, if they so desire, some
opening comments on the subject matter of today's hearing.
Today we are going to talk about the role played by the
credit rating agencies in the subprime mortgage crisis. I asked
the staff a short time ago, just to go back over this, this is
our 13th hearing this year on this subject matter or related
matters to it. We had 35 hearings last year. So it is almost
48, close to 50 hearings since beginning February 7th of last
year. Some of those hearings were conducted by my colleagues
here. I want to thank Jack Reed particularly for doing some of
this last year--in fact, on this very subject matter. And
Senator Shelby, of course, has been deeply involved in these
issues, and we owe him a debt of gratitude for what he has
done. But the Committee has spent an inordinate amount of time
over the last year on this subject matter. Including even when
we had hearings on student loan issues the other day, it was
really related in many ways to the subprime problem. So almost
every other matter we are looking at bears some relevancy to
the subject matter here today. We could have a hearing on
credit rating agencies, but obviously in the context of the
subprime mortgage crisis, it has real relevancy.
Senator Reed, as I mentioned a minute ago, chaired a
hearing of the full Committee on this subject matter, and
Senator Shelby, of course, has been deeply involved in the
subject matter of credit rating agency reform. In fact, during
his tenure or stewardship as Chairman of this Committee, he not
only held hearings on the topic of the credit rating agencies,
but, in addition, the Committee passed legislation. That
legislation, the Credit Rating Agency Reform Act of 2006, was
signed into law on September 29, 2006. It makes important
reforms in the area of capital market reforms, which in my view
were prescient.
Credit rating agencies played a very important role in our
economy and continue to do so. They provide opinions to
investors about the ability of debt issuers to make timely
payments on debt instruments. That may sound like a simple
modest function, but it is an indispensable one. Decisions
about how to invest enormous sums of money are based, at least
in part, on credit ratings. As one commentator has said,
``Credit rating agencies can, with the stroke of a pen,
effectively add or subtract millions from a company's bottom
line, rattle a city budget, shock the stock and bond markets,
and reroute international investment.''
We have seen over the past few months just how influential
a role credit rating agencies play in our markets, and
particularly in the structured finance markets, and not in a
positive sense. Credit rating agencies have played a central
role in the subprime mortgage crisis and, by extension, on the
volatility and illiquidity plaguing our capital markets.
During the past several months, these agencies, which are
technically referred to as nationally recognized statistical
rating organizations, have downgraded their ratings of
thousands of tranches of residential mortgage-backed
securities. Bloomberg recently reported that the three largest
of these organizations began cutting in July and have since
either downgraded or put on review a total of 38,000 subprime
bonds. Moody's and S&P combined have downgraded more than 9,500
of these securities dating from 2005. These downgrades meant
that, with the stroke of a pen, again, assets once seen as safe
and profitable were suddenly something quite the opposite.
Many investors who by Federal or State law must invest in
securities within investment grade ratings were suddenly forced
to sell. Others suddenly found the value of their securities
reduced to a fraction of their previous value. The net result
is that investors have lost tens of billions of dollars.
The impact of these downgrades has spread beyond the
downgraded bonds themselves. Imagine, if you will, using this
analogy, going to a grocery store to buy food for your family.
You are told that almost all of the food in the store is safe
and healthy, but that a small fraction--a small fraction--of
the items contained a very toxic substance that could cause
serious illness or death. It is doubtful that you or anyone
else is going to be doing much shopping in that store without
some assurance that it is free from the taint of any toxic
substance.
In the same manner, the downgrading of some subprime
mortgage securities have sown doubt and fear in investors about
a much larger universe of securities. It has cut investors'
appetite for subprime mortgage securities generally and for a
host of other asset-backed securities. As a result, our credit
markets are experiencing unprecedented levels of volatility and
illiquidity.
These recent rating downgrades have raised serious
questions about the role, function, and performance of credit
rating agencies. For instance, do the credit rating agencies
give ratings that are overly optimistic in order to obtain more
business? Do they sufficiently analyze the data they are given
by clients before issuing ratings? Do they properly manage real
or perceived conflicts of interest with clients who pay for
rating and/or consulting services? And, last, when Congress
acted 2 years ago, it gave the SEC the authority ``to prohibit
or require the management and disclosure of any conflicts of
interest.'' Has the SEC used this authority effectively? Can or
should it do more?
These are some of the important questions that our
witnesses will address this morning. The investing public, of
course, deserves to know that every step is being taken to
protect one of their most basic rights, and that is the right
to sound, reliable, credible information. They deserve to know
that our regulatory agencies will apply and enforce the law
with vigor on their behalf. And they want to see the credit
rating agencies demonstrate that they have learned from their
mistakes and have reformed their practices so that this very
sorry chapter in their history will never be repeated.
I want to welcome Chairman Cox of the SEC to the Committee
once again. We know he is currently working to implement by
rulemaking the new act, and we look forward obviously to
hearing his testimony this morning. Let me also welcome our
other distinguished witnesses who will be here this morning. We
appreciate their willingness to appear before us.
Let me now turn to Senator Shelby and then to other Members
of the Committee for any comments they may have about this
very, very important subject matter.
OPENING STATEMENT OF SENATOR RICHARD C. SHELBY
Senator Shelby. Thank you, Mr. Chairman. Welcome, Chairman
Cox.
Since our last hearing on this subject, the situation in
our financial markets has underscored the role played by the
rating agencies. The past few months have also demonstrated
that the rating agencies were not meeting their
responsibilities. We have witnessed this series of ratings
downgrades particularly in structured finance. It seems that
rating agencies grossly underestimated the risks associated
with these securities. Unfortunately, these products were
widely distributed and held by a broad array of investors and
institutions. The severity of these downgrades sent banks,
pension, and money market funds scrambling for capital.
Plunging investor confidence ultimately led credit markets to
tank worldwide.
The markets for commercial paper, municipal securities, and
auction rate securities have all experienced disruptions in
part because financial institutions no longer trust the credit
ratings of issuers, bond insurers, and other counterparties.
Rather than conduct their own due diligence, too many investors
appear to have relied solely on credit ratings to assess credit
risk. And while credit ratings play and should continue to play
an important part in evaluating risk in our economy, over
reliance on the ratings of just a few firms appears to have
diminished the amount of independent risk assessment undertaken
by market participants.
Because rating agencies underestimated the risk of
subprime-based securities, these securities were allowed to
spread throughout our financial system without their real risk
being detected until it was too late. In our modern economy, we
need not only better ratings but also more market participants
assessing risk to prevent this from happening again.
Before the current crisis began, this Committee worked and
enacted the Credit Rating Agency Reform Act of 2006. This act
sought to improve the quality of ratings and to foster
accountability, transparency, and competition in the industry.
The Securities and Exchange Commission was given broad
authority to enforce this act. Last year, the SEC issued
initial rules governing registration of NRSROs and prohibiting
certain conflicts of interest. These rules have opened up the
process for new firms to become NRSROs, fostering more
competition in the industry. The SEC is now preparing to
propose additional rules to implement the act.
Today, we look forward to hearing Chairman Cox discuss the
types of rules the SEC is considering adopting and what
additional reforms he believes are needed. I believe the SEC
has a chance to help restore confidence in our markets and
establish a more competitive and accountable credit rating
industry. For example, rules that improve the transparency of
the ratings process will make it easier for investors to assess
and compare ratings.
I am also interested in the preliminary finding of the
SEC's ongoing examination of the rating agencies, and we would
like to learn more about the relationship between the agencies
and investment banks. A rating, after all, is only as good as
the information on which it is based.
If there was insufficient due diligence and risk assessment
in the process of creating and underwriting structured
financial products, the ratings will be flawed from their
inception. We found that they were.
Mr. Chairman, given the critical role underwriters played
in this crisis, I hope that our examination--I believe our
examination is incomplete without the participation of the
firms that created these products, and I hope that we will
address those two in the future. The sophisticated underwriters
that structured and sold these securities reaped huge fees for
their efforts, regardless of how the securities performed for
investors. I hope their absence from this discussion is not
permanent, Mr. Chairman.
Chairman Dodd. Well, thank you very much, Senator Shelby.
Just on that point of the investment banks, we had a hearing
last year--in fact, Senator Reed looked into that.
Senator Shelby. We did.
Chairman Dodd. And I am certainly willing to hold an
additional one. As I mentioned, we have had a lot of hearings
on the subject matter, but certainly that is a very legitimate
question that you raise, Senator Shelby, and we will do that.
Senator Reed.
Senator Reed. Mr. Chairman, I will yield to Senator
Menendez. He has----
Chairman Dodd. Fine. Absolutely.
OPENING STATEMENT OF SENATOR ROBERT MENENDEZ
Senator Menendez. Thank you. Let me thank both Senator Reed
and Senator Schumer for their courtesy. I have to chair a
Foreign Relations Subcommittee hearing at 10:30, and I hope to
get back for our second panel, Mr. Chairman. So I appreciate
them both for their courtesy. This is something I have been
following along with the Committee and am very interested in.
Over the last year, we have grappled with a foreclosure
crisis that has swept across our country, devastating families
and neighborhoods and a credit crunch that has spread
throughout our markets with ripple effects throughout our
economy. Within the turmoil, there are many pieces for us to
focus on as we seek to help homeowners, stem any further
spillover into other markets, and work to stabilize our
economy. But the worse mistake I think we could make is not to
learn from what happened and to let the cracks in the system
slip by unfixed. Our credit rating system threatens to possibly
be something that slipped by, and I am glad through your
leadership and the Ranking Member's that we are not letting
that happen.
Last year, we held what I thought was a very important
hearing to examine one of the most severe and overlooked cracks
in the mortgage and the securitization chain. While the credit
rating agencies were not a direct cause of the subprime crisis,
they certainly were a key link in the securitization chain and
had a hand in perpetuating a mortgage process in which no one
asked the right questions. That chain failed in large part
because the very ratings that the market was supposed to rely
on were flawed. And often I think they played the conflicting
roles of referee and coach.
Recently, we witnessed what happens when the whole system
fails. Extenuating circumstances or not, our regulatory system
did not know what hit it when Bear Stearns collapsed. In
addition to the questions I and many of my colleagues have had
about how our regulators missed the warning signs, I have
serious questions about the role that the ratings played or
could have played in helping raise flags earlier. The fact is
credit ratings play an essential role for our markets. Issuers
depend upon them to seek investments. Investors depend upon
them to know the creditworthiness of the investments they are
making. The system as a whole depends upon them to track risk.
But the question is: Who is rating the rating agencies? And
that answer has been clear: No one.
So I want to applaud the SEC for taking seriously the need
to reform this process. I have raised some questions with the
Chairman when he came to visit--I appreciate his visit--of
whether some of the SEC plans go far enough, and I hope we can
find solutions that will increase disclosure and root out the
practices that keep the ratings from being what they should be:
fair, simple, and accurate.
Finally, I hope, Mr. Chairman, we look at the bigger
picture for a moment. This discussion about how to reform the
rating system is largely cleaning up the mess. We are still
mopping up the aisles and trying to figure out what broke and
why. But beyond this, as I spoke to Chairman Cox when he
visited me--and, again, I appreciate that visit--the larger
challenge at hand is getting ahead of the curve. The problem is
not just that the ratings were flawed or that there are
conflicts in the system. It is that what is going on in the
market and on the street is light years ahead often of what is
going on in our regulatory system. How can our regulators watch
for the warning signs and respond if they do not even know what
the signals are? I feel like they are in the same struggle as
parents who cannot keep up with their teenage kids texting back
and forth on their cell phones.
The fact is much of our market operations are taking place
in a language all its own, and we need our regulators to be
fluent in that language as well. And I am looking forward to
the Chairman's proposals in this particular regard and the
Commission's proposals, and I am hoping that we will have a
system that puts us ahead of the curve.
Thank you, Mr. Chairman.
Chairman Dodd. Thank you very much.
Senator Corker.
[No response.]
Chairman Dodd. Senator Reed.
OPENING STATEMENT OF SENATOR JACK REED
Senator Reed. Well, Mr. Chairman, thank you very much, and
I will make some comments.
First, welcome, Chairman Cox. We have been down this road
before. In the wake of Enron, we saw flaws in the credit rating
system. We have tried to address those faults. I want to
commend Senator Shelby for his efforts as Chair last year and
at least giving the SEC some authority and some traction in
this regard. But I think what we have seen in the last 12
months has been another indication that we have to take more
directed action.
Twelve months ago, when at your request I chaired a
hearing, the subprime crisis was seen as a $19 billion
worldwide phenomenon that was already self-correcting. That is
not the case, and so I think we have to do much more. We have
to ensure that the Commission has the authority to adequately
supervise, regulate, or direct the credit rating agencies. We
have to ensure, I think, that the new rules that they are
promulgating really do the job. As I said, we have been down
this road before, and we are still going down it. I think we
want to reach an appropriate conclusion.
We have to, I think, ensure that we have the appropriate
balance between market discipline and good rules and
regulations. That is something, I think, that is out of balance
at this moment.
I will conclude with the comments of Lew Ranieri, who
created the mortgage-backed security years ago, when he said,
``The mortgage-backed security sector was unfettered in its
enthusiasm and unchecked by today's regulatory framework. We
have a quasi-gatekeeper in the rating services, and in the end
the SEC is the regulator of the capital market. It is the one
who can touch this stuff and make a difference.'' And I think
we have to touch this stuff and make a difference now since we
have not in the past.
Chairman Dodd. Thank you very much, Senator. And, again,
for the purpose of the record, all statements, complete
statements of Members and witnesses, will be included in the
record as well.
Senator Allard.
STATEMENT OF SENATOR WAYNE ALLARD
Senator Allard. Mr. Chairman, thank you for holding this
hearing, and also Ranking Member Shelby. I would just make a
few brief comments.
The nationally recognized statistical rating organizations
play an important role in financial markets. Confidence in
these ratings have been shaken following a number of downgrades
of residential mortgage-backed securities. And so this lack of
confidence is of concern to me. I have said this to a lot of
people, I believe. And I just would remind us of a quote from
former Federal Reserve Chairman Alan Greenspan when he said
that people believe that they--meaning the credit rating
agencies--``knew what they were doing, and they don't. What
kept them in place was a belief on the part of those who
invested in that that they were properly priced.''
Now everyone knows that they weren't, and they know that
they can't really be properly priced. And I am anxious to hear
what Chairman Cox might have to say about that particular
statement.
As always, I would like to welcome my friend and former
colleague from the House. It is always good to see you, Mr.
Chairman.
Thank you, Mr. Chairman.
Chairman Dodd. Thank you very much, Senator.
Senator Schumer.
STATEMENT OF SENATOR CHARLES E. SCHUMER
Senator Schumer. Well, thank you, Mr. Chairman. I very much
appreciate the opportunity to have this hearing and very much
appreciate your being here. And I think it is appropriate
because, at least to me, credit rating agencies were the weak
link in the subprime crisis. They, along with mortgage brokers,
are probably more at the center of this than just about anybody
else. And, incidentally, at least until we passed our
legislation--and much of the action occurred before that--
neither the mortgage brokers nor the credit rating agencies had
any real regulation at all. And so it is difficult to ask the
SEC--they now have regulation, and we have met and talked
about--have the ability to look at things like conflict of
interest, but they did not back then. So, to me, the credit
rating agencies are at the heart of this problem, and we need
to do a thorough examination of what is happening. That is why
I appreciate you, Mr. Chairman, and the Ranking Member being so
interested in this issue, which he was when he was Chairman as
well, as well as Senator Reed.
Second, I really regret that the heads of--I want to
commend Fitch's for sending their CEO, but where are the heads
of Moody's and Standard & Poor's? The bottom line, this is
really serious stuff. The whole world is focused on this. And
for the CEOs not to come is very disappointing. They should be
here. And particularly they should be here because I met with
the President of Moody's a while ago, and I asked him, Did
Moody's do anything wrong? And he said no. I would like to know
if he still believes that. He said no, they did nothing wrong.
I was incredulous.
And so, again, I think the credit rating agencies really
have an obligation to send their leaders and to find out what
happened and what is going on here. And I want to register my
disappointment.
Third, to me, the nub of this problem is conflict of
interest. Obviously, when you are paying for a rating, there is
an inherent conflict of interest, and that has to change. And
there was a story in The Wall Street Journal--which I would
just ask unanimous consent to put into the record.
Chairman Dodd. It will be included.
Senator Schumer. It is an article from April 11th that just
documented one instance of a conflict where analysts were
changed because people did not like the rating agency. Here is
a quote from the article: ``On occasion, Moody's agreed to
switch analysts on deals after bankers complained.'' And
another quote: ``There was, rather, a palpable erosion of
institutional support for rating analysis that threatened
market share.'' Moody's decided they would increase market
share in this area, and their standards declined at the same
time.
Conflict is inherent sometimes, and, look, sometimes there
are legitimate reasons to complain: you did not take this into
account; there has to be a dialog between the agency and the
issuer. But disclosure is key, and I asked you, Mr. Chairman,
when we met, would you make sure that this is all disclosed
when an analysis was changed after a complaint or if a rater
was switched? That should be known. Again, you cannot say that
the issuer can never complain. Maybe they missed something. But
at least disclosure would be a prophylactic. And the new
legislation that we supported and Chairman Shelby shepherded
through this Congress allows for that disclosure, and we
eagerly await the regulations that you will have.
One final point I would make here. For somebody to say
nothing is wrong, here is the nub of it: How did no-doc loans,
loans with no documentation that were parts of these packages,
get AAA ratings? Now, when you ask the credit rating agencies
how did no-doc loans deserve AAA ratings, they said, well--not
them but the people analyzing them. They say, well, they
thought housing would go up no matter what. And so, therefore,
it did not matter if the guy could not repay, so you did not
have to look at the loan.
Well, maybe they should have paid one of us. We could have
told them housing prices would go up forever. We did not need
to do any analysis either, or somebody, or the guy on the
street.
So something is really wrong here. Something is really
wrong. I know some of it has been self-corrected already, but
there has to be more to be done, and this hearing is a very
constructive step along that path. And I thank you for holding
it, Mr. Chairman.
Chairman Dodd. Thank you very much, Senator.
We invited the CEO. Today is their shareholder meeting, and
so he--though we could maybe schedule it another time. I did
not know that at the time, and he let us know he would have
been here but for presiding over the shareholder meeting.
Moody's, anyway, I want to include that in the record.
Senator Schumer. Well, I would like an opportunity for them
maybe to come back at some point if we have time, either at the
Committee or the Subcommittee level.
Chairman Dodd. Very good point.
Senator Tester.
OPENING STATEMENT OF SENATOR JON TESTER
Senator Tester. Thank you, Chairman Dodd and Ranking Member
Shelby, for calling today's hearing on credit rating agencies
as another in a series of hearings looking into the turmoil in
the credit markets. I want to welcome Chairman Cox and the
members of the second panel.
My stay here today is going to be limited because I have to
chair on the floor, but this is a topic that is both timely and
critical as issues that surround the credit rating agencies and
their role in the current credit market crisis keep arising. In
Montana, we are confronted with uncertainties in the student
loan market, as the auction rate bond market is no longer
viable, due in no small part, some say, to mistrust in the
credit rating agencies.
As important as it is to delve into the oversight of the
credit rating agencies, I really want to spend my opening
statement today addressing the distinguished witness Chairman
Cox on the possibility that market manipulation led to the fall
of Bear Stearns leading up to its merger with JPMorgan Chase.
Mr. Chairman, you testified before this panel on April 3rd,
along with Chairman Bernanke and other distinguished panelists,
to discuss recent actions of Federal financial regulators as it
related to the Government's role in the Bear Stearns saga. At
the time, I inquired if there is any evidence suggesting that
speculators had bet heavily that Bear Stearns' share price
would fall, known on Wall Street as ``short selling.'' You
responded, and I quote, ``I am a little bit constrained because
the SEC is in the law enforcement business.'' You then
continued to say that the SEC pursues insider trading
aggressively and that your agency was mulling several law
enforcement matters that have not been filed in any U.S. court.
A week after, on April 10th, I sent a letter to you and to
Attorney General Mukasey asking you to immediately and
thoroughly investigate whether illegal insider trading led to
last month's downfall of Bear Stearns. To date, I have not
heard back from your office, nor have I heard back from the
Department of Justice. I understand your response is currently
being drafted and will likely echo the sentiments that you told
me on April 3rd, that you were in the law enforcement business
and cannot confirm nor deny, but you will investigate if any
wrongdoing has taken place.
While I respect that, and I admire the SEC for playing an
integral role in the investigations of securities law
violations, I want you to know that this is not an ordinary
situation, and the events that followed what some view as
market manipulation were unprecedented--a $29 billion loan from
the Government to facilitate a merger of two of the world's
largest banks.
I am not sitting here to criticize the Federal Reserve Bank
of New York for their actions if risking nearly $30 billion of
taxpayer dollars with a limited amount of due diligence was
necessary, but I do want to know if it could have been avoided,
if speculators conducted insider trading to make a buck, a
whole lot of bucks, which led to taxpayers being forced to
stand behind the loan that is big even by Washington, D.C.,
standards, much less the standards of my home State of Montana.
As I stated earlier, I will have to leave very shortly to
go preside on the floor, but we will continue to have a
dialogue. You will continue to hear from me in the coming
months on the need for a thorough investigation. Hopefully that
is going on as we speak, and hopefully it will get to the
bottom of the situation. I have asked other financial
regulators, investors, and knowledgeable individuals their
thoughts, and to a person, they believe fear and speculation
alone did not eat up Bear's significant liquidity position. But
I want to hear it officially from you.
So thank you, Chairman Cox. Thank you, Chairman Dodd.
Chairman Dodd. Thank you very much, Senator.
Senator Bayh has joined us. Senator, do you have any
opening comments you want to make?
Senator Bayh. Thank you, Mr. Chairman. No. I am looking
forward to hearing from Chairman Cox, and I did want to note
my--this takes me back a few years, Mr. Chairman. My corporate
law professor in law school, John Coffee, is here, and I just
wanted to give him my best regards.
Chairman Dodd. Now we are going to really have an
interesting hearing.
Senator Bayh. And for both our sakes, I hope he will not
disclose what my grade in the course was.
Chairman Dodd. I tell you, we expect very tough questioning
from you, though, Senator, of the witness.
Chairman Cox, welcome to the Committee once again. You have
been before the Committee on numerous occasions over the last
year, and we appreciate your being back here today.
Mr. Cox. Thank you. Senator Tester I notice is just
leaving, but----
[Laughter.]
Mr. Cox. Just on your way out, I think you recognize that
both the Department of Justice and the SEC do not confirm
investigations into people for privacy reasons before they have
been publicly identified with wrongdoing. But I also stated at
that hearing that the problem with Bear Stearns was too big to
miss and people should take comfort that the SEC was doing its
job in this area. So I hope to signal by that within the silent
forum that we all must operate in the law enforcement agency
business that that is the case. And beyond that, I look forward
to speaking with you in private to give you the maximum amount
of comfort in that respect.
Senator Tester. You took the words right out of my mouth,
Chairman Cox. I would like to set up a meeting with you, and we
can visit about the issue in private. That would be great.
Thank you.
Mr. Cox. Thank you.
Chairman Dodd. Thanks very much. Mr. Chairman, we look
forward to your statement.
STATEMENT OF CHRISTOPHER COX, CHAIRMAN, SECURITIES AND EXCHANGE
COMMISSION
Mr. Cox. Thank you, Chairman Dodd, Senator Shelby, Members
of the Committee, for inviting me today to discuss the work of
the SEC concerning credit rating agencies.
When Congress passed the Credit Rating Agency Reform Act
and President Bush signed it into law in late 2006, its purpose
was to improve ratings quality by fostering accountability,
transparency, and competition in the credit rating industry.
Prior to the Rating Agency Act, credit rating agencies were
essentially unregulated by the Federal Government, and the SEC
had no authority to make rules governing their business or to
subject them to examinations as nationally recognized
statistical rating organizations. With the passage of the act,
the Commission became their regulator, and since that time, we
have devoted considerable new resources to this responsibility.
Since the end of September 2007, seven credit rating
agencies, including those that were most active in rating
subprime-related products, have been subject to the
Commission's new oversight authority, and subject as well to
our newly adopted rules. In the 6\1/2\ months since the SEC's
authority over CRAs went into effect, the Commission has
aggressively used its authority to examine the adequacy of
their public disclosures, their recordkeeping, and their
procedures to prevent the disclosure of material non-public
information.
The review process has included hundreds of thousands of
pages of the rating agencies' internal records and e-mail. In
addition, the staff are reviewing the ratings agencies' public
disclosures relating to the ratings process for those
securities, and Commission staff have analyzed the ratings
history of thousands of structured finance products. These
extensive examinations have involved approximately 40 SEC
professional staff.
Much has been accomplished already on these examinations,
but there is still much more work to be done. The Commission
expects that the report describing the staff's observations
from the examinations will be issued by early summer. At this
stage, with more examination work to be completed and the
staff's across-the-board inferences yet to be drawn, it is
premature to describe the results. I can say that it appears
the volume of the structured finance deals that were brought to
the credit rating agencies increased substantially from 2004 to
2006, and at the same time, the structured products that the
rating agencies were being asked to evaluate were becoming
increasingly complex, with many employing derivatives such as
credit default swaps to replicate the performance of mortgage-
backed securities.
Meanwhile, the loan assets underlying these securities
shifted from primarily plain vanilla 30-year mortgages to a
range of more difficult-to-assess products, including ARMs and
second-lien loans. We are currently evaluating whether and how
the credit rating agencies adapted their ratings approaches in
this rapidly changing environment. We expect the results of
these staff examinations will provide significant and useful
new information that will help not only the SEC but also
issuers and users of credit ratings to address the problems
that we have seen.
Because the Commission's authority over credit rating
agencies took effect just over 6 months ago, the SEC is already
far along in preparing for a second round of rulemaking. This
second round of rulemaking will be based on information
provided by the staff's ongoing examinations of these firms as
well as the many empirical analyses provided by regulators and
industry groups, academics, and multinational organizations,
including many in which the SEC itself has participated. I
expect the Commission will issue rule proposals for public
comment in the near future. Of course, the internal development
process for these rules within the Commission is still very
much ongoing. So while I am happy to provide you today with an
outline of the rulemaking areas that are under consideration, I
do so with the caveat that what ultimately is included in these
new proposed rules has yet to be decided.
That said, the rules that we are likely to consider will
fall into three broad categories: rules designed to foster
accountability, rules to enhance transparency, and rules to
promote competition in the credit rating agency industry.
These, of course, are the three goals of the Rating Agency Act
itself.
To strengthen accountability, the new rules may include
requirements for enhanced disclosures about ratings'
performance. This would enable market participants to better
compare one NRSRO with another. To ensure NRSRO accountability
for the management of their conflicts of interest, the new
rules could include specific prohibitions on certain practices.
They could also establish requirements to address potential
conflicts of interest that could impair the process for rating-
structured products.
Among the conflicts of interest that could be addressed are
the provision of consulting services by credit rating agencies
to the issuers of the securities that they rate and the rating
of structured securities that the credit rating agency itself
helped to design. The proposed rules may also include
requirements that the firms furnish the Commission with annual
reports describing their internal reviews of how well they
adhere to their own procedures for determining ratings,
managing conflicts of interest, and complying with the
securities laws.
In the second category of enhancing transparency, the
Commission may consider rules to require the disclosure of
information underlying the ratings of subprime-related
products, including, for example, the particular assets backing
MBS, CDOs, and other types of structured products. This would
allow market participants to better analyze the assets
underlying the structured securities and reach their own
conclusions about creditworthiness.
Making this data available to the market could particularly
benefit subscriber-based NRSROs who could use it to perform
independent assessments of the validity of their competitor's
ratings. Other improvements that the new rules could make in
the area of transparency could come from enhanced disclosure
about how NRSROs determine their ratings for structured
products. This new disclosure could include, for example, the
kind of analysis that is done on the degree to which the
mortgages behind asset-backed securities conform with
underwriting standards. Additional disclosure could be required
as well for the firms' procedures for monitoring their current
credit ratings. The Commission may also consider rules to help
investors to readily distinguish the ratings for different
types of securities such as structured products, corporate
securities, and municipal securities.
In the third area of potential rulemaking, promoting
competition, the new rules could include provisions to enhance
disclosure about ratings' performance so that it affords other
credit rating agencies, including newly recognized NRSROs, an
opportunity to identify flaws in their competitive approach or
to demonstrate to investors that their ratings performed
better. The Commission is also reconsidering its extensive
reliance on credit ratings in our own rules. Limiting the use
of credit ratings for regulatory compliance purposes could
encourage investors in the marketplace as a whole to use
ratings for their informational value rather than merely to
satisfy a regulatory requirement. This could induce greater
competition among rating agencies to produce the highest-
quality, most reliable ratings.
Yet another way the new rules might seek to enhance
competition could be to ensure that all NRSROs have access to
the same information underlying a credit rating. In that way,
regardless of whether the NRSRO follows the issuer-pays
approach or the subscriber-based approach, there would be no
competitive advantage or disadvantage based on access to
information on the assets underlying a structured credit
product. And at the same time, NRSROs that were not paid by the
issuers to rate securities could develop their own track record
for rating these products.
To the extent both the issuer-pays and the subscriber-based
models were to flourish in a competitive marketplace, each
could act as a healthy competitive check on the other. Of
course, because this planned proposed rulemaking is ongoing,
there could and undoubtedly will be other subjects covered in
the draft rules that the staff will present to the Commission
for its consideration.
In closing, Mr. Chairman, I want to emphasize that the
Commission is very much open to ideas from the Congress on this
proposed rulemaking, and we especially welcome ideas from this
Committee since you are the authors of the Credit Rating Agency
Act and it is your intent in writing the law that the
Commission is now working to fulfill.
I appreciate this opportunity to provide the Committee with
this update on the SEC's new regulatory responsibilities for
credit rating agencies, and I look forward to answering your
questions.
Chairman Dodd. Well, thank you very much, Mr. Chairman, and
I want to ask the clerk to put up about 7 minutes per Member
here so that we can give everyone a good chance to raise some
issues with you.
First, I was pleased to hear your plans to require greater
clarity of methodologies and to make issuer data available to
all NRSROs and to propose needed reforms. Let me ask you a
couple of sort of underlying questions, and then there is a
series of specific ideas that have been raised, including some
of our witnesses who submitted their testimony and will be
before us a little later this morning.
I guess one question we would have for you, all of us would
up here, putting aside the various ideas, do you need any
additional statutory authority, do you think, for the SEC to
act? And if so, would you share with this Committee what
limitations you have in that regard and what recommendations
you would make if, in fact, there is a gap in terms of what you
think you need to do and the authority that you have been given
either by the legislation we adopted or previous legislation?
Mr. Cox. Mr. Chairman, thank you for the question and for
the opportunity. In connection with this latest proposed round
of rulemaking, we have come upon a number of topics where we
had to ask ourselves, Do we have the authority aggressively to
do this? And thus far, the answers that we have been able to
give are all yes. We do have the authority, not only in the
Rating Agencies Act, but also under the Exchange Act and the
other Commission authorities in combination. And so thus far,
what we have in mind is amply supported by the new legislation
that you have just written.
Chairman Dodd. The second question would be budget. I
expressed in my views and estimates letter to the Budget
Committee of February 25th of this year, I raised concerns as
to whether or not the President's fiscal year 2009 budget
request of $913 million for the SEC would be adequate to
examine and regulate the NRSROs as well as to deal with
enforcement, the subprime crisis, consolidated supervision, and
other issues.
Do you feel, Mr. Chairman, that the amount that you are
going to be given here is enough, will provide enough resources
to effectively oversee the securities markets? And if not,
would you share with the Committee what you believe you are
going to need?
Mr. Cox. Mr. Chairman, as you know, the budget that the
President has put before you is the largest budget that the SEC
would ever have received. It is approaching $1 billion. I think
it would be appropriate for this Committee as authorizers to
consider both the CSE program and the CRA program from the
standpoint of their place within the agency. I think overall
the nearly $1 billion that Congress has provided us in the
latest budget is ample for the overall achievement of our
goals, and I have been able as CEO of the agency to allocate
resources, for example, to credit rating agencies and to our
CSE program.
At the same time, because both of these programs are
relatively new, the CRA program itself very new and it has
never been the subject of extensive consideration, therefore,
on the Appropriations Committee, and because the CSE program is
a voluntary program based on old authority and not itself
authorized in law, I just think it would be very useful if
there were a dedicated funding stream for these two significant
new responsibilities for the agency because they have changed
overall the responsibilities of the SEC.
Chairman Dodd. Well, we will take that into consideration.
Of course, we have some Members of the Appropriations Committee
here, including Senator Shelby, so we can examine that issue
further.
Let me raise a couple of specific suggestions that will be
raised in testimony from some of our witnesses coming along
that I thought were interesting. Again, I agree with your
intent to enable market participants to better compare the
rating agencies. And Professor Coffee has proposed an idea
along these lines that I think has some merit, and I would be
interested in your own reaction to it, and that is, a neutral
website that displays the past ratings for each security rated
by multiple NRSROs so investors could compare the accuracy of
the different rating agencies.
I wonder what your views would be on a proposal such as
that, and could the SEC maintain such a website? Is there
anything that would prohibit you from doing that?
Mr. Cox. Well, the idea of enhancing the transparency of
the ratings themselves and their performance is at the heart of
what we have been talking about, of course. Making the
information as easily available to the public in the most
easily comparable form also is a natural objective. And so
Professor Coffee's proposals in that respect are very much
consonant with at least what I am thinking and I believe what
the Commission staff and perhaps the other Commissioners are
thinking.
As you know, the statute, I think wisely, says that the
Federal Government should not dictate the ratings themselves,
should not tell rating agencies in this competitive market
precisely how they should do it, but there is ample support in
the statute for disclosure around these things. So provided
that the scorecard was disclosure and not indirect regulation--
--
Chairman Dodd. No, I think that is what we are talking
about. Professor Coffee can contradict me when he testifies,
but I think the idea was just to allow--so you would have some
way of looking at the accuracy of this and making judgments.
Mr. Cox. The final point that I would make is that were the
disclosure mandated to be tagged with XPRL data tags in
interactive data form, then almost anyone could put together
their own comparative scorecard, and I think that a lot of
financial intermediaries on the Web would do this probably for
free for consumers and investors in addition to whatever the
SEC might do on its own website.
Chairman Dodd. Yes. Of course, the SEC, that Good
Housekeeping Seal of Approval is a very valued determinant.
Mr. Cox. Yes, of course, and if the SEC requires more
detailed disclosure beyond what already is provided on the
NRSRO, then, of course, all of that data would be official SEC-
filed data.
Chairman Dodd. An additional suggestion from Professor
Coffee would have the SEC temporarily suspend an NRSRO, the
status of a rating agency that consistently errs in rating a
particular type of security over a period of time. What are
your thoughts on that?
Mr. Cox. The authority that you have just given to the SEC
includes not only censure but revocation of the registration of
an NRSRO.
Chairman Dodd. So you have that authority?
Mr. Cox. Yes, we do.
Chairman Dodd. There has been----
Mr. Cox. Now, I should add, not if that authority were to
be used to sanction someone for getting the rating wrong, but
for violating any of the rules or provisions of the statute,
that sanction would be appropriate. And as I mentioned before,
you did not want us, the SEC, to actually regulate the
substance of the ratings. But we would not revoke a
registration for that reason.
Chairman Dodd. There has been a suggestion as well that a
rating agency separate its rating business from its rating
analysis function. We have heard similar arguments in the past
in other related areas of financial services. What is your
reaction to that suggestion?
Mr. Cox. Conflicts of interest of that sort are very much
at the heart of what we are looking at in the proposed new
rulemaking, and I agree both with the mandate in the act very
strongly and with the inferences that have been drawn from it
that conflicts of interest are directly related to some of the
problems that we see in the market.
Chairman Dodd. Let me come back to the previous question. I
heard your answer to it. It is one thing to break rules and
certainly suspend. I understand that, you have that authority.
What I was driving at more is the error in judgment of
consistently--I am not talking about, obviously--and this would
have to be, you know, over a period of time you get just error
after error in judgments and drawing conclusions about various
instruments here. It seems to me there that--well, anyway, you
don't believe that that is an appropriate role for the SEC
where you have a consistent error in judgment on these ratings,
that that would be a justification for suspending that rating
agency's function?
Mr. Cox. Well, what the law contemplates--and I think what
our rules will flesh out when they become final later this
year--is a world in which everyone knows what the rating
agencies are doing and why and how. We know what their internal
procedures are. We know how they deal with conflicts of
interest. We know what the prohibited practices are. If then in
a competitive world their ratings fare less well even though
they are performed exactly according to spec than someone
else's, that fact alone would not be grounds under the statute
for revocation of their registration.
If, on the other hand, the reason that their ratings were
consistently wrong is that they had not followed the procedures
that they described, not disclosed fully what they were doing--
they had, for example, changed their ratings model under
pressure to get business or what have you, or committed any
other kind of error, or worse, in judgment, then I think their
registration under the statute could be revoked by the SEC, and
we would have the authority to do so.
Chairman Dodd. My time has expired, but I suspect my
colleagues may want to pursue this line of questioning with you
a little further.
Senator Shelby.
Senator Shelby. Thank you, Senator Dodd, and thank you for
bringing that up. I want to continue along that line for a
minute.
If some firm or NRSRO is consistently wrong on their
ratings, you know, they rate them, then they are downgraded--
and we have seen this over and over and over consistently--
wouldn't that call in, just common sense, the confidence of
that firm, or whoever it was? And why wouldn't you jerk their
license or whatever they have to do business if they are
consistently wrong, they are ignorant, they are incompetent, or
they do not care, or they are sloppy, they are not diligent? I
think that is what we were getting at, among other things.
Mr. Cox. Well, I think it stands to reason that there would
be a connection in most cases--this is obviously a hypothetical
discussion--between that kind of horrible track record and
failure to follow all of the good hygiene that is mandated in
law and regulation. This has heretofore been an unregulated
industry. Once the regulations are in place and once people
have an opportunity to either be in compliance or not with
them, you will be able to draw a correlation, I imagine,
between compliance with the law and regulation and failure in
the marketplace.
Senator Shelby. But incompetence in the marketplace like
this, especially rating securities that are so important, I
think calls for rigorous enforcement of the rules and whatever
they are. But you take doctors, if they are incompetent, they
jerk their license, you know, lawyers after a while, but why
not something like this that goes to the very heart of our
financial system?
Mr. Cox. Well, I think you may be asking me for advice on
new legislation, because what I am trying to do is interpret--
--
Senator Shelby. I think Senator Dodd asked did you need
anything else.
Mr. Cox. Yes, interpret----
Senator Shelby. You might and you might not. I do not know.
Mr. Cox. Well, I will say that if you want the SEC to
revoke the charter of a credit rating agency simply for being
wrong, even though it has followed all of its own rules and
procedures, fully disclosed them, and fully disclosed the basis
for all of its ratings, then we would need new law to do that
because----
Senator Shelby. Mr. Chairman, we are not asking for being
wrong once. I believe Chairman Dodd used the words
``consistently wrong,'' which would bring about incompetence,
the lack of diligence, and so forth. A lot of these rating
agencies have been consistently wrong on the subprime, and I
think they have contributed greatly to the financial debacle
that we have today. Do you not agree with that?
Mr. Cox. Actually, over a long period of time in a number
of circumstances, I have observed the pattern of ratings of
very high levels, preceding almost by days in many cases
horrible consequences thereafter. There is no question that the
legislation that you provided us and the new authorities that
we are going to exercise are much needed for that reason.
I think, however, that the judgment that you made in
passing the law is a good one, that there is a role for
competition here; that if the Federal Government were to be the
open arbiter of whether ratings approaches were good or bad,
that would probably result in poorer ratings performance over
time because people would not be able to update their models
without regulatory approval. They would always stand to be
second-guessed and so on. I think a system such as the one that
you have designed in which everyone has to be aboveboard about
the approach that they are taking and they are subjecting to
competitive pressures, they are accountable, they are
transparent, is probably most likely to get us the results that
we want to achieve.
Chairman Dodd. Richard, would you let me----
Senator Shelby. Go ahead.
Chairman Dodd. Just on this point, Mr. Chairman, under the
act that was signed in 2006, let me just read the language here
and see if this gives you any pause in terms of your response.
It is entitled, under Section 3, ``Grounds for Decision. The
Commission shall grant registration under this subsection,''
and then there are two or three--or two subparagraphs.
Subparagraph (2), ``unless the Commission finds, in which case
the Commission shall deny such registration, that, one, the
applicant does not have adequate financial and managerial
resources to consistently produce credit ratings with
integrity.'' Now, that would be at the initial granting of a
charter. So we are talking about something a little different
here, and that is to withdraw a charter or to suspend a
charter.
So if you make a decision to grant one based on the ability
to produce results with integrity, it would seem to raise the
question that at least the Commission would have the authority
to suspend that charter if, in fact, that integrity were
compromised.
Mr. Cox. I think that is exactly right. In fact, that is
authority that we intend to use aggressively. That provision,
of course, goes to resources. So I think we are talking
hypothetically here.
What we have done, rather unnaturally, in this hypothetical
discussion is we have isolated just the ratings performance,
and we have imagined that it has nothing to do with lousy
management or violations of rules or procedures or other
things, which undoubtedly in the real world it would. But this
provision that you have just cited concerning failing to
maintain adequate financial and managerial resources goes to
quantity, and if the place spent tens of millions of dollars on
their analysis, they probably would get past this.
In any case, the provision that is the bar to our
regulating the substance of credit rating agencies and the
procedures by--or, pardon me, the substance of credit ratings
or the procedures by which they are adopted is prefaced with
the legislative language, notwithstanding any other provision
of law. So it trumps everything else in the statute, and it
says that what we cannot do is regulate the substance of credit
ratings, and we cannot by regulation prescribe the
methodologies by which they are obtained.
Senator Shelby. Chairman Cox, what are we trying to do with
the rating agencies? Maybe you have a different take than I
have. I hope we are trying to restore confidence in the rating
system, their methodology, how they do what they do, because
there is no trust out there in the market today. I don't know
many people that trust the rating system. They see that as a
big contributor of where we are today and what we are trying to
do, working with the SEC, and I was trying to do when I offered
that legislation, what Chairman Dodd is trying to do now, is to
give the SEC the tools that you need not to do business as
usual. Doing business as usual with a rating agency, that is
gone. So many conflicts of interest, as I see it, always so
many cozy relationships, so much money made if the ratings went
this way and that way.
How do we change that? That is what we are after, is
transparency and so forth, because I think the rating agencies
can play and have played a tremendous positive role in our
financial markets. But today, my gosh, you know, would I buy
bonds that Moody's or S&P rated AAA without looking at them and
having somebody else look at them closely? No. I would be
foolish to do it, wouldn't I be?
Mr. Cox. I think that is exactly right, Senator. And as you
know, as the author of the legislation, the overarching purpose
is to improve ratings quality. The devices of transparency and
accountability and competition are the means to achieving that
result. And, of course, improving the quality of ratings is the
prerequisite to improving confidence in that whole rating
system.
Senator Shelby. Well, you are the Chairman of the SEC.
There have been some recommendations to Secretary Paulson to
change the role of the SEC, to make the Fed, you know, the
great arbiter of everything, which I think would be kind of
dangerous and foolish myself. But if the SEC is not going to do
the job, somebody else will have to do the job. I hope that you
and your leadership and your other Commissioners will do the
job that needs to be done. We are at a crisis here, a crisis of
trust, a crisis of confidence, looking at rating agencies with
so many obvious conflicts of interest. I think it is horrible.
Thank you, Mr. Chairman.
Chairman Dodd. Thank you very much, Senator Shelby.
Senator Reed.
Senator Reed. Well, thank you, Mr. Chairman.
Borrowing an analogy from another field, do you think that
any of these rating agencies were guilty of malpractice, not
meeting the standard that you would expect as the Chairman of
the Securities and Exchange Commission in the execution of
their responsibilities to rate some of these securities?
Mr. Cox. That is one of the questions that we are asking
and in part coming to answers on in our ongoing examination of
the three largest firms. As you know, we have some 40 people on
that project right now, and we expect to report fully to you by
early summer.
Senator Reed. Your typical remedy, again, using this rough
analogy, for malpractice is some type of action against the
individual institutions and restitution or something, or at
least to correct the behavior. And I am trying to sort of
connect the dots here between at least the possibility of not
operating appropriately and any type of sanction. They claim--
and the claim has been, I think, affirmed--that they are
protected under the First Amendment in terms of any type of
legal liability.
How do we get them to behave differently if, in fact, there
is a serious question of misperformance?
Mr. Cox. Well, I think that from the fourth quarter of 2007
forward, we are in a different world, because now we have a
regulated industry with legal standards of conduct, as we were
just discussing. They can be censured. They can be hit with
targeted sanctions. They can get the death penalty. There are
all sorts of regulatory norms that they will now have to comply
with.
In addition, we will have a marketplace that is now much
more competitive, not so oligopolistic. We already have
additional credit rating agencies that have been able to enter
the business as NRSROs as a result of the new legislation. And
the disclosure and transparency that the new regime should
provide will force--I think it is the legislative intent, and
we expect it as well--some quality as a result.
Senator Reed. You said, Mr. Chairman, that you have got 40
individuals working on this analysis. When you implement the
regulations, will you have the dedicated staff of roughly that
size with the expertise to continue to evaluate the performance
under the new regulations?
Mr. Cox. It is not necessary to have the current
examination staff on a permanent basis as part of the CRA
program. But we do have budgeted approximately in the range of
10 to 20 people over the long haul for this purpose.
Senator Reed. It just seems to me that in everything we
read about these products, they are inherently complicated,
complex. In fact, many people will not buy them. Jamie Diamond
has been quoted several times saying, ``They are too
complicated. I do not understand them. I will not buy them.''
And yet you will have about 20 people who are going to overlook
the credit rating agencies. Do you think that is adequate
resources to ensure that they are--unless it is simply
procedural, they check the blocks, you know, we did this, we
did this, we did this, but with no substantive regulation?
Mr. Cox. For purposes of managing the registration and
inspection regime--and, remember, we can use the resources as
we are using now our Office of Compliance, Inspections, and
Examinations for this purpose in the future as well. But for
the ongoing purposes of managing the registration and
compliance regime, I think that that is about the right number.
On the other hand, the reason that I am inviting the views
of this Committee on the size and scope of this program is so
that we can be sure that within the context of the overall SEC,
we have, in fact, right-sized this function. It is a brand-new
function. We want to get it right.
Senator Reed. In that regard, how thoroughly will you
anticipate the staff looking down through--and maybe this is
not exactly correct, but a simple security model of a mortgage-
backed security, where there are actually mortgages and a pool
of mortgages, you sell securities. Then there is the CDOs,
which basically gets more complicated, CDOs, squares, et
cetera. Do you anticipate that your staff would be looking all
the way through independently to the collateral of these
securities, or at least on a spot-checking basis?
Mr. Cox. Most certainly on a sampling basis, and probably
across the board just in terms of the different genres of
products that are being rated.
Senator Reed. You raised in your opening statement the
possibility of less reliance under the SEC rules on ratings.
Can you amplify that?
Mr. Cox. One of the concerns that has been expressed in
several of the multinational fora, including the Financial
Stability Forum and IOSCO, is that that there was insufficient
attention paid to what these rating were and what they were not
and that there was in some cases nearly mindless reliance on
the fact that it said AAA. In order to make sure that the
ratings are understood for what they are and what they are not,
we are going to have a lot of new disclosure. At the same time,
we want to make sure that there is not a check-the-box
mentality, and if the rule says you can do X if you have a AAA
rating, that might induce that kind of behavior.
So we will not be able to purge, by any means, our rules of
references to ratings, but there may be some fine-tuning that
we can do in order to make sure that we do not create that
moral hazard.
Senator Reed. Well, it would seem to me that,
theoretically, as you diminish purposely the role of the
ratings, not that, you know, Good Housekeeping Seal of
Approval, put the onus, I think, either directly by rules for
publicly registered companies to independently evaluate the
ratings that they are either buying or they are arranging to
obtain, that might be appropriate for some larger institutions,
but for small investors, for municipalities, for people who are
looking--do not have the infrastructure, how effective would
that be?
Mr. Cox. Well, I think if the only consequence of a change
were to increase the investor burden, then we would not have
accomplished the objective. We have got to be very, very
sensitive to that. The opportunity, I think, that we have is to
state very clearly in our rules what is the point that we are
trying to establish, what is the objective test that we are
asking people to meet. And if a rating----
Senator Reed. What is that test?
Mr. Cox. It depends entirely on the circumstances. The
ratings themselves are mentioned in rules and in statute, I
should add, in many different contexts. But if we can simply
clearly state, and in plain English, what it is that the law
and the rules are trying to accomplish by referencing these
ratings, we can add a little more context to it so it is not
just a mindless act of I got the rating even if the use of a
rating for that purpose would not be appropriate.
Senator Reed. Thank you.
Chairman Dodd. Thank you very much.
Senator Corker.
Senator Corker. Yes, Mr. Chairman, thank you.
I thank you for your testimony, Mr. Chairman, also, and we
have had you up here many times, and we appreciate very much
the position that you are in.
I have to tell you, the rating agencies do not spend a lot
of time on accounting issues, and we have heard of the fact
that they are not audits, which obviously they are not. But it
seems to me that it would be almost impossible in many cases
for a rating agency to give a rating without at least looking
at some of the basic accounting principles that are being dealt
with.
I spent all day yesterday with the leaders of financial
institutions in New York, and this whole fair-value accounting
system, which needs some kind of updating, there is a huge,
huge issue there. But for them not to at least get into some of
the things that are occurring there as a rating agency and how
judgments are being made as to assets being written down and
that type of thing seems to me a very important--and there are
many others--a very important thing for a rating agency to be
able to do to actually issue a rating. Otherwise, I do not know
how it would occur, and I would love for you to comment on that
if you would.
Mr. Cox. Well, there is no question that particularly in
terms of market pricing, those kinds of accounting judgments
associated with fair value and related issues have had big
impacts. The rating agencies have described on occasion their
purpose and their object as being slightly different than
predicting market prices. They are in the business, they tell
us, of predicting the creditworthiness of the ultimate
instrument on maturity and so on, and that, therefore, gainsays
a lot of the wave motion that might occur, even if the wave
motion capsizes the boat in the short run. And so these issues,
at least in the current market turmoil, have really conflated--
I do not think it is any longer possible to neatly parse what
is the ultimate creditworthiness of the instrument from what is
going on in the marketplace.
If an instrument, for example, is totally illiquid, then it
is reduced to something like worthless for an indefinite period
of time, and it is very difficult for adults to say that,
nonetheless, it is a very valuable security and we want to own
it because ultimately it is going to pay off.
Senator Corker. So then what should be the role? I mean,
should rating agencies not indulge more, if you will, on the
accounting side? I mean, is that not a valuable piece of
information for them, if you will, as to--because there are
people, obviously, that do rely on these ratings; you know,
that is what they have been put in place for in the first
place. Is that not something that rating agencies should be
somewhat involved in as they make these ratings?
Mr. Cox. Yes, I think one of the sources of confusion for
investors has been this big gulf between what they see
happening in the marketplace and what they say with the
ratings. And before the round of significant downgrades, there
was this big gap. Ultimately, many of the downgrades had the
effect of conforming the ratings with the judgment of the
marketplace. And so I think that there is no question what is
going on now inside the rating agencies is informed by this
much different world.
Senator Corker. Talk to us a little bit about the notion
that has been floated--and we have heard a lot from
constituencies about this--of separating out ratings for
structured finance itself and the impact that that might have,
if you will, on those particular instruments. If you would,
expand a little bit on those discussions.
Mr. Cox. Well, the suggestion has been made--and it is
under careful consideration by the SEC and may be included in
our proposed rules--that there be different symbologies for
different kinds of products, whether they be structured
products, corporate or municipal, for example. There is no
question that a AAA rating on a structured product is very
different than on a corporate, and yet the same labels being
applied to all of these things might have caused confusion in
that respect in the marketplace.
So I think this is a very valuable subject for us to
explore in connection with our proposed rulemaking, and I can
confidently predict, even though I do not know what the
ultimate proposed rules look like, that we will expose that
whole idea for public comment.
Senator Corker. And that would obviously have a
tremendously dampening effect on the structured finance market
if that occurred, at least in the short term. Is that not
correct?
Mr. Cox. I do not think that would be the idea at all. The
idea would rather be to let people know exactly what it is,
what a rating means.
Senator Corker. And is there a notion of maybe actually
stamping structured finance as sort of a scarlet letter-type
approach, if you will, to this type of financing?
Mr. Cox. Well, there have been calls from State governments
and State finance officers for different symbology for
municipals as well. They believe that having--at least some
people believe that having different symbology would actually
advantage them because they believe their default rates are
provably lower. And so I do not think it is inherent in the
nature of having unique symbology that you are advantaged or
disadvantaged in the marketplace. It is just simply a way for
people to understand what it is they are talking about.
Senator Corker. Thank you, Mr. Chairman.
Chairman Dodd. Thank you very much, Senator.
Senator Bayh.
Senator Bayh. Thank you, Mr. Chairman, and you, Mr.
Chairman.
It seems to me that the issue we are focused on here, Mr.
Chairman, is that one of the principal factors that has led to
the market failures we have experienced--and I gather Senator
Schumer touched upon this before I arrived--at the originating
end you had brokers who had incentives to just kind of get
these mortgages out the door and not adequately assess the
underlying risks involved, the likelihood of the mortgages
being repaid, you know, that sort of thing. Here we had the
rating agencies giving their blessings to the repackaging and
selling of these loans, again, possibly with incentives to do
that without really digging in and accurately assessing the
risks involved. And markets can't function very well without
access to accurate information, and so here we are today to try
and ensure that we do have accurate information going forward.
Mr. Chairman, it seems to me that this is in some ways
analogous, as I think you mentioned in your testimony, to some
of the accounting issues that we dealt with before. I mean, for
example, can you imagine any reason why a rating agency should
be allowed to pass judgment on products that it has itself
helped to structure? Isn't there just an unavoidable conflict
in such an arrangement? And didn't we decide that in some
respects in the accounting arena?
Mr. Cox. Well, there is an unavoidable conflict in that
arrangement, and there are, in fact, other unavoidable
conflicts that are built into either the issuer-pays or the
subscriber-based models. And so what we are preparing to do in
our proposed rulemaking is in some cases just flat out to
prohibit them if we can see our way clear to doing that without
disrupting markets and the ability of firms to function. And on
the other hand, if you cannot bar a practice altogether without
upending the whole thing, then to come up with approaches to
manage those conflicts very clearly, for example, to make sure
that at a minimum people who are in the business of negotiating
fees do not have anything to do with the ratings process.
Senator Bayh. If we are going to be living going forward in
a more robustly competitive world with new entrants coming into
the rating marketplace, as you were describing, why would a
prohibition not work since there are all the new entrants
coming in and it would presumably be easier to just prohibit
this without it disrupting the marketplace?
Mr. Cox. Well, the only reason I left it open to either
interpretation is I did not say what specific practice it was
that we are talking about prohibiting.
Senator Bayh. Same thing for consulting----
Mr. Cox. Without question, the easier way, even from, I
would think, the firm's standpoint because they know what the
rules are, would be just to flat out prohibit these virulent
practices.
Senator Bayh. Same thing for consulting services by rating
agencies?
Mr. Cox. Yes. I think that is undoubtedly a conflict of
interest, and properly structured, speaking for myself, I do
not see why that could not be prohibited. But I should add once
again, because while I am appearing here as an individual and
as Chairman, that I am part of a five-member Commission, that
there are a lot of issues here, and I do not know what we might
propose in our rules.
Senator Bayh. In the previous enactment of the new law
governing this area and the new robust nature of the
competition you have described that is beginning to take root
in this area, is it your opinion, did real competition exist
among these ratings agencies before this problem we have
encountered here? Was there real competition or not?
Mr. Cox. Barely. If----
Senator Bayh. Somebody used the word ``oligopoly,'' I
think.
Mr. Cox. I think that was I earlier in this proceeding, and
that is my view. This industry needs more competition, and the
legislation that you have passed will help it to mature into a
much more competitive industry. That in turn, I believe, will
improve the quality of the ratings.
Senator Bayh. Do you see a problem between--again,
competition and markets function very well. It is somewhat
dependent upon the incentives that exist in the marketplace. Is
there the potential for a disconnect, a continuing disconnect
between short-term incentives and long-term incentives leading
rational decisionmakers to perhaps make decisions that are in
their own best interests but not in the better interests of the
overall functioning of the marketplace? I will give you an
example here.
Just as the loan originators, the mortgage brokers, were
pushing a lot of this stuff out because they were compensated
in many cases by volume rather than the ultimate accuracy of
the loans they were making, do we have a problem here? I know
at least one of the rating agencies was publicly held directly;
another is held under another publicly held company. In any
event, you might get people who were being compensated because
of their short-term performance, and they get bonuses. If they
have stock in a publicly held company, they can cash their
options or sell their stock as it becomes unrestricted. So they
are making real money in the short run, so there is a real
incentive to do that, even if in the long run, if things go
badly, there may be some risk to the reputation of the firm and
ultimately the long-term value of the stock in the firm, but
the pressure is on now, they are being compensated to perform
now. What about that disconnect? And how does the marketplace
take that into account? Were there such strong, you know,
personal reasons to make a set of decisions today but then in
the long run a rational decisionmaker might not make? What do
we do about that ongoing problem with the way people are
compensated in these businesses? And how does the competitive
model--is there a risk to the competitive model when you have
that kind of disconnect between the short-term incentives and
perhaps long-term factors?
Mr. Cox. Well, that is very much a part of the short-
termism that afflicts our markets overall. There has been on
occasion a call to eliminate quarterly guidance for earnings
for companies, for example, for related reasons. The
compensation structure of a firm, and I would think now credit
rating agencies, since they are now a regulated industry, have
to be thought of in terms of the objects of the regulation and
the objects of the ratings themselves in that particular
industry's case.
So as we look at conflicts of interest, compensation won't
be off limits.
Senator Bayh. Oh, it won't. Very good, because it seems to
me, again, you could have, let's say, with, you know, seven
participants or nine or ten, whatever the number ends up being,
if we have a system that rewards people for making certain
kinds of decisions in the short run, irrespective of their
accuracy in the longer term, we could get, you know, warped
outcomes that affect the entire marketplace and, hence, you
know, the economy and society at large that would not be in
anybody's best interest. So I am glad to hear that you will be
addressing some of those issues as well.
Mr. Chairman, thank you. I am surprised you can do the rest
of your day job. You have been up here with us so many times.
But we do appreciate it very much.
Mr. Cox. Well, if I may say so, Senator, in respect of this
particular part of my job, writing regulations under this
statute, it is of enormous value for me to have these
conversations, these colloquies, because as I said, it is the
intent behind this legislation, which is so fresh, that we are
trying to flesh out with the regulations. And so I am entirely
sincere when I say we want all the ideas we can get. This is
the second round of rulemaking, and it is the most important
one, because it is based on the very recent experience that we
have had in the subprime debacle.
Senator Bayh. Thank you.
Chairman Dodd. Thank you, Senator. Hence, the initial
question: Do you need more statutory authority? What I do not
want to discover here is have you complete this process and
turn around and discover you needed additional authority to do
something and we did not provide it for you when we had an
opportunity to do so. And I heard your answer to the question,
but I assume you will keep us posted if you encounter something
different.
Senator Allard.
Senator Allard. Thank you, Mr. Chairman.
Where I see the greatest conflict of interest might be in
the payment model that has been set up with the credit
agencies; in other words, the one who is being evaluated ends
up paying the credit agency for the outcome. Do you see a
conflict there that concerns you? Or are you comfortable with
that model?
Mr. Cox. Well, it is a necessary conflict of interest that
somebody pay, because whoever pays is going to have some
interest. And so in the issuer-pays model, you get one set of
conflicts. In a subscriber-pays model, there are other kinds of
conflicts that can arise; for example, the people who want to
include certain things in their portfolio might want to have
ratings that permit them to do so. And so there is really no
way out, provided that someone is paying. What you are stuck
with is recognizing and sharply identifying those conflicts and
then managing them.
Senator Allard. Even if the taxpayer pays, there is a
conflict, I guess, in a way. Or is there not?
Mr. Cox. Well, there is a conflict with the Federal budget,
I would imagine at some point.
Senator Allard. Yes.
Mr. Cox. If we nationalize all these functions.
Senator Allard. You know, it is not that that model has not
been used. I think in my own profession where we write a health
certificate, when we write a health certificate, we are an
agent of the Federal Government. We are paid by the--we act as
an agent for the Federal Government, and we also act as an
agent of the State that the animal is being shipped to. But the
one who pays us is the one who is shipping the animal and is
asking for--you know, and he will pay, and obviously we will
pay a fee or whatever. But the consequences of not finding that
you did something that was unethical or whatever can be pretty
severe, and you could lose your license and not be able to
practice in the profession.
Do you think that you have the consequences there that are
severe enough to prevent bad behavior?
Mr. Cox. There is no question, Senator. In addition to the
sanctions that can be applied directly to the rating agencies
under this new law, sanctions can be addressed to their
associated persons. And so every single individual who works in
an NRSRO can be the subject of SEC sanctions as well.
Senator Allard. I want to take you into a hypothetical area
because ``consistency'' has been a term that has been
frequently used by my colleagues here on the panel, you know,
consistent results.
If we were to use--and we have consistency, I guess, in
law. How in the world do you evaluate what consistency is? I
mean, is it 20 percent error? Is it 30, 40, or 50? Or is it
some deviation from the normal from your competitors? Do you
have any idea how we would determine consistency?
Mr. Cox. I think to begin with, it would depend upon the
subject of the rating. It would depend upon the industry and
its volatility and its history. So that coming up with a very
simple rule of thumb that would apply across the board to
everything in the capital markets I think would be impossible.
The best approach to that kind of complexity is to have the
maximum amount of disclosure of all the material information
and then to permit comparison in the marketplace.
Senator Allard. Let me ask a question about disclosure. If
you have foreign securities, is disclosure a problem? You know,
you might, for example, have a business that is partially owned
by some foreign government. And so how do you get an adequate
disclosure as to the background of how you are going to valuate
that security? It kind of gets to the accounting issues, I
think, that we were talking about earlier here on the
Committee. How do you handle those kind of foreign securities?
Mr. Cox. Well, transparency varies dramatically from
jurisdiction to jurisdiction, and in some cases, one is left
with nothing more than a brand name to go on, because there is
so little behind it when making an investment decision.
On the other hand, in some other jurisdictions, there is a
great deal of transparency, and the level of accounting detail
and disclosure about management and ultimate parents and so on
is what we would be accustomed to here in the United States,
ordinarily so.
So it just depends entirely on the jurisdiction in which--
--
Senator Allard. A brand name is pretty subjective, isn't
it?
Mr. Cox. Yes, of course. And I think that people investing
in those parts of the world tend to use diversification at some
protection for the risk that they are taking.
Senator Allard. I have always been under the impression
that what happens in the marketplace as far as rate of return
is somewhat influenced by the risk of the investment. Do you
think that holds up still today? And do you see a correlation
between--I suppose there is because what happens with the rate
of return--I mean, when they do an anticipated rate of return,
I suppose they take into account how the rating agency rated
that particular security.
Mr. Cox. The correlation between risk and return is as
ironclad as the certainty of death and taxes.
Senator Allard. And you see it--and what you are seeing
in--did you see an instance in these securities, particularly
the home mortgage products, was there a higher rate of return
with those more risky mortgages or not?
Mr. Cox. Well, I think that part of the alchemy that led to
what we saw in the subprime turmoil was this sense that there
was a cost-free way to improve the return.
Senator Allard. A cost-free way to----
Mr. Cox. Yes. It turned out not to be the case.
Senator Allard. Yes. But there was an anticipation of
greater return on their investment, would you say?
Mr. Cox. Yes, of course.
Senator Allard. And so----
Mr. Cox. And from the issuer standpoint, the opportunity to
securitize permitted them to borrow at lower rates.
Senator Allard. And that assumption of where that rate of
return came from, do you think it was just the experience of
the investor with the market? Of course, every individual would
have different experiences in that regard. Or was it based
pretty much on credit rating? Or both?
Mr. Cox. Well, I think the overall sense in the short run
was that a better mousetrap had been built, that one of the
working parts of that mousetrap was the credit rating, and even
against the evidence, late in the game people clung to the hope
and the belief that somehow it could be true.
Senator Allard. Thank you, Mr. Chairman.
Chairman Dodd. Thank you very much. You raise a couple of
issues, and I think Senator Schumer may be on his way over as
well to ask a couple of questions before we move to our second
panel.
If I may, we received a letter yesterday from a group of
individuals, the Real Estate Roundtable, the Mortgage Bankers
Association, Commercial Mortgage Securities Association, the
National Association of Realtors. I presume that was all one
letter. Was that one letter?
They were opposed to proposals of the President's Working
Group to differentiate between credit ratings for structured
finance products and other assets. What is your reaction to
that? I will see you get a copy of the letter, too, but I would
be curious what is your reaction.
Mr. Cox. Just a moment, if I may.
[Pause.]
Chairman Dodd. They make a case, by the way, in the letter,
in fairness, they say that the changes would--and I quote them
here--``contribute to greater market volatility and investor
confusion.''
Mr. Cox. Well, I have not seen the letter myself, but this
is a subject that I think we are very interested in. I know
that at IOSCO, regulators around the world are interested in
this topic; within the Financial Stability Forum, it has been
discussed; within the President's Working Group it has been
discussed; at the staff level at the Securities and Exchange
Commission and at the Commissioner level, it has been
discussed.
So I would predict that whatever occurs in our proposed
rules, we would ask the public questions about this and
engender more of that kind of comment on both sides of the
issue.
Chairman Dodd. I would be very interested in hearing if you
do develop that or where maybe the Commission is heading with
that. I would be interested.
Senator Schumer.
Senator Schumer. Thank you, Mr. Chairman. And thank you. I
am sorry that I could not be here for the testimony, but I have
a few questions based--some based on our discussions that we
had, which I want to again reiterate I appreciate your coming
to my office and briefing me on these ahead of time.
Now, we all know now that the SEC has stronger oversight
authority over the agencies since the legislation that Senator
Shelby, Senator Dodd, and others of us endorsed is now the law.
And so you have had examiners at the firms, and my focus is on
the conflict of interest issue.
I know your investigations are ongoing, but can you just
give us a sense of what you found regarding the agencies'
compliance with their stated procedures intended to control
conflict? In other words, the article here that I referred to
earlier, which I found did a very good job, seems to indicate
that before you were given authority, there were conflicts and
nobody paid much attention to them within the credit rating
agencies themselves. Is it getting better? Do they have their
own controls? Does some little buzzer go off when a supervisor
wants to change the person on the job because he is not giving
or she is not giving as good a rating?
Tell me--I am not asking for any specific investigation
about a specific agency. I am asking in general how good are
the agencies at uncovering these conflicts now that it is
against the law to--you know, now that these conflicts are
against the law.
Mr. Cox. Well, I feel very confident in saying that it is
better. It is better for obvious reasons. There is so much
focus on this right now and so much has gone wrong that many
have reacted with alarm, and there is a lot of attention being
paid to it.
We have in the course of our examination thus far found
examples of apparent failure to adequately manage conflicts of
interest, and some instances have even occurred this year.
Senator Schumer. So it would be better, but there are still
lapses, even after all the focus on credit rating agencies. Is
that a fair way to put it?
Mr. Cox. That is a fair way to put it.
Senator Schumer. And what are you doing when you find these
lapses?
Mr. Cox. Well, of course, we are in there with live bodies
in real time, and so anything that is brought to our attention
is dealt with on the spot. But, in addition, we are going to
make broad inferences based on our examination of the three
largest firms and present those publicly, as well as to the
firms in early summer.
Senator Schumer. Good. So we will learn about some of these
lapses.
Mr. Cox. Yes.
Senator Schumer. And that should be somewhat prophylactic
as well in terms of preventing them from doing it again.
Is it that the agencies do not want to? Or is it just so
embedded in their culture? You know, this article points out
when new management came in at Moody's, the whole world changed
because they wanted to increase market share in something that
ended up being risky, although it probably was not thought to
be risky at the time. When you go to the higher-ups in the
firms, do they want to change? Do they want to get rid of these
conflicts of interest? Or do they say, hey, we will lose
business, we better be careful and not do it so fast?
Mr. Cox. Well, it sounds as if you have, as I have, met
with the leaders of these firms, and they certainly express a
strong desire to deal with these problems and to take them
seriously. I think the only proof, however, is going to be in
the pudding.
Senator Schumer. Right, and they are not there yet. OK.
One other question related to this, and I thank you, Mr.
Chairman. People have questioned the agencies' reliance on
information supplied by the issuer to determine their ratings.
You know, I guess the average person, maybe even the average
investor feels that the credit rating agencies do not just take
the information that is given, but go investigate and see if it
is for real, because obviously the issuer is going to put their
best foot forward.
Have you found--shouldn't there be some disclosure on the
amount, or the lack thereof, of the due diligence that is
performed on a bond?
Mr. Cox. Yes.
Senator Schumer. In other words, if they did not
investigate it, if it has another no-doc loan in some other
area, they should say that clearly: This is has no
documentation, and we did not investigate it; or, It has
documentation, and we did not investigate it; or--you know what
I am saying.
Mr. Cox. Yes, that is an important subject for disclosure.
It is one that I mentioned in my testimony that I think may
well be covered in our proposed rules.
Senator Schumer. Right. Now, let me ask you this: Do you
think--this is, again, based on that article, which I guess you
might think from my testimony I am obsessed with, which I am
not.
[Laughter.]
Senator Schumer. But do you think that--it is just a good
article. That is all. I did put it in the record in my opening
statement, Mr. Ranking Member.
Do you think significant changes in market share should
automatically trigger enhanced scrutiny by the SEC over the
rating agency activity? If all of a sudden they rated 20
percent of these bonds and now they are getting 70 percent in a
year, something is up. What do you think of that idea?
Mr. Cox. Well, because you provided it to me, I have had a
chance to read that article, and----
Senator Schumer. Oh, there you go.
Mr. Cox. There is absolutely no question that that kind of
red flag should be a guidepost for an examiner.
Senator Schumer. Good. That is good to hear. And what about
other red flags, such as significant deviation in ratings
performance from historic averages or significant analyst
turnover? In other words, you may not have the specific on this
case, but you are seeing there are a lot of analysts that have
been turned over lately. Should that also provide a similar red
flag?
Mr. Cox. I think so. Obviously, the facts will inform in
any particular examination where the examiners want to go, and
I think over time, as the SEC develops more and more expertise
in this, we will have, either formally or informally, a whole
set of----
Senator Schumer. Right. And we can expect some of these in
the proposed rules that you are going to put out this summer, I
presume.
Mr. Cox. Yes, although what we are talking about right now
is the kind of thing that examiners are going to look to.
Senator Schumer. Right, or guidance to the examiners that
might be made public. We are going to see concrete evidence of
some of these things happening, and it will be sort of out
there publicly that you are doing it.
Mr. Cox. I----
Senator Schumer. Not specifics. The general things.
Mr. Cox. I can undertake to do that, yes.
Senator Schumer. Good.
Mr. Chairman, thank you.
Chairman Dodd. Thank you very much, Senator.
Let me ask one more question. Again, Professor Coffee is
here and obviously is going to be testifying, but he, I
thought, raised a very good issue in his testimony. It goes
beyond the issue of the due diligence and all of the questions
that Senator Schumer, Senator Shelby, Senator Reed, and Senator
Corker raised, and that is the staleness of data. It is one
thing to get it wrong initially, but then to have a conclusion
hanging around for a long time, when information emerges that
would certainly warrant at least someone stepping up to the
plate and saying something--in fact, he calls it the ``gravest
problem'' may be the staleness of the debt ratings. And I guess
the agencies--are agencies timely in updating ratings and
withdrawing obsolete ratings? What standards should they
observe in that process?
On page 8 of his testimony, the professor points out that
major downgrades of CDO securities came more than a year after
the Comptroller of the Currency first publicly called attention
to the deteriorating conditions in the subprime market and many
months after the agencies themselves first noted problems in
the markets. I think it is a very good point, and we have
talked a lot about the front end of this. But the staleness of
data I think is a very good observation. What is the reaction
of the Commission to that?
Mr. Cox. Well, I think that overall, first, I should say
that the contribution that Professor Coffee has made to this
whole discussion has been exceptional, and I want to thank him,
and I am glad you have him on your next panel. I am glad he is
here today. And certainly at the agency, we have spent a good
deal of time taking all of that in.
Second, at least as a matter of pure disclosure, it seems
completely feasible to deal with this issue, to require
disclosure of how often the models are updated and how they do
surveillance, how the rating agencies do surveillance of their
past ratings.
Chairman Dodd. I appreciate that.
Senator Shelby.
Senator Shelby. I want to pick up on something Senator
Schumer brought up. Thank you, Senator Dodd.
How do you measure these rating agencies in a sense?
Transparency will help, but the SEC has to play a role here
because this is such a debacle. They tell us at times--and I
have talked to some of them--well, gosh, you know, we are just
giving our opinion. Are they? Is it more than my opinion? They
say, well, under the First Amendment of the Constitution, we
are just giving our opinion, free speech. But they are selling
this information, and then it is relied on all through our
financial system. So there is something amiss here, Mr.
Chairman. You know, as Chairman of the SEC, I know you are
going to look at all this, but Senator Dodd brought up that the
Comptroller of the Currency has to start asking some questions
about some of the subprime things. Where were the rating
agencies early on in this? I am afraid what they were doing is
continuing to rate a lot of these subprime securities at
investment grade, some even AAA and so forth. And, Mr.
Chairman, I will ask you as you get into this: When did they
start downgrading their ratings? Was it after the whole thing
was in a free fall?
I do not know, but I just know something is amiss here in
all of this. We went through the Enron deal, but, gosh, this is
so much bigger, you know, in many ways than that. And I know
that the rating agencies play an integral role here. The SEC
has to play a big role of oversight here. Make no mistake about
it.
Mr. Cox. Well, I think the fundamental answer to your
question of how you measure their performance is the quality of
their ratings. And, you know, up until very recently, there has
not been a lot of competitive pressure on that quality, and so
whatever----
Senator Shelby. Is that because everybody bought into it,
you know, the euphoria?
Mr. Cox. It is for a variety of reasons, but not least of
which is that there are very few of them, and yet ratings were
by regulation and by law in many cases required. And so they
had a Government-required function. There was no place else to
shop, and so that is not a good competitive climate to begin
with.
The measurement of the quality of ratings is inherently
subjective. It is going to be quantitative, to be sure. It is
going to be analytical. But there are so many things that go
into it, it is going to inherently be subjective. It is the
kind of thing that markets are good at.
Our disclosure system at the SEC when it comes to price
discovery for all sorts of things, like corporate equities,
helps people arrive at a very specific number, the price for a
security, whether people think that that price is the future
discounted cash-flow, the quality of management, new product
introductions, or what have you. Reducing complexity to a
measurement like that is what markets are very good at, and the
SEC is very good at disgorging information to the public and
making sure there is full disclosure so that the public can
make those judgments. I think that is what we are about to do
now with credit rating agencies.
Senator Shelby. Mr. Chairman, if the rating agencies are
the linchpin of our financial markets, our securities market--
and a lot of people believe they are--our linchpin is broken
right now as far as confidence, trust in the financial markets.
And I believe it is going to be--a lot of what you do, and your
other Commissioners, and what we try to help you do is going to
help restore some of that. But the old way of doing business
with the rating agencies, that has got to go. I believe it has
to, and it should have already gone.
Thank you, Mr. Chairman.
Chairman Dodd. Thank you very much, and I am going to leave
the record open, Mr. Chairman, for additional questions that
may come from members here. We have a second panel I want to
get.
Senator Reed and I were chatting and, you know, raised the
issue--which I will not raise with you right now because--
unless Jack wanted to bring it up, but the whole notion of
whether or not--I have had some people say to me, Why even
bother having rating agencies in this day and age? There is a
case to be made for people who are in this world who wonder
whether or not we are just spinning our wheels in a sense by
doing this. I think there is an argument for it. And, second,
why not even consider the possibility of sort of a nonprofit
sort of a credit rating agency, have a colleges approach that
sit and determine whether or not something is a good
institution or not, to take all the conflict out altogether?
Now, that is a bigger question than what we have asked you
to do here, but do you have any quick comments on that at all?
I mean, that is an idea that has been raised by some. We may
never be able to address--as you point out, both from the
subscriber as well as the issuer, the conflicts are going to be
there, no matter which side of this you flip it. And so are we
reaching a point that maybe we ought to be talking about a
different system altogether? To go to the question Senator
Shelby has raised here, if this is the linchpin in all of this.
What are your thoughts on that?
Mr. Cox. I think it is entirely possible to have private
sector entities that are commercial in nature that are,
nonetheless, independent from the securities that they rate and
who do a good job of it.
There is a conflict of interest in virtually every
commercial relationship in the sense that, you know, if you go
to the dentist and it is the dentist's interest to charge as
much money as possible and you do not really know that much
about dentistry, well, the dentist could tell you that you need
all your teeth replaced.
Chairman Dodd. I can sue that guy for medical malpractice.
I cannot sue this guy.
Mr. Cox. Well, I think that that is what is changing. The
result of the legislation that you passed and the regulatory
authority that you have given us, the ability that we now have
to define practices, define what is necessary to manage,
mitigate, or end entirely those kinds of conflicts of interest
makes this all very different.
The authorities that we have under the securities laws will
now apply in like way to participants in this market; not only
the firms themselves but their associated persons will now be
subject to sanction by the SEC. And so the difference between
being completely unregulated, which was the case 6\1/2\ months
ago, and now being a regulated industry is enormous. And I
think there is a good deal of reason to expect that it will do
a lot of good.
Senator Schumer. Mr. Chairman?
Chairman Dodd. Yes.
Senator Schumer. Just one question, if I might. I mean,
there is an intermediate step. Senator Dodd's ideas are, as
usual, intriguing. But do you think there is less conflict in
the investor when the investor pays the agency as opposed to
the issuer paying the agency? I mean, it is conflict from the
other side. It puts the premium on not AAA but maybe failing
grade, you know; or it moves it in one direction rather than
the other, is a better way to put it. But does one make more
sense than the other? Should that be something that is
seriously explored as well?
Mr. Cox. Well, I think that both make more sense in
combination, because each is a check against the other. And
what your legislation has opened the door for now is relatively
easy entry into the market for subscribed-based ratings.
Senator Schumer. And there are few right now.
Mr. Cox. There are already two that we have registered. I
expect there will be more in short order.
Chairman Dodd. Interesting. Mr. Chairman, very good We
thank you immensely, and please stay tuned. And stay in touch
with us, too, on this issue, on that question. If there is
additional statutory authority that your agency thinks you may
need in this area, this Committee would very much want to know
that as soon as possible.
Mr. Cox. Thank you very much, Mr. Chairman.
Chairman Dodd. If our second panel would come up quickly,
and I apologize. You have been waiting a long time.
We hardly need to introduce Professor Coffee. He has been
talked about so often here that he has already been sort of
introduced. But Professor John Coffee, Columbia Law School,
served on distinguished legal bodies, published significant
research, and contributed to the work on Sarbanes-Oxley.
We are pleased to welcome Dr. Arturo Cifuentes, the
Managing Director on the Structured Finance Department of R.W.
Pressprich, and former Managing Director of Global--the global
head of collateralized debt obligation research at Wachovia
Securities.
I want to welcome the representatives of the three largest
NRSROs: Vickie Tillman, Executive Vice President, Standard &
Poor's; Claire Robinson, Senior Managing Director of Moody's
Investor Service; and Stephen Joynt, President and Chief
Executive Officer of Fitch Ratings.
I want to underscore the point that Senator Schumer raised
earlier, and I understand in Moody's case there was a conflict
today because of shareholder meetings. And obviously his
obligation is to be there for that. But for these other rating
agencies, I would very much like to have heard from the heads
of them. We have got people here from these agencies, but,
candidly, it is a little more difficult to expect them to
respond to these questions that we are all going to have for
them.
Senator Shelby. May I say one thing, Mr. Chairman?
Chairman Dodd. Let me turn to Senator Shelby.
Senator Shelby. Mr. Chairman, I want to associate myself
with your remarks. I want to commend the President and Chief
Executive Officer of Fitch Ratings for coming to this hearing.
But I am disappointed, as you are, and Senator Schumer was, and
others, that the other CEOs of Standard & Poor and also
Moody's, regardless of conflict--this is a Senate hearing on
something that I think is very, very important. And we are
going to get them here, I hope, Mr. Chairman, because although
they will have able people here testifying on their behalf, it
is not like having them here themselves.
Thank you.
Chairman Dodd. I appreciate that.
I appreciate your patience. First of all, you have been--I
hope this hearing has been instructive as you have been sitting
there listening to all of this, and helpful to some degree.
Certainly you have heard the expressions expressed by almost
every member here about their concerns about all of this and
the importance of this issue. So let me begin by asking for
your comments, and, again, your testimony will be included in
the record, beginning with you, Professor Coffee, and we will
then turn to Ms. Tillman, Ms. Robinson, Mr. Joynt, and Mr.
Cifuentes.
STATEMENT OF JOHN C. COFFEE, JR., ADOLF A. BERLE PROFESSOR OF
LAW, COLUMBIA UNIVERSITY LAW SCHOOL
Mr. Coffee. Thank you, Chairman Dodd, Ranking Member
Shelby, and it is a pleasure to be back again in front of your
Committee. In order to be brief, let me break my testimony down
into three short segments.
First, and very briefly, what have we long known about the
rating agencies? And I suggest none of this is about to change.
We have long known that they face limited competition, and if
you want, they share an oligopoly.
Two, we have long known that they face very little
liability to investors, and, indeed, they have never been held
liable to investors. That is different than every other
financial gatekeeper, auditors, securities analysts, or anyone
else.
Next, they have a built-in conflict because they are a
watchdog paid by the party they are to watch. Now, auditors are
also, but auditors face real liability.
And, fourth, they have a business model under which they
can make money, even if no one trusts their ratings, because
they are also selling regulatory licenses. Institutional
investors cannot buy debt securities without their rating, and
that protects them even if they are off the mark.
OK. That is what we have always known. What have we learned
recently? We have learned, first of all, that the rise of
structured finance was immensely profitable, but it did
destabilize this industry by aggravating those longstanding
conflict of interest problems. Why? There is no single
corporate issue that has any leverage over a major ratings
agency. But structured finance is controlled by basically five
or six large investment banks. They package these deals on a
monthly basis. They do have some real leverage with respect to
their rating agencies, and thus, that means for the first time
there are clients that have clout with respect to you. That was
different than the past.
Next we have learned that rating structured finance is very
different than rating straight corporate debt. You rate
straight corporate debt largely based on publicly available
financial information, SEC reports, stock market and bond
market prices--all of which tell you a lot. You rate a pile of
mortgages--3,000 mortgages in this pile--it is opaque, it is
non-transparent. It is much more difficult to do, and it is
done basically on a quantitative model--a quantitative model
that has never been in existence that long enough to have been
fully checked out. That is point one.
Point two of what we learned recently, loan originators and
investment banks have learned how to game the model, how to
play with it. This is partly because for a large advisory fee,
the rating agency showed them how their model works. And once
you are shown how it works, you learn how just tweaking it a
little bit and selectively editing the data can get you a
better rating.
Now, I am not saying that rating agencies engage in fraud.
What I am saying instead is that because the rating agencies
are very vulnerable to selectively edited information, and also
to misleading information, we have a system that is defenseless
against loan originators and others who have strong incentives
to try and game the system.
Next point. Because the credit rating agencies do not
perform any due diligence themselves--and some of the Senators
were making this point earlier--because they do not do
verification work themselves, they are almost uniquely
vulnerable, maybe even defenseless, to selective editing and
misinformation given to them by loan originators who have every
incentive to game the system and try to get a higher rating.
OK. The credit rating agencies also appear to have
responded, in my judgment, in a fashion that I have to call
tardy and slow to massive changes in the housing market. The
Comptroller of the Currency and others--and they knew it
themselves--saw major changes in which for the first time home
purchasers were getting 100 percent financing without any
equity stake. They were able to get mortgage loans based on no
documentation. All of these things means that you are
vulnerable to significant problems, and there was a worldwide
market demanding all of the CDO securities that you could sell
if they had that letter rating. All of this meant there was
vulnerability, but it was not until after the crisis broke that
we saw the major fall, the major downgradings, which really as
a major downgrading began in July of 2007.
Now, against that backdrop, what do I suggest most needs to
be done? Well, let me suggest that the single biggest problem
is probably that no one verifies the data. In the world of
structured finance where you are using a quantitative model,
the oldest rule about quantitative models is: Garbage in,
garbage out. And you are going to get people selectively giving
you somewhat incomplete information, and there is little you
can do about it. Even for the future, we have to expect that
loan originators will continue to provide biased or selectively
edited information. It is in their own self-interest.
It is difficult to overstate this. This is almost as if the
credit rating agency were in the position of an accounting firm
that went to the corporate client and said, Give us your data,
your revenues, your costs, your liabilities, and thank you very
much; we won't check this, we will just produce your net income
figures, and we will tell the world what your earnings per
share is.
That is where I think you have to begin, and I suggest that
SEC rules to be meaningful have to introduce some form of
greater verification. Verification is being done. I recognize
it cannot be easily done by the rating agencies because they do
not have the in-house staff to carefully verify thousands of
thousands of securities.
But issuers and underwriters today hire independent, due
diligence firms that go out there and evaluate the quality of
the collateral in the loan pool. That information, I suggest,
should also be provided to the rating agency. And, indeed, the
strongest rule that I would suggest to you is that NRSROs,
Government-licensed rating agencies, should not be able to give
an investment grade rating on structured finance products
without having before them some report from an independent
expert that sampled the loan collateral and reported that the
loan collateral met the following parameters: there were not
more than 10 percent of these mortgages that were without an
equity investment, there were not more than 10 percent that had
no documentation, or there were more, and we will disclose
that. That I think is necessary so that we have a gatekeeper
that really has both the auditing and sampling component as
well as the analytical component.
I would suggest to you today that the credit rating
agencies have a lot of competence, a lot of skill of analysis,
but very little on verifying and gathering data.
Now, two other ideas that are in my testimony, I will be
very brief about these. One, there is the problem of stale
ratings. If you compared the debt rating agency to the
securities analyst--and they are functionally similar--
securities analysts update their ratings on a quarterly basis.
I would suggest there is a lot of harm to the smaller
institutional investor--and you have them in each of your
districts. These are the school boards, the colleges, the
charities, and the endowments who sit there and see that 2-
year-old credit rating, that may no longer be accurate, that
may no longer even be what the agency itself would give under
its newer model. I would suggest something like an annual
revision requirement. You go back, you review it, and you
either renew it, change it, or withdraw your rating. But stale
ratings are dangerous to less than sophisticated institutional
investors.
I would suggest something like an annual revision of your
rating; and, second, when you change your model, make a
material change in your model of methodology, you should go
back and figure how that would change your past ratings,
because today there are ratings that are sitting out there even
though the model has changed and you would not give that same
rating today.
The last thing I would suggest to you besides dealing with
this problem of staleness is the financial scoreboard. I think
there are lots of less than sophisticated institutions that sit
there and know one rating agency or two. They do not know there
might be eight or nine or ten in another year from now. If you
had the SEC giving us one financial scoreboard that showed the
ratings of all of the NRSRO rating agencies, what would they
learn? They would learn that, well, agency one and agency two
gave an investment grades. Some of these newer subscriber-pay
rating agencies are more critical and have given it junk or
intermediate status. You would learn the diversity of opinion.
And when you learn that diversity of opinion, you might decide
to put your short-term money in Government securities or
something else rather than AAA-rated CDOs.
I think the SEC would be the right party to do this because
you do need to standardize some of the terms. Things like
default rates that should be on this website can be computed in
different ways. I do not know which way is best, but I think
the SEC could give us one standardized technique so they could
tell us the default rates on these various classes of products
for each of the major rating agencies that are NRSROs.
I think I have gone over my time, so I will stop there and
answer any questions that you have.
Chairman Dodd. No, that is very helpful and very insightful
as well. Obviously, we want to hear the witnesses.
I hate to interrupt here. Have you voted, Jack?
Senator Reed. No.
Chairman Dodd. All right. What we will do is we will go
over and vote, and whoever gets back here first, just start
right in. So if you would be patient for about 7 minutes, we
will come right back to you. I apologize to you, but we have a
vote on the floor of the Senate. But thank you very much for
your testimony, Dr. Coffee. When we come back, we will hear
from Ms. Tillman.
The Committee will stand in recess.
[Recess.]
Senator Reed [presiding]. If I can ask you to take your
seats, Chairman Dodd asked the first returning member to
reconvene and to begin to take the testimonies. And I believe
we have concluded with Professor Coffee.
Ms. Tillman, please.
STATEMENT OF VICKIE A. TILLMAN, EXECUTIVE VICE PRESIDENT FOR
CREDIT MARKET SERVICES, STANDARD & POOR'S
Ms. Tillman. Thank you, Mr. Chairman, Mr. Ranking Member,
Members of the Committee, and good morning. I am Vickie
Tillman. I am Executive Vice President and head of the ratings
business for Standard & Poor's, and I appreciate the
opportunity to speak before you today.
Senator Menendez. Ms. Tillman, could you put your
microphone toward you.
Ms. Tillman. Oh, sure.
Senator Menendez. Thank you.
Ms. Tillman. You are very welcome.
At Standard & Poor's, a core principle of our business and
key driver of our long track record of analytical excellence is
a constant commitment to improvement. Over the past several
months, rating agencies have been the object of significant
focus, including much critical attention. We have listened to,
and reflected on, the numerous comments and concerns, and we
have focused our efforts to enhance our ratings process,
provide better and more information to investors, and promote
confidence again in our ratings. The result has been a series
of actions that we announced in February earlier this year. But
before I go over those actions, I would like to note that
ratings speak only to creditworthiness, and there have been a
significant number of downgrades, and downgrades, again, are
not defaults. They are movements because things do change in
the environment. But there have been significant downgrades in
the RMBs area and in other structured finance securities. But,
to date, the volume of actual defaults on those securities has
been less than one-fifth of 1 percent of all U.S. RMB assets
Standard & Poor's has rated between 2005 and the third quarter
of 2007. And those numbers at one-fifth of 1 percent are those
that have actually defaulted.
I have attached to my testimony a detailed description of
these actions that we released in February, and they include an
update that we published earlier this month outlining the
significant progress we have made to date in implementing them.
In total, there were 27 different initiatives. I would like
to highlight four broad categories.
The first category of actions relates to our governance
procedures and controls. Notably, initiatives in this category
include: establishing an ``Office of the Ombudsman'' to address
concerns related to, for instance, potential conflicts of
interest; implementing ``look back'' reviews when analysts
leave to work for an issuer; implementing periodic rotations
for lead analysts.
The second area is in analytics. The category of actions
focuses on the substantive analysis we do in arriving at our
ratings opinions. Notable initiatives in this category include:
establishing an independent ``Model Oversight Committee'' to
assess and validate the quality of the models used in our
analysis; complementing traditional credit ratings analysis by
highlighting non-default risk factors that can affect rated
securities, such as volatility of ratings, correlation, and
recovery.
The third area is in terms of information. Notably,
initiatives in this category include: presenting ``what if''
scenario analysis in our rating reports; implementing
procedures to collect more information about the processes used
by issuers and originators to assess the accuracy and integrity
of their data and their fraud detection measures; increased
dissemination of ratings-related data, including default
statistics; developing an identifier to highlight when a rating
is on a securitization or a new type of structure.
And the final area is very important as well, and that is
education. And these actions relate to our efforts to educate
the market about ratings, their role, and their limitations.
Notably, initiatives in this category include: launching a
market outreach program to promote better understanding of
complex securities that Standard & Poor's may rate; working
with other NRSROs to promote ratings quality through the
introduction of best practices and issuer disclosure standards.
We have been working aggressively to implement these
actions. We welcome further suggestions as to how we can
enhance market confidence and continue our tradition of quality
of ratings that offer opinions on creditworthiness to the
market.
In addition to these initiatives, we have been engaged in
discussions with legislators, regulators, market participant in
the United States and around the world. For example, we have
actively been involved with IOSCO as it considers possible
revisions to the model of a code of conduct as it relates to
securitization. Similarly, we have participated in an ongoing
review of rating agencies by CESR and having engaged with the
Financial Stability Forum members in a dialog about their
suggestions. Here at home, we have been working with the SEC as
it conducts its first exam of our ratings process under a
recently established regulatory framework. That exam is still
in progress. Its scope is extensive, and the SEC staff has been
extremely active and thorough in their work.
We look forward to the SEC's completion of its work, and we
are committed to addressing any recommendations that the
Commission may have following its review process.
We are also focused on the work being done by the
President's Working Group. We fully support the group's efforts
to bring transparency, stability, and confidence to the capital
markets, and we look forward to working with them to help drive
the effective functioning of the credit markets.
In conclusion, I would like to thank you for the
opportunity to participate in this hearing, and I would like to
let you know that we are committed to improving on our
analytical excellence and our desire to continue to work with
the Committee as it explores developments affecting the capital
markets, and I would be happy to answer any questions that you
may have. Thank you.
Chairman Dodd [presiding]. Thank you.
Ms. Robinson, welcome.
STATEMENT OF CLAIRE ROBINSON, SENIOR MANAGING DIRECTOR, MOODY'S
INVESTORS SERVICE
Ms. Robinson. Good morning, Chairman Dodd and Members of
the Committee. I am pleased to be here on behalf of my
colleagues at Moody's Investor Service to discuss our views of
some of the recent developments in the credit markets and the
initiatives underway to address them, both at Moody's and
across the industry.
As you are well aware, the global credit markets have seen
incredible turmoil over the past year. That turmoil has been
driven by many causes, one of which is the deterioration in the
U.S. housing sector resulting from an unprecedented confluence
of factors. These include a sharp erosion in mortgage
underwriting standards, misrepresentations in the mortgage
application process, the steep decline in home prices, and a
sharp contraction in credit available for refinancing.
The rating agencies are one of many players with
historically well-defined roles in the credit and structured
finance markets. We believe that addressing the current
challenges in the credit markets, including the general loss of
confidence among many individuals and institutions, will
require action on the part of all market participants. We are
eager to work with the Congress, regulators, and other market
participants to that end.
Over the past several months, Moody's has been working
constructively with various global authorities, policymakers,
and others to identify and begin implementing initiatives that
can enhance confidence in the global credit markets. We have
been cooperating fully with the SEC in its review of these
issues. That review has been extensive, and it is continuing.
The President's Working Group in the U.S. and the Financial
Stability Forum internationally also have examined the current
market turmoil and developed a series of recommendations for
addressing it.
We believe that implementing these measures globally can
have a positive impact in helping to address some of the
current issues in the credit markets. And we have already begun
to adopt many of these recommendations.
Moody's has always been committed to continuously improving
our ratings processes and analytic capabilities. We have
recently undertaken several significant initiatives to enhance
the quality of our analysis, address concerns in the
marketplace, and further improve the usefulness of our credit
ratings to investors. These measures include steps to: enhance
our analytical methodologies, enhance our review of the due
diligence process conducted by originators and underwriters,
provide more clarity about the credit characteristics of
structured finance ratings, promote objective measurement of
ratings performance, continue effectively managing potential
conflicts of interest, and enhance investors' understanding of
the attributes and limitations of our ratings.
Let me elaborate on two of these initiatives.
Moody's has implemented several measures to further
demonstrate the independence of our rating process. These
include formalizing the separation of our ratings-related and
non-rating businesses, enhancing our credit policy function,
and codifying our existing policies about analysts'
communications with issuers.
We are also implementing a lookback review to confirm the
integrity of analysis performed by any analyst who goes to work
for an issuer or issuer's agent that he or she covered while at
Moody's.
We have also undertaken a review of our rating system for
structured securities. We have proposed five different
potential alternatives to the current structured finance rating
scale and asked market participants for their reactions to
these proposals. Those alternatives could include moving to a
completely new rating scale, adding a modifier to ratings on
the existing scale to identify them as structured finance, or
adding a suffix to the existing rating scale to indicate rating
volatility risk.
Finally, recent events show how rapidly and dramatically
markets can change in today's global economy. That is why we
believe improvements to all market practices, including
improvements to credit analysis, must be pursued vigorously to
restore confidence in credit markets. We are firmly committed
to the effectiveness, integrity, and transparency of our rating
methodologies and practices. In this regard, we look forward to
continuing our dialog with the authorities and market
participants to help strengthen confidence in the financial
markets.
I am happy to respond to any questions.
Chairman Dodd. Thank you very, very much. We appreciate
your being here.
Mr. Joynt, we thank you very much.
STATEMENT OF STEPHEN W. JOYNT, PRESIDENT AND CHIEF EXECUTIVE
OFFICER, FITCH RATINGS
Mr. Joynt. Thank you, Mr. Chairman, for inviting me. My
name is Steve Joynt. I have been President and CEO of Fitch
Ratings. I have been with the company for 18 years, and I have
been President for 12 years, so I have a good degree of
experience in the industry. I am happy to answer your questions
after some brief remarks.
The past 10 months have seen continuing deterioration in
first the U.S. and then in global fixed-income markets. Severe
asset quality deterioration in the U.S. subprime market and
related CDO securities initially caused large market price
declines that required revaluations of these securities by
financial institutions because ultimate credit losses are now
expected to be far greater than anyone anticipated.
Today's market stresses, however, have become more broad
based and emanate from a global reassessment of the degree of
leverage and the appropriateness of short-term financing
techniques inherent in today's regulated and unregulated
financial institutions. Deleveraging is dramatically reducing
liquidity and contributing to price volatility. Many financial
market participants today are seeking ways to enhance stability
in the system.
Fitch's contribution to a better functioning market
requires a reassessment of the changed risk environment, rating
changes that reflect these changes in risk, ratings that are
more stable and reliable, an improvement in our analysis and
modeling techniques, and, finally, full transparency so
investors and all market participants can understand and use
our ratings to supplement their own risk analysis and their own
decisionmaking.
Like all of the major rating agencies, our structured
finance ratings have not performed well and have been too
volatile. We have downgraded large numbers of structured
finance securities, particularly in the subprime mortgage and
CDO areas, in many cases by multiple rating notches. While we
still expect almost all AAA securities to pay off, we have
downgraded many, and some previously highly rated securities
are at risk of incurring losses in the future.
While we were aware of, and accounted for, the many risks
posed by subprime mortgages and the rapidly changing
underwriting environment in the U.S. housing market, we did not
foresee the magnitude or velocity of the decline in the U.S.
housing market nor the dramatic shift in borrower behavior
brought on by the changing practices in the market. We also did
not foresee and are surprised by the far-reaching impact the
subprime crisis has had on markets throughout the world.
Understandably, the rating agencies have lost some
confidence of the market for which I am disappointed. I think
it will be a long and difficult road to win back confidence. We
have, however, aggressively started down that road, and we
believe we are making progress, although slowly.
To win back investor confidence, we simply must do a better
job with our structured finance ratings and all our ratings.
Our structured finance ratings must be more predictive and
stable. Our research and analysis must be more forward thinking
and insightful. We must tell investors about what might happen
tomorrow instead of just what has happened yesterday. We, of
course, remain committed to ensure that our work is of the
highest integrity and objectivity.
We have reevaluated our ratings across all structured
finance areas and the financial services industry broadly as
the credit turmoil has progressed. We are working hard to
anticipate what might come next. Fitch has also been busy
reassessing our structured finance criteria and models,
changing them to reflect what we have observed in this turmoil.
It has been our belief that we best serve the market by
concentrating our efforts on improving our ratings, our
criteria, and our models before doing anything else.
As we conduct this work, we have decided to stop rating new
issues in some structured finance markets that have experienced
some of the greatest turmoil, such as CDOs. We will remain out
of these markets until we can assure the market and ourselves
that we have adequately updated our models and criteria to
reflect what we have observed during this turmoil.
The world's financial infrastructure has become
increasingly interconnected, and it seems as a result that
credit ratings have become increasingly important to all market
participants. Unfortunately, we have come to learn that ratings
have been used in some cases as a proxy to measure liquidity
and market risk, which ratings were never designed to address.
Accordingly, we must do a better job at providing ratings and
additional tools that allow investors to better assess risk in
this increasingly complicated environment.
We have been busy working with the other rating agencies as
a group to increase transparency and the quality of ratings and
to address the many varied concerns of regulators around the
world. Here in the U.S., we have worked with the SEC
extensively in their extensive examination of us. They began
their formal examination last September. They have been
conducting a thorough examination. We believe that will prove
constructive to the SEC as it undertakes the important work
Chairman Cox described, considering new rules for credit
ratings and credit rating agencies. We support their efforts to
improve transparency, integrity, and quality of ratings, and we
believe their work will aid our efforts to win back investor
confidence.
We have been actively meeting with the staff of this
Committee and the staff of the House Financial Services
Committee, who have both taken a leadership role in
understanding this market turmoil. And since last spring, we
have been meeting with the Treasury Department, many bank
regulators, State insurance commissioners, and many State and
local officials, as well as the broad base of investors to
share our perspective and gain insight from them.
Thank you. I am happy to answer your questions.
Chairman Dodd. Thank you very much.
Mr. Cifuentes, thank you for being here.
STATEMENT OF ARTURO CIFUENTES, Ph.D., MANAGING DIRECTOR, R.W.
PRESSPRICH & CO.
Mr. Cifuentes. Good afternoon, Chairman Dodd, Senator
Shelby, Members of the Committee. My name is Arturo Cifuentes.
I am an investment banker based in New York. Thank you very
much for the opportunity to be here. I am really honored to
have my opinion considered in the matter at hand.
I submitted yesterday a long sort of statement with my
recommendations and my views. I am not going to read it here. I
am just going to make a couple of points which I think are
relevant. As I said in my testimony, just for the sake of
clarity, my opinions here are my own opinions, for good or for
bad. I do not intend to represent anybody but myself.
One thing that we have here and I think is important to
realize is that the press and in general there has been a view
that the U.S. is having a credit crunch or a subprime crisis.
Actually, I happen to believe that that is true, but actually
the situation is far more serious than that.
What we really have is the collapse of the alternative
banking system, and by that I mean the system of finance that
was created with securitization and credit derivatives, and
that is very unfortunate because that was a big engine of
growth behind the U.S. economy, and now there is a limit of
trust. The market does not seem to be really convinced that the
structured finance ratings are accurate, and that has impacted
that market; for example, the asset-backed commercial paper is
impacted. That is a very serious problem. So I think there is
an issue of trust here.
The other thing we need to keep in mind, this is a very
global market; 50 percent of the participants in the fixed-
income market are outside the U.S., and they trust the market
because they trust the transparency of the market and they like
the ratings. We are at the risk of losing that right now.
Now, we are going to march into the right thing. I think it
should--I mean, we have this view that the rating agencies are
getting quite a few things wrong, but I think it is important
to realize the nature of the problem. The rating agencies,
unfortunately, initially rated the mortgages wrong. So there
was a mistake there. For whatever reason, the rating of the
mortgages was wrong.
Then there was a second mistake, and I am going to use a
term that sounds a little bit technical here, but,
nevertheless, we have to mention it: CDOs of ABS. So I included
a diagram here at the end of my testimony to clarify the issue
of what is a CDO of ABS. But the point is there was a second
mistake there. So, in addition to the wrong ratings on the
mortgages, we had the wrong ratings on the CDOs of ABS. These
were securitizations that included already the mortgages.
And the further reality is also this happened at the same
time, so they all got it wrong at the same time, which really
magnified the problem. Now, I am not trying to suggest that the
rating agencies acted in coordination to give the wrong
ratings, but I think the system somehow encourages that kind of
outcome.
I just want to make one issue that might sound a little bit
theoretical, but I think we need to keep it in mind, and then I
am going to make a couple of recommendations.
If you remember, initially the ratings were created with
the whole purpose of giving investors information regarding
credit risk. That is it. Information for investors. And that
was fine. Later, the regulators sort of took advantage of that
and decided to use the rating as a proxy for other things, I
mean, for example, capital requirements, where the bank needs
enough resources, et cetera.
Now, so we have two constituencies right now. We have the
regulators using the ratings, and then we have here the
investors. It is not obvious to me that both have the same
goals in mind. It is not that they have contradictory goals,
but a rating which is good for the regulator, presumably a
rating that needs to be more stable, is not necessarily a
rating useful for somebody who is an active participant trading
securities in the secondary market. I mean, there is a little
bit of--I do not want to say conflict of interest, but it is
not clear that these two things are the same. So something to
think about there.
I want to also mention something that, unfortunately, in my
opinion, has taken a great deal of attention in the press and
everywhere, and I think it is the wrong issue, and it is not a
good idea to spend a lot of time talking about that because it
is not the main problem. There is a much more serious deal.
We have talked about the conflict of interest because
allegedly the investment banker pays the fee to the rating
agency. I believe that is not the case. In reality, what
happens, an investment banker raises money. You issue the
securitization bonds, and at the same time, everybody gets
paid--the rating agencies, the lawyers, the trustee, et cetera.
So I do not believe there is a link there between the--I mean,
paying attention to that potential conflict of interest, in my
opinion there is no problem there.
In addition to that, there has been the thought that the
rating agencies have somehow been involved in designing this
concept. Having been on all sides of this business, that is
simply not the case. You have the regular give-and-take between
what could have been an architect that wants to build a
building and the city engineer telling him what he can and
cannot do. So that is not really a serious problem.
What I do believe is a serious problem--and if you remember
one thing of my testimony, I think that is probably the key
point here. We need to have a Chinese wall. We have gone
through this road before. This is the same situation we had
when we had the issue with the research in investment banking
before. You remember at that time there was a gentleman or a
lady writing research on the research side, and then there is
the business side. So there is a very serious conflict of
interest here. Evidently, if you have a rating analyst who is
rating something and the person who is supervising the analyst
is more concerned about credit risk, creates a very serious
problem of interest.
Now, this is more serious than investment banking or any
other activity because if I do not like the research that the
bank writes, I just do not read it, or I toss it and do not pay
any attention to it. But the rating agencies have regulatory
power. So the opinion of the research analyst or the credit
analyst is very, very relevant. So I think the idea of having a
Chinese wall in which analysts will be protected, I think it is
something that we should think about.
The other point that I believe--I made a few points there,
some correlated highly with what Professor Coffee said, so I am
not going to expand on that. But one thing that puzzles market
participants at this point, because they do not believe very
much in ratings--and when I say ratings, I mean just for the
sake of clarity, I am talking about structured finance ratings.
I am not talking about corporate debt or emerging market or any
security. Well, it seems like the rating agencies got it very
wrong in the structured products, and so people wonder what
else do they need to do in order to prevent from backing the
security--yes, I mean, it seems like it could be unfortunate.
We might be in a situation in which we have only three ratings
agencies, and there is nobody on the horizon. That is why I am
a little bit concerned about the 3-year requirement in terms of
operating as a rating agency before you are approved. And I
would pose that it is difficult to operate as a rating agency
and making any money if you are not allowed to issue real
ratings. I mean, you would need venture capital or somebody
willing to finance you for 3 years.
One final point that I would like to make, and, again, it
might sound a little bit academic here, but we have been
talking about ratings, and we have been talking about mortgages
and subprime, et cetera, et cetera. Well, that is fine, but in
my view, that is 50 percent of the problem. The other 50
percent of the problem are the CDOs of ABS, which in my view
were rated using wrong assumptions.
If you look at what happened in 2007, CDOs of ABS I believe
accounted for more than 90 percent of CDOs downgrade. As I show
in my diagram here, this is the securitization, so using
information that is probably contaminated or something like
that.
So, I mean, that is something to look into. That market
obviously is completely paralyzed today, but just a casual
inspection of the morals and assumption that were done for CDOs
of ABS, it seems to me that maybe they were a little bit too
relaxed. I mean, that is my impression based on some
preliminary observations.
The only thing--I think I am going to stop here. The only
comment that I would like to make just in response to some of
the things that have been said is that, well, maybe the reason
we are seeing this massive amount of downgrades is because
there is a unique situation, and the U.S. housing market maybe
is having an extraordinarily bad time. Well, there is some
truth in that, but I happen to believe that the argument is a
little bit circular, because we would not be having this
situation if somebody would have said initially, look, I am not
going to allow you to put these mortgages in these CLOs because
they are really bad. So I would make the case that perhaps the
situation was exacerbated--in fact, it was not stopped because
some of the ratings in these mortgages were not particularly
accurate.
So I think I am going to stop here. I thank Chairman Dodd
and Senator Shelby, and I really appreciate being here, and I
would be happy to answer whatever questions you may have.
Chairman Dodd. Well, thank you very, very much, and I will
ask the clerk to keep the clock on about 5 minutes here so we
can get around. We have kept you a long time this morning, and
we will probably have a lot of additional questions to raise
with you.
Let me, if I can, jump right in. Ms. Robinson, let me begin
with you on the due diligence issue, if I could. It has been
raised earlier. You have heard the conversation. I think
Senator Corker was sort of talking about it to one degree.
Senator Shelby raised it and others have as well. And I am
looking at Moody's Code of Professional Conduct, and let me
quote it. It says, ``Moody's has no obligation to perform and
does not perform due diligence with respect to the accuracy of
information it receives or obtains in connection with the
rating process. Moody does not independently verify any such
information.'' That is of June 2005. Now, it may have changed.
Maybe that has changed since then. If it has, you will correct
me.
Obviously, when you have got a proliferation of liar loans,
as we know about, the no-doc loans going forward here, how do
you answer the question that obviously you probably have been
asked before, that a rating agency should not be required to
perform some due diligence when you are branding these bundles
as being AAA, and yet not a heavy due diligence would have
informed one that these products were anything but investment
grade. I mean, these were products here that were very shaky,
and how do you make that case that there is not a requirement
here since so many people are relying--their long-term
financial security, the security of a municipality,
foundations, colleges, all these things depending upon that,
and that there is no requirement for any due diligence and a
Code of Conduct of ethical conduct when so much has been at
risk here and so much lost for people, how do you address that?
Ms. Robinson. Well, Mr. Chairman, the accuracy of the
information that we receive is central in importance to our
analysis. And we rely on the work of other parties to verify
and establish the accuracy of that information. So, first of
all, it is primarily the responsibility of the issuer and the
loan originator to provide information to rating agencies and
others that is accurate.
Furthermore, the underwriters, the investment bankers who
market the securities have an obligation to perform due
diligence on the loans included in the securitization. And,
finally, information presented in the offering documents
associated with those securities is vetted by accounting firms.
And so I agree that the accuracy of the information is very
important, but there are others whose role it is to check and
verify that information.
Chairman Dodd. But people are relying--I mean, people are
saying this is--Moody's puts its Good Housekeeping Seal of
Approval on this. That is what people are counting on. I am
counting on when you say this is AAA, Senator Dodd, this is a
good product here, I am saying, you know, Moody's told me so,
Moody's gave me that advice. And you are suggesting to me here
that you do not bear any responsibility to me as someone who is
counting on you here to do any kind of work at all to let me
know that something is not--I understand that others have
obligations, but what is the obligation of the rating agency if
not to do some homework on this, so that when I count on you to
give me that recommendation, that there has been some work that
has caused you to draw that conclusion, not some lesser
conclusion about it?
Ms. Robinson. Well, Mr. Chairman, our role as a rating
agency is to provide our best opinion about the credit risk
associated with the securities that we rate. And our opinion
really goes to the creditworthiness of the securities.
Chairman Dodd. Well, let me--I do not want to--I have
limited time here. Let me go to you, Mr. Joynt. We thank you
for coming today.
Mr. Joynt. Sure.
Chairman Dodd. All my colleagues, we express our gratitude
to you.
Mr. Joynt. Thank you.
Chairman Dodd. And I raised earlier Professor Coffee's--
which I thought was a very, very good point, the staleness of
data. And I do not recall your exact statement in your opening
remarks, but you talked about an obligation to sort of be
current. At least that is how I read your statement. And yet
here you have downgrades that did not occur--here you had--they
came after the Comptroller of the Currency had drawn his
conclusions. They come after, months after the agencies
themselves know there are problems in the markets. You know,
you say you were surprised by this. When we met here last year
and asked the Federal Reserve staff, when did you have any idea
this problem was becoming--I was stunned as the new Chairman of
this Committee to learn it had been 3\1/2\ years earlier that
they began to identify a problem. Now, that is a separate
issue. But the fact of the matter is how could you be stunned
if you--it seems to me if you were doing your work in this
area, one, how do you get stunned by it as a rating agency?
And, second, what about the staleness of the information? Why
can't we do a better job here? When you are getting the
Comptroller of the Currency and regulatory bodies acting and
yet still the downgrades do not occur until months after that
occurs, I mean, the credibility has been shot here.
Mr. Joynt. Yes. So I think the awareness of the problem
from subprime loans in the first instance was most obvious, to
us at least, in the beginning of last year and only started
being reflected in the delinquency data that we were seeing in
the securities that we were looking at in a way in which we
could incorporate that new information into our modeling.
Chairman Dodd. Did you pay any attention when the
Comptroller of the Currency--I mean----
Mr. Joynt. Of course. Also, we need to recognize that we
reflected, as did others, that the underlying loans were quite
poor quality, and so when we are speaking about giving high
ratings, behind those high ratings was a large amount of
subordination. So there was a recognition that the loans were
not--were subprime, were very weak. So it obviously was not
enough recognition in hindsight, but it was not like we were
unaware of these being weak loans. We were not aware, it is
certainly true, and did not do the due diligence function of
trying to recognize whether there was fraud involved in the
origination of loans. That is certainly true. And I believe
that has become one of the biggest accelerants for why there
have been problems so across the board in the mortgage markets
itself, so extending to all, and even prime mortgages now. So
that part we did not do. But we were aware of the weakness of
the loans. We were aware that the securities in being put
together were tranched so that senior classes were supported by
junior classes. But I----
Chairman Dodd. But did any of you think--were any of you
facing liability that someone could sue you for not being
forthcoming with information, that that might change the
reaction of the agencies, the fact that you are sort of
protected under the First Amendment--and I see my good friend
Floyd Abrams here, who I respect immensely as a good First
Amendment lawyer, and I have great respect for the First
Amendment. But the whole idea you are insulated in a sense--
anyone else gives me bad information like that, I can sue them.
I can take them to court.
Mr. Joynt. Yes. So that is difficult for me to answer. I am
not a lawyer. But I would say that our reputation is as
important to us as the money that might come from a lawsuit,
and that has been damaged. So by not being able to be accurate
and forward thinking about our ratings, then you--you are
holding us accountable, not you but everyone, accountable for
that and reflecting on how good the credit ratings are. And so
I think that is pretty significant. We treat that seriously.
Chairman Dodd. Senator Shelby.
Senator Shelby. All this is troubling to me, the role of
the rating agencies, lack of due diligence and so forth.
Professor Coffee, thank you again for coming here to bring some
light to this subject, and I mean this sincerely. You were
succinct about what you believe needs to be done.
Were the rating agencies basically blinded by events? In
other words, the subprime situation was going on. They were
pumping them out, the assemblers of it, and they were rating
them, and they were all making a lot of money. But this product
was a new product, as I understand it, the packaging and
slicing and so forth and rating of subprime loans as opposed to
the old method of very few defaults and so forth. Were they
blinded by greed? Were they blind to the situation? Were they
blinded by the fact that they were telling themselves and
others were telling them that, gosh, their opinion--they just
gave their opinion, it did not mean anything, yet as I said
earlier, it seems to be the linchpin of the financial industry.
What is your comment there?
Mr. Coffee. I do not know----
Senator Shelby. Turn your microphone on.
Mr. Coffee. I cannot tell you whether they knew these
ratings were false. I do not happen to believe that. I happen
to believe that in a bubble market and a time when everyone
sees prices rising and the world getting better and great
profit being received, you do not look too carefully at whether
the data you are receiving is phony. And you are structurally
in a position where you are relying upon the loan originator to
tell you everything because you yourself do not have the in-
house capacity to do that verification.
As we go forward, I think the answer is to try to find ways
to bring third-party verification into the credit rating
process.
Senator Shelby. Do you believe that credit rating agencies
should have some responsibility for what they rate and how they
rate it because so many people rely on it in the marketplace?
Mr. Coffee. Absolutely. They are the unique financial
gatekeeper in that they do not have liability--and I am not
pushing liability remedies as the answer.
Senator Shelby. We know.
Mr. Coffee. But they do not have anything like the risks
and exposure of accountants or securities analysts. And they
are functionally a securities analyst for debt markets. So I
think we should look at the reforms that Congress and the New
York Stock Exchange and the NASD have recently imposed on
securities analysts to reduce conflicts of interest. That would
involve Chinese walls around the rating agency, less consulting
income, and other ways----
Senator Shelby. Consulting income, conflicts?
Mr. Coffee. Well, you have heard Chairman Cox say he is
thinking seriously about this, and I congratulate them, because
I think that deserves a serious look.
Senator Shelby. Do you believe that the SEC--and you teach
law and you are into all this very deeply. Do you believe the
SEC can help remedy this situation?
Mr. Coffee. Well, they can certainly help remedy it. I
think there are some ways in which they have to take maybe some
bolder steps than I have yet heard----
Senator Shelby. Absolutely.
Mr. Coffee [continuing]. About both verification,
staleness, and some way that you can ultimately tell a rating
agency that it no longer is an NRSRO without having to prove
they were personally at fault. If you have to show that they
were personally at fault, we are talking about 5 years of
litigation because they will get challenged in court.
I think the real outlier, the rating agency that has a 50-
percent default rate when the next highest default rate is 20
percent, should not continue to be an NRSRO because too many
people are relying upon them.
Senator Shelby. Ms. Robinson, Moody's 12-month downgrade
rate for global structured finance products reached a historic
high of 7.4 percent in 2007. In a recent Wall Street Journal
article dated April 11th of this year, Moody's President, your
President, Brian Clarkson, was cited as saying that the top
thing that could get a Managing Director fired was inaccurate
ratings. Is this report correct, that inaccurate ratings are
the top thing that can get somebody fired? And if so, what
steps has Moody's taken to hold its executives accountable for
its poor ratings?
Ms. Robinson. Well, the accuracy of our ratings are a
primary concern, and, you know, we are a learning institution,
you know, we like to say, and we are constantly reevaluating
our analysis and our methods to make sure that we incorporate
all of the information that is available to us at the time. You
know, our business really rests on our reputation and the
confidence----
Senator Shelby. And your reputation is in tatters right
now, wouldn't you think, in the financial world, all the rating
agencies, or challenged deeply now? You wouldn't agree to that?
Ms. Robinson. Oh, yes, we are challenged at the present
time.
Senator Shelby. Ms. Robinson, in your written testimony,
you also stated that Moody's tracks debt for more than 11,000
corporate issuers, 26,000 public finance issuers, and 110,000
structured finance obligations--110,000. How often does Moody's
review and, if necessary, update each rating, or do you do that
when you see pandemonium in the marketplace?
Ms. Robinson. Well, to take an example of the RMBS market,
we receive data monthly on all of the mortgage-backed
securities that we rate. We have a separate surveillance team
that is charged with reviewing those ratings, and we review
that data every month.
Senator Shelby. Professor Coffee, what would you say that
the SEC, and perhaps this Committee as the Committee of
jurisdiction, needs to do to make sure as best we can that we
can restore some confidence in the rating agencies and what
they do?
Mr. Coffee. I think there----
Chairman Dodd. The microphone again.
Mr. Coffee. I think there are a number of things. You
already heard me talk about the need for getting some kind of
verification----
Senator Shelby. Absolutely.
Mr. Coffee [continuing]. A mandatory element before you
give an investment grade rating on structured finance. You
heard me talk about currency, and I would say there should be
at least the requirement that you annually reaffirm, republish,
reduce, or withdraw your rating, not just get information but
state it again: I am reaffirming this because I believe this,
or I am upgrading, downgrading, or withdrawing it.
I also would say when you change your model, you have got
to, within 90 days, say we are going to reduce ratings on every
model that would produce different results had it been used
back when these ratings were given. That is what I talked about
earlier with the financial scorecard.
Beyond that, I would tell you that you probably should
disclose all fees. When you give a rating, there is today a
problem of what I will call the hidden advisory fee. You get a
fee as a consultant and as advisor, and you get a fee when you
give the rating. This produces an incentive for what I will
call ``forum shopping.'' You can find out from five agencies
what their fee will be and get it from only the one or two that
give you the highest rating. Forum shopping is a problem. One
way to discourage forum shopping is to require rating agencies
to disclose any fee they have received from an issuer or a
structured finance offering, even if they did not give the
rating, and that could show up on the SEC's website, because
they could show you that for this offering there were four
ratings, two other agencies that got fees but did not rate.
That would tell you there is something funny here that they got
a fee and didn't give a rating. So forum shopping is one of the
problems.
I have also suggested in prior testimony that there is an
SEC rule called Regulation FD which effectively exempts the
credit rating agencies and thereby permits selective
disclosure. There are new agencies coming in that are
subscriber-paid. I wish them well. It is a new form of
competition. But they are not going to get cooperation from any
issuers or underwriters because they prefer dealing with the
agencies that they pay because they can predict what will
happen with the agencies----
Senator Shelby. We have got to change the rules, have we
not?
Mr. Coffee. You have got to change Regulation FD so that
all rating agencies get access to the same data.
Senator Shelby. I hope the SEC is listening.
Chairman Dodd. They are listening.
Senator Shelby. Thank you, Mr. Chairman.
Mr. Coffee. Thank you.
Chairman Dodd. Senator Reed.
Senator Reed. Thank you very much, Mr. Chairman.
Professor Coffee, one of the themes that is constant in
Chairman Cox's comments and the questions of my colleagues is
accountability. Today, before this new regulation is proposed,
other than shareholders, who are the rating agencies
accountable to in a material sense?
Mr. Coffee. It is easier----
Chairman Dodd. Again, you have got to----
Mr. Coffee. It is easier to say ways in which they are not
accountable. They do not have private liability. There is no
regulatory agency like the NASD or PCAOB for accountants that
has jurisdiction over them. It is only now that the SEC is
proposing rules.
Sure, they have a reputation, but in a world in which for
the past there have only been three agencies, it is not a world
where reputation counts as much. And reputation means less when
you are also selling a regulatory license. So even if the
market does not trust you, they will still pay you a fee to get
that regulatory license.
They are left in a position where they are only very weakly
accountable and less accountable than the other major financial
gatekeepers.
Senator Reed. Chairman Cox suggested that when these new
rules are rolled out, there will be a new world, a world in
which the presumptive immunity from even a suit for negligence
would be overturned. Can you comment on that? What is your
sense of this newer world that is emerging?
Mr. Coffee. I have a great respect for Chairman Cox, but
the devil is always in the details. And I do not know what
these new rules will say. I think that there are areas in which
we need some strong rules, and while I thought he gave us a
strong statement, much of it was a little opaque on exactly
what the rules are going to look like. And I cannot evaluate
rules until I see them.
But I do not think absent some kind of either liability
risk of possibility of suspension or forfeiture that we are
going to have the same governmental oversight powers over the
rating agencies that we have over the accounting profession or
the securities analysts.
Senator Reed. A final question, because you have all been
very patient, and I am not picking on Professor Coffee, nor
anyone else. Thank you all, ladies and gentlemen, for your
testimony. But it would seem--I mean, I think the system could
be described as there is absolutely no incentive for an
investment bank that is putting together an issuance and going
to a credit agency to then come back and say you gave us a
lousy rating, because what they are trying to buy is the best
rating, and when they get it, they have got what they paid for.
So there is nothing in the system today for any one
individual to come back and say you did not do the job. And
that goes back, I think, to the same point about it is easier
to list the lack of accountability than the points of
accountability.
Mr. Coffee. And as a result, this market has collapsed. No
longer are there any real estate mortgage-backed
securitizations. There are also very few commercial mortgage
securitizations. Thus, I think the industry does have a common
interest with the regulators. This market is not going to come
back, and there are not going to be fees for rating
securitizations that do not happen, unless we can make the
rating agency credible again.
So I want to focus prospectively, and I think the industry
as well as regulators have to find a way to create confidence,
because without it there are not going to be fees.
Senator Reed. Thank you very much.
Thank you, Mr. Chairman.
Chairman Dodd. Just to make that point--I think I made it
the other day in a hearing here. In the commercial mortgage-
backed security area, last year that industry did $230 billion
worth of business in 2007. And this year, as of late April,
they have done $5 billion worth of business, just by
comparison, to give a sense of the magnitude of the problem.
Senator Corker.
Senator Corker. Thank you, Mr. Chairman. It is another
great hearing, and I just want to emphasize something that Ms.
Robinson said. And, by the way, I want you all to know I am
loath to sort of pile on after the fact. You know, it is not
really what I like to do. I will say in this case that it is
hard not to, OK? But obviously you guys have lost reputation,
credibility. I know recently I called about a specific thing. I
remember the broker saying, ``Oh, this is AAA rated,'' and now
I guess all of us in the world are realizing, What difference
does it make if it is AAA rated?
I would just go back and say to the Chairman that we make
people use these folks, and I think that is something that we
need to look at. We make people use these folks. And then if
they do not use them, in essence, they cannot issue securities.
So that whole situation is something we need to certainly look
at.
But, Mr. Coffee, I really enjoyed your testimony. I have
never taken any law courses. Yours is one I actually wish I had
taken. But what would the third party do that you mentioned
earlier? You talked about a third party being involved in some
verification. What exactly would they do?
Mr. Coffee. Well, what they do today, and they do this for
the underwriters. The underwriters will hire a so-called due
diligence firm--the best known is Clayton Holdings, Inc.--and
they will send a team of investigators out to look at this
mortgage pool. There may be in the old-fashioned real estate-
backed securities, there may have been 5,000 mortgages in this
pool. They will sample it, and they will do the kind of
sampling that is similar to what an auditor might do to say
they are reasonably confident that no more than 10 percent of
these loans lack documentation, no more than 10 percent of
these loans had no equity stake, no more than 10 percent of
these loans had a credit score below the minimum level that the
bank or underwriter wants.
So they will tell you how many of these loans are exception
loans, outside the normal lending criteria. And if you hear
there are 30, 40, or 50 percent, which was the statistics that
were occurring in 2007 and 2006, you now have a warning signal
that tells you this is really dangerous.
I think if you give that information to the rating
agencies, they will respond by downgrading or not giving an
investment grade rating. The underwriters overlooked this in
some cases because they thought their lawyers could write
boilerplate that would protect them from any fraud liability.
But I think the rating agency would be more sensitive to this
if they got the information.
Chairman Dodd. Can I ask a simple question?
Senator Corker. Yes.
Chairman Dodd. Why wouldn't you have the rating agency--why
hire a consultant? Why not just do it?
Mr. Coffee. Because--you heard the numbers--there might be
100,000 securitizations out there that they have to perform
ratings on. The underwriters are already doing this and bearing
the cost. If you give this information to the rating agency,
whether it is the bank that gives it to them or whether it is
the third-party firm, I think you have a way that will work,
and it is more feasible, given the small in-house staff.
Chairman Dodd. Thank you, Bob.
Senator Corker. Of course, no, no. Of course, the fact is
the underwriter is driven to get this product out, too. So
there are actually conflicts there, too.
Mr. Coffee. Lots of conflicts in this business.
Senator Corker. OK. So I would go back--actually, that was
my next question.
Chairman Dodd. How about having the underwriter have some
skin in the game, too? That may increase the likelihood of
accountability, I think.
Mr. Coffee. The underwriter does have liability and is
somewhat better deterred, but the underwriter's liability is
for fraud, and if he puts in a lot of boilerplate disclosures,
it will say, ``We told the market that there was this
problem.''
Senator Corker. There is not, I do not think, any
meaningful liability there. But going back to the liability
issue--and obviously I think all of us are really puzzled to
realize that there is just absolutely zero liability. You would
have to perform--I am talking to the rating agencies now. You
would have to perform lots of due diligence to take on
liabilities, and just sort of the flip side of this is
obviously fees would be very different if you were taking on
liability--is that correct? Rating agency charges would be
much, much higher, much different if you were taking on
liability. Is that correct?
Mr. Joynt. Yes.
Senator Corker. OK. And is it reasonable for us sitting
here, realizing the meltdown that has occurred was reliance
upon--and the reliance that was placed in structured finance
being rated AAA, AA, whatever. Should you have liability? I
mean, would that be a good step forward for your various
companies? Obviously, it would change your entire business
model, but is that something you would actually advocate?
Mr. Joynt. I would say no. I do not think in the case of
the responsibility for the due diligence, which we assume
someone else is doing, that we would have to structure
ourselves in such a way that we would be organized to do that
and charge for it appropriately. But it is not really our main
business function. It is not the business that I really want to
be in. I certainly would not want to take on that business in
order to take on liability in that way. We would prefer to have
a business model that is opinion oriented, and so that is the
business model that we have now.
Ms. Tillman. If I may add, you know, I do not totally
disagree with what Professor Coffee was saying, because in some
of the leadership actions that we are proposing, I think it is
important that we get better disclosure and make more of an
effort on the quality of the information that we are receiving.
Some of the things that we are looking at is just the--yes, the
obligation for the due diligence is on the bankers, it is on
firms like Clayton that do it for the bankers. They have a
whole different business model. But at the same time, we as a
rating agency can request and require a certain level of reps.
and warranties and/or a certification or comfort level that the
types of due diligence that is required to ensure that the
quality of the information we get is, in fact, at that level.
We do look at--I mean, a lot of the comments here make it
appear that we do not do anything. In fact, we do do a lot. We
do take all the loans that are in the pool. We run them through
our models. We make our model assumptions available to
everybody. We publish our scenario analysis. We publish our
criteria.
But I do think that there is an important element around
the veracity and the integrity of the data quality, and I think
that is something the market and the rating agencies need to
deal with.
Senator Corker. I know we have a bit of an interchange. I
just have many questions, and I will wait until after Senator
Menendez. But the last question in this round, Mr. Cifuentes--I
may have pronounced that incorrectly.
Mr. Cifuentes. No. You pronounced it correctly.
Senator Corker. Good. Well, I will not try again.
Mr. Cifuentes. AAA for pronunciation.
Senator Corker. Thank you very much for being here. The
structured finance is basically over.
Mr. Cifuentes. I hope not.
Senator Corker. Well, it----
Mr. Cifuentes. It is in a state of semi-paralysis right
now.
Senator Corker. OK. If you could give us a vision of how
this--whatever potion is going to be used to basically cause it
to move out of paralysis and how you see the industry being, if
you will, in 6 months.
Mr. Cifuentes. Well, your statement was 90 percent correct.
It is not totally paralyzed, but it is very paralyzed.
Basically, what we have right now is asset-backed
commercial paper that is very much--CBOs of ABS, that is
totally gone, and we slowly see a recovery of CLOs, which are
CDOs supported by bank loans. So, broadly speaking, yes, the
structured finance market, it is pretty paralyzed right now. I
hope that is not forever, because as I said, we are talking
about the market in the trillions of dollars. So it is a very
significant amount when it comes to financing.
As I said initially, the only way people and investors are
going to recover the confidence is the confidence in the
ratings. I mean, that is the end of it. I mean, there is
nothing beyond that.
Now, I just want to make a brief comment, if I may, if I
can elaborate on an important point.
Senator Corker. Let me just----
Mr. Cifuentes. Sure.
Senator Corker. That is a big statement. I mean, that is
kind of like the market will return once people believe in the
ratings.
Mr. Cifuentes. That is my hope.
Senator Corker. But based on the scenario that has just
been laid out as to how the ratings occur, how could there be
faith in the ratings when there is no accounting activity,
there is no audit activity, there is no understanding of how
these are really put together. How could there be?
Mr. Cifuentes. But let me--I think that is a very valid
point, and let me elaborate on that, because I work rating CDOs
so I do not want to give the impression that we just feed data
into the computer.
Just to give you an idea regarding--I work rating CDOs, so
I am not familiar with the process of mortgages as to how--but
I will tell you about the due diligence, and I think it is
something that should be recovered, I guess, if it was lost.
I rated, for example, the first French CLO, a CDO done with
French bank loans. So the bank came to us, they told us what
they wanted to do. The first thing we did, I took a plane and I
went to Paris with a colleague of mine. We met with the CEO of
the bank. We looked him in the eye. They showed the loans they
had. They had an internal rating system from 1 to 6. Obviously,
we said fine. We took a sample of those loans, and we gave it
to the people at Moody's who rated bank loans, because I have
no idea how to rate a bank loan. They gave sort of a
correspondence between the internal rating of the bank and what
Moody's had, and then after some verification of the data, we
used that to proceed.
My understanding, my recollection, whenever we did CLOs at
that time, that is the way it was done. The data was verified.
So the point that Professor Coffee made I think is very valid.
A rating which is not based on verification of the data, what
is it? I mean, basically it's some input that somebody told me
that I put into the computer program, and then it comes out OK.
Now, I am not very optimistic about all these things about
disclosure and conflict and things like that because, at the
end of the day, as it was pointed out, the rating is no longer
an opinion. It is an opinion with regulatory power, and you do
not have any choice. I mean, you have to use the rating.
So I think at the very least there should be an element of
serious due diligence, making sure the quality of the data you
are being presented, there is some integrity there. I mean,
what it will define statistical processes so somebody gives you
a pool with, say, 1,000 loans, you can take a small sample, do
some analysis, and at least have a rough idea of how good or
how bad they are. Apparently that was not the case.
Senator Corker. Thank you.
Chairman Dodd. Thank you.
Senator Menendez.
Senator Menendez. Thank you, Mr. Chairman. Thank you for
the hearing, and I appreciate all of our panelists here.
Let me ask the three rating agencies here, in November and
December, all three of your agencies downgraded various Bear
ratings slightly, but they were still investment grade. The day
the collapse was announced on that Friday, you all downgraded
Bear. S&P downgraded them to the second lowest investment
grade. Moody's downgraded them to three levels above junk,
which is non-investment grade. And the question is: Did you
make any changes prior to March? Or is that the right
timeframe? You did something in November and December, but they
were all still investment grade. And you did not do anything
until the collapse. Is that a fair statement?
Mr. Joynt. I do not have the answer off the top of my head
on Bear Stearns. I could look into it for you.
Ms. Tillman. I would say the same thing. I do not have the
answer off the top of my head. I would have to check back and
get back to you.
Senator Menendez. I would have thought that on one of the
biggest challenges we have had that has spurred $29 billion by
the Federal Reserve to prop up JP, you would have thought maybe
that question would have been asked of you. But----
Ms. Tillman. Yes, but I would like to give you the accurate
information.
Senator Menendez. All right. I will look forward to the
accurate information. But from all my research, the answer is
you did not. And in my mind, how is it possible that Wall
Street seemed to know that Bear was in trouble since they
started pulling out and demanding their money back before the
collapse, yet the regulators seemed unaware of the looming
downfall, and we got no signs from the ratings--you know, which
to my knowledge, as I said--you can correct me if your facts
are different, but remain unchanged until after the collapse.
This goes to the heart of this problem. You know, because
at the end of the day, as I listened to Professor Coffee and
Dr. Cifuentes, you know, this is about valuating the underlying
debt, the underlying instruments. And if you do not have a good
sense of that valuation, I do not know how you give these
ratings. And if you do not look at the transition over time, as
Professor Coffee has suggested, how do you continue to maintain
a rating in the midst of Wall Street acting in a different way,
the regulators then following up and nothing changing from the
rating agencies?
Ms. Tillman. If I may, I do know that we had put out
articles and made comments on the prime mortgage market, the
brokerage market, the securities industry. What I cannot tell
you exactly is the chronology in terms of the rating action.
Senator Menendez. But just take for a moment, Ms. Tillman
my facts for a given, just for argument's sake. And I am pretty
sure you will find them to be the case. If those are the facts,
isn't something wrong? Isn't something wrong that you did
absolutely nothing in making the appropriate downgrades until
after the collapse? What good is it to the investors at the end
of the day to have that information after the collapse?
You know, I know you all--I hear you say that you are
listening. I wonder whether you are--you are hearing. I am
wondering whether you are listening. I did not hear anything in
the testimony that leads me to believe that you are ready to
make the fundamental changes that I think need to be made and I
hope the Securities and Exchange Commission, Mr. Chairman, is
going to make, and then this Committee will hopefully instigate
them to move in that direction.
Let me ask you another question. Recently, the example of
MBIA, Fitch downgraded MBIA's rating from AAA to AA citing a
lack of capital. It also called MBIA's outlook ``negative.''
However, S&P and Moody's both kept MBIA's ratings at the
highest level. Before Fitch's announcement, MBIA decided it did
not want to be rated by Fitch anymore.
Now, Professor Coffee, is that an example of rating
shopping?
Mr. Coffee. It may be, but I cannot tell you. I cannot
point the finger and tell you the answer to that question, but
it could be.
Senator Menendez. Clearly, if MBIA saw it was going to get
downgraded, it basically could have said, well, let me pull the
plug and say thanks, but no thanks, because at the end of the
day, there is a consequence to it. And so this whole effort of
transparency and openness that some of us have advocated for
the SEC is incredibly important because it would give people
across the spectrum to say, you know, we went to an agency, we
decided not to take their rating, and that pretty much gives us
at least a cautionary flag at the end of the day.
You know, I do not quite understand how ratings without
valuation with an uncertainty--it is almost like, you know, you
put--whatever you put into a process, it is what you are going
to get out. And if at the end of the day we have ratings
without valuations of the underlying instruments and the change
of these instruments--these instruments have dramatically
changed over time, so understanding the nature of those
instruments and what their underlying values are is incredibly
important. Otherwise, I do not quite understand how a rating
means anything other than the fact that you are largely the
only game in town. There may be a couple other rating agencies,
but last year, at the end of 2007, of the 356,000 asset-backed
securities for which there were ratings, you three did all but
1,000 of them. So that pretty much makes it the only game in
town. And when that game is wrong, there is a real consequence
to the investors in this country. And that is what is at stake.
So, Mr. Chairman, I look forward to working with you to
make sure that we are more aggressive than what I have heard
the agencies are willing to pursue themselves.
Chairman Dodd. Thank you, Senator, very much.
Let me just follow up on the forum shopping issue to you,
Ms. Tillman and Ms. Robinson and Mr. Joynt. You have heard
Professor Coffee talk about how this works. Do you have
anything to add to that discussion? And is the suggestion about
how this works, do you think, a legitimate point?
Ms. Tillman. Well, I think at least from Standard & Poor's
perspective, we certainly do not like the practice of ratings
shopping, if that is what you mean by forum shopping. That is
what I am assuming what you are talking about.
Chairman Dodd. But it is ongoing.
Ms. Tillman. We believe that it does, in fact, happen, yes.
But the difficulty, for instance, at Standard & Poor's is we
know that when someone comes to Standard & Poor's and requests
a rating and then does not choose to have that rating. What we
do not know is then who they eventually go to. So if there is a
way that there is some kind of disclosure that is involved that
can indicate, you know, and let there be transparency around
who does give ratings and who does not and who went to the
rating agency and not, that is certainly something that
Standard & Poor's would feel comfortable with.
Chairman Dodd. How about you, Ms. Robinson? How do you feel
about that?
Ms. Robinson. I think our view is that rating shopping does
exist, and I think issuers naturally wish to obtain the best
rating they can obtain.
One of the ways in which we feel that we can kind of
counterbalance that tendency of issuers is we feel it is very
important that we make sure that investors understand what
Moody's rating approach is and what Moody's point of view is,
because ultimately investors are the users of our ratings. So
although issuers obtain the ratings, it is really ultimately
investors' comfort level and satisfaction with those ratings.
Chairman Dodd. Well, doesn't it help the investor to feel
more comfortable if, in fact, they know that maybe they have
tried to get a rating from someone else and did not get one? As
an investor, aren't I in better shape to be more comfortable if
I know that?
Ms. Robinson. Oh, well, we are fully supportive of efforts
to provide more disclosure in this area.
Chairman Dodd. So you would agree with that. How about you,
Mr. Joynt?
Mr. Joynt. I have a slightly different view than that. I
disagree with Ms. Tillman on the topic of the rating shopping
and the disclosure of the ratings. In the case of MBIA, they
asked to withdraw the rating from Fitch. We have maintained it
so far and subsequently changed the rating to what we thought
was the accurate rating. But they have suggested that we may
not have enough information to keep an accurate rating, and we
are dialoguing and debating that ourselves. That would be based
on public information, the ability to rate on public
information.
Several month ago, a large financing, Texas toll road
financing, a several-billion-dollar financing, we also were
asked not to rate that financing because we thought they were
taking on additional debt load, and our rating was falling
below the A category into BBB, where the other two rating
agencies were continuing ratings at A. So we have been asked
and have had to, because we do not have the information, to
withdraw the rating in that case.
So I think there is an important job for the SEC and
others, in the case of the public finance market, to make sure
there is adequate information outstanding for any of the rating
agencies to do an appropriate rating.
On the second point, I believe that if you force people to
disclose the fact that they have gone to rating agencies and
subsequently not accepted their rating, they will limit their
initial approach to rating agencies, and our view is probably
to the largest and dominant rating agencies so that they do not
have to disclose that they went to others and got more
conservative rating opinions. So that would discourage
competition, I think, in a very significant way.
Chairman Dodd. Professor Coffee, how do you answer that?
Mr. Coffee. Well, first of all, the new agencies are
subscriber-paid agencies, and they are not going to be getting
a fee. So we are not going to have forum shopping to them.
There may be contact and you could ideally disclose any
application or any forum submitted. But I do not think that the
subscriber-paid agencies are going to be deterred by rules that
seek to disclose forum shopping.
Mr. Joynt. Well, we are not sure all the new agencies will
only be investor-oriented, though. Yes?
Chairman Dodd. Well, that is why you have got to apply the
same rules to everybody.
Mr. Joynt. Correct. I think the simple remedy, as I was
suggesting earlier, is if the SEC had one website and you had
all the NRSROs up there, you could say there are three agencies
that gave a rating, two that got an advisory fee but did not
rate, and one that got an application for a rating but it was
withdrawn. You could show that all on one simple chart.
Chairman Dodd. Yes. That was, I think, my first or second
question to the Chairman. And I think he sort of endorsed the
idea. I thought he did, anyway. It was unclear.
Mr. Joynt. I thought he was at least sympathetic to it, but
the devil is in the details.
Chairman Dodd. That was the opaque answer I think you
talked about earlier.
This has been most fascinating, and it is a very important
hearing. Just the hearing itself I think could be helpful to
enlighten, obviously, our colleagues and the Committee and
others who are following this, but also I think important for
the SEC to hear from the office of the legislation about the
direction we would like it to move in. And it is very
enlightening for us to understand how this works and how we can
get it right, because it is a critical component in all of
this. And while we have made some recommendations and
suggestions on how to deal with the underlying problems of
foreclosure, which I think we have got to address, if we do
that and do not also structurally address these issues, then
these problems can recur again. So it is an important issue to
look at.
I thank you all very, very much for being here. I will
leave the record open because there were Members who were not
able to be here this morning who may have questions and others
who were here may have additional questions for you. And I
would ask you to respond in a timely fashion, if you could. But
I am very grateful to all of you for your presence here this
morning.
The Committee will stand adjourned.
[Whereupon, at 1:38 p.m., the hearing was adjourned.]
[Prepared statements, responses to written questions, and
additional material supplied for the record to follow:]
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
RESPONSE TO WRITTEN QUESTIONS OF CHAIRMAN DODD FROM CHRISTOPHER
COX
Q.1. Conflicts--Separate Analysts from Business? Dr. Cifuentes'
testimony contains a recommendation by Mr. Mark Froeba that a
rating agency be required to separate its rating business
function from its rating analysis function. Has the Commission
considered whether a significant conflict exists in this area
and whether it should be addressed by regulation?
A.1. Yes. The Commission recently proposed rule amendments that
would prohibit issuance of a credit rating if the NRSRO or an
affiliate of the NRSRO made recommendations to the obligor or
the issuer, underwriter, or sponsor of the security about the
corporate or legal structure, assets, liabilities, or
activities of the obligor or issuer of the security. The
amendments would also prohibit a person within a NRSRO who
participates in determining credit ratings, or in developing or
approving procedures or methodologies used, from participating
in any fee discussions or arrangements.
Q.2. Timeliness of Updates of Ratings. Professor Coffee's
written testimony states ``the gravest problem today may be the
staleness of debt ratings.'' What standards should NRSROs
observe in updating ratings and in withdrawing obsolete ratings
for the benefit of investors and the integrity of markets?
A.2. The Commission believes credit ratings should reflect
current assessments of the credit worthiness of an obligor or
debt security. Consequently, NRSROs should have policies and
procedures for monitoring and reviewing existing credit
ratings. Furthermore, the Commission recently proposed new
rules and rule amendments to require greater disclosure about
the NRSROs' procedures and methodologies for monitoring
existing ratings, including how frequently ratings are reviewed
and whether different models are used in the initial rating and
monitoring processes. This proposal is designed to provide the
market with sufficient information on the surveillance
processes of the NRSROs to allow for comparisons with respect
to how actively they monitor and review existing ratings.
Q.3. Due Diligence. You testified that ``The Commission's
intent is to promote greater due diligence by market
participants.'' Would the quality of ratings improve if NRSROs
themselves performed some form of checking or due diligence on
the data they receive before issuing ratings?
A.3. Because of the sheer volume of securities they rate,
credit rating agencies may be less suited to performing due
diligence than issuers and underwriters. But this should not
relieve credit rating agencies of the responsibility to ensure
that their ratings are based on reliable information, even if
the due diligence is performed by others. The Commission
recently proposed new rules and rule amendments that would
require disclosure as to the level of verification performed by
issuers and underwriters and NRSROs, and how the NRSROs take
that verification into account when determining credit ratings.
------
RESPONSE TO WRITTEN QUESTIONS OF SENATOR SHELBY FROM
CHRISTOPHER COX
Chairman Cox, in his written testimony, Professor Coffee
notes that because only a limited number of investment banks
underwrite structured finance products, they have leverage over
the rating agencies. If they don't like the ratings they get
from one rating agency, they can go to another rating agency
that has lower standards. Since a few investment banks control
which ratings agencies receive the large revenues that come
from rating structured finance products, rating agencies may be
compelled to lower their ratings to remain competitive.
Q.1.a. Do you agree with Professor Coffee about the market
power of investment banks over rating agencies?
A.1.a. This is one of the issues the Commission is reviewing as
part of its examination of the role of credit rating agencies
in the credit market turmoil. The Credit Rating Agency Reform
Act and our proposed new rules recognize that this could
happen, and therefore provide broadened competition and
transparency as a remedy. At this date, the Commission has not
reached any final conclusions as to whether investment banks
unduly influenced the rating process.
Q.1.b. Has the SEC's investigation of the rating agencies
revealed evidence that (1) the rating agencies compromised the
integrity of their ratings in order to increase their profits,
(2) there is a relationship between securities that have been
downgraded and the investment banks that underwrote them or the
credit rating agency that rated them, or (3) investment banks
actively steered business to the rating agencies with lower
standards?
A.1.b. The Commission will be making a formal report to you of
its examination findings on this question very soon. (1)
Preliminary observations suggest the credit rating agencies
were in fact focused on how the ratings they issued influenced
their market share. (2) The staff's preliminary evaluations
have not found any significant relationship between the
securities whose ratings were downgraded and the investment
banks that issued those securities. In addition, examiners have
not found a link between the downgraded securities and certain
credit rating agencies. (3) The ongoing reviews have not found
indications that investment banks actively steered business to
the rating agencies with lower standards.
Chairman Cox, many institutional investors can purchase
only securities rated by the rating agencies listed in the
investment guidelines that govern their funds. Because S&P and
Moody's have historically dominated the ratings market, the
investment guidelines for many investment firms list only one
or both of those firms. It has been suggested that the fact
that investors do not regularly update or re-consider which
rating agencies are specified in their investment guidelines
places new rating agencies at a competitive disadvantage. Even
if a new firm produces better ratings than S&P and Moody's,
investors may still have to use S&P and Moody's ratings due to
the requirements of their investment guidelines.
Q.2.a. As matter of good business practice, should
institutional investors regularly review their investment
guidelines and conduct due diligence to determine which credit
rating agencies' ratings their guidelines should require?
A.2.a. Yes. As with any number of institutional investors'
screening methods and evaluation criteria, it is prudent for
those investors to periodically review their guidelines that
incorporate credit ratings. In conducting a review,
institutional investors should consider the reliability of the
agencies on whose ratings they may rely and consider available
alternative rating firms. Moreover, for many institutional
investors, a security's rating likely would operate only as a
starting point in a reasonable due diligence process. Further,
where modeling is a significant part of the rating process,
institutional investors should develop an understanding of the
credit rating agencies' models. For example, that understanding
could include the various risks those models seek and do not
seek to capture.
Q.2.b. If institutional investors reviewed more regularly the
rating agencies listed in their investment guidelines, would it
provide an additional incentive for rating agencies to produce
high quality ratings?
A.2.b. Yes. To the extent they do not do so today,
institutional investors' periodic review of the efficacy and
adequacy of their investment guidelines, including the
reliability of credit ratings and the firms that issue them,
could provide an additional incentive for credit rating
agencies to provide higher quality ratings. In addition, it
would be useful if issuers seeking ratings and the rating
agencies themselves were fully aware of investors' perceptions
of, and perspectives on, both those agencies and the ratings
they issue.
Chairman Cox, S&P, Moody's, and Fitch indicated in their
testimony that they are taking steps to make it easier for
investors to understand the methodologies used in rating
different types of securities. However, they have stopped short
of proposing that different symbols be used to distinguish
ratings on corporate, structured finance, and municipal
securities.
Q.3. Would having different ratings symbols for each rating
category provide investors with useful information about the
nature of those ratings?
A.3. Yes. However, there are also questions about the costs of
such a requirement, which the Commission is carefully
evaluating. Given the reliance of some investors on ratings of
subprime securities, the Commission has proposed requiring
NRSROs to provide investors and other users of credit ratings
with more useful information about credit ratings and processes
used by credit rating agencies to determine credit ratings. An
amendment proposed by the Commission would require a NRSRO to
attach a report each time it publishes a credit rating for a
structured finance product that describes the rating
methodology used to determine the credit rating and how it
differs from the determination of a rating for any other type
of obligor or debt security, and how the credit risk
characteristics associated with a structured finance product
differ from those of any other type of obligor or debt
security. A NRSRO would not be required to attach that report
if the rating symbol identifies the credit rating as relating
to a structured finance product as distinct from a credit
rating for any other type of obligor or debt security.
Recognizing that market participants have a range of views on
the symbology approach and whether it would be effective,
particularly from a cost-benefit analysis, the Commission looks
forward to the public's comments on this proposal.
Q.4. Chairman Cox, presently if an investor wants to compare
the accuracy of the ratings of different rating agencies, could
an investor easily obtain the necessary information? How would
the proposals you outlined in your testimony, if adopted, make
it easier for investors, analysts, and scholars to analyze the
accuracy of ratings?
A.4. Currently making comparisons across NRSROs is difficult.
For that reason, the Commission recently proposed new
disclosure requirements designed to assist investors and others
in comparing the performance of NRSROs. Under the proposed new
rules a NRSRO would need to provide transition statistics for
each asset class of credit ratings for which an applicant is
seeking registration broken out over 1, 3, and 10 year periods.
Both upgrades and downgrades would have to be included in these
statistics. In addition, default statistics would show defaults
relative to the initial rating and incorporate defaults that
occur after a credit rating is withdrawn. These new rules would
make it easier for academics, investors, and others to compare
how different NRSROs initially rated a security, and whether
they subsequently changed the rating.
Q.5.a. Chairman Cox, during our last hearing on rating
agencies, this Committee heard testimony that the use of
ratings by NRSROs in financial regulation creates artificial
demand for NRSRO ratings. Because financial institutions must
obtain NRSRO ratings to satisfy regulatory requirements, there
is a demand for ratings even if they are inaccurate. Does
demand for NRSRO ratings for regulatory purposes reduce the
incentive for credit rating agencies to produce accurate
ratings?
A.5.a. Not necessarily, but it could reduce the incentives of
investors to be critical users of the ratings. Of course,
ratings are used for a variety of purposes. One of the major
uses of ratings is by issuers to give confidence to buyers that
the debt instrument offered for sale is of high quality. The
reputation of the rating agency is critical for that purpose.
The Commission staff does not have any evidence to suggest that
the coincident use of ratings for regulatory purposes reduces
the NRSROs' incentive to protect their reputations by producing
accurate ratings. To deal with the problem of regulatory over-
reliance on credit ratings as a shorthand for achieving other
regulatory objectives, we will soon consider a rule proposal to
provide alternative means of meeting those objectives.
Q.5.b. Should steps be taken to eliminate or reduce the use of
NRSROs in financial regulation? If so, how could this be
accomplished?
A.5.b. Yes. Financial regulators, including the SEC, should
consider the extent to which the use of ratings for regulatory
purposes induces investors to over-rely on ratings. The
Commission is currently reconsidering the use of NRSRO ratings
in its own rules. The Commission proposed new rules on June 25
designed to ensure that the role assigned to ratings in
Commission rules is consistent with the objective of having
investors make an independent judgment of risks and of making
it clear to investors the limits and purposes of credit ratings
for structured products.
Q.6. Chairman Cox, NRSRO ratings are widely embedded in our
economy. We heard testimony at the hearing about the great
weight investors and regulators place on ratings. Do investors
and regulators overly rely on ratings by NRSROs? Has over-
reliance on ratings reduced the amount of due diligence and
risk assessment undertaken in our economy?
A.6. The fallout from the credit market turmoil indicates some
investors relied too heavily on credit ratings for structured
products rather than conducting their own assessment of the
credit quality of the product. While, many of the financial
institutions impacted in the turmoil had devoted substantial
resources to establishing internal risk assessment functions
(some of which ultimately failed to protect them), there is no
question that there is a connection between over-reliance on
ratings and the level of due diligence and risk assessment. The
Commission proposed new rules on June 25 designed to ensure
that the role assigned to ratings in Commission rules is
consistent with the objective of having investors make an
independent judgment of risks and of making it clear to
investors the limits and purposes of credit ratings for
structured products.
Q.7. Chairman Cox, the use of NRSRO ratings for financial
regulation appears to multiply the impact inaccurate ratings
can have on our economy. For example, NRSRO ratings are used in
capital requirements and the SEC's money market rules. This
means inaccurate ratings can allow financial institutions to
hold too little capital, or force them to sell assets that no
longer satisfy regulatory requirements. The need for financial
institutions to raise more capital or re-allocate assets
following large scale ratings downgrades could significantly
affect the economy. If our financial regulatory system had
relied less on NRSRO ratings, would our economy have been
better prepared to weather the impact of the recent large scale
ratings downgrades?
A.7. Yes, because the ratings for subprime-related securities
were categorically wrong due to a variety of methodological
factors that the agencies have since acknowledged. The large
number of subsequent credit rating downgrades played a role in
the credit market turmoil. However, as noted previously,
issuers purchase credit ratings to make their securities
marketable and many of the investors demanding that securities
be rated are not subject to regulations that use credit
ratings. The link between the use of credit ratings in
Commission rules and investor over-reliance on credit ratings
is difficult to quantify with precision.
Chairman Cox, the Credit Rating Agency Reform Act of 2006
sought to increase competition among rating agencies by making
it easier for new firms to become NRSROs. The Act favors no
particular business model. Two firms that use an ``investor
pays'' model have registered as NRSROs. Some have argued that
the ``investor pays'' model has fewer conflicts of interest
than the ``issuer pays'' model because it makes the rating
agency directly accountable to investors.
Q.8.a. How do we go about fostering innovation and further
reducing the conflicts of interest in the credit rating
industry?
A.8.a. The Rating Agency Reform Act of 2006 was designed to
achieve these goals through requirements that promote
accountability, transparency and competition in the credit
rating agency industry. The Commission recently proposed new
rules and rule amendments that are designed to further these
goals in the context of structured finance, including by
requiring more comparable performance statistics, the
disclosure of ratings history, and greater disclosure of the
assets underlying structured finance products and the
methodologies used to determine and monitor structured finance
ratings. The goal is to make it easier for the market to assess
the quality of NRSRO ratings.
Q.8.b. What competitive barriers still entrench S&P and Moody's
in their dominant market positions?
A.8.b. In the past, S&P and Moody's widespread market
acceptance has given them an advantage because issuers and
investors were familiar with their rating record and
reputation. Issuers were inclined to use their services because
they helped issuers sell their securities. Following the
enactment of the Credit Rating Agency Act, it will be easier
for competitor firms to become NRSROs and for the users of
credit ratings to become comfortable with NRSROs other than S&P
and Moody's. Ultimately, the test of their quality and value in
the marketplace will be whether users are willing to pay for
ratings from these other organizations. At the same time, S&P,
Moody's, and the other NRSROs will need to provide
significantly more information to the public to demonstrate the
quality of their ratings and ratings processes. The Commission
recently proposed new rules and rule amendments designed to
require NRSROs to provide more information with which investors
and other market participants could evaluate the NRSROs'
performance record.
------
RESPONSE TO WRITTEN QUESTIONS OF SENATOR MENENDEZ FROM
CHRISTOPHER COX
RATING SHOPPING
Without a doubt, one of the most worrisome practices that
undercuts the accuracy and reliability of ratings is rating
shopping. I am pleased to hear your proposal include
improvements to disclosure, but I am concerned that it doesn't
go far enough to ensure rating shopping cannot occur.
Q.1.a. Would the proposed rules eliminate rating shopping?
A.1.a. This is exactly what the proposed rules are intended to
do. Specifically, the proposed new rules and rule amendments
are designed to target the problem of rating shopping by making
it easier to track the ratings of NRSROs, and by making it
easier for competitor NRSROs to issue unsolicited ratings for
structured products that would allow investors to compare these
ratings. Currently, the information necessary to determine an
initial rating for structured products typically is not made
widely available. If this information were made available to
all NRSROs, those that are not hired to determine the credit
rating could nonetheless issue a credit rating. This approach
is designed to eliminate the potential harm of rating shopping
by promoting unbiased ratings.
Q.1.b. The fundamental problem is that people can still get a
preliminary rating and then decide to go elsewhere. Would
issuers have to disclose if they received a preliminary rating?
A.1.b. The agreement recently reached between the three largest
NRSROs and the New York Attorney General to change the payment
structure in the industry attacks this problem in a slightly
different way, by ending the practice of free preliminary
ratings. Now, issuers will have to pay even if they do not
obtain a rating. The Commission staff believes this will also
avoid the problem that arises if issuers forego approaching
NRSROs in order to defeat a disclosure requirement.
Q.1.c. On the disclosure proposals you outlined--are you going
to require issuers to share material non-public information
with all NRSROs if they provided the same information to one
NRSRO? Can you describe in detail the proposal for disclosure?
A.1.c. The Commission recently proposed new rules and rule
amendments to require the disclosure of information about the
assets underlying a structured finance product that are used by
an NRSRO to determine a rating. The goal is to provide
information to NRSROs that were not hired to determine the
credit rating so they would have an opportunity to issue a
credit rating. The details of this proposal are described in
the attached rule release.
BEAR STEARNS
The Bear Stearns collapse signaled a few problems in our
system, one of which was that we seemed to have no idea how
faulty Bear's assets were until it was too late.
Q.2.a. In the days and weeks leading up to Bear Stearns'
collapse, it appears we received no signals from the ratings,
which to my knowledge were unchanged until after the collapse.
What does this say to you about the reliability of these
ratings? Is this an example of a broader problem, in your mind?
A.2.a. Credit ratings issued by the NRSROs are intended to be
an indicator of the credit risk associated with particular
instruments or issuers. The extremely rapid deterioration of
the financial position of Bear Stearns highlights the
limitations of credit ratings and demonstrates the importance
of considering the total mix of facts and circumstances in
evaluating a firm, rather than relying on any single indicator
of firm health.
Q.2.b. Shouldn't we be able to use the ratings as some sort of
guide about the overall health and the risk of a firm's assets?
Would you say that in the case of Bear, the ratings failed?
Even if this was under extraordinary circumstances, shouldn't
the ratings better reflect the actual risk at hand?
A.2.b. While the NRSROs would argue their definitions of credit
risk reflected by their ratings accurately described the case
of Bear Stearns even though it approached bankruptcy, you are
right that users implicitly expect a correlation between
ratings and performance. While the Credit Rating Agency Reform
Act of 2006 prohibits the Commission from regulating the
substance of credit ratings, our proposed new rules are
intended to increase the accuracy of ratings through better
disclosure, transparency, and competition. As noted previously,
the events at Bear Stearns demonstrate the importance of
considering the full range of information about a firm and
broader market conditions in making judgments about the health
of any firm.
Q.2.c. Is the SEC looking at the credit rating history for Bear
Stearns, specifically the relationships between Bear Stearns
and the rating agencies?
A.2.c. Yes, one of several areas covered by our examination of
the three largest credit rating agencies is the relationship
between issuers and the rating agencies. As a matter of
enforcement policy, the Commission does not confirm or deny the
existence of any ongoing enforcement investigation.
UPDATING RATINGS
Q.3.a. I'd like you to comment on a proposal by Professor
Coffee for the rating agencies to periodically update ratings,
as is done by securities analysts. Is this feasible? Do you
think the SEC could require this within its existing authority?
A.3.a. The NRSROs generally have policies and procedures in
place to monitor each rating and update it as necessary. The
Commission has proposed new rules and rule amendments that
would require greater disclosure about the NRSROs' procedures
and methodologies for monitoring existing ratings including how
frequently ratings are reviewed and whether different models
are used in the initial rating and monitoring processes.
Q.3.b. Do you think the issue of ratings becoming ``stale'' is
a concern? Could we argue that the ratings on Bear were in fact
``stale''?
A.3.b. Yes, this is a concern. The Commission believes credit
ratings should reflect current assessments of the credit
worthiness of an obligor or debt security. As described above,
our proposed new rules tackle this problem through new
disclosure requirements. And although the Commission is
statutorily prohibited from second guessing credit rating
decisions made by the NRSROs, the Commission may evaluate
whether an NRSRO followed its stated methodologies. We intend
to do this through regular examinations.
Q.3.c. Will the proposed rules provide sufficient assurance to
the markets that the ratings are current?
A.3.c. The proposed new rules that require greater disclosure
about the NRSROs' procedures and methodologies for monitoring
existing ratings such as how frequently ratings are reviewed
and whether different models are used in the initial rating and
monitoring processes are designed to provide the market with
sufficient information on the surveillance processes of the
NRSROs to allow for comparisons with respect to how frequently
and actively they monitor and review existing ratings.
Q.3.d. Is the SEC looking at providing additional
interpretation regarding what it means for a credit rating
agency's ratings to be ``current assessments''? Is this an area
the Commission should be looking at, in your opinion?
A.3.d. Yes. As part of the notice and comment process for our
proposed new rules, we expect to receive useful information on
this question. The Commission believes credit ratings should
reflect current assessments of the credit worthiness of an
obligor or debt security, and we will continue to explore ways
to effectuate this principle. The Credit Rating Agency Act of
2006 defines a ``credit rating'' as ``an assessment of the
creditworthiness of an obligor as an entity or with respect to
specific securities or money market instruments.'' Under this
definition an ``assessment'' must reflect the NRSRO's current
view of creditworthiness of the obligor or debt security.
------
RESPONSE TO WRITTEN QUESTIONS OF SENATOR BUNNING FROM
CHRISTOPHER COX
Q.1. Should Reg FD be amended to give all investors access to
the same information that the rating agencies have so as to be
able to judge for themselves whether the agency's opinions are
valuable?
A.1. This is the purpose of the proposed new rules that would
require the disclosure of information about the assets
underlying the structured finance products that the NRSROs
rate. This would allow market participants to better analyze
the assets underlying structured securities, and reach their
own conclusions about their creditworthiness. However, these
new rules are not an amendment to Reg FD, which was designed to
address the problem of issuers making selective disclosures of
material nonpublic information to persons who were likely to
use that information to their advantage in securities trading.
Q.2. Should Reg FD be amended to allow all NRSROs the same
access to information if any NRSRO gets access to that
information?
A.2. While not styled as an amendment to Reg FD, this is the
purpose of the Commission's recently proposed new rules to
require the disclosure of information about the assets
underlying the structured finance products that the NRSROs
rate. This data availability could particularly benefit
subscriber-based NRSROs, who could use it to perform
independent assessments of the validity of the ratings by their
competitors who use the ``issuer pays'' model.
Q.3. If issuers pay the rating agencies for the ratings, how
should investors be protected from rating shopping? Wouldn't it
be better if the users of the ratings paid for them so that
rating agencies that did a bad job and issued inflated ratings
would be punished by the users in the form of lost market
share? Doesn't the current structure reward the softest graders
with increased business?
A.3. The issuer-pay and subscriber-pay models are subject to
different types of potential conflicts. Consequently, I believe
the users of credit ratings are served by having NRSROs that
operate under both models as they serve as a check on the
other. In addition, the Commission recently proposed new rules
and rule amendments that would make it easier for NRSROs to
provide unsolicited ratings for structured products. The goal
is to create a mechanism to expose whether an NRSRO is
employing less conservative methodologies than other NRSROs to
determine ratings in order to increase market share.
------
RESPONSE TO WRITTEN QUESTIONS OF SENATOR SHELBY FROM JOHN C.
COFFEE, JR.
Q.1. If these reports are true, what duties would investment
banks have violated under the securities laws?
A.1. In the case of an offering registered under the Securities
Act of 1933, any investment banks that ignored warnings from
their agents (i.e., the due diligence firms) would almost
certainly face private liability under Sections 11 and 12 of
the Securities Act of 1933 (plus, of course, liability in SEC
enforcement actions based on Section 17 of that Act and Rule
10b-5). The provisions of Section 11 entitle investors who
purchased in the offering to sue for any material omission,
unless the underwriter can establish its due diligence defense
under Section 11(b)(3) that it ``had, after a reasonable
investigation, reasonable ground to believe and did believe, at
the time such part of the registration statement became
effective, that the statements therein were true and that there
was no omission to state a material fact required to be stated
therein or necessary to make the statements therein not
misleading.'' (Similar provisions are also found in Section
12(a)(2) with only modest differences). I seriously doubt that
either affirmative defense could be satisfied if the investment
were on notice that a significant percentage of the loans in
the structured finance product were outside usual lending
guidelines and these facts were not clearly and specifically
disclosed.
In the case of offerings done by means of a private
placement or other exemption from registration, the above
sections will not apply, but the investment banks would still
face liability under Rule 10b-5 if they made a materially false
statement or omitted to make a statement necessary to be made
in order to make the statement made, in light of the
circumstances under which they were made, not misleading.
Again, I think the investment banks who withheld material
information will face a high risk of liability (but a variety
of legal defenses are possible).
Q.2. Should credit rating agencies, as gatekeepers responsible
for monitoring the quality of securities offered in our
markets, conduct an independent assessment of asset-backed
structured finance markets?
A.2. Ideally, yes, because this is what gatekeepers normally
do. But logistically, it may be very difficult for the major
ratings agencies to gear up to take such a step. Thus, a
second-best alternative would be to require that ``NRSRO''
rating agencies not confer an ``investment grade'' rating on a
structured finance product in the absence of receipt of a
verification from an independent expert that the latter had
conducted an investigation, using such sampling or similar
procedures as the rating agency deemed reasonable for these
purposes, and had reached specified conclusions about the
quality of the collateral underlying the security. These
specified conclusions might include that not more than a
defined percentage of the loans were outside traditional
lending criteria (i.e., such as that the borrower had an equity
stake of at least [20] percent in the home).
The point of this alternative is that the underwriters (and
not the rating agency) could bear the cost of this ``due
diligence'' investigation, but the rating agency would get an
independent certification from an expert firm (which should
have both civil and criminal liability for fraud for any
knowing misstatements). Other techniques can also be imagined
by which the rating agencies receive verification from parties
other than the loan originators (and that is what is
important--not that the rating agencies do it themselves).
Q.3. How often should ratings be reviewed and, if needed,
updated by rating agencies?
A.3. All other gatekeepers on whom investors rely for
evaluations--e.g., securities analysts and auditors--do
regularly update their evaluators. With credit rating agencies,
updates are the exception, not the rule. I would suggest two
principles:
First, debt ratings should be reviewed and updated at least
annually. The rating agency could at this periodic moment re-
affirm, change or simply withdraw its rating. But such a
withdrawal would be public and would alert investors not to
continue to rely on a ``stale'' rating.
Second, whenever the rating agency either (a) updates its
model or methodology or (b) realizes that there has been an
error in its model, it should promptly inform the market of the
change that the new model (or the discovery of the error) would
produce. Recent press reports have suggested that Moody's
discovered a computer error in some ratings on European
offerings that resulted in ratings that were three levels too
high--and it did nothing! That is the kind of culpable omission
that should be impermissible.
Q.4. If ratings were required to be updated more frequently,
would it significantly increase the cost of ratings?
A.4. There would be an increase in the cost, which would be
largely passed onto the issuer in all likelihood (given the
weak level of competition in the ratings industry). But the
costs of updating should not approach the cost of the original
rating. In the typical case, the rating agency would already
have developed its methodology and the issue would be largely
whether any information input had changed (for example, had the
default rate on mortgages in a particular location risen
materially?). These will be relatively exceptional events.
Where the agency changes its model, it should be able to
generate all the implications for prior ratings in a single
computer run (using the new methodology). Thus, while I
acknowledge that there would be some cost increases, I do not
believe they would approach the cost increases that Sarbanes-
Oxley imposed (justifiably) on the accounting profession.
Q.5. Would you please comment on Chairman Cox's testimony about
the areas of rulemaking that the SEC is considering?
A.5. It is difficult to comment (and possibly unfair to do so)
until proposed regulations are released. I did hear Chairman
Cox testify that he favored rules restricting the currently
pervasive conflicts of interest in this field, and that is
desirable.
Nonetheless, I do not yet believe that the SEC is
considering rules to require increased verification of
information in the rating process or to address the staleness
problems. No ``solution'' can be adequate until these problems
are addressed. Complex and sophisticated as any computer model
may be, the first rule in this field is: ``GIGO: garbage in,
garbage out.'' If loan originators are not subject to close
scrutiny in terms of the data they provide, the process will
inevitably produce distorted and optimistically biased
conclusions.
------
RESPONSE TO WRITTEN QUESTIONS OF CHAIRMAN DODD FROM VICKIE A.
TILLMAN
Q.1. Ratings Scoreboard. What are your views on the
recommendation that has been made for the creation of a central
website which investors could access and on which they could
compare the accuracy of past ratings by the different NRSROs
for the same types of securities?
A.1. We agree that rating agencies should work towards greater
transparency and disclosure. At S&P, we are regularly
considering new ways to do so. In our experience, the most
effective way to measure ratings performance is through
historical measures such as default and transition studies.
These studies can demonstrate effectively the existence (or
lack) of a correlation between ratings assigned by an NRSRO and
the likelihood of default. At the same time, it is important to
note the broad disparities in rating definitions, criteria and
methodologies used by various rating agencies that help foster
competition in the industry. Meaningful differences exist among
rating agencies, not only in the way ratings are defined, but
also in the way defaults and other relevant credit events are
determined and measured, all of which can affect reported
results. We would not object to having each rating agency make
available--in a central repository--information about its
performance history but would caution that such a repository
must note these differentiations among rating agencies.
Investors can then judge ratings performance and determine for
themselves the value to them of a particular rating agency's
opinions.
Q.2. Due Diligence. Did S&P undertake to verify the information
it used to decide ratings on the structured finance products
that were subsequently downgraded? In recent years, there has
been widespread awareness, through the press and otherwise,
about the proliferation of so-called ``liar loans''--mortgage
loans with little or no documentation required and on which
borrowers ultimately have stopped paying. Do you feel that
NRSROs should have performed some investigation or due
diligence on structured debt that contained these ``liar
loans''?--Are there circumstances under which NRSROs should be
required to perform some form of due diligence before issuing a
rating?
A.2. The information concerning the collateral for the
securitizations that we rate typically comes from the
participants in the transaction being rated: the issuers and
underwriters. S&P is very specific about the data it requires
in its rating process. For example, with respect to U.S.
Residential Mortgage-Backed Securities (``RMBS''), S&P
publishes a detailed list of approximately 70 data points that
it requires issuers to submit with respect to each loan in each
pool that it is asked to rate. We also publish a detailed
glossary of definitions that the issuer must utilize when
providing data to S&P.
S&P does not go on-site to review individual loan files
held by originators and servicers, or perform an independent
verification of the information provided to it in connection
with its ratings analysis. As others at the hearing noted, we
are not auditors and are generally not in the position to
verify underlying data. The participants in the transactions we
rate understand that S&P relies upon them for the accuracy of
the data they provide. These participants issue representations
and warranties in the operating documents for the transaction
we are asked to rate with respect to the loan level data,
regulatory compliance, and other issues and make disclosures
about the collateral in the prospectus.
However, in light of recent events, we determined that S&P
can take steps on our own to improve disclosure of information
on collateral underlying structured securities, and as I
testified, S&P has announced and is implementing a
comprehensive set of new measures designed to improve the
ratings process. In addition, S&P has begun to implement
procedures to collect more information about the processes used
by issuers and originators to assess the accuracy and integrity
of their data and their fraud detection measures so that we can
better understand their data quality capabilities.
Q.3. Timeliness of Updates of Ratings. Professor Coffee in his
testimony pointed out that major downgrades of CDO securities
``came more than a year after the Comptroller of the Currency
first publicly called attention to the deteriorating conditions
in the subprime market and many months after the agencies
themselves first noted problems in the markets.'' His testimony
also states ``the gravest problem today may be the staleness of
debt ratings.'' What is S&P doing to update ratings in a timely
manner and eliminate stale ratings? What standards should
NRSROs observe?
A.3. S&P continually strives to balance the twin goals of
updating its ratings in a timely fashion while also adhering to
its criteria and taking action when and only when it has the
data to support a change in its rating opinion. In response to
recent events, we have increased the frequency of our reviews
of rated transactions. As part of our recently announced
Actions (discussed at greater length in my testimony), we have
undertaken several additional steps to improve the
effectiveness and speed of our surveillance process. These
include:
increasing resources dedicated to surveillance;
continuing to separate our new rating and rating
surveillance functions;
expanding our use of search and market based
tools;
incorporating new capabilities we have gained as
part of our acquisition of iMake, a leading global provider of
structured cash flow models and data; and
developing an early warning indicator to
investors that a key credit quality attribute (e.g.,
delinquencies or losses) of an issue or issuer differs from our
expectations and has or may trigger a full review by S&P
surveillance.
We believe strongly that these steps will improve our
surveillance process and help provide the market with timely
and appropriate ratings updates.
Q.4. Separate Ratings from Business? Dr. Cifuentes' testimony
contains a recommendation that a rating agency separate its
rating business function from its rating analysis function.
What are your views on how NRSROs should address this analyst
independence concern?
A.4. S&P shares Dr. Cifuentes' belief in the importance of
analyst independence and has long sought to protect the
integrity of its ratings analysis and opinions through policies
and procedures designed to promote that independence. For
example, analysts are not involved in negotiating fees. Nor can
S&P personnel who are responsible for negotiating fees vote in
ratings committees. Additionally, we specifically structure our
analysts' compensation so that it is not dependent on the
revenue generated by the ratings they assign. Moreover, S&P's
Analytic Firewalls Policy imposes numerous requirements and
responsibilities on both ratings analysts and other employees
of S&P and the McGraw-Hill Companies in order to ensure that
ratings analysts ``have the freedom to express their respective
opinions free from the improper influence of other Standard &
Poor's/McGraw-Hill employees and free from the influence of the
commercial relationships between Standard & Poor's/McGraw-Hill
and third parties.''
Additionally:
ratings analysts are prohibited from
participating in consulting or advisory services;
ratings analysts are prohibited from cross-
selling of credit ratings or ancillary ratings products and
services with any other S&P or McGraw-Hill product or service;
and
ratings employees are prohibited from joint
selling or calling ratings customers with other S&P or McGraw-
Hill employees.
Q.5. Ratings Shopping. We have heard concerns about ``ratings
shopping,'' where an underwriter or an issuer goes to the NRSRO
that it feels will give it the highest rating, even if it is
not necessarily the most accurate. Is ratings shopping a
problem? How should the negative aspects of it be addressed?
A.5. We believe that ratings shopping is an issue to be
considered by the markets as a whole. One suggested measure to
address the concern is to require structured finance issuers to
disclose whether they have approached rating agencies other
than the ones providing a rating on the applicable transaction.
Q.6. Professional Analyst Organization. Dr. Cifuentes'
testimony suggests ``the creation of a professional
organization, independent of the rating agencies, to which
ratings analysts must belong and which sets forth ethical,
educational and professional standards.'' Please share your
thoughts on the potential merits of such an organization.
A.6. At S&P, we believe that all rating agencies should have
systematic procedures to help ensure that their analysts are
able to identify, understand, and analyze information relevant
to the issues and issuers they rate. In assessing the
competence of analysts, S&P considers their level of education;
experience within sectors, industries and geographic regions;
experience with particular transactions and asset classes and
other specialty areas; analytical ability; decision making;
professionalism; time management ability; leadership; teamwork;
and their written and verbal communication skills. S&P has
adopted and continues to enhance policies and procedures
designed to ensure that its analysts receive sufficient
training and support to facilitate the generation of
independent, objective and credible rating opinions. A major
emphasis of the action plan that S&P announced in February is
strengthening analyst training.
However, given the importance of rating agency independence
and the value of having a diversity of opinions in the market,
we do not believe that Congress or the SEC should impose
minimum standards for analyst training, background, experience,
or other characteristics. Standards along these lines would
replace independent judgments with those of the government. The
result would be more homogeneity, and less innovation.
------
RESPONSE TO WRITTEN QUESTIONS OF SENATOR SHELBY FROM VICKIE A.
TILLMAN
Q.1. Ms. Robinson, Ms. Tillman, and Mr. Joynt, do you think
that it is easy for investors to compare the accuracy of the
ratings of the different credit rating agencies? If not, do
S&P, Moody's, and Fitch favor the SEC issuing rules to require
enhanced disclosure of ratings performance as Chairman Cox
outlined in his testimony?
A.1. We have a long-standing tradition at Standard & Poor's
(``S&P'') of publishing significant amounts of information
about the default and transition history of our ratings. We
believe the studies we publish assist issuers and investors in
their evaluation of the quality of our rating opinions. And we
are always open to considering new ways to inform the public
about what we do and the excellent track record of S&P's
ratings.
We would support having each rating agency make available--
in a central repository--information about its performance
history. Investors can then judge that performance and
determine for themselves the value, to them, of a particular
rating agency's views. I will note, however, that our rating
opinions represent an analytic judgment based on a wide range
of factors, many of which are assessments of future
developments. Rating agencies employ different definitions of
ratings and have different criteria. We believe that although
this diversity of approaches is beneficial to the markets, it
makes an ``apples to apples'' comparison of ratings difficult.
Q.2. Ms. Robinson, Ms. Tillman, and Mr. Joynt, would you please
explain the process by which you obtain the information you use
to rate structured finance securities?
How much of the information is from issuers, underwriters,
or other sources?
Do you ever seek to verify the accuracy of the information
you receive?
A.2. The information concerning the collateral for the
securitizations that we rate typically comes from the
participants in the transaction being rated: the issuers and
underwriters. S&P is very specific about the data it requires
in its rating process. For example, with respect to U.S.
Residential Mortgage-Backed Securities (``RMBS''), S&P
publishes a detailed list of approximately 70 data points that
it requires issuers to submit with respect to each loan in each
pool that it is asked to rate. We also publish a detailed
glossary of definitions that the issuer must utilize when
providing data to S&P.
We at S&P are not auditors and are generally not in the
position to verify underlying data. Currently, the participants
in the transactions we rate understand that S&P relies upon
them for the accuracy of the data they provide. These
participants issue representations and warranties in the
operating documents for the transaction with respect to the
loan level data, regulatory compliance, and other issues and
make disclosures about the collateral in the prospectus.
As I testified, S&P has announced and is implementing a
comprehensive set of new measures designed to further
strengthen the ratings process, including steps to improve the
quality and integrity of information we collect. We are working
with market participants to improve disclosure of information
on collateral underlying structured securities. Specifically:
On transactions closing after May 1, 2008, we are
requesting updated loan level performance data from issuers on
a monthly basis, consistent with data customarily sent to
Trustees and third party data vendors in the U.S. RMBS market.
We are in the process of revamping criteria for assigning
overall mortgage originator rankings based on operational
process and procedures. New criteria should be established by
mid-2008.
We are evaluating various fraud tools and detection
policies used by originators for improved data integrity and
will be incorporating these evaluations in the criteria to be
established by mid-2008.
Q.3. Do you have any reason to believe that inaccurate or
fraudulent data contributed to the poor performance of your
ratings on structured finance securities over the last few
years? If yes, please provide supporting evidence.
A.3. Published reports indicate that data quality and fraud are
among the factors that may have impacted loan performance for
the vintages that have seen worse-than-expected performance, as
well as a host of other potential factors. Certain published
reports also suggest a significant increase in fraud with
respect to recent vintages. For example, the Mortgage Asset
Research Institute, commissioned by the Mortgage Bankers
Association to conduct a mortgage fraud study in 2006, reported
``findings of fraud were in excess of previous industry
highs.'' It noted that key risk variables that have
historically influenced default patterns, such as FICO, LTV and
ownership status were proving less predictive.
Ms. Robinson, Ms. Tillman, and Mr. Joynt, according to
testimony provided by Chairman Cox, Moody's has downgraded 53
percent and 39 percent of all its 2006 and 2007 subprime
tranches; S&P has downgraded 44 percent of the subprime
tranches it rated between the first quarter of 2005 and the
third quarter of 2007; and Fitch has downgraded approximately
34% of the subprime tranches it rated in 2006 and the first
quarter of 2007.
Q.4. What steps have each of your companies taken during the
past three years to hold accountable its executives and
analysts for the poor performance of its ratings? Has your
company dismissed or otherwise disciplined any of the
executives or analysts responsible for overseeing or producing
its ratings of structured finance products? Please provide a
complete list of disciplinary actions.
A.4. Ratings transitions, even significant transitions, do not
reflect errors in our initial analysis as they could be caused
by a multitude of unforeseen factors such as the unprecedented
market conditions we are currently experiencing. Moreover, the
downgrades of our 2006 and 2007 subprime tranches do highlight
the success of our surveillance procedures in place at S&P as
we adapt to turbulent market conditions.
Additionally, S&P considers personnel actions to be
confidential and does not--as a rule--discuss publicly reasons
for promotions/demotions/dismissals.
We have been listening to and learning from the concerns
and criticisms raised about our industry. We take very
seriously our responsibility to implement whatever measures we
can to improve the way we do business consistent with our role
in the financial markets. As a result of our ongoing commitment
to improve our rating process, we recently announced that we
are adopting wide-ranging set of new measures to increase
responsiveness and accountability at S&P from top to bottom.
Among these numerous initiatives, we are increasing the
annual training requirements for our analysts, expanding the
scope and the course offerings of our training programs,
including increasing our focus on policy requirements and
compliance, and are establishing an analyst certification
program in partnership with an academic institution. We also
recently created and filled two new executive positions in the
areas of risk oversight, criteria management and quality
assurance. These changes add strength and depth to our ratings
leadership and capabilities, and demonstrate S&P's commitment
to serving the broad and growing needs of the global credit
markets. Among other things, we named a new Executive Managing
Director of Ratings Risk Management, who will be responsible
for identifying, assessing and mitigating potential internal
and external risk exposures in our ratings business. Also we
split Quality and Criteria governance responsibilities into two
separate functions to further strengthen their respective
independence and effectiveness.
Consistent with these Actions and with our ongoing efforts
to do what is best both for our company and the financial
markets we serve, we will continue to evaluate the performance
of all of our employees and take action where we believe it to
be appropriate.
Q.5.a. Ms. Robinson, Ms. Tillman, and Mr. Joynt, during the
last 3 years, did your firm notice a decline in underwriting
standards for mortgages being used to create residential
mortgage-backed securities? If so, did you alter your ratings
process in anyway to account for this decline in underwriting
standards? Did you disclose to investors that there was a
decline in underwriting standards?
A.5.a. S&P repeatedly and publicly voiced concerns about the
subprime market and the deteriorating credit quality of RMBS
transactions as far back as April 2005. These warnings included
discussion of the various ``affordability'' mortgage products
employed in the subprime market and the risks they entailed,
including the risk of loosening underwriting standards. For
example:
In an April 4, 2005 article entitled S&P Comments On Risk
In Newer Mortgage Products, As Discussed At Industry Event, we
noted that ``there is growing concern around the increased
usage of [interest-only, negative amortization, and 40-year
amortization] mortgages in new RMBS securitization, which may
pose significant credit risk. . . . [S]ome of the inherent
risks that may arise include payment shock due to interest rate
increases, coupled with the addition of principal repayment,
undercollaterlization with regard to negative amortization, and
home price depreciation.''
In an April 20, 2005 article entitled Subprime Lenders:
Basking in the Glow of a Still-Benign Economy, But Clouds
Forming on the Horizon, we stated that we ``remain concerned
about how these subprime lenders will perform in a prolonged
rising interest rate environment.'' We observed that increased
competition among subprime lenders threatened a relaxation in
underwriting standards and warned that the growing popularity
of ``affordability'' mortgage products ``suggests that Standard
& Poor's concerns are justified.'' We singled out interest-only
mortgages as ``[e]specially worrisome,'' noting that ``these
loans are more likely to feature adjustable rates . . . setting
borrowers up for potential problems should mortgage rates rise
dramatically.''
On July 10, 2006, in an article entitled Sector Report
Card: The Heat Is On For Subprime Mortgages, we noted that
downgrades of subprime RMBS ratings were outpacing upgrades due
to ``collateral and transaction performance.'' The article also
identified ``mortgage delinquencies'' as a ``potential hot
button,'' and noted that such delinquencies ``may become a
greater concern for lenders and servicers.''
On February 14, 2007, we took the unprecedented step of
placing on CreditWatch negative (and ultimately downgrading)
transactions that had closed as recently as 2006. As we
informed the market in the accompanying release: ``Many of the
2006 transactions may be showing weakness because of
origination issues, such as aggressive residential mortgage
loan underwriting, first-time home-buyer programs, piggyback
second-lien mortgages, speculative borrowing for investor
properties, and the concentration of affordability loans.'' In
a February 16, 2007 Los Angeles Times article, S&P's
announcement was described as `` `a watershed event' because it
means S&P is now actively considering downgrading bonds within
their first year.'' See S&P to Speed Mortgage Warnings, Los
Angeles Times, Feb. 16, 2007.
In a February 28, 2007 article entitled RMBS Trends: U.S.
Subprime Market Continues Correction As Issuers Strengthen
Underwriting Standards, we observed that: ``Recent-vintage
loans continue to pay the price for loosened underwriting
standards and risk-layering in a declining home price
appreciation market, as shown by early payment defaults and
rising delinquencies. Lenders have reported tightened
underwriting standards during the industry consolidation, with
weaker players exiting the origination business or being
acquired by larger entities, most prominently investment banks.
Although evidence of improved underwriting standards has been
represented in loan documentation data, other measures such as
LTV have not fully supported the reports. However, as there is
a lag between loan origination and securitization, we may begin
to see more evidence in the coming quarters.''
In an April 27, 2007 article entitled Special Report:
Subprime Lending: Measuring the Impact, we stated: ``The
consequences of the U.S. housing market's excesses, a topic of
speculation for the past couple of years, finally have begun to
surface. . . . Recent-vintage loans continue to pay the price
for loosened underwriting standards and risk-layering in a
declining home price appreciation market, as shown by early
payment defaults and rising delinquencies.''
In a July 25, 2007 teleconference, we observed that the
``poor performance'' in U.S. RMBS ``results from a combination
of factors including but not limited to an environment of loose
underwriting standards, pressure on home prices, speculative
borrowing behavior, risk layering, very high combined loan to
value, financial pressure on borrowers resulting from payment
increases on first-lien mortgages and questionable data
quality.''
Q.5.b. Did you alter your ratings processes in any way to
account for this decline in underwriting standards?
A.5.b. In response to deterioration in the sub prime mortgage
market, which was attributable to a number of factors, we
tightened our criteria through changes in our
LEVELS' model targeted to increase the credit
enhancement requirements for pools with subprime loans. As
noted above, in February 2007, we also took the unprecedented
step of placing on CreditWatch negative (and ultimately
downgrading) transactions that had closed as recently as 2006.
We continued taking downward action through as recently as this
week. We increased the severity of the surveillance assumptions
we use to evaluate the ongoing creditworthiness for RMBS
transactions issued during the fourth quarter of 2005 through
the fourth quarter of 2006 and downgraded those classes that
did not pass our heightened stress test scenario within given
time frames. In addition, we modified our approach for ratings
on senior classes in transactions in which subordinate classes
have been downgraded. We also announced that, with respect to
transactions closing after July 10, 2007, we would implement
changes that would result in greater levels of credit
protection (collateral) for rated transactions and would
increase our review of lenders' fraud-detection capabilities.
No one can see the future. The point of these articles and
actions, however, is to highlight our reaction to increasing
subprime deterioration--looking, as we always do, to historical
or paradigm-shifting behaviors to help analyze long-term
performance. Consistent with our commitment to transparency we
repeatedly informed the market of our view that the credit
quality of subprime loans was deteriorating and putting
negative pressure on RMBS backed by those loans. And,
consistent with our commitment to analytical rigor, we revised
our models, took action when we believed action was
appropriate, and continue to look for ways to make our
analytics as strong as they can be.
Q.6. Ms. Robinson, Ms. Tillman, and Mr. Joynt, when each of
your companies tries to attract new customers, how do you
distinguish your ratings from the ratings of other rating
agencies?
Do you have empirical data that demonstrates that your
ratings are better than the ratings of other companies? If yes,
please provide documentation supporting your answer.
Do you compete more on price or ratings accuracy? Please
provide documentation supporting your answer.
A.6. S&P began its credit rating activities 90 years ago, and
today is a global leader in the field of credit ratings and
credit risk analysis. We vigorously protect our reputation and
we believe--recent criticism notwithstanding--that our
excellent historical track record of providing the market with
independent and rigorous rating opinions and information is
widely recognized in the market. Investors attach value to our
ratings because of this track record, and this is and of itself
diminishes any leverage that underwriters may be perceived to
have over the process. It is also this track record, along with
our commitment to innovation and improvement, that we sometimes
discuss with potential new customers. We believe, as well, that
S&P's proven role as a market leader will continue to
distinguish us from our competitors even as the credit rating
industry expands.
Q.7. Ms. Robinson, Ms. Tillman, and Mr. Joynt, what are the
idealized default rates for each of your ratings?
A.7. There are no idealized default rates for our ratings,
since ratings are not mapped to particular expected default
rates. Instead, we arrive at our rating opinions by applying
our published assumptions, methodologies and criteria to the
best available information in our possession.
Over the last 30 years (through May 16, 2008), S&P's
cumulative default rate by original rating class for all
structured finance ratings has been as follows:
------------------------------------------------------------------------
Initial rating Percent of default
------------------------------------------------------------------------
AAA........................................... 0.14
AA............................................ 0.60
A............................................. 1.46
BBB........................................... 3.53
BB............................................ 5.21
B............................................. 4.78
------------------------------------------------------------------------
Q.8. Ms. Robinson, Ms. Tillman, and Mr. Joynt, in his written
testimony, Professor Coffee notes that because only a limited
number of investment banks underwrite structured finance
products, they have leverage over the rating agencies. If they
don't like the ratings they get from one agency, they can go to
another with lower standards.
Has your firm ever felt pressure to lower your rating
standards in order to attract business?
How do you attract customers if your ratings use the most
stringent standards? Will issuers and underwriters simply go to
other firms with less demanding standards?
A.8. At S&P, we do not permit issuers to dictate any aspect of
our analytical process. Our analytics are driven by our
criteria, and we do not compromise that criteria to meet a
particular issuer's needs or agenda. We have refused to rate
whole categories of transactions that do not meet our criteria
and we believe that we have lost numerous RMBS deals for this
reason.
As noted, we believe our reputation and integrity are our
most valuable long-term assets. It would be contrary to our
best interests to sacrifice these qualities by providing
anything other than what we believe to be our best opinion of
creditworthiness. In some instances, this may mean that an
issuer will take its business elsewhere, but that is a risk we
are willing to take in order to preserve the far more valuable
asset that is our reputation for independence.
------
RESPONSE TO WRITTEN QUESTIONS OF SENATOR MENENDEZ FROM VICKIE
A. TILLMAN
Q.1. During the hearing, I asked you for specifics on the
ratings your agency provided on Bear Stearns in the months
leading up to the collapse. Please provide the Committee with a
detailed explanation of the ratings for Bear Stearns from
November 2007 and March 2008. In addition, please answer the
following questions.
A.1. On November 15, 2007, S&P downgraded the Bear Stearns
Companies' (``Bear'') long-term counterparty credit rating from
``A+'' to ``A'' and affirmed the short-term rating of ``A-1''.
This action followed our decision on August 3, 2007 to
revise our outlook on Bear from stable to negative based in
part on the reputational harm suffered by Bear in the wake of
problems with its managed hedge funds, as well as its material
exposure to holdings of mortgages and MBS, the valuations of
which, we said, remained under ``severe pressure.'' We further
noted on August 3 that Bear had exposure to debt it had taken
up as a result of unsuccessful leveraged finance underwritings
and had significant further underwriting commitments. We
observed that Bear had a relatively high degree of reliance on
the U.S. mortgage and leveraged finance sectors, and its
revenues and profitability would be especially affected if
there were an extended downturn in those markets. We continued
this negative outlook in our November 15 rating action.
The November 15 rating action followed Bear's announcement
that it would take a fourth-quarter writedown on its CDO and
subprime exposure of $1.2 billion. We considered the writedown
to be comparatively less than that of its peers, particularly
given Bear's substantial business concentration in the U.S.
mortgage market. We noted that the potential for further
writedowns remained given continued dislocation in the mortgage
market but considered the company's remaining CDO and subprime
exposure to be manageable. Nevertheless, we warned that
additional writedowns could further impair the company's future
earnings performance, particularly in light of Bear's
relatively greater revenue reliance on fixed income markets,
which were experiencing a general slowdown.
The negative outlook on the ratings reflected our
continuing concern that the general slowdown in Bear's core
fixed income businesses could have a negative impact on its
earnings performance in the near to medium term. We also
remained concerned that long-term lingering effects (including
litigation) of the widely publicized problems in the company's
managed hedge funds would have a negative impact on performance
in the company's asset management unit. We noted that the
ratings could be lowered if earnings failed to stabilize at a
satisfactory level beyond the next few quarters, which we
expected to be difficult ones for the firm. We observed that in
contrast, if Bear were able to overcome current challenges and
affect a more rapid earnings recovery than we currently
anticipated, the outlook could be revised to stable.
On March 14, 2008, S&P downgraded Bear's long-term
counterparty credit rating from ``A'' with negative outlook to
``BBB'' and its short-term rating from ``A-1'' to ``A-3''. At
the same time, we placed the long- and short-term ratings on
CreditWatch with negative implications.
This rating action followed Bear's announcement that its
liquidity position had substantially deteriorated in the two
days prior to the rating action. The severe impairment of
Bear's liquidity had resulted in the negotiation of a 28-day,
Fed-backed secured loan facility with JP Morgan Chase
(``JPMC'') that was designed to ease Bear's liquidity pressures
until it could implement a longer term funding structure.
We noted that Bear had been experiencing significant stress
during the week of March 10 because of concerns regarding its
liquidity position. Although the firm's liquidity at the
beginning of the week had held steady with excess cash of $18
billion, ongoing pressure and anxiety in the markets resulted
in significant cash outflows toward the week's end, leaving
Bear with a significantly deteriorated liquidity position at
end of business on Thursday, March 13, 2008.
We observed that our ratings were based on our expectation
that Bear would find an orderly solution to its funding
problems. We noted, however, that although we viewed the
liquidity support to Bear as positive, we considered it a
short-term solution to a longer term issue that did not
remediate Bear's confidence crisis. We also remained concerned
about Bear's ability to generate sustainable revenues in an
ongoing volatile market environment.
Finally, we stated that we expected to resolve the
CreditWatch in the coming weeks, as more concrete, longer term
solutions to Bear's liquidity and confidence crisis were
fleshed out. We warned that the ratings could be lowered
further if there were a failure to stabilize liquidity or to
achieve a satisfactory longer term funding structure.
On March 17, 2008, we placed our ``BBB'' long- and ``A-3''
short-term counterparty credit ratings on Bear on CreditWatch
with developing implications.
This rating action followed the announcement that JPMC had
agreed to acquire Bear in an all-stock transaction.
We noted that we considered the acquisition of Bear by JPMC
as positive, as it would permit Bear to meet its obligations
through funding sources obtained directly from its new parent.
We observed that we expected that JPMC would assume all of
Bear's obligations when the transaction closed. We stated that
we would resolve the CreditWatch placement when details about
the integration of Bear's activities became tangible.
We warned that if the acquisition by JMPC were not to close
as expected, the ratings on Bear would come under renewed
pressure. Conversely, if the acquisition was to proceed as
expected and Bear's businesses were successfully integrated
into JPMC, the ratings on Bear could be equalized with those on
its new parent.
On March 24, 2008, we raised the counterparty credit
ratings on Bear to ``AA-/A-1+'' and removed them from
CreditWatch Developing where they had been placed on March 17.
We determined that the outlook for the ratings was stable.
This rating action recognized the strengthened immediate
guarantee by JPMC of all of Bear's counterparty obligations. We
noted that JPMC was also to assume Bear's debt obligations upon
completion of the acquisition.
We observed that in our view, the price increase for the
transaction and the anticipated increase in the amount of
shares controlled by JPMC raised the probability that the deal
would be completed. We warned that on its own, Bear's viability
was uncertain, and that if the deal were to be amended in any
way, we would review the circumstances at that time.
We stated that we expected the acquisition by JPMC to be
completed under the revised terms by mid-May. In light of the
guaranty and our expectation that Bear's debt would be assumed
by JPMC, we believed that Bear's creditors benefited from
JPMC's creditworthiness and participated in the outlook for
JPMC. Therefore we equalized the ratings and outlook with those
on JPMC.
Our press releases for each of these ratings actions are
attached.
Q.2. Were any of the ratings downgraded between December 2007
and March 14, 2008?
A.2. As noted, on November 15, 2007, S&P downgraded Bear's
long-term counterparty credit rating from ``A+'' to ``A,''
which followed our decision in August 2007 to change our
outlook on Bear from stable to negative.
On March 14, 2008, S&P downgraded Bear's long-term
counterparty credit rating from ``A'' with negative outlook to
``BBB'' and its short-term rating from ``A-1'' to ``A-3.'' At
the same time, we placed the long- and short-term ratings on
CreditWatch with negative implications.
Q.3. Were any of the ratings downgraded during the week of the
collapse (March 10-14)?
A.3. As noted, on March 14, 2008, S&P downgraded Bear's long-
term counterparty credit rating from ``A'' with negative
outlook to ``BBB'' and its short-term rating from ``A-1'' to
``A-3.'' We also placed Bear's long- and short-term ratings on
CreditWatch with negative implications.
Q.4. Can you explain from your agency's point of view how
Bear's collapse unfolded and the role the ratings may have
played?
A.4. While we are continuing to review the factors that led to
the sudden and severe weakening of Bear's liquidity situation,
a number of factors are clear at this time. As noted in our
published reports, Bear faced (i) material exposure to CDOs and
subprime investments, as well as the general slowdown in its
fixed income businesses; (ii) significant dislocation in the
mortgage market; and (iii) severe, ongoing reputational harm
that eventually led to a crisis in confidence. The damage to
Bear was hastened, in our view, by its inability to effect a
rapid earnings recovery in the face of these challenges.
As noted, Bear's situation deteriorated rapidly in March
2008 when it announced that its liquidity position had
substantially and rapidly deteriorated over a two-day period,
which resulted in part from significant cash outflows, as well
as ongoing pressure and anxiety in the markets. It was also
reported that some hedge funds suddenly withdrew billions of
dollars in assets from Bear, which was unexpected and would
have contributed to the bank's rapid decline. The sudden loss
in confidence in Bear was critical in our view, and was widely
unexpected. It became clear that Bear required a long term
solution to its liquidity problems, which eventually arose in
the form of JPMC's announced acquisition.
We do not believe that S&P's ratings had a role in causing
these events to occur. Rather, consistent with our long-
standing reputation for independence and objectivity, our
ratings simply reflected our current opinion of the
creditworthiness of Bear based on the best facts available to
us at the time.
Q.5. Do you think the lack of changes to the Bear Stearns'
ratings is an example of a unique event in the markets or an
indication of larger flaws in the structure of the ratings?
A.5. We believe the speed with which Bear deteriorated was a
unique event in the market and broadly unanticipated.
Q.6. Under ideal circumstances, would you agree that the
ratings should have been downgraded to more accurately reflect
Bear's risk?
A.6. As with all of our ratings, we downgraded Bear's rating
when we concluded that the data available supported such a
move. It is always easy in hindsight to look back and question
whether certain ratings should have or could have been
different if we knew then what we know now. That, however, is
different than the reality of our work, which is to take the
information available to us at the time and try as best we can
to project what is likely to happen in the future. That is what
we did with our ratings on Bear.
Q.7. What lessons do you think we should take from the Bear
Stearns collapse as it relates to the credit ratings?
A.7. As noted, our rating on Bear was based on the best
information available to us at the time, including statements
by management and regulatory filings. We had concerns about
Bear's exposure to CDOs and subprime investments, as well as
the consequences of continued harm to its reputation, among
other things, and made those concerns public. Of course, we are
always looking for ways to refine our analytical processes and,
in this case, are continuing to assess the factors that led to
Bear's rapid decline.
Q.8. What are your thoughts on a proposal Professor Coffee
discussed at the hearing for rating agencies to periodically
update ratings?
A.8. At S&P we believe that timely monitoring of our rating
opinions is a key component of the value we bring to investors
and the market. We are constantly looking for ways to enhance
our surveillance process and have made improvements--including
increasing the frequency of our reviews of rated transactions
and the amount of resources dedicated to the process--response
to recent events.
While we believe that credit ratings should be ``current''
assessments of creditworthiness, we do not believe a mandated
fixed schedule of periodic reviews in the manner that Professor
Coffee suggests would improve the surveillance process. Ratings
are subjective in nature and are typically formulated and
disseminated after deliberation of whatever duration is
appropriate to assess the particular issue or issuer being
considered. The necessary frequency and scope of any ratings
review may vary considerably based on issue and issuer-specific
factors as well as the original method of analysis. Any rule
attempting to impose a specific, fixed period during which
ratings must be updated would divert attention away from
surveillance based on risk identification and assessment, and
would by its nature be arbitrary, burdensome and, we believe,
ineffective.
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
RESPONSE TO WRITTEN QUESTIONS OF CHAIRMAN DODD FROM CLAIRE
ROBINSON
Q.1. Ratings Scoreboard. What are your views on the
recommendation that has been made for the creation of a central
website which investors could access and on which they could
compare the accuracy of past ratings by the different NRSROs
for the same types of securities?
A.1. Moody's would support the establishment of a centralized
repository, such as an industry portal, for rating performance
studies. Indeed, we believe such a repository could enhance the
ability of users of credit ratings to compare and contrast
rating agency performance in a more efficient manner. We also
would support a centralized repository that lists the ratings
of each NRSRO for a particular security, so long as such
requirements do not intrude on our rating methodologies or the
content of our ratings.
Q.2. Timeliness of Updates of Ratings. Professor Coffee in his
testimony pointed out that major downgrades of CDO securities
``came more than a year after the Comptroller of the Currency
first publicly called attention to the deteriorating conditions
in the subprime market and many months after the agencies
themselves first noted problems in the markets.'' His testimony
also states ``the gravest problem today may be the staleness of
debt ratings.'' What is Moody's doing to update ratings in a
timely manner and eliminate stale ratings? What standards
should NRSROs observe?
A.2. Our initial ratings on these securities reflected our
expectation of the asset performance in the pools. As always,
we monitored our published ratings and took rating actions
accordingly and when warranted by performance data and when our
expectation of future performance changed due to changes in
market conditions.
Professor Coffee's comments seem to suggest that rating
actions on a security should be taken as soon as any
significant changes in market conditions are observed. Based on
our ongoing conversations with investors, issuers and
regulators, many market participants have a strong preference
for credit ratings that are not only accurate but stable. They
want ratings to reflect enduring changes in credit risk because
rating changes have real consequences--due primarily to
ratings-based portfolio governance rules and rating triggers--
that are costly to reverse. Market participants, moreover, do
not want ratings that simply track market-based measures of
credit risk. Rather, ratings should reflect independent
analytical judgments that provide counterpoint to often
volatile market-based assessments.
However, we do believe that there are additional steps that
can be taken to enhance the quality and efficiency of our
surveillance activities. For example:
Further enhancing our surveillance function: We
are continuing to expand the resources allocated to wholly
separate monitoring teams within our Structured Finance Group.
This initiative began before the RMBS and CDO downgrades. More
recently, as part of our commitment to enhancing surveillance,
Moody's created and filled the position of Global Structured
Finance Surveillance Coordinator. The executive in this role is
working with the surveillance managers and departmental heads
throughout our global Structured Finance Department to enhance
our surveillance processes and make them more efficient.
Enhancing the automated review of data: Moody's
has also been implementing a number of automated processes and
systems, including proprietary applications, to routinely sift
through entire databases of transactions, updating performance
statistics and flagging rating outliers. The rollout of these
initiatives began before the RMBS and CDO downgrades.
Enhancing market communication: Moody's has
allocated more resources to the function of communicating our
monitoring activities to the market.
Enhancing review of internal process and market
trends: As part of the CRA industry initiative, Moody's has
committed to evaluating our internal processes and market
trends regularly so that we maintain the operational
flexibility to enable us to dedicate the resources needed to
monitor existing ratings and conduct reviews on a timely basis.
Q.3. Separate Ratings from Business? Dr. Cifuentes'
testimony contains a recommendation that a rating agency
separate its rating business function from its rating analysis
function. What are your views on how NRSROs should address this
analyst independence concern?
A.3. Moody's would not object to a clearer distinction
between the business arm of a credit rating agency and the
analytical work that it conducts. For our part, Moody's Code
sets forth our policies that govern the roles and
responsibilities of our rating agency employees, with the
primary goal of ensuring that our analytical activities remain
appropriately distanced from the commercial management of our
business. In particular, the following provisions are relevant:
2.11 Reporting lines for Employees and their compensation
arrangements will be organized to eliminate or effectively
manage actual or potential conflicts of interest. Analysts will
not be compensated or evaluated on the basis of the amount of
revenue that [Moody's] derives from Issuers that the Analyst
rates or with which the Analyst regularly interacts.
2.12 [Moody's] will not have analysts who are directly
involved in the rating process initiate, or participate in,
discussions regarding fees or payments with any entity they
rate.
Implementation of these and other standards in the Moody's
Code is subject to oversight by our internal Compliance
Department as well as external examination by authorities such
as the SEC.
Furthermore, while we believe that we operate with a high
degree of independence and clarity around analytical
remuneration, greater clarity regarding our policies and
practices to protect analysts' independence may be beneficial
for the market. We recently have implemented, or are in the
process of implementing, several other measures to further
demonstrate the independence of our rating process. These
include:
Formalizing the separation of ratings-related
businesses: Moody's recently reorganized its operating
businesses to formalize the separation of our ratings-related
and non-rating activities into two different business units.
Enhancing the Credit Policy function: The Credit
Policy function at Moody's has long been independent from those
parts of the rating agency with revenue-generating
responsibility, and we have taken steps to further separate
this function. The Chairman of Credit Policy now has a
reporting responsibility to the President of Moody's. The
performance incentives for Credit Policy personnel are based
exclusively on the effectiveness of the rating process and the
analytical quality of their oversight. The measurement of the
unit's performance is wholly independent of the financial
performance of the company or any business unit.
Codifying the existing policies about analyst
communication with issuers: In order to enhance market
confidence in the appropriateness of communications between
Moody's analysts and issuers or advisors, we are codifying our
existing practice that such communications are limited to
communications about credit issues.
Implementing ``look-back'' reviews to confirm
integrity of analysis: Moody's has adopted a new policy related
to employees who leave Moody's to work for another market
participant. When we learn that an issuer or a financial
intermediary representing the issuer has hired a Moody's
employee who has served as lead analyst for that issuer, we
will now review the analyst's work related to the issuer and
its securities over a six-month ``look-back'' period to confirm
the integrity and rigor of that analyst's work.
Also, as part of the CRA industry initiative, we have
committed to conduct formal and periodic, internal reviews of
compensation policies and practices for analysts and other
employees who participate in rating committees to ensure that
these policies do not compromise the rating process.
Q.4. 4. Ratings Shopping. We have heard concerns about
``ratings shopping,'' where an underwriter or an issuer goes to
the NRSRO that it feels will give it the highest rating, even
if it is not necessarily the most accurate. Is ratings shopping
a problem? How should the negative aspects of it be addressed?
A.4. Ratings shopping occurs in the structured finance market
because there is relatively little information available
publicly about the transactions prior to their issuance.
Arguably, market disciplinary forces and transparency around
NRSRO methodologies already operate to some extent to address
the negative aspects of rating shopping. For example, because
rating methodologies are published, market participants can
compare various CRA's different approaches to the same sectors
or asset classes. A potential user of an NRSRO's credit ratings
can decide for itself whether or not the NRSRO's methodology is
more or less stringent than another NRSRO's methodologies and
take this into account in deciding whether to attach any weight
to the NRSRO's ratings. On the other hand, this same
transparency allows issuers to assess the conservatism of a
particular NRSRO and ratings shop at the outset of the rating
process.
We believe that the appropriate way to deal with the
negative aspects of ratings shopping in the structured finance
market is to have issuers publicly disclose in a comprehensive
and standardized manner:
the characteristics of each asset in the asset
pool;
the structure of the transaction and performance
data for each asset in the asset pool;
the validation process used to verify the quality
of the information provided and all pertinent representations
and warranties; and
Servicer and Trustee reports prepared after the
issuance of the transaction.
Presently, because of the generally limited data in the
public market about structured securities prior to their
issuance, neither investors nor CRAs that have not had
sufficient contact with the issuer are able to formulate an
informed opinion on the securities. However, if robust
information about structured finance products were publicly
available once the details of the transaction had been
finalized,\10\ both the investors and credit rating agencies
could form higher quality opinions, regardless of whether or
not an issuer has directly contacted them. As a result, in many
circumstances market participants would have the benefit of
multiple and potentially diverse opinions about the same
transaction. Finally, and most importantly, having the
underlying data published by the issuers or originators would
allow investors to form their own opinions about the strengths
and weaknesses of a particular transaction, which could support
authorities' efforts to discourage the use of ratings for
purposes other than an objective opinion about relative credit
risk.
---------------------------------------------------------------------------
\10\ Given their complex and mutable nature, structured finance
products may not lend themselves to unsolicited ratings before that
time.
---------------------------------------------------------------------------
Some policymakers and market commentators have suggested
that ratings shopping can be addressed by requiring CRAs to
disclose the names of issuers who provide data to a credit
rating agency and ask for a preliminary assessment but then
choose to publish the rating of another credit rating agency.
We believe that such a solution would be unworkable and likely
would fail to resolve concerns about the practice because:
The CRA would not necessarily know if the issuer
contracted with another CRA for a final rating of the same
structure.
Even if the CRA tried to monitor the conduct of an issuer,
the CRA could not know with certainty that it had identified
all cases requiring disclosure under such a provision because
it might not have access to all relevant information. Moody's
believes it is inappropriate to impose a disclosure obligation
on an entity that cannot, as a practical matter, control the
means by which it acquires information that triggers that
obligation.
Ratings shopping would simply occur at an earlier
point in time.
Moody's believes that requiring credit rating agencies to
disclose cases of ratings shopping might change the nature of
the practice but would not eliminate it. Some originators,
underwriters and sponsors of structured securities who wished
to avoid being identified by CRAs as ratings shoppers likely
would get around the disclosure trigger by withdrawing earlier
in the process. Others might simply refrain from approaching
CRAs that were believed to have more conservative methodologies
or were less-well established, and whose methodologies were not
well-understood or well-tested in the market.
Q.5. Professional Analyst Organization. Dr. Cifuentes in his
testimony suggested ``the creation of a professional
organization, independent of the rating agencies, to which
ratings analysts must belong and which sets forth ethical,
educational and professional standards.'' Please share your
thoughts on the potential merits of such an organization.
A.5. In evaluating the necessity of a ``professional analyst
organization'', we believe the following points should be
considered.
We believe that a rigorous credit ratings process
involves the expertise of a combination of professionals--
including lawyers, MBAs, accountants and others--who bring
their respective and possibly divergent points of view to bear
upon an analysis. Consequently, establishing a single body that
establishes one set of standards for credit analysts, in our
view, may diminish some of the advantage of having employees
with different educational and professional credentials.
Many of our professionals already belong to
professional organizations, such as state bars. These
organizations impose continuing education requirements and
ethical standards. Establishing yet another professional
organization may be overly burdensome on the analysts.
Moody's Code already has extensive provisions
dealing with, among other things, the quality, integrity and
independence of the rating process. There are provisions
directed specifically to analysts.
Moody's also has an extensive professional
development and training program that has been designed to
enhance the quality of Moody's rating analysis and analysts'
understanding of relevant policies and procedures, including
those relating to ethics. Moody's analysts are required to meet
annual continuing education requirements by completing Moody's
courses or approved, external courses.
As part of the CRA industry initiative, we intend
to incorporate into the Moody's Code an explicit commitment to
adopt and maintain a continuing education program appropriate
to the nature of our business. Doing so will make our policies
and practices in this area subject to monitoring by the SEC.
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
RESPONSE TO WRITTEN QUESTIONS OF SENATOR SHELBY FROM CLAIRE
ROBINSON
Q.1. Ms. Robinson, Ms. Tillman, and Mr. Joynt, do you think
that it is easy for investors to compare the accuracy of the
ratings of the different credit rating agencies? If not, do
S&P, Moody's, and Fitch favor the SEC issuing rules to require
enhanced disclosure of ratings performance as Chairman Cox
outlined in his testimony?
A.1. We agree that new ``requirements for enhanced disclosures
about ratings performance'' is an important area for
consideration as the Securities and Exchange Commission
(``SEC'') contemplates new rules for our industry. We believe
that consideration of ratings performance data would create
objective criteria for assessing whether a credit rating
agency's (``CRA's'') ratings are suitable for use in
regulation.
For Moody's part, we publish and make freely available a
wealth of data on ratings performance so that users of our
ratings, as well as regulators, can judge the performance of
our ratings. For your information, we have provided in this
packet our ``Guide to Moody's Default Research: March 2008
Update'', which lists our performance, default, transition and
loss severity research reports in reverse chronological order
and broken down by topic. Moody's believes there is substantial
value in encouraging agencies to present their analysis of
their ratings performance in the ways they believe to be most
relevant, since there is no single agreed upon approach.
One reason why there is no unique way to measure ratings
performance is that different users of ratings place different
value on different characteristics of the relationship between
ratings and credit risk and as such different rating agencies
seek to measure different attributes. For example, users may be
concerned with one or more of the following:
the relationship between ratings and defaults;
the relationship between ratings and expected
credit losses (which are the product of default probabilities
and loss severity rates in the event of default);
ratings stability;
the relationship between ratings and ``mark-to
market'' risk;
the information in rating outlooks rather than
just the ratings alone;
the ability of ratings to rank relative credit
risk at a point of time; and
the ability of ratings to rank relative credit
risk over time.
We agree in concept that presenting data in a standardized
format would facilitate ratings performance comparisons. We
believe, however, that such a standard may be difficult to
implement in practice for a number of reasons, including:
There may be differences of opinion on the most
meaningful way to present the data. Any given presentation may
advantage or disadvantage one rating agency compared with
another.
Rating agencies do not all define their ratings
in the same way, which may result in standardized performance
reports not being perfectly comparable. For example, Moody's
ratings are intended to be opinions of expected loss whereas
some other rating agencies may intend their ratings to measure
other indicators of credit risk, such as just probability of
default. Therefore, a standardized format that focuses on
default experience alone may not effectively capture the
overall predictive content of a Moody's rating.
Rating agencies have different approaches to
dating defaults on structured finance securities because events
of default are more subjective in structured finance than in
corporate finance. In particular, many securitizations are
structured as pass-through securities and, as such, are not at
risk of payment defaults in a strict contractual sense until
they reach their legal final maturity dates (in perhaps 30
years) even though cash flows to investors may cease many years
prior to maturity.
Q.2. Ms. Robinson, Ms. Tillman, and Mr. Joynt, would you please
explain the process by which you obtain the information you use
to rate structured finance securities?
A.2. At Moody's, the analyst or analysts assigned to a
particular structured finance transaction begin the credit
analysis by assembling relevant information on the transaction.
Information about the specific transaction may come from the
originator or a market intermediary in meetings or other
communications with the analyst(s). Our analysts compare this
transaction-specific information with data we have regarding
past transactions, deals effected by other market participants,
thematic research generated by Moody's analysts regarding
industry trends, credit research generated in other rating
departments (e.g., regarding the creditworthiness of the
financial institutions participating in the securitization) and
macro-economic trend research generated by Moody's
Analytics.\1\
---------------------------------------------------------------------------
\1\ Moody's Analytics is a subsidiary of our corporate parent,
Moody's Corporation. It is legally and operationally separate from
Moody's Investors Service, the rating agency, and is a provider of
research, data, analytic tools and related services that are distinct
from credit ratings.
---------------------------------------------------------------------------
We have provided in this packet the Moody's Investors
Service Code of Professional Conduct (``Moody's Code''), which
presents the various policies that we have in place to address
issues of (1) quality and integrity of the rating process; (2)
independence and management of conflicts of interest; (3)
responsibilities to investors and issuers; and (4) enforcement
and disclosure of the Code of Professional Conduct and
communication with market participants. Of particular relevance
to this question is the following provision:
1.4 Credit Ratings will be determined by rating committees
and not by any individual Analyst. Credit Ratings will reflect
consideration of all information known, and believed to be
relevant, by the applicable [Moody's] Analyst and rating
committee, in a manner generally consistent with [Moody's]
published methodologies. In formulating Credit Ratings,
[Moody's] will employ Analysts who, individually or
collectively, have appropriate knowledge and experience in
developing a rating opinion for the type of credit being
analyzed.
Our primary contact for transaction-specific information is
typically the intermediary (``Arranger'' ``Underwriter'' or
``Investment Banker'') that chooses the assets to be included
in the transaction and sets up the structure of the
transaction, divides the structure into different classes of
securities (``tranches'') and markets the tranches. We may also
deal with and obtain information from:
one or more ``Originators'', which either
originate the underlying assets in the course of their regular
business activities or source them in the open market;
the ``Servicer'', which collects payments and may
track pool performance;
in managed transactions, an ``Asset Manager'',
which may assemble the initial pool and may subsequently buy
and sell assets in the transaction; q
the ``Trustee'', who oversees cash distributions
to investors and monitors compliance with transaction
documentation;
a ``Financial Guarantor'', who may provide
guarantees on principal and/or interest payments to, or may
sell credit default swaps on, particular tranches; and
in asset-backed commercial paper (``ABCP'')
programs, an ``Administrator'' of the ABCP conduit that funds
several asset pools.
In addition, it is not unusual for the Arranger to ask us
to communicate directly with the transaction lawyer in order
for us to get a better understanding of the transaction
structure.
Q.2.a. How much of the information is from issuers,
underwriters, or other sources?
A.2.a. The relative proportions of the information we obtain
from the different sources vary depending on the asset class
and the transaction in question. In general terms, our primary
source of transaction-specific information is the Arranger (or
its agents). As noted above, however, our analysts compare this
transaction-specific information with data we have regarding
past transactions, deals effected by other market participants,
thematic research generated by Moody's analysts regarding
industry trends, credit research generated in other rating
departments (e.g. regarding the creditworthiness of the
financial institutions participating in the securitization) and
macro-economic trend research generated by Moody's Analytics.
Q.2.b. Do you ever seek to verify the accuracy of the
information you receive?
A.2.b. Our analysis takes into consideration and compares data
from a variety of sources. Moody's Code of Professional Conduct
Provision 1.7 states:
1.7 [Moody's] will invest resources sufficient to carry
out high-quality credit assessments of Issuers or obligations.
When deciding whether to rate or continue rating an obligation
or Issuer, [Moody's] will assess whether it is able to devote
sufficient personnel with appropriate skills to make a proper
rating assessment, and whether its personnel likely will have
access to sufficient information needed in order to make such
an assessment.
When rating a corporate issuer, we receive audited
financial data and regulatory filings. When rating a structured
finance product, the Originator and/or Arranger of the
structured product make representations and warranties to the
other parties in the transaction as to the quality of the loan
level data describing the collateral. With respect to the
publicly offered securities in the structured finance market,
the prospectus also contains information that must be provided
to investors in accordance with U.S. securities laws. The named
underwriter performs due diligence on the security being issued
to help verify the accuracy of information in the prospectus.
These underwriters frequently hire a due diligence firm to
examine the underlying loans. Accounting firms also are
frequently hired by underwriters to verify that the summary
information about the loan pools matches the information in the
related loan files.
As part of the credit rating process, we do consider, among
other factors:
(a) the source of the data we receive;
(b) the track record of the source in providing quality
data;
(c) the predictive powers associated with the data; and
(d) whether or not the data (such as financial
information) has been subject to review by a third party.
In addition, as noted in our response to the preceding
question, we also assess the transaction-specific information
in the context of the much broader and deeper data sets and
other information we possess as a result of our credit rating
and credit-related research activities. However, others in the
market (e.g. auditors, issuers and underwriters) are far better
positioned--given their expertise and resources--to certify the
accuracy of data.
Our experience over the decades that we have been rating
structured securities has been that most of the issuers
operated in good faith and provided reliable information to us,
and we have relied upon them to do so. Nevertheless, our
analysts seek to exercise skepticism in our analysis of
information provided to us. Furthermore, if we believe we have
inadequate information to provide an informed credit opinion to
the market, we will exercise our editorial discretion and
either refrain from publishing an opinion or withdraw our
published credit rating.
In light of recent market difficulties, we believe that the
due diligence process conducted by the parties who originate,
arrange, and/or service residential mortgage backed securities
(``RMBS'') needs to be further strengthened. We have proposed a
series of measures to improve transparency, data integrity and
accountability in U.S. residential mortgage securitizations,
including \2\:
---------------------------------------------------------------------------
\2\ For more information, please see the enclosed ``Moody's
Proposed Enhancements to U.S. Residential Mortgage Securitizations:
Call for Comments''.
---------------------------------------------------------------------------
Stronger representations and warranties;
Independent third-party pre-securitization review
of underlying mortgage loans;
Standardized post-securitization forensic review;
Expanded loan-level data reporting of initial
mortgage pool and ongoing loan performance; and
More comprehensive originator assessments.
We believe that these measures taken together will provide
more standard and reliable information on RMBS transactions
than currently available.
Q.3. Do you have any reason to believe that inaccurate or
fraudulent data contributed to the poor performance of your
ratings on structured finance securities over the last few
years? If yes, please provide supporting evidence.
A.3. While the sharp decline in home prices and contraction of
mortgage credit availability across the U.S. have been key
factors contributing to the current market turmoil, numerous
market sources have identified certain market practices--
including lenient lending practices by mortgage originators and
misrepresentations by certain mortgage brokers, appraisers and
some borrowers themselves--as also contributing to the
unexpectedly poor payment performance of recent subprime
mortgage loans. This is why we are supporting the strengthened
due diligence measures noted in our response to question 2a
above.
Q.4. Ms. Robinson, Ms. Tillman, and Mr. Joynt, according to
testimony provided by Chairman Cox's testimony, Moody's has
downgraded 53 percent and 39 percent of all its 2006 and 2007
subprime tranches; S&P has downgraded 44 percent of the
subprime tranches it rated between the first quarter of 2005
and the third quarter of 2007; and Fitch has downgraded
approximately 34% of the subprime tranches it rated in 2006 and
the first quarter of 2007.
What steps have each of your companies taken during the
past three years to hold accountable its executives and
analysts for the poor performance of its ratings? Has your
company dismissed or otherwise disciplined any of the
executives or analysts responsible for overseeing or producing
its ratings of structured finance products? Please provide a
complete list of disciplinary actions.
A.4. Moody's is committed to providing the most accurate,
objective and independent credit assessments available in the
global credit markets. As in any company, Moody's regularly
evaluates the performance of its employees, including its
executives and analysts. The assessment of the performance of
each employee, which is measured by the ability of the employee
to perform his/her job function, is an assessment that is
distinct from ratings performance. Moody's does make
disciplinary decisions, including employment termination
decisions, based on poor performance in a job function. Moody's
individual personnel decisions, however, are confidential and
the Company is therefore not in a position to provide more
detailed information on specific personnel actions.
Also, as you know, the Securities and Exchange Commission
(``SEC'') has been conducting a non-public examination of
certain NRSORs, including Moody's. Under the Credit Rating
Agency Reform Act of 2006 (``Reform Act'') and related rules
the SEC is entitled to inspect Moody's books and records,
including those relating to Moody's compliance function, credit
policy function and human resources function. Moody's has been
cooperating fully with this extensive examination.
We recognize that the unprecedented financial turmoil that
has developed in the past year has caused a great deal of
anxiety and uncertainty in the markets. While examination of
the root causes of the situation reveals multiple points of
market failure, we believe the speed and extent of rating
downgrades have been one contributor to the loss of confidence
in the credit markets and undermined the credibility of credit
rating agencies.
For Moody's part, we have been and will continue working
hard to respond quickly and sensibly to rapidly changing market
conditions, and we continue to refine our practices to improve
our performance in the future, based on what we have observed
from this confluence of events. We can and must always strive
to improve the quality of our work. Lessons from the recent
market turmoil highlight opportunities for improvements in
assessing the quality of information used in our rating
process, the modeling and explanation of risk factors, and the
application of multi-disciplinary analysis to even the most
highly specialized instruments.
Q.5. Ms. Robinson, Ms. Tillman, and Mr. Joynt, during the last
3 years, did your firm notice a decline in underwriting
standards for mortgages being used to create residential
mortgage-backed securities? If so, did you alter your ratings
process in any way to account for this decline in underwriting
standards?
A.5. Yes. Beginning in 2003, Moody's observed an increase in
the risk profile of subprime mortgage portfolios that we were
asked to review prior to assigning ratings and adjusted our
ratings standards accordingly. Our response to these increased
risks can be categorized into three broad sets of actions:
(1) We began warning the market starting in 2003.
We provided early warnings to the market, commenting
frequently and pointedly over an extended period on the
deterioration in origination standards and inflated housing
prices. We frequently published reports on these issues
starting in July 2003 and throughout 2004, 2005 and 2006.\3\ In
January 2007, we published a special report highlighting the
rising defaults on the 2006 vintage subprime mortgages \4\ and
we have continued to publish on similar trends in the
market.\5\
---------------------------------------------------------------------------
\3\ Please see Annex I, which sets out in a table excerpts from our
publications on this issue. We have also provided you with all of the
documents referenced in Annex I.
\4\ Please see ``Early Defaults Rise in Mortgage Securitization,''
Moody's Special Report, January 18, 2007. We have included this
document in Annex II, which also provides a list of the updates we
provided to the market as well as the actual published research for
your information.
\5\ Please see Annex II.
---------------------------------------------------------------------------
(2) We tightened our ratings criteria.
In response to the increase in the riskiness of loans made
during the last few years and the changing economic environment,
Moody's steadily increased its loss projections and levels of credit
protection for each rating level we looked for on pools of subprime
loans. Our loss projections and credit protection (or ``enhancement'')
levels rose by about 30% over the 2003 to 2006 time period, and as a
result, bonds issued in 2006 and rated by Moody's had significantly
more credit protection than bonds issued in earlier years.
(3) We took rating actions as soon as warranted by the
performance data.
As illustrated by Figure 1, the earliest loan delinquency
data for the 2006 mortgage loan vintage was largely in line
with the performance observed during 2000 and 2001, at the time
of the last U.S. real estate recession. Thus, the loan
delinquency data we had in January 2007 was generally
consistent with the higher loss expectations that we had
already anticipated. As soon as the more significant collateral
deterioration in the 2006 vintage became evident in May and
June 2007, we took prompt and deliberate action on those
transactions with significantly heightened risk.
Figure 2 shows the significantly higher loan delinquencies
in the 2006 vintage, as of July 2007. For example, at 10 months
of seasoning, 8.6% of the underlying loans in the 2006 vintage
were seriously delinquent, nearly twice the level of
delinquencies of the 2001 vintage 10 months after closing.
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Moody's observed the trend of weakening conditions in the
subprime market and adjusted our rating standards to address
the increased risk. Along with most other market participants,
however, we did not anticipate the magnitude and speed of the
deterioration in mortgage quality (particularly for certain
originators), the rapid transition to restrictive lending that
subsequently occurred or the virtually unprecedented national
decline in home prices.
Q.5.a. Did you disclose to investors that there was a decline
in underwriting standards?
A.5.a. Yes. Please see Annex I, including in particular, the
excerpts quoted from the following publications:
(1) 2003 Review and 2004 Outlook: Home Equity ABS (January
20, 2004);
``Moody's expects relatively high defaults and
losses for these mortgage types and has set credit enhancement
levels to offset the risks.'' (Page 5)
``Potentially indicating deteriorating credit
quality, the percentage of full documentation loans in subprime
transactions continues to decline as borrowers choose more
expensive low and no doc alternatives to minimize the time and
scrutiny taken by lenders to underwrite new loans.'' (Page 6)
``Not only are borrowers susceptible to payment
shock in a rising interest rate environment, but at the end of
the TO period borrowers will again suffer payment shock with
the introduction of principal in their monthly payment. Because
of the shorter amortization period, that principal amount will
also be significantly higher.'' (Page 6)
(2) 2004 Review and 2005 Outlook: Home Equity ABS (January
18, 2005); and
``Because these loans are generally underwritten
based on lower initial monthly payments, many subprime
borrowers may not be able to withstand the payment shock once
their loans reset into their fully indexed/amortizing schedule.
The resulting higher default probability, which may be
exacerbated with slowing home price appreciation, could have a
very negative effect on home equity performance in the
future.'' (Page 3)
``The increase in reduced documentation in the
subprime sector is particularly worrisome because for borrowers
with weaker credit profiles the need for establishing repayment
capability with stronger asset and income documentation becomes
even more important.'' (page 6)
``Moody's increases credit enhancement on such
loans to account for the lower borrower equity and the higher
borrower leverage'' (page 6)
(3) 2005 Review and 2006 Outlook: Home Equity ABS (January
24, 2006).
``Full documentation levels fell by almost 10
percent on average per transaction from the beginning of 2004
to the end of 2005. Therefore, in 2005 not only did we see a
proliferation of riskier 'affordability' products, but also a
gradual weakening of underwriting standards.'' (Page 5)
``Moody's loss expectations on the interest-only
mortgages are about 15%-25% higher than that of fully
amortizing mortgages.'' (Page 6)
``In Moody's view, credit risk for this product
is approximately 5% higher than the standard 30 year fully
amortizing product, all other credit parameters being equal.''
(Page 6)
``Moody's considers hybrid ARM loans to be
riskier than equivalent fixed-rate loans primarily because of
the risk of payment shock associated with adjustable-rate
products.'' (Page 6)
Q.6. Ms. Robinson, Ms. Tillman, and Mr. Joynt, when each of
your companies tries to attract new customers, how do you
distinguish your ratings from the ratings of other rating
agencies?
A.6. When endeavoring to attract new customers for our credit
rating services, Moody's seeks to distinguish our rating
services from those of other CRAs on the basis of a number of
factors, including the following:
Moody's overall reputation for trustworthiness
and credibility in the market place, based on the aggregate
performance of our ratings over time, our objectivity and
independence, the depth and breadth of our research and ratings
coverage, transparency and the quality of services.
The analytical capabilities of the rating teams
that cover the particular sector or asset class in question.
The transparency and analytical rigor of our
rating methodologies for the sectors or asset classes in
question.
The depth and breadth of our research and ratings
coverage for the particular sector, geographic region and/or
asset class in question.
The quality of the services we provide to users
of our credit ratings, as well as to issuers and their agents
in the credit rating process. We believe that both users of our
credit ratings and issuers and their agents appreciate the
efficiency, effectiveness, accessibility, and courtesy of our
rating teams, issuer relations teams and investor services
teams.
The ability of Moody's analysts to access and use
the research, data and analytic tools produced by Moody's
Analytics, an operationally and legally separate business unit
within Moody's Corporation.
Moody's credit ratings seek to opine on
``expected loss'', which reflects an assessment of both
probability of default and loss given default. This approach is
a distinctive feature of our credit ratings and differs from
our competitors.
Q.6.a. Do you have empirical data that demonstrates that your
ratings are better than the ratings of other companies? If yes,
please provide documentation supporting your answer.
A.6.a. No. We do not possess the comprehensive, comparative
ratings data histories for each CRA that would be needed to
undertake such analyses. Moreover, we believe performance
comparisons should be made by others, not the ratings agencies
themselves, because ratings agencies naturally have an interest
in the subject matter of such comparisons. However, if we were
to become aware of performance comparisons made by others that
we believed were incorrect or subject to misinterpretation, we
would try to correct the resulting misunderstandings.
Having said that, we would also note that credit default
studies show that our ratings have been remarkably consistent
and reliable predictors of default over many years and across
many economic cycles. The predictive quality of credit ratings
is empirically verifiable and has been evaluated by Moody's and
independent third parties. We would refer you to our Guide to
Default Research, which is attached. Examples of rating
performance reports that we publish include:
Quarterly global and regional reports on
corporate bond rating performance, both with respect to rating
accuracy and rating stability.
Semi-annual reports on global structured finance
rating performance, both in the aggregate and disaggregated by
asset class sub-sectors.
Annual reports on corporate and structured
finance default rates, loss given default rates and rating
transitions.
Periodic reports on default and loss
characteristics of bonds, bank loans and preferred stocks for
specific company sectors and regions.
In addition to publishing issuer or obligation-specific
rating actions and credit opinions, Moody's also publishes our
rating methodologies and various studies relating to the
historical, aggregate performance of our credit ratings. These
and other publications facilitate the assessment of our
ratings' relevance and usefulness by potential users of our
credit ratings as well as other third parties.
As noted in question 6 above, however, our credit ratings
seek to offer an opinion on expected loss, which differs from
what some of our competitors attempt to address, which
consequently makes direct comparison difficult. We believe that
we serve users of our credit ratings best by being as
transparent as possible about our rating methodologies, the
reasoning in support of our credit opinions and the aggregate
performance of our ratings.
Q.6.b. Do you compete more on price or ratings accuracy? Please
provide documentation supporting your answer.
A.6.b. Moody's seeks to compete on the basis of the quality of
our products (including credit ratings and related research),
the trustworthiness of our reputation, and the quality of the
services we provide to users of credit ratings and the people
and firms with which we interact as part of the credit rating
process. We believe that the aggregate performance of our
credit ratings over time is a very important factor in the
assessment of the quality of our work. In this regard, we refer
you to our Guide to Default Research.
Q.7 Ms. Robinson, Ms. Tillman, and Mr. Joynt, what are the
idealized default rates for each of your ratings?
A.7 Moody's does not target specific default or loss rates for
its ratings.\6\ That is to say, Moody's credit rating scale
does not measure ``absolute'' credit risk; rather it provides
an ordinal ranking of credit risk.
---------------------------------------------------------------------------
\6\ Moody's primary objective for its ratings is to provide an
informative ordinal ranking of credit risk at each point in time. As
such, in our view the most appropriate measure of Moody's accuracy is
the ``power'' of its ratings, the information content of their rank
orderings at specific points in time with respect to expected credit
losses (the product of default probability and expected loss severity)
as realized over a long horizon. Credits that have low ratings today
should on average prove to be more risky than credits that have high
ratings today.
In addition to a relative ranking of risk at a point in time, some
investors desire a consistent relative ranking of credits across time,
so that the riskiness of a credit today can be compared to similarly
rated debt instruments in the past. To measure the accuracy of Moody's
ratings across time, the most appropriate metric is the ``power'' of a
pool of ratings assigned to multiple credits, and possibly even the
same credits, observed at different points over time.
Moody's believes that as a consequence of its relative rating
approach, the meaning of its ratings should be highly consistent over
time. Since the relative creditworthiness of bond issuers does not, on
average, change rapidly, there should not generally be any need to
change average rating levels sharply over time. As a practical matter,
therefore, Moody's does not manage its ratings to achieve cardinal
accuracy or to maintain constant default rates by rating category.
Doing so would require Moody's to change its ratings en masse in
response to changes in cyclical conditions. Rather, ratings are changed
``one-at-a-time'' as needed in order to improve the current rank
ordering of credit risk.
---------------------------------------------------------------------------
When, however, we need to associate specific default or
loss rates with ratings for quantitative modeling purposes, we
refer to a table of idealized expected credit loss rates.
(Expected credit loss rates are the products of default
probabilities and expected amounts of loss suffered if defaults
occur.) These idealized loss rates are broadly consistent with
the long-term average historical loss rates of securities that
carry the same ratings and are used for associating modeled
expected losses of both structured and corporate securities
with corresponding ratings.
Some models require default rates, rather than expected
loss rates, as inputs. In those cases, we are able to derive
idealized default rates from the idealized loss rates simply by
dividing every value in the idealized loss table by an
appropriate expected loss severity rate. To derive an idealized
default rate for senior unsecured corporate bonds, for example,
we could assume an average expected loss severity of 55%. For
secured bonds and loans, we would typically assume a lower
severity rate, and for subordinated bonds a higher severity
rate. For structured securities, expected loss severity rates
(and hence idealized default rates) have varied by asset class
and potentially other features of the security.
Since Moody's first began rating municipal securities in
1920, municipal securities have been rated on a separate scale
that places greater weight on default risk than expected loss
severity. This rating scale has been associated with lower
overall credit risk by rating category than comparably rated
corporate and structured securities. For municipal securities,
we have developed a similar idealized default rate table that
is sometimes used to model expected portfolio defaults on a
pool of municipal securities. Given the very limited number of
defaults in the municipal sector and secular changes in credit
risk profiles in the sector, the derivation of this table is
less closely tied to historical data and is more likely to be
reviewed from time to time.
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Q.8. Ms. Robinson, Ms. Tillman, and Mr. Joynt, in his written
testimony, Professor Coffee notes that because only a limited
number of investment banks underwrite structured finance
products, they have leverage over the rating agencies. If they
don't like the ratings they get from one agency, they can go to
another with lower standards.
Has your firm ever felt pressure to lower your rating
standards in order to attract business?
A.8. Issuers, arrangers, underwriters, investors and other
users of credit ratings naturally have strong incentives to try
to influence CRAs' credit rating analysis and decisions, both
when a credit rating is first issued and over the lifetime of
the securities in question. It is not surprising, therefore,
that from time to time, various members of these groups try to
exert pressure on us, e.g. to: (a) change our methodologies,
models or assumptions; (b) reach a decision on a rating that
favors their interests; and/or (c) make a rating decision
faster or slower than we consider appropriate in light of the
information available. Since various market participants and
users of credit ratings often have diverging interests, we are
accustomed to our actions being unpopular with one group or
another.
However, Moody's reputation and long-term success are
critically dependent on the market's confidence in our ethics,
objectivity and credit judgments. Consequently, we have long
had in place strong policies and procedures to ensure the
independence and objectivity of our ratings. (For a more
detailed descriptions in Section 2 of Moody's Code and our
annual reports on implementation of the Moody's Code.\7\) For
example:
---------------------------------------------------------------------------
\7\ We make these documents publicly available on the Regulatory
Affairs webpage at moodys.com and have included them in Annex III to
this response.
ratings are decided on by committees, not
individuals;
analyst compensation is unconnected to either
ratings or fees;
a separate surveillance team reviews the
performance of most structured transactions;
a separate and independent credit policy group
within Moody's is responsible for reviewing and vetting
methodologies and models; and,
perhaps most significantly, our methodologies,
models and processes are publicly available and transparent so
all market participants can assess our integrity and rigor.
Q.8.a. How do you attract customers if your ratings use the
most stringent standards? Will issuers and underwriters simply
go to other firms with less demanding standards?
A.8.a. In our view, the best mechanism to discourage rating-
shopping is investor confidence in our ratings. If investors
believe that our ratings are thoughtful opinions about the
credit quality of a security, they ultimately will demand that
issuers seek our ratings. Alternately, if investors believe
that the models, assumptions and methodologies from Moody's or
another CRA are inappropriately conservative or lax and
therefore fail to produce predictive ratings, over time, we
believe investors, issuers and their agents will prefer the
ratings of another CRA whose ratings appear to be better
predictors of credit quality.
As noted above, Moody's long-term success is critically
dependent on the market's confidence in our ethics, objectivity
and credit judgments.
Q.9. Ms. Robinson, in your written testimony you stated that
Moody's tracks debt for more than 11,000 corporate issuers,
26,000 public finance issuers, and 110,000 structured finance
obligations.
How often does Moody's review and, if necessary, update
each rating?
A.9. The frequency with which Moody's periodically reviews the
creditworthiness of issuers and obligations varies across
sectors and asset classes based on the unique characteristics
of each.\8\ In very general terms, the frequency of our
regular, periodic reviews typically is associated with the
frequency with which new information about the issuer or
obligation is made available. (Ratings may also be reviewed
between these regular, periodic reviews when information
indicates that the creditworthiness of a security could be
materially affected.)
---------------------------------------------------------------------------
\8\ See Moody's Code of Professional Conduct. Provision 1.9 states
our policy regarding monitoring of credit ratings:
``Except for Credit Ratings that clearly indicate that they do not
entail ongoing surveillance, once a Credit Rating is published, Moody's
will monitor the Credit Rating on an ongoing basis and update it by:
1.9.1 periodically reviewing the creditworthiness of the Issuer or
other relevant entity or debt or debt-like securities;
1.9.2 initiating a review of the status of the Credit Rating upon
becoming aware of any information that might reasonably be expected to
result in a Credit Rating action (including termination of a Credit
Rating), consistent with the applicable rating methodology; and
1.9.3 updating on a timely basis the Credit Rating, as
appropriate, based on the results of such review.
---------------------------------------------------------------------------
For example, the frequency of our regular, periodic reviews
for structured finance securities typically is determined by
the scheduled payment dates for the rated securities. This is
the case for two reasons. First, the receipt of transaction
underlying asset performance information from the Trustee or
the Servicer is driven by these payment dates. Second, until
the performance information is received, it will not be clear
whether there has been any deterioration in underlying asset
performance and thus whether a rating adjustment needs to be
considered. Consequently, Moody's structured finance monitoring
process typically occurs either monthly or quarterly, depending
on the frequency with which the trustees or servicers generate
and provide information to us. If we receive performance data
or other information between scheduled payment dates that
indicates material deterioration or improvement in the
creditworthiness of securities, we would take appropriate
action. The transaction performance data is further informed by
Moody's analysis of macroeconomic conditions.
With corporations and financial institutions, analysts for
the issuer in question typically conduct periodic reviews that
are timed to coincide with the publication of financial
statements and other key, periodic filings with authorities
(e.g. on a quarterly, semiannual or annual basis, depending on
the filing and jurisdiction in question). They may also listen
to investor briefings organized by the issuer, monitor the
business and specialized industry press and relevant
authorities' websites. In addition, if they identify
information from the issuer or other sources that would
indicate material deterioration or improvement in the
creditworthiness of securities, they take appropriate action at
that time. Furthermore, rating teams conduct regular (e.g.
annual) portfolio reviews. In a portfolio review, all of the
analysts involved in rating issuers or securities belonging to
a particular sector, together with their supervisors, credit
officers for the sector and possibly related sectors, and
relevant specialists (e.g. corporate governance, accounting and
risk management analysts) are invited to participate in a
meeting where the credit ratings of all issuers in a sub-sector
are considered relative to each other and in light of Moody's
methodology for the sector and outlook for the industry as a
whole.
Within a given sector or sub-sector, there can be
differences in the frequency with which issuers are brought to
a committee for review. For example, all things being equal, an
issuer whose ratings are under review for possible upgrade or
downgrade likely will be brought to a rating committee within a
shorter period of time than an issuer to whom Moody's has
assigned a ``stable'' outlook.
Q.9.a. Does Moody's review municipal ratings as often as it
reviews corporate and structured finance ratings?
A.9.a. As indicated above, the frequency of our review will
depend upon the specific characteristics of each sector and
asset class. In our public finance group the level of issuance
activity in a particular sector, the level of issuance activity
by a particular issuer, the rating level of a particular issuer
(lower rated credits are reviewed more frequently) and the
overall volatility in that issuer's sector are important
factors in determining the frequency of reviews. There are
certain issuers in the public finance sector who are very
active in the debt market, who are not highly rated and who are
in a more credit-sensitive sector. These issuers generally will
have their ratings reviewed on a more frequent basis than those
who, in contrast, are small issuers in less volatile sectors
who access the market very infrequently and whose credit
characteristics are not as complex as some of the larger
issuers.
Consequently, the frequency of our review is directly
linked to the complexity of the credit, the volatility of the
sector, and the susceptibility of the credit to change.
------
RESPONSE TO WRITTEN QUESTIONS OF SENATOR MENENDEZ FROM CLAIRE
ROBINSON
Q.1. During the hearing, I asked you for specifics on the
ratings your agency provided on Bear Stearns in the months
leading up to the collapse. Please provide the Committee with a
detailed explanation of the ratings for Bear Stearns from
November 2007 and March 2008.
A.1.
November 14, 2007: Moody's placed the long-term
ratings of The Bear Stearns Companies Inc. (``Bear'') and its
subsidiaries on review for possible downgrade. This rating
action was taken in response to Bear's announcement that it
expected to post a net loss in the fourth quarter of 2007,
resulting from $1.2 billion in net market valuation losses on
its exposures to subprime mortgage-related assets and CDOs. The
loss also followed weak third quarter 2007 earnings. Moody's
stated in its rating action that it ``expected that on a go-
forward basis, Bear's exposure to these specific assets would
pose only modest downside risk but that during its review
Moody's would evaluate Bear's firm-wide exposures and
valuations in other asset classes.'' Moody's also expressed the
opinion that Bear's performance through the market inflection
and dislocation was more challenged than some competitors,
which reflected not only tough markets, but certain risk and
strategic decisions made by the firm. Specifically, we stated
that ``[A]lthough Bear had improved its earnings
diversification over the past five years, its level and scale
of product and geographic diversification still lagged that of
its peers and had not provided a sufficient buffer to offset
write-downs in mortgages and leveraged lending.''
December 20, 2007: Moody's lowered the long-term
senior debt rating of Bear to A2 from A1 and changed the rating
outlook to stable, ``concluding the review for downgrade that
was initiated on November 14, 2007'' (see above). Factors
considered included the sizeable write-downs on its mortgage
and CDO portfolios, Bear's elevated risk appetite at the time,
and Moody's ongoing concern regarding Bear's corporate
governance, including board oversight of management's strategic
risk decisions and leadership succession planning. The A2
rating and stable outlook factored in Moody's expectations at
that time for the future risk of loss posed by Bear's net
exposures, as well as Moody's expectations for a reduced, but
acceptable, level of operating profitability in 2008. Moody's
also indicated that Bear's ratings benefited from an ample
capital position and strong liquidity profile. Bear had
recently announced a partnership agreement with CITIC
Securities Co. Ltd., which included a $1 billion preferred
stock investment in Bear, further bolstering its capital
position.
March 14, 2008: Moody's lowered the long-term
senior debt rating of Bear two notches to Baa1 from A2 and its
short-term ratings to Prime-2 from Prime-1, and placed the
company's long-term and short-term ratings on continued review
for a possible downgrade. The rating action was in response to
the rapidly deteriorating liquidity position of Bear, which
necessitated an emergency secured funding line from JPMorgan
Chase & Co. (``JPMorgan'') back-stopped by the Federal Reserve
Bank of New York. The 28-day funding facility represented a
temporary liquidity respite for Bear as it looked to identify a
long-term resolution to its liquidity problems.
The rating action reflected Moody's opinion that Bear's
customer franchise had been hurt by the crisis, and would
continue to erode if a long-term stabilizing solution was not
quickly achieved. The review would focus on the financial and
strategic alternatives under consideration by Bear and the
likelihood for a timely resolution.
Given the fluidity of the situation, Moody's stated that it
would re-address its ratings within 7-10 days. Importantly,
Moody's indicated that Bear had a number of attractive
franchises that could facilitate a strategic solution.
On March 17, 2008: Moody's placed Bear's ratings
on review for possible upgrade in response to the announcement
by JPMorgan that it would acquire Bear with assistance from the
Federal Reserve.
March 28, 2008: Moody's announced it was
continuing its review for possible upgrade of Bear's Baa1
ratings and those of its rated subsidiaries. This rating action
followed revisions to the original March 16, 2008 merger
agreement and operating guaranty from JPMorgan.
In addition, please answer the following questions.
Q.1.a. Were any of the ratings downgraded between December 2007
and March 14, 2008?
A.1.a. Response: Yes.
December 20, 2007: Moody's downgraded the
following ratings of Bear Stearns and its subsidiaries:
-- long-term senior unsecured debt to A2 from A1
-- issuer rating to A2 from A1
-- subordinated debt to A3 from A2
-- trust preferred stock to A3 from A2
-- preferred stock to Baa1 from A3
March 14, 2008: Moody's downgraded the following
ratings of Bear Stearns and its subsidiaries, and placed the
ratings on review for possible further downgrade:
-- long-term senior unsecured debt to Baa1 from A2
-- commercial paper to Prime-2 from Prime-1
-- issuer rating to Baa1 from A2
-- subordinated debt to Baa2 from A3
-- trust preferred stock to Baa2 from A3
-- preferred stock to Ba1 from Baa1
Q.1.b. Were any of the ratings downgraded during the week of
the collapse (March 10-14)?
A.1.b. Yes. Please see our response to the above question. On
March 14, 2008, Moody's lowered Bear's long-term senior debt
rating two notches to Baa1 from A2, and placed the company on
review for further downgrade. We expressed the opinion that the
liquidity crisis was the result of sudden diminishing market
confidence in Bear by its counterparties and customers,
compounded by persistently negative market conditions. The
downgrade also reflected our opinion that Bear's franchise had
been hurt by the liquidity crisis and would continue to erode
if a long-term, stabilizing solution was not quickly achieved.
Moody's also noted that Bear had a number of attractive
franchises that could facilitate a strategic solution--which is
what ultimately occurred.
Q.1.c. Can you explain from your agency's point of view how
Bear's collapse unfolded and the role the ratings may have
played?
A.1.c. During the week of March 10, 2008, the market was
flooded with rumors about liquidity problems at Bear. Although
Bear did not face any sizeable net writedowns or credit losses,
and the bulk of its franchises were intact, rampant rumors
about its liquidity position, compounded by persistently
negative market conditions, further eroded confidence in Bear
by its counterparties and customers. Investor concerns over the
impact that the failure of the Peloton and Carlyle hedge funds
would have on Bear added to the pressure.
Because market participants value both accuracy and
stability in credit ratings, Moody's manages its ratings so
that they are changed only in response to changes in relative
credit risk that we believe will endure, rather than in
response to market rumors, transitory events or shifts in
market sentiment. We recognize, however, that rumors about
liquidity problems at a financial institution can, in and of
themselves, contribute to liquidity problems and that liquidity
problems for such an institution can have an enduring impact on
creditworthiness. Consequently, Moody's analysts were actively
reviewing Bear's evolving liquidity position on a daily basis
throughout the week.
It is our understanding that Bear's liquidity situation
declined precipitously between March 12 and March 14, 2008.
What was originally market perception and rumors had become
reality. This sudden erosion in liquidity severely constrained
Bear's financial and operating flexibility. Prime brokerage
clients pulled cash and investment balances out of the firm,
haircut requirements rose on Bear's short-term collateralized
funding and an increasing amount of short-term collateralized
funding failed to roll at maturity. As a result, Bear's
liquidity pool, which had started the week at about $18
billion, rapidly declined to around $5 billion by the end of
Thursday, March 13. On March 14, we downgraded Bear's long-term
senior unsecured debt ratings from A2 to Baa1 and short-term
debt ratings from Prime-1 to Prime-2 and left those ratings on
review for further downgrade. We expressed the opinion that the
liquidity crisis was the result of diminishing market
confidence in Bear by its counterparties and customers,
compounded by persistently negative market conditions. The
downgrade also reflected our opinion that Bear's franchise had
been hurt by the liquidity crisis and would continue to erode
if a long-term, stabilizing solution was not quickly achieved.
Moody's also noted that Bear had a number of attractive
franchises that could facilitate a strategic solution--which is
what ultimately occurred.
Q.1.d. Do you think the lack of changes to the Bear Stearns'
ratings is an example of a unique event in the markets or an
indication of larger flaws in the structure of the ratings?
A.1.d. Moody's believes that our credit ratings of Bear and its
securities were appropriate in light of the information
available to us throughout the relevant time period. Moreover,
although the company's equity suffered a dramatic loss in value
as a result of this crisis, Moody's maintained our credit
rating on Bear's debt at investment grade, in part because Bear
had a number of attractive franchises that could facilitate a
strategic solution--which is what ultimately occurred. As noted
earlier, Moody's ratings speak to whether a debt investor who
holds the securities to maturity will be made whole, and not
whether a company's equity will retain its value.
Ultimately, the issue with Bear was a severe and extreme
crisis of confidence based on a liquidity problem that arose
suddenly and materialized in a matter of days. This crisis of
confidence denied Bear's access to short-term secured
financing, even when the collateral consisted of agency
securities with a market value in excess of the funds to be
borrowed. Confidence sensitivity was expected to be less of an
issue in the secured funding markets, particularly where
franchise impairment was limited. (Notably, Bear had survived
prior crises utilizing many of the same tools that were at its
disposal this time.) However, access to the secured funding
markets, which had operated smoothly throughout many previous
market crises, evaporated over the span of week. The market
dislocation was so extreme that Bear could not borrow against
high-grade collateral. This is a situation that Bear--or any
other securities firm--would find difficult to protect against,
and as a result the Federal Reserve was prompted on March 16,
2008 to provide liquidity to the securities firms.\9\
---------------------------------------------------------------------------
\9\ Specifically, by the implementing the primary dealer credit
facility.
---------------------------------------------------------------------------
Our analysis suggested that Bear was more vulnerable than
the other major securities firms because it had slightly weaker
liquidity, was less diversified and had concentrations in
stressed asset classes. Bear's long-term ratings were lower
than those of its peers, reflecting this risk. However, it also
appears to us that a high degree of risk avoidance by market
participants (due to persistently negative market conditions
and market-wide opacity with respect to counterparty exposures)
may have led to the very unusual situation where market
participants refused to accept high-grade collateral at any
haircut. In addition, during the week of March 10 the departure
of client balances that had financed prime brokerage lending
contributed to Bear's liquidity difficulties.
Q.1.e. Under ideal circumstances, would you agree that the
ratings should have been downgraded to more accurately reflect
Bear's risk?
A.1.e. We believe that our credit ratings of Bear and its
securities appropriately reflected the credit risks of which we
were aware in light of the information available to us at the
time. It is important to note that while Bear Stearns suffered
a severe crisis of confidence, it has not defaulted on any of
its debt instruments, and its ratings are currently on review
for upgrade in connection with its pending acquisition by
JPMorgan. In hindsight, a lower rating on such instruments
would have overstated the risk of default.
Q.1.f. What lessons do you think we should take from the Bear
Stearns collapse as it relates to the credit ratings?
A.1.f. Moody's credit ratings intend to offer an opinion on the
risk of default and severity of loss in the event of default.
As stated above, we believe that our credit ratings of Bear and
its securities appropriately reflected the risks of which we
were aware given the information available to us at the time.
We also believe that, more generally, Moody's long-term credit
ratings strike the appropriate balance between accuracy and
stability. Our conversations with investors, issuers and
regulators have led us to conclude that they have a strong
preference for credit ratings that are both accurate and
stable. They want ratings to reflect enduring changes in credit
risk because rating changes have real consequences--due
primarily to ratings-based portfolio governance rules and
rating triggers--that are costly to reverse. Market
participants, however, do not want us to provide ratings that
simply track market-based measures of credit risk (although
such measures can be a useful supplementary source of
information in the investment decision-making process). They
want our credit ratings to reflect independent analytical
judgments that provide a counterpoint to often volatile market-
based assumptions.
Having said that, the recent market turmoil has highlighted
a vulnerability of securities firms, namely the loss of access
by a solvent firm to secured funding, even when secured by high
quality collateral. This scenario had not previously occurred
in the history of the industry. The SEC is also now more
focused on this vulnerability, as SEC Chairman Cox recently
noted: ``We are discussing with each of the firms various
stress scenarios that include not only impairment of unsecured
funding but also of secured funding. We now live in a post Bear
Stearns reality.'' (Reuters, May 26) In addition, the increased
complexity of these firms and of the financial instruments in
which they deal have elevated the analytic challenge. Moody's
is and will continue to evaluate the appropriateness of our
rating methodology for securities firms in light of the recent
events.
Q.1.g. What are your thoughts on a proposal Professor Coffee
discussed at the hearing for rating agencies to periodically
update ratings?
A.1.g. Moody's believes that credit rating announcements should
be made when the credit rating agency has new or relevant
information to share with the market. Generally, these
instances are either: a change in rating (the rating opinion
has changed); or a rating affirmation (there is a significant
event in the market and investors are unsure whether the rating
remains unchanged).
Moody's does not believe that publishing rating
announcements according to a prescribed timetable or schedule
would prevent mass downgrades or improve the appropriateness of
existing ratings for the following reasons:
Ratings are already monitored on an ongoing basis
and Moody's changes our ratings when our opinion about the
fundamental creditworthiness of the obligation changes.
A requirement to announce on a quarterly, semi-
annual or annual basis that our rating has not changed would
saturate the market with redundant and potentially confusing or
obfuscating information.
Arbitrary review dates could inappropriately
focus investor and issuer attention on those dates, rather than
on credit-relevant events and thereby inadvertently conceal
significant rating actions.
Paradoxically, publishing more information could
reduce the usefulness of the rating and impair transparency.
------
RESPONSE TO WRITTEN QUESTIONS OF CHAIRMAN DODD FROM STEPHEN W.
JOYNT
Q.1. Ratings Scoreboard. What are your views on the
recommendation that has been made for the creation of a central
website which investors could access and on which they could
compare the accuracy of past ratings by the different NRSROs
for the same types of securities?
A.1. In our Joint Response to the Technical Committee of the
International Organization of Securities Commissions'
(``IOSCO'') Consultation Report on the Role of Credit Rating
Agencies in Structured Finance Markets, dated April 25, 2008,
the participating rating agencies (Fitch, AM Best, DBRS,
Moody's and Standard & Poor's) stated our commitment to create
a centralized, industry portal to house our ratings performance
studies and other relevant data.
I note that the SEC will soon publish its recently
announced proposed rules (the ``Proposed Rules''), which will
include a requirement that certain performance statistics be
made publicly available to facilitate comparisons among rating
agencies. Fitch looks forward to working with the SEC in the
context of the Proposed Rules to enhance the availability of
performance data for users of ratings.
Q.2. Due Diligence. Did Fitch undertake to verify the
information it used to decide ratings on the structured finance
products that were subsequently downgraded? In recent years,
there has been widespread awareness, through the press and
otherwise, about the proliferation of so-called ``liar
loans''--mortgage loans with little or no documentation
required and on which borrowers ultimately have stopped paying.
Do you feel that NRSROs should have performed some
investigation or due diligence on structured debt that
contained these ``liar loans''? Are there circumstances under
which NRSROs should be required to perform some form of due
diligence before issuing a rating?
A.2. The principal contacts at the initial stages of the rating
process are with the originator, the issuer and/or the
arranger. Fitch will also receive information and documentation
from the transaction lawyers. These parties will typically
provide an overview of the transaction and the originator, as
well as a detailed term sheet setting out the main features of
the legal and financial structure. The arranger often acts as
the conduit between Fitch and the originator for information on
the underlying assets and their historic performance. It may
also act as a conduit for outside opinions from other experts,
such as accountants.
Where relevant, Fitch will meet the originator to conduct
an on-site servicer review, the purpose of which is to
understand the asset origination process, the way the assets
are administered and what steps are undertaken in the event of
non-performance (e.g., of individual loans within a consumer
loan portfolio). This also represents an opportunity for Fitch
to resolve any outstanding questions about the data that it has
already received. Following this review any further questions
on the origination, underwriting or administration process are
addressed directly to the originator or via the arranger.
Fitch's own lawyers (internal or external) may discuss
legal and structural aspects of the transaction with
transaction counsel, to better understand the transaction and
whether and how legal risks relevant to our credit analysis
have been mitigated. However, in all cases, these reviews are
not designed to supplant or replace the legal analysis
performed by transaction counsel, and are instead undertaken
simply to understand the legal analysis provided by transaction
counsel. In cases where Fitch receives reports and information
from other external advisors or experts, such as auditors,
actuaries and consultants, we may discuss these reports and
information with such third parties to understand their impact
on our credit analysis. Fitch also utilizes data gathered from
servicers, trustees and data services in the course of
monitoring existing transactions to evaluate new transactions.
As part of this process, we consider, among other factors,
(a) the source of the data we receive; (b) the track record of
that source in providing quality data; (c) the predictive
powers associated with any one piece of data; and (d) whether
or not the data (such as financial information) has been
subject to review by a third party.
However, as we make clear in our Code of Conduct and other
documents and publications, Fitch does not audit or verify the
truth or accuracy of any information provided or available to
it. This responsibility is not one which it is feasible or
appropriate for rating agencies to discharge, and one that, in
a clear, statutory context, already exists for other parties.
We do agree with the SEC's position, in the Proposed Rules,
that it would be very helpful to users of our ratings for us to
disclose the extent to which we rely on the due diligence of
others to verify the assets underlying structured products. In
addition, we will be amending our Code of Conduct, in line with
IOSCO's recently amended Code of Conduct Fundamentals for
Credit Rating Agencies, to state that we will adopt reasonable
steps to assess that the information provided to us for use in
ratings is of sufficient quality to support credible ratings.
Indeed, we have already introduced additional measures
aimed at reviewing the plausibility of data used in the rating
process. In November 2007, we announced a reassessment of the
risk management processes of originators, conduits and/or
issuers for product being securitized going forward. Beginning
in January 2008, our RMBS rating process has incorporated a
more extensive review of mortgage origination/acquisition
practices, including a review of originator/conduit/issuer due
diligence reports, and a sample of mortgage origination files.
Additionally, Fitch is studying how a more robust system of
representation and warranty repurchases could help to provide
more stable RMBS performance. Fitch will not rate subprime RMBS
without completion of the review process.
Q.3. Timeliness of Updates of Ratings. Professor Coffee in his
testimony pointed out that major downgrades of CDO securities
``came more than a year after the Comptroller of the Currency
first publicly called attention to the deteriorating conditions
in the subprime market and many months after the agencies
themselves first noted problems in the markets.'' His testimony
also states ``the gravest problem today may be the staleness of
debt ratings.'' What is Fitch doing to update ratings in a
timely manner and eliminate stale ratings? What standards
should NRSROs observe?
A.3. We believe that a number of recent steps will improve our
timeliness.
In order to better signal concerns about potential ratings
pressure, Fitch is rolling out the use of Rating Outlooks for
all structured finance securities. Outlooks indicate those
securities for which the risk of rating actions is heightened,
but has not yet reached the level of Rating Watch.
Additionally, Fitch has made substantial investments in
automation to provide for more frequent in-depth analysis of
the large portfolios of rated RMBS, CDO and other structured
products. This allows for the ability to more quickly
communicate the impact of fast-moving events on large
portfolios.
More broadly, as with many other market participants, we
have learned lessons from the precipitous changes in
performance and environment for several asset classes and are
adding additional review steps to the process by which criteria
assumptions are determined. These will not guarantee that
future assumptions will always be replicated in actual events--
no process could realistically assure this--but they will
incorporate recent experience regarding origination standards,
product correlation and risk-layering.
We have introduced structural changes to a number of
groups, from senior management rotation down to increased
resources devoted to dedicated surveillance work. We have also
added Credit/Risk Officers to each of the ratings groups, to
bring enhanced analytical oversight, experience and training to
these groups. The Credit and Risk Officers will work with each
group to identify important trends and to ensure that Fitch's
analytical process is both rigorous and balanced.
At the same time, we are conscious of the need to manage
expectations of the degree to which the timing of rating
actions will ever meet universal acclaim. Ratings are a
``single-point'' representation which inevitably will be
subject to change as different risks crystallise and others
recede. Particularly when market conditions are volatile,
rating efficacy can also be measured in terms of the swiftness
with which the ratings are revised to reflect a change in
circumstances, rather than their absolute ability to have
predicted a series of unexpected events. It is in this spirit
that we continue to place significant focus on the timeliness
of our continued surveillance.
Q.4. Separate Ratings from Business? Dr. Cifuentes' testimony
contains a recommendation that a rating agency separate its
rating business function from its rating analysis function.
What are your views on how NRSROs should address this analyst
independence concern?
A.4. Fitch acknowledges and addresses the potential conflicts
of being an issuer-paid rating agency in four primary ways.
One, we have separated business development from credit
analysis, to keep each group focused on its core task. Two, we
have relocated all of our non-rating operations into a separate
division, Fitch Solutions, which operates behind a firewall.
Three, we have established and enforce a Code of Conduct and
related policies to address these conflicts. Four, no analyst
or group of analysts is directly compensated on the revenues
related to their ratings. To that end, we are in agreement with
the SEC's Proposed Rule that would prohibit anyone who
participates in determining a rating from negotiating the fee
that is paid for such rating.
Q.5. Ratings Shopping. We have heard concerns about ``ratings
shopping,'' where an underwriter or an issuer goes to the NRSRO
that it feels will give it the highest rating, even if it is
not necessarily the most accurate. Is ratings shopping a
problem? How should the negative aspects of it be addressed?
A.5. We understand the concerns surrounding ``rating
shopping'', which have arisen most recently in the context of
structured finance ratings. To address these concerns, Fitch
has consistently advocated greater transparency regarding
transaction data.
As it stands today, generally there is limited data in the
public market about structured securities prior to their
issuance such that neither investors nor rating agencies who
lack direct contact with the issuer are able to formulate an
informed opinion on structured securities. However, if robust
information about structured finance products were publicly
available once the details of the transaction had been
finalized, agencies could provide ratings, regardless of
whether or not an issuer requested a preliminary rating.
The dissemination of unsolicited ratings, where possible,
likely would reduce the frequency of rating shopping, since
rating opinions could be disseminated into the market
regardless of whether the issuer specifically contracted with
the agency or not. As a result, in many circumstances market
participants would have the benefit of multiple and potentially
diverse opinions about the same transaction.
Additionally, and most importantly, as mentioned above
under Response I, having the underlying data published by the
issuers or originators would allow investors to form their own
opinions about the strengths and weaknesses of a particular
transaction, which could support authorities' efforts to
discourage the use of ratings for purposes other than an
objective measure of relative credit risk. Voluntary efforts
currently in progress being coordinated by the American
Securitization Forum will potentially provide much more
standardized data to all participants in the U.S. RMBS market.
Q.6. Professional Analyst Organization. Dr. Cifuentes in his
testimony suggested ``the creation of a professional
organization, independent of the rating agencies, to which
ratings analysts must belong and which sets forth ethical,
educational and professional standards.'' Please share your
thoughts on the potential merits of such an organization.
A.6. Fitch typically is sympathetic to any industry initiative
which seeks to support analysts from rating agencies, and other
institutions, in their professional development. At the same
time, we note that recent market feedback to the Committee of
European Securities Regulators (``CESR''), with which CESR
concurred,\1\ was that there was no need to impose educational
and professional qualifications upon the staff of rating
agencies.
---------------------------------------------------------------------------
\1\ Paragraph 165 of CESR's Second Report to the European
Commission on the compliance of credit rating agencies with the IOSCO
Code and The role of credit rating agencies in structured finance, May
2008.
---------------------------------------------------------------------------
Membership of such an organization would also have to be
voluntary--it is unlikely we could compel membership by our
employees. Equally, it would be important for operational,
compliance, and regulatory reasons that the formal, mandatory
policies of each individual agency, including our policies on
the management of conflicts of interest, be understood as the
standard by which employee behavior is judged.
------
RESPONSE TO WRITTEN QUESTIONS OF SENATORS SHELBY FROM STEPHEN
W. JOYNT
Q.1. Ms. Robinson, Ms. Tillman, and Mr. Joynt, do you think
that it is easy for investors to compare the accuracy of the
ratings of the different credit rating agencies? If not, do
S&P, Moody's, and Fitch favor the SEC issuing rules to require
enhanced disclosure of ratings performance as Chairman Cox
outlined in his testimony?
A.1. In our Joint Response to the Technical Committee of the
International Organization of Securities Commissions'
(``IOSCO'') Consultation Report on the Role of Credit Rating
Agencies in Structured Finance Markets, dated April 25, 2008,
the participating rating agencies (Fitch, AM Best, DBRS,
Moody's and Standard & Poor's) stated our commitment to create
a centralized, industry portal to house our ratings performance
studies and other relevant data.
I note that the SEC will soon publish its recently
announced proposed rules (the ``Proposed Rules''), which will
include a requirement that certain performance statistics be
made publicly available to facilitate comparisons among rating
agencies. Fitch looks forward to working with the SEC in the
context of the Proposed Rules to enhance the availability of
performance data for users of ratings.
Q.2. Ms. Robinson, Ms. Tillman, and Mr. Joynt, would you please
explain the process by which you obtain the information you use
to rate structured finance securities?
How much of the information is from issuers,
underwriters, or other sources?
Do you ever seek to verify the accuracy of the
information you receive?
A.2. The principal contacts at the initial stages of the rating
process are with the originator, the issuer and/or the
arranger. Fitch will also receive information and documentation
from the transaction lawyers. These parties will typically
provide an overview of the transaction and the originator, as
well as a detailed term sheet setting out the main features of
the legal and financial structure. The arranger often acts as
the conduit between Fitch and the originator for information on
the underlying assets and their historic performance. It may
also act as a conduit for outside opinions from other experts,
such as accountants.
Where relevant, Fitch will meet the originator to conduct
an on-site servicer review, the purpose of which is to
understand the asset origination process, the way the assets
are administered and what steps are undertaken in the event of
non-performance (e.g., of individual loans within a consumer
loan portfolio). This also represents an opportunity for Fitch
to resolve any outstanding questions about the data that it has
already received. Following this review any further questions
on the origination, underwriting or administration process are
addressed directly to the originator or via the arranger.
Fitch's own lawyers (internal or external) may discuss
legal and structural aspects of the transaction with
transaction counsel, to better understand the transaction and
whether and how legal risks relevant to our credit analysis
have been mitigated. However, in all cases, these reviews are
not designed to supplant or replace the legal analysis
performed by transaction counsel, and are instead undertaken
simply to understand the legal analysis provided by transaction
counsel. In cases where Fitch receives reports and information
from other external advisors or experts, such as auditors,
actuaries and consultants, we may discuss these reports and
information with such third parties to understand their impact
on our credit analysis. Fitch also utilizes data gathered from
servicers, trustees and data services in the course of
monitoring existing transactions to evaluate new transactions.
As part of this process, we consider, among other factors,
(a) the source of the data we receive; (b) the track record of
that source in providing quality data; (c) the predictive
powers associated with any one piece of data; and (d) whether
or not the data (such as financial information) has been
subject to review by a third party.
However, as we make clear in our Code of Conduct and other
documents and publications, Fitch does not audit or verify the
truth or accuracy of any information provided or available to
it. This responsibility is not one which it is feasible or
appropriate for rating agencies to discharge, and one that, in
a clear, statutory context, already exists for other parties.
We do agree with the SEC's position, in the Proposed Rules,
that it would be very helpful to users of our ratings for us to
disclose the extent to which we rely on the due diligence of
others to verify the assets underlying structured products. In
addition, we will be amending our Code of Conduct, in line with
IOSCO's recently amended Code of Conduct Fundamentals for
Credit Rating Agencies, to state that we will adopt reasonable
steps to assess that the information provided to us for use in
ratings is of sufficient quality to support credible ratings.
Indeed, we have already introduced additional measures
aimed at reviewing the plausibility of data used in the rating
process. In November 2007, we announced a reassessment of the
risk management processes of originators, conduits and/or
issuers for product being securitized going forward. Beginning
in January 2008, our RMBS rating process has incorporated a
more extensive review of mortgage origination/acquisition
practices, including a review of originator/conduit/issuer due
diligence reports, and a sample of mortgage origination files.
Additionally, Fitch is studying how a more robust system of
representation and warranty repurchases could help to provide
more stable RMBS performance. Fitch will not rate subprime RMBS
without completion of the review process.
Q.3. Do you have any reason to believe that inaccurate or
fraudulent data contributed to the poor performance of your
ratings on structured finance securities over the last few
years? If yes, please provide supporting evidence.
A.3. The very high delinquency and default performance of
recent vintage subprime RMBS and Alternative-A RMBS has a
variety of causes, including declining home prices and the
prevalence of high-risk mortgage products such as stated-income
loans and 100% combined-loan-to-value loans. However, as
indicated in your question, these factors do not fully account
for the large number of early defaults that are occurring. Many
industry observers have noted that poor underwriting, together
with borrower/broker fraud, also appear to be playing a role in
high defaults.
For example, for an origination program that relies on
owner occupancy to offset other risk factors, a borrower
fraudulently stating intent to occupy will dramatically alter
the probability of the loan defaulting. When this scenario
happens with a borrower who purchased the property as a short-
term investment, based on the anticipation that the value would
increase, the layering of risk is greatly multiplied. If the
same borrower also misrepresented his income, and cannot afford
to make the payments, the loan will almost certainly default
and result in a loss, as there is no type of loss mitigation,
including modification, which can rectify these issues.
It is not possible to confidently make a broad statement of
how pervasive these problems are across the range of
originators and issuers in Fitch's rated portfolio. However,
given the combination of our review of historical loan
performance, the level of problems identified in recent Fitch
studies and the findings of third-party reviews, Fitch believes
that poor underwriting quality and fraud may account for as
much as one-quarter of the underperformance of recent vintage
subprime RMBS. More details on this can be found in our
November 28, 2007 report ``The Impact of Poor Underwriting
Practices and Fraud in Subprime RMBS Performance'', a copy of
which is attached to this letter.
Q.4. Ms. Robinson, Ms. Tillman, and Mr. Joynt, according to
testimony provided by Chairman Cox's testimony, Moody's has
downgraded 53 percent and 39 percent of all its 2006 and 2007
subprime tranches; S&P has downgraded 44 percent of the
subprime tranches it rated between the first quarter of 2005
and the third quarter of 2007; and Fitch has downgraded
approximately 34% of the subprime tranches it rated in 2006 and
the first quarter of 2007.
What steps have each of your companies taken during the
past three years to hold accountable its executives and
analysts for the poor performance of its ratings? Has your
company dismissed or otherwise disciplined any of the
executives or analysts responsible for overseeing or producing
its ratings of structured finance products? Please provide a
complete list of disciplinary actions.
A.4. Like all of the major rating agencies, our structured
finance ratings have not performed well and have been too
volatile, but we have found no evidence of violations of our
policies or procedures which would indicate disciplinary action
is either warranted or appropriate.
We have, however, seen merit in making a number of changes
to the senior management team to introduce additional
perspectives into the work of our structured finance groups. On
January 22, 2008, Fitch appointed successors to the positions
of Global Head of Structured Finance Ratings, responsible for
all structured finance ratings globally, and Global Head of
Structured Credit Ratings, responsible for all CDO ratings
globally. In making these and other appointments, we have
reflected a belief that adding senior managers with experience
of corporate and financial sector assets is an important
addition to the robustness of the rating process.
Q.5. Ms. Robinson, Ms. Tillman, and Mr. Joynt, during the last
3 years, did your firm notice a decline in underwriting
standards for mortgages being used to create residential
mortgage-backed securities? If so, did you alter your ratings
process in any way to account for this decline in underwriting
standards?
Did you disclose to investors that there was a decline in
underwriting standards?
A.5. Some degree of decline was apparent in the migration to
higher risk products such as ``no-money-down'' and ``no
documentation'' loans. The rating process accounted for these
factors by assuming higher default and loss rates for these
mortgages than for other, less risky mortgages. We described to
investors the risks of various mortgage products in our
criteria reports, and we discussed the trends to higher risk
products in numerous investor presentations and special
reports, e.g., the 2006 and 2007 Global Structured Finance
Outlook reports.
Fitch did not change the rating process until it became
apparent that not only the underwriting standards, but the
underwriting processes and controls, had become so weak that
RMBS became exposed to very high-risk loans, in many instances
exhibiting evidence of borrower and broker fraud. In response
to these developments Fitch announced an enhanced originator
review process described in Response B above.
Q.6. Ms. Robinson, Ms. Tillman, and Mr. Joynt, when each of
your companies tries to attract new customers, how do you
distinguish your ratings from the ratings of other rating
agencies?
Do you have empirical data that demonstrates that
your ratings are better than the ratings of other companies? If
yes, please provide documentation supporting your answer.
Do you compete more on price or ratings accuracy?
Please provide documentation supporting your answer.
A.6. While we can point to occasions where we believe our
methodologies and rating actions have demonstrated greater
prescience than those of our competitors, at a very high level,
it is difficult to argue conclusively that one set of ratings
is demonstrably ``better'' than another. Our aim has always
been to provide a valid, independent opinion that investors can
use as one additional data point to include in their own
analysis, and that can be judged on its own merits, based on
the quality of our rating commentary, accompanying research and
the published performance data.
The decision by an entity as to which CRA to approach is
based on a variety of factors, including the efficiency of the
rating process, the quality of the analysis and the
accompanying research reports, the relative cost and, most
importantly, the reputation of the agency with investors.
Ultimately, the long-term success or failure of an agency is
measured in terms of the latter, which, in Fitch's case, has
resulted in investors--voluntarily and at their own
initiative--incorporating Fitch in their investment guidelines
over the past five years, on an equal footing with the two
larger agencies. This greater recognition--based on the quality
of our work and not the level of our ratings--has been the
greatest spur to increased business opportunities for our
agency.
Q.7. Ms. Robinson, Ms. Tillman, and Mr. Joynt, what are the
idealized default rates for each of your ratings?
A.7. Given the ordinal nature of ratings--that is, ratings are
a relative ranking, rather than a specific percentage
prediction--Fitch has not historically benchmarked individual
ratings to cardinal default expectations. In our public
transition and default studies, we have measured our
performance using a variety of traditional measures, including
comparisons of cumulative default rates, Gini coefficients and
Lorenz curves.
Q.8. Ms. Robinson, Ms. Tillman, and Mr. Joynt, in his written
testimony, Professor Coffee notes that because only a limited
number of investment banks underwrite structured finance
products, they have leverage over the rating agencies. If they
do not like the ratings they get from one agency, they can go
to another with lower standards.
Has your firm ever felt pressure to lower your
rating standards in order to attract business?
How do you attract customers if your ratings use
the most stringent standards? Will issuers and underwriters
simply go to other firms with less demanding standards?
A.8. We do not view the nominal concentration of investment
banks active in the capital markets as representing increased
leverage such as to threaten the objectivity of our work. The
banks in question operate across multiple asset classes, in
dozens of geographies. In this work, the decision on which
rating agency to approach and ultimately to engage is not
steered centrally by any one individual, or any one group of
individuals within any of the banks. The decision to hire or
not hire a given agency is based on the variety of factors
outlined above in Response F, rather than a narrow
consideration of the treatment of that bank's prior
transaction.
Where the level of credit enhancement is also used by banks
as a determining factor, we believe that our track record amply
demonstrates many segments where our market share was lower in
part because of the credit view which Fitch took. We understand
this as a natural part of the business of being an independent
rating agency, and believe, as noted above in Response F, that
ultimately the long-term success or failure of our agency will
be measured relative to our reputation with investors, not
short-term gains in market share.
Q.9. Mr. Joynt, Chairman Cox outlined in his testimony the
rulemaking areas the SEC is considering.
The outline contains several ideas for improving
competition in the ratings industry.
Are there any additional measures the SEC should consider
to foster competition?
A.9. Fitch supports competition in the marketplace and has been
working diligently to provide an alternative global, full-
service rating agency capable of successfully competing with
Moody's and S&P across all products and market segments. We
believe that one of the Proposed Rules--requiring the public
disclosure of the information a rating agency uses to determine
a rating on a structured product--would be very constructive in
furthering competition. Such disclosure would also have the
added benefit of assisting investors in conducting their own
investment analysis process. However, it may be most practical
that this disclosure requirement should apply to the sources of
the information--i.e., originators, arrangers and issuers--
rather than the receivers of the information.
Q.10. Mr. Joynt, Chairman Cox indicated in his testimony that
the SEC may consider rules that would require all NRSROs to
have access to the information underlying credit ratings.
Would this make it easier for your company and the other
smaller rating agencies to compete against S&P and Moody's?
A.10. Fitch supports competition in the marketplace and has
been working diligently to provide an alternative global, full-
service rating agency capable of successfully competing with
Moody's and S&P across all products and market segments. We
believe that one of the Proposed Rules--requiring the public
disclosure of the information a rating agency uses to determine
a rating on a structured product--would be very constructive in
furthering competition. Such disclosure would also have the
added benefit of assisting investors in conducting their own
investment analysis process. However, it may be most practical
that this disclosure requirement should apply to the sources of
the information--i.e., originators, arrangers and issuers--
rather than the receivers of the information.
------
RESPONSE TO WRITTEN QUESTIONS OF SENATOR MENENDEZ FROM STEPHEN
W. JOYNT
Q.1. During the hearing, I asked you for specifics on the
ratings your agency provided on Bear Stearns in the months
leading up to the collapse. Please provide the Committee with a
detailed explanation of the ratings for Bear Stearns from
November 2007 and March 2008. In addition, please answer the
following questions.
Were any of the ratings downgraded between
December 2007 and March 14, 2008?
Were any of the ratings downgraded during the
week of the collapse (March 10-14)?
Can you explain from your agency's point of view
how Bear's collapse unfolded and the role the ratings may have
played?
Do you think the lack of changes to the Bear
Stearns' ratings is an example of a unique event in the markets
or an indication of larger flaws in the structure of the
ratings?
Under ideal circumstances, would you agree that
the ratings should have been downgraded to more accurately
reflect Bear's risk?
What lessons do you think we should take from the
Bear Stearns collapse as it relates to the credit ratings?
What are your thoughts on a proposal Professor
Coffee discussed at the hearing for rating agencies to
periodically update ratings?
A.1. The ``A+'' long-term Issuer Default and senior debt
ratings reflected Fitch's view that Bear Stearns' capacity for
payment of financial commitments was strong, but more
vulnerable to changes in circumstance or economic conditions
than higher rated obligors. Positive rating considerations
included leading franchises in clearing and securities
settlement and fixed income and equities securities sales and
trading. The company had a conservative market and credit risk
culture as regards proprietary trading and investment banking
relative to peers. Senior executives at Bear Stearns had
established a culture of no surprises and accountability which
had served them well, demonstrated by a very long history of
good and steady profits. There was minimal turnover at high
ranks and material employee ownership indicated a degree of
alignment of the firm's interests with those of its customers.
Fitch analysts meet with broker dealer issuers several
times a year to assess business risk and strategies as well as
review principal ratings factors already listed above. We also
maintain an open dialogue through regular conversations, pre-
earnings calls and regular information requests on business and
balance sheet conditions.
Fitch published a special analysis on liquidity in August
2007, following market liquidity pressures in July and August.
Information requests also increased once the volatile markets
began in earnest in August 2007. Bear Stearns and other issuers
provide us with updates on exposures and commitments to
leveraged loans, commercial real estate, ABS CDOs, mortgage
inventory, counterparty credit relationships to financial
guarantors and hedge funds. We also obtain regular updates on
liquidity and market volatility trends.
Bear Stearns' funding structure was similar to peers
although net adjusted leverage was slightly lower. Bear Stearns
assumed significant operational and reputation risk from its
global clearing and prime brokerage business but had managed
this risk very well historically. Strategic expansion was
thoughtful and carefully balanced against its expenses
resulting in reduced revenue diversification as compared to
peers. Product expansion typically lagged industry trends.
Diversification was limited by this both geographically and on
the product side. The firm had more limited revenues outside
the U.S. The firm had recently been investing in its asset
management business to build higher fee revenues often seen as
ballast against trading results. Its support of a managed hedge
fund in June 2007 was a marked departure for Bear Stearns.
On November 14, 2007, Fitch downgraded a number of Bear
Stearns' ratings. Full rating histories are attached. At the
parent company level, the Short-term Issuer Default Rating of
BSC was lowered to ``F1'' and the Individual rating (a measure
of standalone financial strength) was lowered to ``B/C''.
Additionally, the rating outlook was revised to ``Negative''
from ``Stable''. The downgrades reflected Fitch's view that
Bear Stearns' near term profitability was expected to be weak,
pressured by its exposure to the U.S. mortgage market as a
whole. Its global clearing and equities businesses were
performing well, however; Fitch believed financial performance
in 2008 would remain challenging given the scale of Bear
Stearns' fixed income business and more limited international
scope. Liquidity had been managed well and remained adequate
but deteriorating conditions in the capital markets were
considered a potential threat to Bear Stearns' financial
flexibility. Fitch highlighted that future downgrades could
result from declines in earnings, severe negative valuation
adjustments, an increased risk profile, diminished liquidity,
rising leverage and/or tangible equity erosion.
In December, Fitch continued the Negative Outlook following
Bear Stearns' earnings release of FY07 results in a published
press release. Bear Stearns posted its first quarterly loss in
its history and was mildly profitable for FY07. The firm took
$1.9 billion in mortgage inventory write-downs. Liquidity
measures had improved during the last half of 2007. Ratings
were not downgraded further during the period March 10-13,
2008.
On March 14, 2008, Fitch lowered all ratings on Bear
Stearns and its subsidiaries rated by Fitch. The parent company
Long-term Issuer Default Rating was lowered to ``BBB'' from
``A+'' and the Short-term Issuer Default Rating was lowered to
``F3'' from ``F1''. The Individual rating was lowered to ``C/
D'' from ``B/C''. All issue ratings were also placed on Rating
Watch Negative. The Support rating was raised from ``5'' to
``3'', reflecting the secured loan agreement concluded with JP
Morgan Chase.
Bear Stearns suffered a rapid decline in liquidity over a
24-hour period. In February and early March, there was
unprecedented spread widening in all credit and particularly in
mortgage products as the failure of several high profile hedge
funds pressured prices. Liquidity had dried up in almost the
entirety of the domestic mortgage-backed securities market,
including unprecedented credit spread widening in ``AAA''-rated
US Agency paper.
Bear Stearns had a capital base that was the smallest of
the bulge bracket and had the highest percentage of its
securities inventory in mortgage based assets. It was the
lowest rated broker dealer at Fitch. As indicated above, Fitch
had downgraded its Short-term ratings to ``F1'' in November and
the Negative Outlook indicated a probability that its rating
may face further downgrades. Bear Stearns also possessed a high
market share in providing financing to fixed income hedge
funds. Fitch believes these factors all led to increasing
reluctance by investors to hold its paper, particularly as
their quarter end was approaching.
Fitch believes that market conditions were highly volatile
for several weeks preceding the Bear Stearns' failure. Unique
elements include the unprecedented spread widening in products
that had been highly liquid for years and through multiple
stress scenarios. While similarities can be drawn between this
period and market conditions in the fall of 1998, there are
numerous unique elements including the turmoil in domestic RMBS
markets, the absence of Fannie Mae and Freddie Mac in active
purchases of mortgages due to portfolio caps, the existence of
the mortgage-based ABX indexes allowing greater speculation and
accumulation of short positions, and the increase in hedge fund
and statistically-based program trading in fixed income and
equities. Fitch believes these market conditions likely
resulted in the acceleration of the rate of deterioration.
It is very difficult to attribute ``what if'' scenarios to
the operations of financial markets since human reaction can be
so unpredictable. Ratings consider the diversification of
sources, tenor and types of unsecured funding as well as its
reliance and ability to withstand periods of illiquidity.
Treasury management is an integral part of the culture and
management of these firms and risk mitigation takes multiple
forms including short term limits to rollover risk, investor
concentrations and availability of unencumbered assets.
Contingency funding plans are detailed and make various
assumptions on the firm's ability to shift from an unsecured to
a secured environment.
Fitch noted a shift in industry trends since 1998. The
industry and Bear Stearns, in particular, reduced reliance on
unsecured credit sources, emphasizing the extension of long-
term funds, the use of bank charters to support certain
businesses and increased reliance on secured bank funding
agreements to support the growing inventory of illiquid assets.
Bear Stearns' liquidity ratios were on target with peers.
Its funding coverage of less liquid assets was the strongest of
peers having limited credit granted to investment banking
clients, merchant bank and private equity funds and generally
conservative posture in expanding its balance sheet. It also
maintained a relatively conservative capital structure with
minimal levels of hybrid capital issuances versus peers.
While we strive to incorporate a prospective view in our
ratings, Fitch believes that the evolving credit stress has
elements of great severity and rapidity not previously
foreseen. We are evaluating this scenario, as well as the
recent programs modified by the U.S. Federal Reserve in the
ongoing ratings assessment of the other U.S. based brokers.
Three of the remaining four are presently assigned a Rating
Outlook of Negative.
Our ratings are subject to continuous review, other than
where expressly disclosed as point-in-time in nature. This
means that any material event can cause a rating action for any
rating at any time. Fitch is staffed to ensure that sufficient
analysts of appropriate experience are available to attend
whenever committees need to be called.
The topic of ``bunching'' actual rating actions to meet
pre-determined dates has also recently been discussed in the
context of regulatory consultations in Europe, and we
understand that the overwhelming majority of rating users do
not see a benefit in such an idea.
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
RESPONSE TO WRITTEN QUESTIONS OF SENATOR SHELBY FROM ARTURO
CIFUENTES, PH.D.
Q.1. Dr. Cifuentes, in your testimony you point out that rating
agencies were originally created to provide investors with
information about credit risk. Later, ratings were used by
regulators for a variety of other purposes including
determining capital requirements and establishing investment
guidelines for financial institutions.
Would you please elaborate on your view that ratings that
are useful for investors may not be necessarily as useful for
regulators?
A.1. A regulator is mostly concerned with the soundness
(solvency) of financial institutions; to be precise, whether a
financial institution has enough capital. In that sense, a
regulator prefers conservative ratings (that is, ratings that
err in the direction of ``safety'') and stability (that is,
ratings that do not change frequently for nobody wants to check
the solvency of an institution on a daily basis).
An investor or portfolio manager, on the contrary, benefits
from accurate (that is, neither conservative not aggressive)
and unbiased ratings. Moreover, to the extent that investors
trade securities, they would benefit from timely changes in
ratings. These objectives lend themselves to ratings that
should be more ``dynamic'' (change more frequently).
Therefore, the needs of these two constituencies (stable
versus dynamic ratings; conservative versus accurate and
unbiased ratings) are not one-hundred per cent aligned.
A final observation: Some naive commentators think that the
ratings agencies secretly welcome the use of ratings by
regulators because this would help the agencies to secure a
reliable and steady source of revenue. Actually, the opposite
is true. In fact, T. J. McGuire, a former Moody's Executive
Vice President, gave a speech to the SEC in 1995 in which he
expressed the view that this practice would eventually erode
the integrity and objectivity of the ratings system. (See T. J.
McGuire speech delivered on April, 1995 at the Fifth Annual
International Institute for Securities Market Development, U.S.
Securities and Exchange Commission, available from
www.Moodys.com.)
Q.2. Dr. Cifuentes, the credit rating industry is highly
concentrated with S&P and Moody's each rating more than 90
percent of rated corporate bonds. Fitch is the only other firm
that has significant market share. Chairman Cox has indicated
that the SEC is considering rules to require disclosure about
the performance of the ratings of each rating agency.
Do you believe that disclosure of ratings performance data
would help new credit rating agencies compete with more
established firms?
A.2. No. I don't think so. Ratings performance data are
available already and have done little to help potential new
agencies to compete with the existing ones.
In my opinion, the most significant obstacles faced by a
new agency are a bit different: (i) the three-year ``waiting''
period (essentially, they need to survive for three years while
selling ratings that are not accepted by regulators: an
incredibly tough barrier to entry); (ii) the fact that the
existing rating agencies have not been sanctioned for their bad
performance; and (iii) the fact that the new agencies do not
have access to historical data (bond default frequencies,
recovery values for defaulted securities, etc.) that Moody's
and S&P have.
Q.3. Would this disclosure also make rating agencies more
accountable to investors?
A.3. I don't believe so. Rating agencies, to the extent that
they can protect themselves under the First Amendment, are
accountable to nobody. And worse, whether they are accountable
to investors or not, it is in a sense an academic issue: being
approved by the SEC is the only thing that matters.
Additional Material Submitted for the Record
As Housing Boomed, Moody's Opened Up
The Wall Street Journal, Friday, April 11, 2008
By AARON LUCCHETTI
Bond-rating agency Moody's Investors Service used to be an ivory
tower of finance. Analysts were discouraged from having a drink with a
client. Phone calls from bankers went unanswered if they rang during
intense, almost academic debates about credit ratings.
A decade ago, as the housing market was just beginning to take off,
Moody's was a small player in analyzing complex securities based on
home mortgages. Then, Moody's joined Wall Street and many investors in
partaking of the punch bowl.
A firm once known for a bookish culture began to focus on the
market share that affected its own revenue and profit. The rating firm
became willing, on occasion, to switch analysts if clients complained.
An executive overseeing mortgage ratings went skydiving with a client.
By the height of the mortgage-securities frenzy in 2006, Moody's had
pulled even with its largest competitor, rating nine out of every 10
dollars raised in these instruments. It gave many of the bonds its
coveted triple-A rating.
Profits at the 99-year-old firm, which John Moody started to rate
railroad bonds, rose 375 percent in six years. The share price
quintupled.
Now, Moody's and the other two major rating firms, the Standard &
Poor's unit of McGraw-Hill Cos. and the Fitch Ratings unit of Fimalac
SA, are under fire for putting top ratings on securities that
ultimately collapsed in value. Investors, many of whom relied on
ratings to signal which securities were safe to buy, have lost more
than $100 billion in market value. The credibility of the ratings
system is in tatters as new downgrades of mortgage securities come
almost weekly. Investigators from Congress, the Securities and Exchange
Commission and several state attorneys general are examining the rating
firms' practices.
Moody's acknowledges it sometimes got things wrong in judging
mortgage bonds, but says these were honest mistakes and not the result
of efforts to garner market share. It says it has maintained its rigor
and objectivity in a rating process that is still adversarial toward
big investment banks.
Of the three big rating agencies, Moody's underwent the deepest
cultural change amid the housing boom. At the heart of the firm's
gradual transformation into a player in the mortgage game was Brian
Clarkson, 51 years old, who joined the company as an analyst in 1991
and became president last August. Mr. Clarkson maintains that his focus
on making Moody's friendlier to Wall Street was what the company needed
early this decade. ``We're in a service business,'' he says. ``I don't
apologize for that.''
When Mr. Clarkson first joined Moody's, the agency was known as a
place where analysts often didn't even promptly pick up their phones,
much less talk extensively to companies whose bonds they were rating. A
magazine story in the mid '90s attempted to answer the question ``Why
Everyone Hates Moody's.''
Mr. Clarkson himself had dealt with Moody's as an outsider, and
been frustrated with its manner. As he began to rise within the firm,
he set out to make it more client-friendly and focused on market share.
Firms like Moody's are hired by companies, governments and other
organizations that seek to sell bonds. The firms rate bonds based on
the likelihood they'll default and, in Moody's case, also based on how
much of their principal bondholders are likely to get back.
Top-rated triple-A bonds rarely miss payments, and even if they do,
investors can expect to get nearly all of their money back. Bonds rated
B and C are more likely to lose money for their owners. To compensate
for the added risk, they pay higher interest rates. Bond buyers depend
heavily on the ratings, and conservative investors often buy only
triple-A bonds.
Bond issuers, knowing that a higher rating means they pay a lower
interest rate, have an incentive to shop around among rating agencies.
And they have clout as they shop: They're the ones paying the bill.
Moody's toughness gave issuers reason to go elsewhere, and back in the
mid-1990s, Fitch and S&P were both rating more mortgage bonds than
Moody's, in large part because their standards were considered easier.
For instance, in commercial mortgage-backed securities, Moody's trailed
its two main competitors by 30 percentage points in market coverage in
1996.
That year, Mr. Clarkson took over the group at Moody's that
analyzed such securities. The firm added new analysts and overhauled
its ratings approach, allowing for higher ratings in the area. Within a
year, Moody's moved ahead of both Fitch and S&P in the sector. Rivals
said Moody's had cut its standards. Mr. Clarkson was quoted as calling
this ``sour grapes.'' He says now that the change in the ratings
approach was the right call.
In 1999 Mr. Clarkson took over the part of the firm's ``structured
finance'' business that oversaw bonds and complex securities based on
home mortgages. Moody's rated just 14 percent of high-quality ``prime''
bonds in that area in the year before he took over, compared with 51
percent that Fitch rated and 89 percent that S&P rated, as calculated
by the publication Asset-Backed Alert. (The same bond often gets a
rating from two different firms.)
Moody's top home-mortgage analyst at the time, Mark Adelson, took a
cautious approach that resulted in fewer triple-A ratings. Mr. Clarkson
shook things up, firing or reassigning about two dozen analysts and
hiring new ones who started giving higher grades under a new
methodology. Mr. Adelson left for an investment bank. In 2001, Moody's
market coverage was up to 64 percent. Mr. Adelson says ``the world
thought differently than I did'' about mortgage bonds in 1998 and 1999.
He isn't critical of Mr. Clarkson's management. Mr. Clarkson ``is what
Moody's needs,'' Mr. Adelson says. ``He's very smart, capable and
driven.''
By 2001, Moody's was an independent company. It had long been
tucked inside financial publisher Dun & Bradstreet Corp., but D&B spun
it off as a new public company in 2000. Just before it did so, Warren
Buffett saw the growth and profitability of Moody's business and had
his Berkshire Hathaway Inc. raise its stake in D&B. Berkshire is now
Moody's biggest shareholder, with a 19 percent interest. In some areas,
Moody's continued to make it hard to get a high rating, with the result
that it didn't do much business in those areas; these areas included
the riskier part of home-mortgage bonds and products known as net-
interest margin securities.
Mr. Clarkson encouraged his people to be more responsive picking up
the phone when in the office and to find ways deals could get done
within Moody's methodologies. Customer-service coaches gave sessions on
improving relationships with bond issuers and investors.
``Brian (Clarkson) created a dialogue between Moody's and the
Street that was good,'' says Paul Stevenson, a former Moody's executive
who now works at BMO Financial Group. But ``the most recent problem,''
he says, ``is that the rating process became a negotiation.''
Consider a Bank of America mortgage deal in early 2001. As in most
such deals, the vast majority of the securities based on the pool of
mortgages would be rated triple-A. The question was how big a chunk
would be rated lower paying a higher interest rate and bearing the
brunt of any defaults that occurred.
A rating committee at Moody's voted to require that the issuer put
about 4.25 percent of the deal's value in the lower-rated section, to
provide extra protection for buyers of the top-rated section. But after
Bank of America complained and said it might go with a different rating
firm, Moody's reduced the size of the lower-rated chunk slightly saving
the issuer some interest costs according to people with knowledge of
the matter.
Linda Stesney, a Moody's managing director who was then co-head of
mortgage-backed securities, says she doesn't recall the deal. She says
Moody's reconsidered its view on deals when issuers presented new
information affecting credit quality. She adds that Moody's mortgage
ratings at the time held up well.
In 2002, Mr. Clarkson's realm extended to the fast-growing business
of CDOs. In this complex product, already-sliced-up bonds are further
sliced into new pieces, based on risk and potential return. Moody's was
already rating 90 percent of the dollar value of CDOs. Mr. Clarkson
told an analyst he didn't want bad service to cause that to slip, say
people familiar with the matter.
``There was never an explicit directive to subordinate rating
quality to market share,'' says Mark Froeba, a former Moody's analyst
who recently started a bond valuation company that may compete with
rating firms.
``There was, rather, a palpable erosion of institutional support
for rating analysis that threatened market share.'' An example would be
raising too many legal issues on deals, slowing them down
unnecessarily.
Mr. Clarkson says the goal was maintaining consistency about the
issues Moody's raised on deals. ``I have no problem losing deals for
the right reasons,'' he says. ``We don't change methodology to garner
market share.''
Some supporters say that while Mr. Clarkson cared about market
share, he cared more about the quality of Moody's ratings. Bill May, a
Moody's managing director, recalls Mr. Clarkson warning him in 2002
about the things that could get a managing director fired. He says
inaccurate ratings topped the list, followed by ``arrogant or rude''
behavior toward market participants.
On occasion, Moody's agreed to switch analysts on deals after
bankers complained. Among banks that requested that a different analyst
look at their deals were Credit Suisse Group, UBS AG and Goldman Sachs
Group Inc., according to a person familiar with the matter. The banks
declined to comment. Mr. May says analysts were switched on ``rare''
occasions to accommodate such a request.
Mr. Clarkson stressed relationships, in a break with tradition at
the firm, whose office in Lower Manhattan is adorned with sepia-toned
pictures of its founder. John Bohn, Moody's president from 1989 to
1996, says he used to tell recruits that Moody's was a ``special
business'' where ``you can't go out for beers'' with friends who worked
for investment banks.
Mr. Clarkson's view is that ``it's important to socialize.'' The
onetime mountain climber and recreational weightlifter met with
investment-bank officials and gave speeches at industry conferences
peppered with movie quotes and references to television shows like
``Survivor.''
When Moody's sought to rate more deals for GMAC's residential-
finance unit in the late 1990s, Moody's officials traveled to the
company's Minneapolis offices several times. Mr. Clarkson and several
others from Moody's accepted an invitation to go skydiving with
officials of the GMAC unit. ``We paid our own way,'' Mr. Clarkson
recalls.
Some analysts say they occasionally would attend the dinners that
celebrated the launch of a new CDO Moody's had just rated. Moody's says
it has rules to prevent conflicts, including a $50 limit on gifts, and
that building better relationships with Wall Street officials was part
of its effort to be more transparent in its rating methodologies.
As Moody's staff grew to accommodate the surging mortgage market,
Mr. Clarkson arranged off-site meetings for employees to get to know
each other better. At one, he sung as a Blues Brother, while at
another, two Moody's executives entertained by wrestling in fat suits.
Mr. Clarkson's structured-finance group grew to account for about
43 percent of Moody's revenue in 2006, up from 28 percent in 1998. By
2006, the firm had more revenue from structured finance $881 million
than its entire revenue had been in 2001.
Employees, though paid a fraction of what they could earn on Wall
Street, sometimes grew wealthy from Moody's surging share price and
their stock options. According to a regulatory filing, Mr. Clarkson's
compensation totaled $3.8 million in 2006. The firm's chief executive,
Raymond McDaniel, earned $8.2 million that year, more than twice what
his predecessor made in 2000. Moody's says the rise in their
compensation reflected growth in the overall business, not just the
mortgage area, and that much of the rise came from the increasing value
of stock options that had been granted years before.
By early 2007, some Moody's analysts were growing worried about the
market for securities backed by subprime mortgages. But Mr. McDaniel
told a group of investors in May 2007: ``The good-news story for us''
includes ``very strong growth coming out of our largest business, which
is the structured-finance business. It is both large and a significant
growth engine for the company.''
Despite some analysts' concerns, Moody's rated about 94 percent of
the $190 billion in mortgage-related and other structured-finance CDOs
issued in 2007, the second busiest year ever. Many of those CDOs have
since been downgraded, some from triple-A to levels that suggest
investors will have significant losses. Moody's says some bonds it
rated were backed by fraudulent loans. It also notes that it wasn't
alone in being surprised by the depth of the housing decline. ``We were
preparing for a rainstorm and it was a tsunami,'' Mr. Clarkson says.
Since becoming Moody's president in August, he is spending up to
half of some weeks dealing with regulators. ``They want the same things
we do,'' he says. Some options that Moody's is considering to improve
its process such as adding new labels to structured-finance ratings to
convey the products' unique attributes and risks were earlier raised by
regulators.
Mr. Clarkson says analysts have kept their ``adversarial''
approach, but adds, ``One of the things we have to do going forward is
be more skeptical.''