[Senate Hearing 111-397]
[From the U.S. Government Publishing Office]
S. Hrg. 111-397
SECURITIZATION OF ASSETS: PROBLEMS AND SOLUTIONS
=======================================================================
HEARING
before the
SUBCOMMITTEE ON
SECURITIES, INSURANCE, AND INVESTMENT
of the
COMMITTEE ON
BANKING,HOUSING,AND URBAN AFFAIRS
UNITED STATES SENATE
ONE HUNDRED ELEVENTH CONGRESS
FIRST SESSION
ON
EXAMINING THE SECURITIZATION OF MORTGAGES AND OTHER ASSETS
__________
OCTOBER 7, 2009
__________
Printed for the use of the Committee on Banking, Housing, and Urban
Affairs
Available at: http: //www.access.gpo.gov /congress /senate/
senate05sh.html
----------
U.S. GOVERNMENT PRINTING OFFICE
56-262 PDF 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 JIM BUNNING, Kentucky
EVAN BAYH, Indiana MIKE CRAPO, Idaho
ROBERT MENENDEZ, New Jersey BOB CORKER, Tennessee
DANIEL K. AKAKA, Hawaii JIM DeMINT, South Carolina
SHERROD BROWN, Ohio DAVID VITTER, Louisiana
JON TESTER, Montana MIKE JOHANNS, Nebraska
HERB KOHL, Wisconsin KAY BAILEY HUTCHISON, Texas
MARK R. WARNER, Virginia JUDD GREGG, New Hampshire
JEFF MERKLEY, Oregon
MICHAEL F. BENNET, Colorado
Edward Silverman, Staff Director
William D. Duhnke, Republican Staff Director
Dawn Ratliff, Chief Clerk
Devin Hartley, Hearing Clerk
Shelvin Simmons, IT Director
Jim Crowell, Editor
______
Subcommittee on Securities, Insurance, and Investment
JACK REED, Rhode Island, Chairman
JIM BUNNING, Kentucky, Ranking Republican Member
TIM JOHNSON, South Dakota JUDD GREGG, New Hampshire
CHARLES E. SCHUMER, New York ROBERT F. BENNETT, Utah
EVAN BAYH, Indiana MIKE CRAPO, Idaho
ROBERT MENENDEZ, New Jersey BOB CORKER, Tennessee
DANIEL K. AKAKA, Hawaii DAVID VITTER, Louisiana
SHERROD BROWN, Ohio MIKE JOHANNS, Nebraska
MARK R. WARNER, Virginia
MICHAEL F. BENNET, Colorado
CHRISTOPHER J. DODD, Connecticut
Kara M. Stein, Subcommittee Staff Director
William H. Henderson, Republican Subcommittee Staff Director
Randy Fasnacht, GAO Detailee
(ii)
C O N T E N T S
----------
WEDNESDAY, OCTOBER 7, 2009
Page
Opening statement of Chairman Reed............................... 1
Prepared statement........................................... 30
Opening statements, comments, or prepared statements of:
Senator Bunning.............................................. 2
WITNESSES
Patricia A. McCoy, George J. and Helen M. England Professor of
Law, and Director, Insurance Law Center, University of
Connecticut School of Law...................................... 3
Prepared statement........................................... 30
George P. Miller, Executive Director, American Securitization
Forum.......................................................... 5
Prepared statement........................................... 41
Andrew Davidson, President, Andrew Davidson and Company.......... 7
Prepared statement........................................... 51
J. Christopher Hoeffel, Executive Committee Member, Commercial
Mortgage Securities Association................................ 8
Prepared statement........................................... 62
William W. Irving, Portfolio Manager, Fidelity Investments....... 10
Prepared statement........................................... 69
Additional Material Supplied for the Record
Statement submitted by the Mortgage Bankers Association.......... 74
(iii)
SECURITIZATION OF ASSETS: PROBLEMS AND SOLUTIONS
----------
WEDNESDAY, OCTOBER 7, 2009
U.S. Senate,
Subcommittee on Securities, Insurance, and Investment,
Committee on Banking, Housing, and Urban Affairs,
Washington, DC.
The Subcommittee met at 2:35 p.m. in room SD-538, Dirksen
Senate Office Building, Senator Jack Reed (Chairman of the
Subcommittee) presiding.
OPENING STATEMENT OF CHAIRMAN JACK REED
Chairman Reed. The Committee will come to order. I want to
welcome everyone and particularly thank our witnesses for
making themselves available today.
This hearing will examine a key activity within our
financial markets--the securitization of mortgages and other
assets--and will build on previous hearings this Subcommittee
has held to address various aspects of regulatory
modernization, including hedge funds, derivatives, corporate
governance, SEC enforcement, and risk management at large
financial institutions.
Securitization is the packaging of individual loans or
other debt instruments into marketable securities to be
purchased by investors. At its core this process helps free
lenders to make more loans available for families to purchase
items like homes and cars and for small businesses to thrive.
But we have learned from the financial crisis that
securitization or how it is conducted can also be extremely
harmful to the financial markets and families without
appropriate diligence and oversight. Arguably, many of the
basic requirements needed for effective securitization were not
met over the course of the last several years.
Today's panel will discuss how in recent years the
securitization process created incentives throughout the chain
of participants to emphasize loan volume over loan quality,
contributing to the buildup and collapse of the subprime
mortgage market and the broader economy.
Today we find ourselves in the opposite position from a few
years back with hardly any issuances in key markets that could
help return lending to responsible levels. So this afternoon's
hearing is about how to strengthen the securitization markets
and enact any needed changes to ensure that securitization can
be used in ways that expand credit without harming consumers
and the capital markets.
I have asked today's witnesses to address a number of key
issues, including the role securitization played in the
financial crisis, the current conditions of these markets, and
what changes may be needed for Federal oversight of the
securitization process.
Unfortunately, a number of the banks who issue these
securities could not find anyone in their workforce who was
willing to testify today, but we are lucky to have experts
here, both academic and business experts. I welcome you all and
look forward to your testimony.
Let me now turn it over to Senator Bunning for his remarks.
STATEMENT OF SENATOR JIM BUNNING
Senator Bunning. Thank you, Mr. Chairman.
All it takes is a short amount of time studying the market
for asset-backed securities to realize really how complicated
it is. Right now there is no basic private securitization
market, especially for mortgages. I hope this hearing will help
us all get a better understanding of the market and what we can
do and should be done to make it work better.
In theory, securitization is a great idea that brings more
capital to the financial markets, leading to more loans for
individuals and businesses. Done properly, that is a good
thing. But as we saw last year, if it is done wrong, it can
lead to disaster.
The natural first question is whether the problems we saw
were a result of a bad theory or bad execution. For several
reasons, I think what happened was bad execution as a result of
other bad policies and regulations.
Probably the biggest factor that led to the problems in the
securitization market were artificial demand created by bank
capital rules favoring highly rated securities over whole
loans. That artificial demand found a home in residential
mortgage securities thanks to the GSEs' loose underwriting and
easy money. And the rating agencies enabled it all. We should
start by fixing those problems.
Once the bad incentives and artificial demand are taken
away, real risk analysis can be done, and price can be based on
real value. The Government will not have to solve all the
problems because investors will demand more protections from
the issuers.
For example, the model where issuers were paid by the
number of deals closed and loan originators passed on all
responsibility and collects profits up front will not be
tolerated by investors in the future. That will lead to a
solution tailored to a particular asset and flexible enough to
be changed as the market evolves.
I hope our witnesses will comment on these ideas and
provide some of their own, because we really need them.
Thank you.
Chairman Reed. Thank you, Senator Bunning, and I would
welcome any comments by my colleagues Senator Corker or Senator
Gregg.
[No response.]
Chairman Reed. Thank you very much. Now let me introduce
our witnesses.
Our first witness is Professor Patricia A. McCoy, the
Director of the Insurance Law Center and the George J. and
Helen M. England Professor of Law at the University of
Connecticut Law School. Professor McCoy specializes in
financial services law and market conduct regulation. Prior to
her current role, Professor McCoy was a partner in the law firm
of Mayer Brown in Washington, DC, and specialized in complex
financial services and commercial litigation. Thank you,
Professor McCoy.
Our next witness is Mr. George P. Miller. Mr. Miller is the
Executive Director of the American Securitization Forum, an
association representing securitization market participants
including insurers, investors, and rating agencies. Mr. Miller
previously served as Deputy General Counsel of the Bond Market
Association, now SIFMA, where he was responsible for
securitization market advocacy initiatives. Prior to that, he
was an attorney in the corporate department at Sidley, Austin,
Brown & Wood, where he specialized in structured financial
transactions, representing both issuers and underwriters of
mortgage and asset-backed securities. Thank you, Mr. Miller.
Mr. Andrew Davidson is the President of Andrew Davidson &
Company, a New York firm which he founded in 1992 to specialize
in the application of analytical tools to mortgage-backed
securities. He is also a former managing director in charge of
mortgage research at Merrill Lynch.
Mr. Christopher Hoeffel is an Executive Committee member of
the Commercial Mortgage Securities Association, the trade
association representing the commercial real estate capital
market finance industry. Mr. Hoeffel is also the Managing
Director of the investment management firm Investcorp
International, responsible for sourcing, structuring,
financing, underwriting, and closing new debt investments for
the group. Mr. Hoeffel joined Investcorp from JPMorgan Bear
Stearns where he was a senior managing director and global
cohead of commercial mortgages.
Our final witness is Dr. William Irving, a portfolio
manager for Fidelity Investments. Dr. Irving manages a number
of Fidelity's funds, including its mortgage-backed security
Central Fund, Government Income Fund, and Ginnie Mae Fund.
Prior to joining Fidelity, Dr. Irving was a senior member of
the technical staff at Alpha Tech in Burlington, Massachusetts,
from 1995 to 1999 and was a member of the technical staff at
MIT Lincoln Laboratory in Lexington, Massachusetts, from 1987
to 1995.
Welcome, all of you. Professor McCoy, would you please
begin?
STATEMENT OF PATRICIA A. McCOY, GEORGE J. AND HELEN M. ENGLAND
PROFESSOR OF LAW, AND DIRECTOR, INSURANCE LAW CENTER,
UNIVERSITY OF CONNECTICUT SCHOOL OF LAW
Ms. McCoy. Thank you. Chairman Reed, Ranking Member
Bunning, and Members of the Subcommittee, thank you for
inviting me here today.
In the run-up to the crisis, Wall Street financed over half
of subprime mortgages through private label securitization.
When defaults spiked on those loans and housing prices fell,
securitization collapsed in August 2007. It has been on life
support ever since. When private label securitization comes
back, it is critical to put it on sound footing so that it does
not bring down the financial system again. The private label
system had basic flaws that fueled the crisis.
First, under the originate-to-distribute model, lenders
made loans for immediate sale to investors. In addition,
lenders made their money on up-front fees. Both features
encouraged lenders to ``pass the trash.'' Lenders cared less
about underwriting because they knew that investors would bear
the brunt if the loans went belly up. In addition, to boost
volume and fees, lenders made loans to weaker and weaker
borrowers. In fact, when I have examined the internal records
of some of the largest nonprime lenders in the United States, I
have often found two sets of underwriting standards: lower
standards for securitized loans and higher ones for loans held
in portfolio.
Second, securitizations spread contagion by allowing the
same bad loan to serve as collateral for a mortgage-backed
security, a collateralized debt obligation, and even the CDO of
CDOs. It further spread contagion because investors used
tainted subprime bonds as collateral for other types of credit,
such as commercial paper and interbank loans. This shook
confidence in the entire financial system because investors did
not know where the toxic assets were located.
Last, securitization resulted in a servicing system that
creates thorny barriers to constructive workouts of distressed
loans. We have had too many foreclosures as a result. In this,
there were three victims: borrowers, who were steered into
bafflingly risky mortgages, often at inflated interest rates;
investors, who were forced to rely on ratings because
securities disclosures were deficient and securitizations were
so complex; and, finally, the public, who had to pay to clean
up the mess.
So how do we fix these problems going forward? There are
two aspects: lax underwriting and loan workouts.
First, fixing underwriting. One group of proposals seeks to
realign incentives indirectly so that mortgage actors do
careful underwriting. These include requiring securitizers to
retain risk, higher capital requirements, better compensation
methods, and stronger representations and warranties along with
stiff recourse.
I applaud these measures, but they are not enough to ensure
good underwriting. I doubt, for example, whether prohibiting
issuers from hedging their retained risk is really enforceable.
Banks are adept at evading capital standards, and the Basel II
standards are badly frayed. And stronger reps and warranties
are only as good as the issuer's solvency. Consider the fact
that most nonbank subprime lenders are out of business and 128
banks and thrifts have failed since the crisis began.
Another group of proposals focuses on better due diligence
by investors and rating agency reform. This, too, is badly
needed. However, memories of this crisis eventually will grow
dim. When that happens, query whether investors will really
take the time to do careful due diligence when a high-yield
investment is dangled out in front of them.
For these reasons, we need to finish the work the Federal
Reserve Board began last year and adopt uniform Federal
underwriting standards for mortgages that apply to all mortgage
actors across the board. A brand-new study by researchers at
UNC-Chapel Hill just found that States with similar laws had
lower foreclosure rates than States without those laws. And a
2008 study found that State assignee liability laws did not
reduce access to credit.
Then one last thought: facilitating loan workouts. Here I
propose amending Federal tax laws to tax securitized trusts
unless they provide ironclad incentives to do loan workouts
when cost effective.
Thank you, and I welcome any questions.
Chairman Reed. Thank you very much, Professor.
Mr. Miller, please.
STATEMENT OF GEORGE P. MILLER, EXECUTIVE DIRECTOR, AMERICAN
SECURITIZATION FORUM
Mr. Miller. Chairman Reed, Ranking Member Bunning, Members
of the Subcommittee, on behalf of the American Securitization
Forum, I appreciate the opportunity to testify today.
Securitization plays an essential role in the financial
system and the broader U.S. economy. It is a mainstream source
of credit and financing for individuals and businesses and
finances a substantial portion of all consumer credit.
Currently there is over $12 trillion of outstanding securitized
assets, including mortgage-backed securities, asset-backed
securities, and asset-backed commercial paper.
The size and scope of securitization activities reflects
the benefits and value it has historically delivered to the
financial system and economy. Restoration of greater function
and confidence to this market is a particularly urgent need
today, in light of capital and liquidity constraints currently
confronting financial institutions and markets. With the
process of bank de-leveraging and balance sheet reduction still
underway, and with increased bank capital requirements on the
horizon, it is clear that the credit and funding capacity
provided by securitization cannot be replaced by deposit-based
financing or other alternatives.
Simply put, the recovery and restoration of confidence in
securitization is a necessary ingredient for economic growth to
resume and for that growth to continue on a sustained basis
into the future.
The U.S. securitization markets experienced substantial
dislocation during the ongoing financial market turmoil. While
there are signs of recovery in certain market sectors, others--
most notably, private residential mortgage-backed securities--
remain dormant, with other asset classes remaining
significantly challenged.
Although tightened lending standards are one important
reason for a broader constriction in the supply of credit, the
impairment and reduction in securitization activity plays an
equal, if not more important role.
Certain Government programs, including direct support for
Government-guaranteed mortgage securitization and the TALF
program for certain asset-backed securities, have been
successful in supporting financing and the liquidity needs in
part of this market. However, these programs are temporary, and
a larger challenge remains to create a stable and sustainable
private capital market platform for future securitization
activity.
To accomplish this essential goal, a number of weaknesses
and deficiencies of securitization revealed by the financial
market crisis must be addressed. ASF and the broader industry
are working actively to pursue and implement certain critical
reforms, and we will continue to work constructively with
policymakers on others. I would like to offer several
overriding perspectives on these reform measures.
First, many of the problems that have been identified are
not inherent in securitization per se. Instead, they relate to
the manner in which securitization was used. As a general rule,
the amount of risk inherent in a securitization transaction is
equal to the risk that is embedded in the securitized assets
themselves. However, ancillary practices and strategies, such
as the excessive use of leverage and undue reliance on short-
term funding for long-term liabilities, poor credit
underwriting, or the absence of effective risk management
controls, can amplify and concentrate these risks. This does
not, however, mean that securitization itself is inherently
flawed.
Second, any reform measures should be targeted carefully to
address specific and clearly identified deficiencies. Equal
care should be taken to consider the individual and combined
effects of various policy reforms to ensure that they do not
inadvertently stifle otherwise sound and desirable
securitization activity. We are very concerned that some reform
measures currently being pursued or under consideration--most
particularly, the combined effect of accounting standards
changes and proposed regulatory capital rules--are
counterproductive policy responses that are not reasonably
targeted to address identified problems. Such reforms may
render it prohibitively expensive to securitize a wide range of
consumer and business assets. In turn, this could blunt the
ability of the financial system to originate and fund consumer
and business credit demand that finances jobs and investments,
just as the broader economy begins to recover. We believe that
this is an important matter that would benefit from Congress's
further attention.
Finally, from an industry perspective, ASF is focused
primarily on devising and implementing concrete steps to
improve the basic securitization market infrastructure in
response to specific deficiencies identified in preexisting
practices. Grouped broadly under the heading of ``Project
Restart,'' these reforms will substantially improve and
standardize information and data that is captured and reported
to investors in securitized products, including, in the case of
residential mortgage-backed securities, extensive and detailed
loan level data. With these data enhancements broadly in place,
securitization risks will be more transparent and capable of
evaluation by investors and other market participants. At the
same time, these data and standardization improvements will
support higher-quality rating agency, due diligence, quality
assurance, valuation, and other processes that depend on
accurate and reliable underlying data.
And, finally, and briefly, another important goal of
Project Restart is to enhance and standardize representations
and warranties that originators of mortgage loans typically
provide. Much like a defective product is returned to a store
from which it was sold, a mortgage loan that does not meet
specified underwriting criteria should be returned to the
originator through its removal from a securitization trust for
cash. We believe that more effective representations and
warranties will result in a full retention of economic risk by
originators of defective loans consistent with the policy goal
of requiring those who originate assets for securitization to
retain a meaningful and continuing economic stake in the
quality of those loans.
I thank the Subcommittee for the opportunity to testify
today.
Chairman Reed. Thank you very much.
Mr. Davidson, please.
STATEMENT OF ANDREW DAVIDSON, PRESIDENT, ANDREW DAVIDSON AND
COMPANY
Mr. Davidson. Good afternoon, Chairman Reed, Ranking Member
Bunning, Members of the Subcommittee.
More than 2 years since the collapse of the Bear Stearns
high-grade structured credit enhanced leverage fund, its name a
virtual litany of woes, we are still in the midst of a
wrenching economic crisis, brought on at least in part by the
flawed structure of our securitization markets. I appreciate
the opportunity to share my views on what regulatory and
legislative actions could reduce the risk of such a future
crisis.
I believe that securitization contributed to the current
economic crisis in two ways:
First, poor underwriting led to unsustainably low mortgage
payments and excessive leverage, especially in the subprime and
Alt-A markets. This in turn contributed to the bubble and
subsequent house price drop.
Second, the complexity and obfuscation of some structured
products such as collateralized debt obligations caused massive
losses and created uncertainty about the viability of key
financial institutions.
Now to solutions. Boiled down to the essentials, I believe
that for the securitization market to work effectively,
bondholders must ensure that there is sufficient capital ahead
of them to bear the first loss risks of underlying assets; that
the information provided to them is correct; that the rights
granted to them in securitization contracts are enforceable;
that they fully understand the investment structures; and that
any remaining risks they bear are within acceptable bounds.
If these conditions are not met, investors should refrain
from participating in these markets. If bondholders act
responsibly, leverage will be limited and capital providers
will be more motivated to manage and monitor risks.
If this is the obligation of investors, what then should be
the role of Government?
First, Government should encourage all investors and
mandate that regulated investors exercise appropriate caution
and diligence. To achieve this goal, regulators should reduce
or eliminate their reliance on ratings. As an alternative to
ratings, I believe regulators should place greater emphasis or
reliance on analytical measures of risk, such as computations
of expected loss and portfolio stress tests.
Second, Government should promote standardization and
transparency in securitization markets. While the SEC, the ASF,
and the rating agencies may all have a role in this process, I
believe that transforming Fannie Mae and Freddie Mac into
member-owned securitization utilities would be the best way to
achieve this goal.
Third, Government can help eliminate fraud and
misrepresentation. Licensing and bonding of mortgage brokers
and lenders, along with establishing a clear mechanism for
enforcing the rights of borrowers and investors for violations
of legal and contractual obligations, would be beneficial to
the securitization market. However, I believe that there are
superior alternatives to the Administration's recommendation of
retention of 5 percent of credit risk to achieve this goal.
I would recommend an origination certificate that provides
a direct guarantee of the obligations of the originator to the
investors and the obligation of the originator to the borrowers
coupled with penalties for violations even in good markets and
requires evidence of financial backing. This would be a more
effective solution.
If the flaws that led to the current crisis are addressed
by Government and by industry, securitization can once again
make valuable contributions to our economy.
I look forward to your questions. Thank you.
Chairman Reed. Thank you, Mr. Davidson.
Mr. Hoeffel, please.
STATEMENT OF J. CHRISTOPHER HOEFFEL, EXECUTIVE COMMITTEE
MEMBER, COMMERCIAL MORTGAGE SECURITIES ASSOCIATION
Mr. Hoeffel. Thank you. I am testifying today on behalf of
the Commercial Mortgage Securities Association. CMSA represents
the collective voice of all market participants in the
commercial real estate capital market finance industry,
including lenders, issuers, investors, rating agencies, and
servicers, among others. These participants come together to
facilitate a transparent primary and secondary market for
commercial mortgages.
I am also an investor in CMBS, but I have more than two
decades of experience as a commercial lender and a CMBS issuer.
I would like to thank the Committee for the opportunity to
share our views on securitization, which is crucial to borrower
access to credit and our overall economy.
This afternoon, I will focus specifically on securitized
credit markets for commercial real estate, focusing on three
issues: first, the enormous challenges facing the $3.5 trillion
market for commercial real estate finance, of which about $850
billion is securitized; second, the unique structure of CMBS
and the need to customized regulatory reforms accordingly to
support recovery; and, finally, the need to restore the CMBS
market to meet significant borrower demand.
Today the commercial real estate market is facing a perfect
storm based on three interconnected and pressing challenges.
First, there is no liquidity or lending. In 2007, there were
approximately $240 billion in CMBS loans made, approximately
half of the total real estate lending market. CMBS issuance
fell to only $12 billion in 2008, despite strong credit
performance at the time and high borrower demand. It has now
been well over a year since a new CMBS deal has been done.
Second, there are significant loan maturities through 2010.
In fact, hundreds of billions of dollars is coming due in the
next 2 years. Capital refinance these loans is largely
unavailable, and loan extensions are difficult to achieve.
Third, the downturn in the U.S. economy persists.
Commercial real estate is greatly impacted by the macroeconomic
factors: high unemployment, low consumer confidence, poor
business performance, and falling property values. This last
point is especially important to highlight. Remember,
commercial real estate did not cause the current liquidity
crisis. It has been negatively affected by it now, 2 years into
the crisis. Second, even within the commercial real estate
finance industry, CMBS or securitization did not cause stress.
In fact, nonsecuritized loans are now underperforming CMBS and
are experiencing in some cases greater defaults. Ironically,
securitization may be ultimately an exit strategy for these
troubled loans.
As financial policymakers, including the current and
previous Administration, have rightfully pointed out, no
recovery plan will be successful unless it helps restart the
securitization markets. The IMF also asserts that
securitization will assist withdrawal of Government
interventions, employing private capital to fuel private
lending.
Today many recovery efforts in the commercial real estate
market, such as TALF and PPIP, have been helpful. But they are
in a nascent and delicate stage, as discussed in my written
testimony. So it is important that regulatory reforms,
including accounting changes, as George mentioned, must work to
strengthen the securitized markets and to give private
investors who bring their own capital to the table certainty
you and confidence.
Above all, in the commercial real estate context, there is
a real concern that some of the reform proposals will be
applied in a one-size-fits-all manner that could actually
impede recovery. Specifically, there are a number of important
distinctions between CMBS and other asset-based securities
markets, and the upshot of these distinctions is that they help
the CMBS market avoid problems of poor underwriting or
inadequate transparency. These significant differences are in
four major areas:
First, the borrower. In CMBS, the borrower in most cases is
a sophisticated business within income-producing property and
contractual revenues from tenants as opposed to some situations
in the subprime residential mortgage where a loan may have been
underwritten for a borrower who could not document his income.
Second, the structure of CMBS. There are only about 100 to
300 loans in a typical CMBS deal as opposed to thousands of
loans in residential deals. This enables greater due diligence
and analysis to be performed on CMBS pools by several different
parties, including rating agencies and investors.
Third, the existence of a third-party investor or B-piece
buyer in the securitization process. Unlike other asset
classes, CMBS has an investor who purchases a first loss
position and conducts extensive due diligence as a result,
which includes sit visits to every property. This investor also
re-underwrites proposed loans in a potential pool, and they can
negotiate to kick out any loans in which they do not wish to
invest.
Finally, greater transparency. CMBS market participants
have significant access to loan, property, and bond level
information at issuance and on an ongoing basis. In fact, the
CMSA investor reporting package is used as a model for
transparencies by other types of ABS markets.
It is from this unique perspective that we approach
regulatory reform proposals that will undoubtedly change the
CMBS market. We do not necessarily oppose some of these
proposals despite the fact that they will address practices
that were typical in the subprime and residential
securitization markets, not CMBS. Instead, we ask that
policymakers ensure that such reforms are tailored to address
the specific needs of each securitization asset class and to
recognize the many safeguards that already exist in the CMBS
market today.
In this regard, two aspects of regulatory reform are of
utmost interest to CMSA: a requirement that securitizers--that
is, bond issuers and underwriters--retain at least 5 percent of
the credit risk in any securitized loan pool; and a restriction
of the ability of issuers to protect against or hedge this 5-
percent retained risk.
As is explained in more detail in my written testimony, the
basic concern we have about both of these proposals is whether
they will be applied in a one-size-fits-all manner. While we
agree that it is important for the appropriate parties to keep
skin in the game, CMBS deals are already structured to do this
in a way that has worked well for the market and for the
overall economy for years and can continue to serve the policy
objective that is sought here.
As discussed earlier, first loss buyers conduct their own
extensive credit analysis on the loans, examining detailed
information concerning every property before buying the
highest-risk bonds in the CMBS securitization. If these reforms
are not applied in a tailored fashion, the danger is that the
reforms will end up hampering the ability of CMBS lenders to
originate new loans, thereby limiting capital and the flow of
credit at a time when our economy desperately needs it.
Thank you.
Chairman Reed. Thank you very much.
Dr. Irving, please.
STATEMENT OF WILLIAM W. IRVING, PORTFOLIO MANAGER, FIDELITY
INVESTMENTS
Mr. Irving. Good afternoon, Chairman Reed, Ranking Member
Bunning, and Members of the Subcommittee. Thank you for the
opportunity to participate in today's panel.
I have a very simple three-part message that I want to
convey today.
First, securitization can be a very effective mechanism for
channeling capital into our economy to benefit the consumer and
commercial sectors.
Second, as a result of the financial crisis, the
residential mortgage-backed security market and the asset-
backed market are sharply bifurcated. As I will describe, some
are performing well, some less so.
And then, finally, third, there are four broad areas of
reform worthy of pursuit to help the securitized markets
function better. In my remaining time, I will elaborate on
these three points.
One of the most important benefits of the securitization
process is that it provides loan originators an additional
funding source as an alternative to conventional retail
deposits. As an example, I manage the Fidelity Ginnie Mae Fund,
which has doubled in size in the past year to over $7 billion
in assets. The mortgage-backed security market effectively
brings together shareholders in this Ginnie Mae Fund with
individuals all over the country who want to purchase a home or
refinance a mortgage. In this manner, securitization breaks
down geographic barriers between lenders and borrowers, thereby
improving the availability and cost of credit across regions.
Second, to provide further insight into the value of
securitization, consider what happened to the consumer ABS
sector. From 2005 through 2007, auto and credit card ABS
issuance was roughly $170 billion per year. However, after the
collapse of Lehman Brothers in September of 2008, new issuance
came to a virtual halt. As a result, the interest rate on new
car loans provided by finance companies increased by about 5
percentage points between July of 2008 and year end. Issuance
did not resume until March of this year, when the TALF program
began. Thanks to TALF, between March and September, there was
$91 billion of card and auto ABS issuance. Coincident with the
resumption of a functioning auto ABS market, new car financing
fell back into the 3 percent range.
I will now turn to the agency mortgage market, which is
also performing well, thanks to the extraordinary Government
intervention over the past year. This intervention has had two
parts. First, in September of 2008, Fannie Mae and Freddie Mac
were placed into conservatorship, thus reassuring tens of
thousands of skittish agency MBS investors that the Government
stood behind their investments.
Second, the Federal Reserve pledged to purchase $1.25
trillion of agency MBS by the end of 2009. So far, the Fed has
purchased just over $900 billion, thus reducing significantly
the spread between the yields on agency MBS and Treasuries. As
of this week, the conforming balance 30-year fixed mortgage
rate is approximately 4.85 percent, which is very close to a
generational low. Furthermore, the agency MBS market is deep
and liquid.
In contrast, the new issued private label mortgage market
has received no Government support and has effectively shut
down. From 2001 to 2006, issuance in this market had increased
almost fourfold, to $1.2 trillion. But when the financial
crisis hit, the issuance quickly fell to zero. Virtually the
only source of financing for mortgage above the conforming loan
limit, so-called ``jumbo loans,'' is a bank loan, and generally
the available rates are not that attractive.
At first glance, the higher cost of jumbo financing may not
seem to be an issue that should concern policymakers. But
consider the following. If the cost of jumbo financing puts
downward pressure on the price of homes costing, say, $800,000,
then quite likely there is going to be downward pressure on the
homes costing $700,000 and so forth. So in my opinion, at the
same time that policymakers deliberate the future of Fannie Mae
and Freddie Mac, they should consider the future of mortgage
finance in all price and credit quality tiers.
To help improve the functioning of the securitized markets,
I recommend that regulatory and legislative efforts be
concentrated in four key areas. First, promote improved
disclosure to investors at the initial marketing of
transactions as well as during the life of a deal. For example,
there should be ample time before a deal is priced for
investors to review and analyze a full prospectus, not just a
term sheet.
Second, strengthen credit underwriting standards in the
originating process. One way to support this goal is to
discourage up-front realization of issuers' profits. This issue
is complex and likely will require specialized rules tailored
to each market sector.
Third, facilitate greater transparency of the methodology
and assumptions used by the rating agencies to determine credit
ratings. In particular, there should be a public disclosure of
the main assumptions behind rating methodologies and models.
Finally, support simpler, more uniform capital structures
in securitization deals. This goal may not be readily amenable
to legislative action, but should be a focus of industry best
practices.
Taking such steps to correct the defects of recent
securitization practices will restore much-needed confidence to
this critical part of our capital markets, thereby providing
improved liquidity and capital to foster continued growth in
the U.S. economy.
Thank you, and I look forward to answering your questions.
Chairman Reed. Thank you very much, Dr. Irving.
In fact, I wanted to thank all the witnesses for their not
only very insightful, but very concise testimony. I appreciate
it very much. All of your written statements will be made a
part of the record and any of the statements that my colleagues
wish to be submitted will be made part of the record.
Let me pose a question to all of you, which in some cases
will allow you to elaborate on your initial comments. We have
seen a--I am getting to the point now where I can say lifetime,
and that is a long time--shift from a very small secondary
market for loans to a well-functioning market, now to one that
has basically seized up. I think some rough numbers that I have
seen, that loans on bank balance sheets, roughly $3.5 trillion,
compared to securitization products, about $7.1 trillion, and
that market has sort of collapsed.
So the issue is how do we--or what are the key factors that
are stalling this market and that have to be addressed by us?
And again, I think you have alluded to some of them, but let me
start with Professor McCoy and go down the row.
Ms. McCoy. Thank you. The problem right now on the investor
end is lack of investor trust. Investors were not getting
useful disclosures up front. They simply weren't. They weren't
given information on the individual loans in the loan package
so they could figure out whether the underwriting was good or
bad. The due diligence done on those deals by investment banks
left a lot to be desired, and in some cases, I fear, was
tantamount to fraud to the investors.
When I have looked at securitization prospectuses for
mortgage-backed securities, often they would say, here are our
underwriting standards. But many of the loans in the loan pool
were exceptions to these standards, and there is no further
description of the exception loans or how many of the loans in
the loan pools are exception loans. In some cases, it was more
than half, and I guarantee you they did not exceed the
underwriting standards. They fell far below. So this is a pig
in the poke, and for starters, that needs to be fixed.
My additional concern is that investors' interests are not
always protective of borrowers. We also need to rebuild
securitization so that it does not saddle borrowers unknowingly
with products that they cannot afford to repay, and that is a
separate issue.
Chairman Reed. Mr. Miller? Thank you.
Mr. Miller. Senator, I think the reasons are interrelated.
There are a number of them. I think I would agree, overall, if
I had to characterize it, it would be a significant lack of
confidence in various parts of securitization market activity.
I think that it certainly relates to withdrawal of confidence
from investors who are in parts of the market for the kind of
data and data integrity and reliability to give them comfort
that they are able to evaluate--make meaningful evaluations of
securitized instruments. I think it clearly relates to similar
lack of confidence in certain rating agency methods and
processes.
Having said that, I think it is important to note that
while there are clearly parts of the securitization market that
are dormant and significantly challenged, there are other parts
of the market that are functioning to some reasonable degree of
normalcy, and while I think you can also point to Government
programs, for example, TALF playing a significant role, TALF
has also been beneficial in that it has brought back non-TALF
issuance in investors for products that aren't directly
supported by Government loans. And I think what that reveals is
that it is not something that is endemic to securitization as a
whole, but there are specific and identifiable deficiencies
that need to be addressed.
And so that is why, again, from ASF's perspective, things
that lie more perhaps within the industry's control are areas
where we are focusing to rebuild the securitization
infrastructure, improve the quality, comparability,
standardization, and reliability of data, and then finally to--
and I think this goes to some of Andy's comments, which I agree
with--to help rebuild confidence in the operational processes
and controls so that protective measures that are there to
protect investors and ensure that their rights and entitlements
as promised are delivered, that those protective measures
actually work. And I think there is some significant work and
effort that is needed in that area, but all directed at helping
to rebuild and restore confidence.
Chairman Reed. Mr. Davidson.
Mr. Davidson. Addressing the current illiquidity, I would
focus sort of on two different areas. One is the area of
uncertainty. We still have a tremendous amount of economic
uncertainty and regulatory uncertainty, and that just takes
some investors out of the market because they need the risks to
know a little bit better. And the other area is just the lack
of availability of leverage to certain types of instruments.
Without leverage, many instruments have to trade at very
discounted prices, and so the institutions who hold those now
and do have leverage are not willing to transact at the all
equity price as opposed to the leveraged price. And I think
that is why some of the Government programs, like TALF, have
been so effective is because they have reinstituted leverage
into these markets.
In thinking about the solutions, we have to consider what
is the appropriate amount of leverage and make sure that that
can be delivered through those markets because that will be an
important part of their future success.
Chairman Reed. Thank you.
Mr. Hoeffel, you can also focus in on the commercial loan--
--
Mr. Hoeffel. Yes. I will definitely have a bent to that. I
will look at it from both the investor and the lender point of
view, or the originator or issuer point of view.
For investors, one of the issues, as Mr. Davidson mentions,
is their inability to finance their investments, so that has
caused spreads to be very volatile and, in general, trend
toward higher spreads than what we had seen when the market was
healthier. TALF has certainly helped that from a secondary
point of view, some of the existing securities, and may help in
some new issues if we get some deals done, but that is yet to
be seen.
There are certainly concerns about the rating agencies and
the rating process, not that they were necessarily wrong, but
as I think several people have mentioned, we need greater
transparency so investors can understand what the rating
process is and delve in and do their own critical analysis of
what the ratings mean.
And third, from an investor point of view, I think there
are concerns about the continuous changes in accounting and
regulatory policies. They don't know what the potential
ramifications of investing in a security might be down the road
because some of the FASB rules keep changing and there is a
certain amount of uncertainty there that is unpalatable.
From a lender point of view, because, again, of the
volatility of credit spreads, in order to make a loan work, you
would have to originate it at a pretty high spread today and
that is not competitive to the few people that are active in
the markets. Some banks and life companies are making new
commercial mortgages.
Second, there is a big challenge in the commercial real
estate space to aggregate collateral. In a healthy market, it
takes 3 to 6 months to aggregate sufficient loans to do a
securitization. Today's market, it would probably be longer.
Typically, lenders would hedge their positions against movement
in credit spreads or interest rates during that aggregation
period through a number of different derivative options that
don't exist right now, or are so uncorrelated to the market
that they can't really use them to effectively hedge. So people
are unwilling to take the balance sheet risk to aggregate loans
solely for securitization.
And similarly, lenders or aggregators are concerned about
ongoing accounting changes in the market and how that might
affect them while they are aggregating, but before they sell.
Chairman Reed. Thank you.
Dr. Irving, finally.
Mr. Irving. I will make four comments. First of all, I
think uncertainty about home prices and how borrowers behave
when they are underwater on their mortgage, when the loan-to-
value ratio is greater than 100, has increased the risk premium
in the market.
And the second facet of uncertainty which is causing
skittishness about these securities is just uncertainty about
Government policy. The Government in some sense has been in the
position inadvertently of picking winners and losers in terms
of which investments do well and which do not. Those that get
the Government support perform better than those that do not,
so it becomes less of an intrinsic relative value of the cash-
flows and more an assessment of how the Government policy is
going to go.
The third would be the equity-like price volatility that we
have seen exhibited in many of these marketplaces, again causes
there to need to be an increased risk premium, that is, prices
go down.
And then finally, the complexity. We have sort of a rule of
thumb on our trading room floor that for every additional
sentence I need to describe to my boss the structure of the
security I am buying, the price has to be lower by about a
point, and----
Chairman Reed. That would be terrible here.
[Laughter.]
Chairman Reed. Thank you, Dr. Irving. And one point, I
think, emerges, and I am going to turn it over to Senator
Bunning, is as we proceed forward on financial reform
legislation, that will provide one way of at least an
additional degree of certainty and calculation of the market,
so that might contribute to, in a small way, to expanding this
market.
Senator Bunning.
Senator Bunning. Thank you, Mr. Chairman.
I am going to start on the other side. Without reform of
bank capital standards, rating agencies, and housing subsidies
like the GSEs, is there any way the private asset-backed
security market will ever return?
Mr. Irving. So first of all, we do have evidence that--from
the TALF program for a number of----
Senator Bunning. That is Government-backed, though.
Mr. Irving. No, but where I am going to go with that is
that certain high-quality issuers are no longer relying on the
TALF program. They can do issuance without the benefit of the
Government subsidy.
More generally, though, I would say that the past year's
experience would suggest that in the residential mortgage
market, some sort of a Government guarantee is probably going
to be required, and the evidence that I would put forth to that
would be if you look at the striking difference between the
performance of the agency market, even before the Fannie Mae-
Freddie Mac conservatorship, and a nonagency market, where in
the nonagency market, even prime jumbo responsible loans with a
loan-to-value ratio of 70 were priced at, like, 80 cents on the
dollar, there was so much furor in the marketplace and so much
concern, that I think that that evidence suggests to me that in
times of tremendous stress, at least, there needs to be some
sort of a Government backstop. That is not to say necessarily
you need to have organizations with large retained portfolios,
but some sort of a Government guarantee or credit guarantee, in
my opinion.
Senator Bunning. Another question. Which problems that
surfaced in the asset-backed securities markets can be solved
by market participants on their own, and which need Government
action? All of them, or just some of them?
Mr. Irving. Well, for instance, I think that in terms of
aligning the interests, one of the key principles is to align
the interests of the investors and the issuers. For instance,
there is the proposal of issuers retaining a 5-percent slice of
the security. I would say that that is far too blunt an
instrument, and what we need instead is to take a step back and
set up an overall regulatory environment and then let that
regulator work with a trade organization like ASF to--and CMSA
and come up with more detailed rules that are tailored to each
particular sector of the market.
So, for instance, in that case of aligning the interests of
the investors and the issuers, I think that is something that a
regulator should do by working closely with the organization to
tailor solutions for each individual marketplace.
I think maybe one area where legislative could help would
be in terms of disclosure in the rating agencies, so we have
fuller transparency on their methods and quicker turnaround
when there is changing to their methodology or when they
discover errors in their process.
Senator Bunning. Mr. Miller, you mentioned that there were
about $12 trillion worth of assets. How much would you say of
that is near or under water?
Mr. Miller. Well, from a, I think--and we can look into
this and get back to you with specific detail, but the $12
trillion refers to the amount of securitized assets currently
outstanding----
Senator Bunning. That is correct. I understand that.
Mr. Miller. Right. I think a very small minority of that
would be in technical default, so that the securitized
instruments are not paying as promised----
Senator Bunning. We have approximately five million
homeowners that are in foreclosure or are--obviously, their
houses are worth less than their mortgages.
Mr. Miller. Right. And I am distinguishing here--I am
speaking at the security level, so the mortgage----
Senator Bunning. I understand that, but those securitized
mortgages were the things that were sold as AAA rated, and that
is where we got into all kind of the devil is in the details.
And I find that the rating agencies were right in the middle of
all that. In other words, they were the ones that were selling
those as AAA quality to not only other banks, but the same
banks that had sold them the mortgages in the first place, and
all around the world. And that is why when the bubble burst, it
didn't just burst here in the United States, it burst in Europe
and other places.
Mr. Miller. I would certainly agree with that, and to the
extent that rating agencies were overly optimistic or
miscalculated in terms of their assessment of credit----
Senator Bunning. Do you think they did due diligence in
finding out exactly what kind of mortgages they were
securitizing?
Mr. Miller. I don't think the rating agencies traditionally
have performed due diligence on the underlying mortgages. I do
think--I guess the point that I was going to make is that to
the extent they did get it wrong, I think, number one, it
emphasizes the critical importance of avoiding undue reliance
on rating agencies by all parties.
And then, second, as I indicated in my earlier testimony, I
think one of the core features of reform that will assist
issues and problems with the rating agencies and many other
issues and deficiencies that we have identified is simply
having access to better data that can then support better due
diligence, better quality assurance, better rating agency
processes in a much more transparent way. I think part of the
problem is that judgments made by rating agencies and others
were really not easily capable of similar evaluation by others.
Senator Bunning. Meaning, in other words, being able to
distinguish the mortgages that were in the portfolio----
Mr. Miller. Yes.
Senator Bunning. ----that they were doing.
Mr. Miller. That is certainly part of it.
Senator Bunning. Are all of you familiar with the 1994 law
that the Congress passed giving the Federal Reserve the
jurisdiction over all banks making mortgages and also the
mortgage brokers that were making mortgages? They were
empowered with oversight--the Federal Reserve was--to see that
they were doing their job. In other words, they were watching
the store. And it was exactly 14 years from the day that we
passed that bill that the first regulation was written, and
that was 2 years into Chairman Bernanke's oversight, the first
regulations were promulgated on mortgages. So we went 14 years
without a regulation. Would someone like to comment on that?
Ms. McCoy. Senator, I am very familiar with that history.
Senator Bunning. OK.
Ms. McCoy. I am actually writing a book on it.
[Laughter.]
Senator Bunning. I have spoken enough to write a book on
it, so----
[Laughter.]
Ms. McCoy. I was on the Consumer Advisory Council for the
Federal Reserve from 2002 to 2004. We begged the Federal
Reserve to exercise that power. We were aware of the burgeoning
problems with the subprime market at that time, and I was
privately told by Governor Gramlich that he very much supported
that rule, but it would never fly with the Board.
Senator Bunning. Oh, really?
Ms. McCoy. Yes.
Senator Bunning. Well, it is funny, but the Congress of the
United States gave that power to the Federal Reserve and
expected them to completely fulfill their obligation in
oversight of the mortgage market, whether it be the bank or
whether it be the mortgage broker.
Ms. McCoy. When we would talk to Federal Reserve staff
during that time period, we were told that we only had
anecdotes to offer, that we could not produce proof of a
deleterious effect on the macroeconomy, and that, therefore,
the Board would not take action.
Senator Bunning. Well, I can tell you when Chairman
Greenspan and Chairman Bernanke came before this Banking
Committee as a whole, they were all warned about it, especially
early in the early 2000s, that we were getting ourselves into a
potential bubble situation like we did in the dot-com bubble,
and we couldn't get action out of the Federal Reserve. I am
just wondering if anybody here was aware of that. No one here
was aware that the Fed had that power except the person who was
in direct contact with the Federal Reserve?
Ms. McCoy. Yes.
Senator Bunning. OK. Thank you very much.
Chairman Reed. Thank you, Senator Bunning.
Senator Corker.
Senator Corker. Since I was in another hearing, I am going
to let Senator Gregg go and I will go after him.
Chairman Reed. Senator Gregg.
Senator Gregg. That is very kind of you, Senator.
First off, I thought your testimony was exceptional and
very, very helpful and constructive, everyone's, and the fact
that you were concise and had specific thoughts and ideas as to
what we should do is extremely useful.
My opening thought, though, however, as I listened to all
of you, was does any of this need to be legislated? It sounds
to me like almost every specific proposal you have suggested
should fall to a regulatory agency to do, and most of it went
to underwriting and better underwriting standards, it seemed
like. So I would ask anybody on the panel, is there anything
here that needs legislation to accomplish it versus just having
the proper regulatory agencies noticed that this is the way we
should approach these issues?
Ms. McCoy. Senator, if I may, I have jotted down eight
different things, and we can divide them between the private
market and Government intervention.
I think representations and warranties, recourse clauses,
standardizing products, and having a functioning resale market
for mortgage-backed securities is probably a private sector
function, although the Government might convene discussions
along those lines.
But for Government action----
Senator Gregg. I am talking about Congressional action,
not----
Ms. McCoy. Yes. Yes. I believe that better disclosures to
investors can be handled by the SEC directly and Congress does
not need to intervene there.
Better underwriting standards, I think, do need
Congressional action because the Fed is still not sufficiently
aggressive and there is very strong legislation in both
chambers along those lines.
Higher capital standards, I believe banking regulators will
address.
Rating agency reform may very well need Congressional
attention.
Senator Gregg. I would just note that I think if you are
going to have a uniform underwriting standard, you don't want
that written into law if you want to have flexibility on how--
--
Ms. McCoy. Yes, but I believe----
Senator Gregg. That is going to require some mutation.
Ms. McCoy. The authorization needs to come from Congress
and then delegated, I have proposed, to the new agency.
Senator Gregg. You don't think that power already exists
within the Fed or----
Ms. McCoy. Well, the power may exist within the Fed, but
the Fed is not exercising it effectively.
Senator Gregg. OK. So does anybody else have Congressional
action that is required?
Mr. Davidson. Senator Gregg, certainly in the area of
Fannie Mae and Freddie Mac, which is central to the mortgage-
backed securities market----
Senator Gregg. Yes, I accept that.
Mr. Davidson. ----Congressional action is necessary. And
then that would have a number of spillover effects, depending
on how that process went, that may or may not require further
Congressional action.
Senator Gregg. Does anybody else have anything? You know,
this does come down to underwriting. Everybody used that as an
example of where the problem lies. Should we move toward a
system like the Australians have, where you basically have to
put a certain percent down--in Australia, I think it is 20
percent--then you have recourse on mortgages. Or should we
continue with the system of the Congress telling everybody in
America that they have a right to have a loan to buy a house,
no matter whether they can pay it back or not, through the CRA?
Or is there someplace in between?
Mr. Hoeffel. Senator, I don't think you need to regulate
underwriting per se. I think you need to make sure that
potential investors who might be impacted by the underwriting
are fully aware of what they are investing in, so that if the
underwriting has been poor, it is not glazed over by a rating
or a structure. They have all the information they need to make
the proper assessments.
Mr. Miller. I would agree. I don't think it is desirable to
legislate or regulate underwriting standards per se. I do think
it is important, though, for those involved in credit
underwriting functions, and I am thinking specifically in the
residential mortgage market, for those involved in those
activities--mortgage lenders, brokers, and others--to be
subject to the same type of regulation so that you have a level
playing field and consistent standards that apply to all who
are engaged in those functions.
Ms. McCoy. I am forced to disagree. We saw a situation in
which the residential mortgage lending industry was unable to
organize self-regulation, and, in fact, engaged in a race to
the bottom in lending standards, which was aided and abetted by
our fragmented regulatory system which, as Senator Bunning
noted, refused to impose strong standards. That is how we got
in this mess, and I think the only way that we prevent that
from happening is to have some basic common sense standards
that apply to all lenders in all States from the Federal
Government.
To my mind, the most important one is require borrowers to
produce documentation that they have the ability to repay the
loan at inception. That is common sense. We don't have to
obsess about down payment requirements. But that, to me, is
essential.
Senator Gregg. I don't want to--doesn't that go to
recourse? I mean, should there be recourse?
Ms. McCoy. Against the borrower?
Senator Gregg. Right. Should that be a standard that we
subscribe to in this country, which we don't now?
Ms. McCoy. Well, some States do subscribe to it. It depends
on the State.
Senator Gregg. Well, is it a good idea or bad idea?
Ms. McCoy. I think right now, it is causing people who have
already lost their houses to be pushed further into crisis and
it is not helping the situation right now.
Senator Gregg. And didn't this push to the bottom--wasn't
the shove given by the Congress with the CRA and the way it set
up Fannie Mae and Freddie Mac as basically guaranteed entities?
Ms. McCoy. Actually, CRA loans have turned out to perform
pretty well, and one of the reasons is that banks held them in
portfolios so that those higher underwriting standards actually
applied to CRA loans. They have been a success story among
different classes of loans.
Fannie Mae and Freddie Mac, I agree, they cut their
underwriting standards, but they joined the bandwagon late. The
private label nonconforming loans created a strong competitive
threat that they felt necessary to meet, and so they were not
the cause of the problem, although they did join the bandwagon.
Senator Gregg. Thank you.
Chairman Reed. Senator Corker.
Senator Corker. Thank you, Mr. Chairman, and I am sorry I
missed part of the end of the testimony going to another
hearing, but I got the general idea.
Focusing on commercial real estate right now, I know there
has been a lot of discussion. We were just in New York, lots of
people concerned about this huge amount of indebtedness that is
coming due, huge amounts of loans done 10 years ago. You had
10-year term, 30-year ARM. In essence, you kind of sold the
project at that time because it was almost--you almost got full
value because underwriting was so loose, so you kind of
wondered, what is the problem? These have got to roll over, and
the developer kind of sold the deal on the front end.
But I guess as we--and I know that is not the case in every
case. But what is the key? Some organization that wants to
begin originating commercial real estate loans again and
securitizing them from just doing those things and market needs
to make those be sold by keeping recourse or doing other kinds
of things? I just don't get it, really. The real estate values
are dropping. You are underwriting at lower levels. The bond
holders today are going to take a haircut to get financed out.
The developer is going to have a little bit different deal or
lose his property, but what is to keep the private market from
just functioning right now? I really don't get it, and I don't
understand why the focus is on us.
Mr. Hoeffel. Well, there are a couple of different
responses. One is that there are loans being made by insurance
companies and some banks that are holding those loans----
Senator Corker. Right.
Mr. Hoeffel. So that business is happening. Unfortunately--
--
Senator Corker. And it is heating up a little bit, isn't
it?
Mr. Hoeffel. It has picked up, but there is just not a
capacity for banks and insurance companies to fill the void
that is left by the absence of securitization. They just don't
have the balance sheet strength or size to write all the loans
that need to be written. But it can be done.
For the securitized, and you may have missed these
comments, there is a mechanical issue in that there are a lot
of people who would like to go and make loans to securitize,
but it takes a great deal of time to aggregate a sufficient
pool to go out and create a pool to securitize. It was 3 to 6
months. It may be 6 to 12 months today just because the market
has slowed down.
Traditionally, issuers would hedge their positions against
movements in credit spreads or interest rates during that
aggregation period, but there really aren't any instruments to
do that now. There is no efficient way for them to warehouse
their lines while they are--warehouse their portfolios while
they are aggregating or hedge those specific interest rate
risks--not credit risk, not credit of the underlying asset, but
just movements in market spreads. And until that really exists,
people are not willing to take on the balance sheet strain of
aggregating a billion dollars' worth of new commercial
mortgages, even if they are underwritten to lower values and
better standards.
Plus, there is so much uncertainty on what the ultimate
execution might be for those securitizations. It is kind of a
chicken and the egg. Once a few securitizations get done, an
index will be able to be created so people can use that to
hedge their positions. But until that happens----
Senator Corker. Let me ask you, so I would assume there
are, like, trillions of dollars of legacy securitizations that
already are pulled together.
Mr. Hoeffel. Mm-hmm.
Senator Corker. People have an operating history on those
portfolios. So there would be no risk in aggregation. Those
exist. So why isn't there a market to at least deal with the
legacy issues? Why aren't people cranking that up and going in
and writing those assets down? The operating history is there.
I don't understand why that is not occurring and why somebody
isn't willing just to put up some recourse liability to make
that get done and move on.
Mr. Hoeffel. There is a market for both legacy loans and
legacy securities. The securities market has been helped by
TALF as an ability to finance those acquisitions, but there has
been both TALF-financed and non-TALF-financed trading of
mortgage securities.
For whole loans, there is a market, as well, but those
loans are being purchased based on new values and that requires
the seller to recognize a loss, and many times sellers don't
want to recognize that loss if they don't have to. So if a loan
is written to $100 and the market value based on what you think
the property is worth is $70, to sell the loan, you would sell
it for $70 or less and then the owner of the loan would have to
recognize a $30 loss.
Well, if the mortgage is performing, it is a 10-year loan
and there is sufficient cash-flow today to service that loan,
the seller is going to forestall that sale until they
ultimately have to, and hopefully between now and the time that
loan matures, the value of the underlying asset may improve. So
there hasn't been a lot of impetus for holders of whole loans
to sell.
Senator Corker. So back to the securitization--is it OK if
I continue?
Chairman Reed. Go ahead.
Senator Corker. Back to the securitization piece, I assume
that what is happening on that side is the loans are just being
extended, and if you happen to own some of those securities,
you are just in them longer than you anticipated being in those
securities.
Mr. Hoeffel. That decision is being made by the servicers
alone on a case-by-case basis. In some cases, they are being
extended. In some cases, they are being foreclosed or otherwise
worked out. So there has been resolution, but there is just
such a wave of requests for work-outs and modifications, it is
going to take time to get through that.
Senator Corker. What role should--you know, there is a lot
of discussion here about covered bonds, and I realize that at
the volume levels we are talking about, it is not going to
certainly supplant the need for securitizations down the road,
but what level of faith should we as policymakers have in the
cover loan process here in our country as it relates to
commercial real estate?
Mr. Hoeffel. I think we want to not do anything that
precludes commercial mortgages from being eligible to be in a
covered bond issue. I don't think it is going to be the
solution. It can be another tool to provide liquidity to the
commercial real estate market, but because banks or the issuer
has to keep those assets on their balance sheet and there will
be regulatory capital requirements against those assets, it is
a tool, but it is not going to be a sufficient tool to fill the
void.
Senator Corker. Let me just ask one more question.
Chairman Reed. Take your time.
Senator Corker. Do you think there is a sense among a lot
of the larger players that we are going to do something here? I
am hearing that from some of the larger players, and so instead
of going ahead and taking some of these write-downs and moving
on and sort of taking the pain, they are waiting, thinking that
either through TARP or some other mechanism here, we are going
to create a solution.
Mr. Hoeffel. There is hope that something will happen, and
some players are--may be waiting. I think it would be difficult
to justify to sit around and wait for something to happen if
you don't know it is going to happen, but that may be, in fact,
the case.
Senator Corker. Would it be a good signal to the market to
let everybody know that TARP is over at the end of the year,
that the circumstances that created the need for it are
different and not there today, and would that help the market
sort of move along versus this hope that there is a possibility
that there won't be as great a loss and, therefore, let us hold
on and not do the write-downs now?
Mr. Hoeffel. I think TARP and the TALF financing for
commercial real estate has been a help. It has created
liquidity and it has created trading volumes. Certainty, I
think, is always beneficial. If people know absolutely when
something is going to start and when it is going to stop, the
market can react to it, and the market may not always react
favorably, but it will react one way or the other. And I think
part of the problem with some of these programs is they have
had fits and starts, and people think it is going to go one way
and then it goes another or dates aren't certain. So the market
will react one way or another to certainty and I think
certainty is beneficial for everybody.
Senator Corker. Would anybody like to respond to the
certainty of people knowing that this is a private sector
issue? We may do some regulatory reform down the road, and I
know Mr. Miller had some concerns about what some of those
might be, and I guess in another setting we will probe those,
but does anybody else want to respond to the people who are
involved in commercial real estate financing knowing that
nothing else is going to occur? Would that alone not help move
along the process to some degree?
Mr. Davidson. You know, the Government had become, through
various programs, both the Fed, Treasury, TARP, TALF, PPIP,
very involved in the financing of a wide variety of financial
instruments, and I think rapidly removing all of those at once
will certainly be detrimental to the market because there are
no other mechanisms in place now. So I agree with the idea that
certainty is important, but I also believe that there needs to
be a transition period, given how extensive Government's
involvement currently is in financing.
Mr. Hoeffel. One clarification. It is not only certainty
what Congressional action will be, but certainty of what the
regulators are going to do and what the accountants are going
to do, because all of those things have been interplaying and
some of the good work that is done here is undermined by work
that is done elsewhere in Washington or in Connecticut. So I
think you need certainty on all fronts.
Senator Corker. And I would just close by saying that in
the event we did end TALF as, I think, everybody had hoped
might happen at the end of the year, the programs that are
funded right now, TALF and others, they would continue on until
they ran out. So it wouldn't be like all the Government
assistance that is occurring today would end at the same time.
It is just there would be no more commitments.
Thank you all for your testimony.
Chairman Reed. Thank you, Senator Corker.
Let me begin a second round with a question that Mr.
Hoeffel and others have raised, which is the FASB's role in the
securitization process, particularly Statements 166 and 167,
but all of the FASB rules affect this. Let us start with Mr.
Hoeffel. Can you comment about how that might be inhibiting and
what might be due to help FASB?
Mr. Hoeffel. OK. Well, FAS 166 and 167 get rid of the QSP,
the qualified special purpose entity, that was the vehicle
through which many securitizations were done. On a going-
forward basis, I think we can work with that, but one of the
key issues is that it is retroactive, so that people who have
invested a small part of a securitization pool, maybe the
bottom five to 10 percent, will be forced to consolidate all of
the assets and all of the liabilities for that transaction onto
their balance sheet, which could give them rate cap issues or
low covenance, if they are a private company, on their
financing. So it is a real challenge.
It is almost impossible for these companies to get audited
after the fact because they would have to consolidate
everything down to the individual loan level, which may or may
not be feasible given the terms of the loans themselves. So it
is a significant challenge to the market.
And further, given that there is some weakness in property
markets, if a certain class got wiped out through recognized
losses or realized losses, you could have the next bond holder
have to consolidate. So you could theoretically have a BBB or a
single-A investor suddenly have to consolidate, and that is not
something they had envisioned at all when they bought those
bonds.
That will be specific to certain issues, but it is a
challenge, both the lack of QSP for new issue and the
consolidation that would happen for existing debt that is out
there.
Chairman Reed. Any other comments? Mr. Miller.
Mr. Miller. Thank you, Mr. Chairman. Just extending those
comments a little bit to other parts of the securitization
market, at a macro level, the outcome of the 166 and 167
accounting standards changes will be to require a large volume
of securitized assets in many different product sectors to be
put back onto balance sheets or to prevent them from moving off
balance sheet.
And our position--I want to be clear about this--is we are
not for or against on- or off-balance sheet accounting. We
simply think the accounting should be appropriate in light of
exposure to risks or entitlement to assets. We think FASB's
outcome in these standards is to--will result in an over-
consolidation of many of these vehicles where the consolidating
party really does not have meaningful entitlements to the
benefits of those assets or exposure to the risks.
Having said that, the standards have been--will be adopted.
They will generally take effect in January. I think our bigger
concern at this point, as I mentioned earlier, is the ripple
effects of those accounting standards changes. If you picture a
very large volume of assets coming back onto bank balance
sheets exactly at a time when those balance sheets are already
very constrained, the larger asset side of the balance sheets
will attract higher regulatory capital charges. They will
factor into leverage ratio calculations. They will attract loan
loss reserves.
And our concern, and we have heard this very forcefully
from our members across a wide range of markets, is that that,
coupled with other steps being considered, may constrict the
ability of financial institutions to use their capital base to
support new lending. And so we are very concerned, both about
the accounting standards changes, but even more so now about
the downstream impacts that those may have.
Chairman Reed. Any other comments on this issue?
Mr. Hoeffel. One other thought is that the 5-percent
retention that is being discussed in the regulatory reform
proposals will--this sort of flies in the fact of that, because
if you are required to retain 5 percent, now you are going to
have to consolidate. So it exacerbates some of the issues that
Mr. Miller mentioned.
Chairman Reed. Let me raise another issue with Mr. Hoeffel,
and anyone else, and that is, there seems to be one distinction
between residential mortgage-backed securitization and remedies
and commercial, which is commercial bankruptcy code is
available to the individual mortgages if they default. Is that
a difference that makes a difference in terms of the commercial
market versus the residential market?
Mr. Hoeffel. I think it will impact the resolution of
workouts, and we have seen that forestall some of the workouts
that have happened in some very large securitizations to date.
We are still waiting to hear what the outcome of some of those
cases are.
So, again, I think that process, because commercial
borrowers can file for bankruptcy--we have tried to avoid that
through recourse carveouts for bankruptcy, but even that seems
to be not as enforceable as maybe some people had thought. So
it does create, again, more uncertainty, which is a challenge
for investors going forward.
Chairman Reed. But at least in the commercial context, the
access to bankruptcy was clearly understood before the
securitization process took place. And I guess the question
would be if someone has the ability to work it out, it is
understood beforehand. That is anticipated by the investors.
Does that facilitate the process at all or is it sort of
neutral?
Mr. Hoeffel. Well, investors always knew that bankruptcy
was an option and always has been. We tried structurally to
limit a borrower's ability to file for a bankruptcy by putting
most securitized loans into special purpose entities where you
needed unanimity of all the directors to file for bankruptcy,
and there you had independent directors that would not file for
bankruptcy on a solvent entity. Again, there have been some
court challenges to that, and many of us are waiting to find
out what will happen.
Nobody thought that commercial real estate, even in an SPE,
was completely bankruptcy proof, but we did think that there
were enough hurdles to that to provide protections for
investors.
Chairman Reed. Thank you.
Professor McCoy, you have suggested that borrowers be given
an affirmative claim against assignees, a violation of Federal
lending standards. Can you elaborate on that? Then I would ask
others to comment on that proposal.
Ms. McCoy. Yes, I would be glad to. We are in a situation
right now where in the majority of States, if a borrower's loan
is sold, generally through securitization, they lose, without
their consent, their defenses to collection and their ability
to sue the holder of the loan for consumer protection
violations and fraud. And where the rubber really hits the road
is when that borrower is sued for foreclosure.
If the loan has been securitized, let us say the borrower
was defrauded originally, the loan later goes into foreclosure,
under State law the borrower cannot raise the fraud as a
defense to foreclosure. They lost that, and they lost that
through a process over which they had no say.
In addition, because the borrowers can really only sue
their lender, or their mortgage broker, it means that we do not
have the threat of making the borrower whole that investment
banks have to care about; that investors have to care about
when they think about will we do due diligence or just rely on
the rating agency.
And I feel in order to bring rationality and consistency to
the entire mortgage process, we need to allow borrowers to
bring claims of fraud and consumer protection violations
against whoever holds their loan.
Now, there are ways you can structure this liability that
rating agencies can rate and that securitization can function
with. Economists and I and other coauthors studied the effect
of similar laws in nine States, and what we found is in six of
those States, access to subprime credit actually increased,
holding everyone else constant, despite assignee liability.
In three of the States, depending on the indicator, the
results were mixed, but in no State was there an affirmative
drop in access to credit.
Chairman Reed. And I am going to ask others to comment on
this, obviously, but to follow up, would this be a way to
complement or displace the requirement of the Administration to
hold 5 percent of a mortgage or 5 percent to give the
originator sort of some skin in the game or----
Ms. McCoy. I view it as a complement. I am supportive of
the Administration's 5-percent retention measure, but my
concern is even though the Administration would prohibit
hedging it, I do not think that that prohibition is enforceable
because often hedges are taken on a broad variety of positions.
And there is a lot of devil in the details with respect to that
proposal, so I would have assignee liability as well.
Chairman Reed. Mr. Miller, and anyone else who wants to
jump in on this issue.
Mr. Miller. Certainly borrowers should have remedies and
defenses against fraud that may relate proximately to a
foreclosure action against them. The details, though, of any
assignee liability mechanisms are very, very important and
should be addressed at that level of detail. But, broadly
speaking, the securitization industry would have very
significant concerns about broad-based assignee liability.
While, again, the interests of the borrowers here are primary,
at the same time those borrower interests are also served by
having investors who are willing to commit capital to the
mortgage-backed securities markets. And if those investors are
potentially subject to downstream claims by borrowers for
origination defects over which they have absolutely no ability
to perform diligence upon or to verify, they are not sitting at
the loan closing table, my fear is that and I think the
industry's fear is that if those types of assignee liability
provisions are broadly introduced, it will significantly
curtail, if not dry up completely, the willingness of investors
to take that risk. If it is not a risk that they can manage, I
do not believe that it is a risk that they are broadly going to
undertake. So there are some very significant competing
considerations that would weigh against broad-based assignee
liability.
Chairman Reed. Just a follow-up. Would one of aspect of
this might be that those investors would be much more careful
about what they are buying and what they are investing in?
Because they would like to make sure that the originator was
doing their job in underwriting and that would be a market
solution to this problem.
Mr. Miller. I think they certainly want to be and will be
more careful. I think the issue, though, is whether they really
are in a position even with the extreme time and effort and due
diligence to be able to know whether, in fact, fraud was
committed. I do not think they can be in that position, and so
there, I do not think that at least it is a universal solution
or market-based response that could work.
Chairman Reed. Thank you.
Mr. Davidson, do you have a comment?
Mr. Davidson. Sure. In both my written statement and oral
statement, I mentioned this idea of an origination certificate.
Chairman Reed. Right.
Mr. Davidson. And the idea of that is an alternative to
both the assignee liability and the current way that
representations and warranties travel through the system. And
the idea there is to say that these are the obligations of the
originator and that that tracks along with the loan, or whoever
the investor is, and stays as an obligation back to the
borrower, and that we also track through a bonding system or
other capital system capital of that originator or lender so
that in the case there are violations of representations and
warranties or there is fraud against the borrower, there is
money to go after. And so this puts the responsibility in the
hands of the person who created the problem rather than other
parties who really, as Mr. Miller has said, cannot really know
exactly what happened.
Chairman Reed. All right. And, Mr. Miller, you have also
suggested a unique identification number for loans. How would
that work? And would it work in conjunction with Mr. Davidson's
proposal? Or what other aspects would it help?
Mr. Miller. I think that it would, and just building on
what Mr. Davidson just indicated, I think also the
representations and warranties and enhancements there are
really, I think, very consistent with what he was stating in
terms of creating an ongoing economic responsibility. His
proposal is a bit of a variation on that theme.
I think the unique loan identifier, which ASF has recently
announced, will broadly assist the process of being able to
drill down to the individual level of the mortgage loan as that
makes its way into the secondary and debt capital market so
that no matter what type of securities structure--it could be a
whole loan sale, it could be a mortgage-backed securitization,
it could be another type of instrument down the road--investors
and other parties would be able to identify the specific loans
underlying that instrument and coupled with the other data,
standardization enhancements through Project Restart, be able
to perform analytics at a very deep level of detail, providing
investors and other market participants with a much better
window into the performance characteristics and risk profiles
of those loans and, thus, the securities that they are a part
of.
Chairman Reed. Thank you. One final area of questioning,
and that is, many of these securitizations depend on REMIC, the
real estate conduit tax treatment. And there has been some
discussion that because of the structure of these vehicles, it
is very difficult to modify mortgages held in them because in
some cases it requires unanimous consent, which is hard to get
if you are at the lowest tranche. And I am wondering as we go
forward, should we consider conditioning this favorable
treatment on an agreement to modify loans that are financially
appropriate? I mean, you know, not subsidized loans, but if the
modification will have a value more than a foreclosure, then
that should be done? Professor McCoy, and then anyone else who
wants to comment.
Ms. McCoy. Yes, I think this is essential. There seem to be
three impediments right now. One is that perhaps the REMIC
rules themselves discourage workouts, although the IRS has been
trying to soften that.
The second problem is servicer compensation often is more
lucrative if you go to foreclosure. That is a separate problem
that needs to be fixed.
But, last, servicers do have some justifiable fear that
they will be sued by one set of tranche holders if they benefit
another in the process of doing a good-faith workout. And I
think we can use the REMIC rules to say the trust will not
receive Federal tax favored treatment unless these problems are
solved, so that when workouts are cost effective, that they go
forward, and the servicer has the incentive to do it and is not
worried about lawsuits.
If I could work in one other thing?
Chairman Reed. Yes, please.
Ms. McCoy. Which is with respect to assignee liability,
again and again we hear this claim that investors will not come
to the table if there is carefully crafted assignee liability
that does not expect investors to do the impossible. In fact,
in States that had carefully crafted standards, investors did
fund those loans.
What drove them away was the failure of securitization. So
it is ironic to talk about assignee liability driving them away
when securitization was able to do that just fine on its own.
Chairman Reed. Any other comments, particularly on this
REMIC question? Mr. Miller.
Mr. Miller. Yes, I do not think that the REMIC regulations
are themselves an impediment at all to loan modifications that
are otherwise contractually permitted. I think it is really
quite well established that under the REMIC regulations, if a
mortgage loan is in default or that default is reasonably
foreseeable, which covers, I think, a lot of territory, that a
loan modification can be pursued-- again, subject to any
contractual requirements in the securitization itself.
Having said that, I do not believe that it would be
advisable public policy to condition REMIC qualification or
continuing REMIC qualification on requirements to perform
modifications or to do that in a certain way. Again, I think
that would threaten the tax treatment that is provided through
the REMIC regulations in a way that would, again, chill or
inhibit participation and create distortions in the
marketplace.
To the extent that there are solutions or improvements to
the loan modification process, I think we should address those
frontally and head on. Part of Project Restart looking forward
prospectively is to support changes and develop standardized
provisions governing loan modifications and loss mitigation for
future securitization transactions to address uncertainties or
ambiguities in the way that that language is currently
constructed in those transactions. So I think the better way
would be to address it directly and not indirectly through the
Tax Code.
Chairman Reed. Just a follow-up question. The point that
Professor McCoy makes about the incentives for services
financially, in some cases--not all, obviously--that
foreclosure provides them more income than a modification,
which takes time, et cetera, and that seems to be a classic
case of the obvious benefit to one person but socially a cost
to all of us because as more and more mortgages go into default
and foreclosure, it is hurting the economy grievously.
So is that something that we can correct or should correct?
Mr. Miller. Well, I think the answer that I would give to
that is, regardless of incentives of any of the parties who are
involved in that circumstance, again, at least in the
securitization context, the duties and obligations and
responsibilities of servicers are laid out in the contracts,
they are and should be held to those standards by investors and
others. So regardless of any potential incentive that they may
have--and I personally think that some of the arguments about
servicing incentives to foreclose as opposed to, you know,
taking reasonable workout strategies, especially where that can
yield a greater net present value, I think some of those
statements are overstated or exaggerated. But, again, you know,
I think that really is something that is determined and
dictated by contract and the parties should be held to their
contractual obligations.
Chairman Reed. Just a final point. You are talking about
prospectively fixing this system. But currently we are looking
at estimates range from 4 million to 6 million foreclosures
next year, which is a huge drag on the economy and which may,
in fact, be sufficient drag to cutoff or at least to deflect
the growth and the prosperity we are all hoping for.
So I think we are confronting--I applaud your efforts to go
forward prospectively, but we have a huge problem with what we
have to deal with right now.
Anyone else who has a comment on this topic?
[No response.]
Chairman Reed. If not, let me thank you all again for
excellent testimony. I think Senator Gregg said it very well:
great insights together with very specific suggestions and done
in a very concise and understandable way. So thank you all for
your wonderful testimony.
The hearing is adjourned.
[Whereupon, at 4:11 p.m., the hearing was adjourned.]
[Prepared statements and additional material supplied for
the record follow:]
PREPARED STATEMENT OF CHAIRMAN JACK REED
I want to welcome everyone and thank our witnesses for appearing
today.
This hearing will examine a key activity within our financial
markets--the securitization of mortgages and other assets--and will
build on previous hearings this Subcommittee has held to address
various aspects of regulatory modernization, including hedge funds,
derivatives, corporate governance, SEC enforcement, and risk management
at large financial institutions.
Securitization is the packaging of individual loans or other debt
instruments into marketable securities to be purchased by investors. At
its core, this process helps free lenders to make more loans available
for families to purchase items like homes and cars and for small
businesses to thrive.
But we have learned from the financial crisis that securitization,
or how it is conducted, can also be extremely harmful to financial
markets and families without appropriate diligence and oversight.
Arguably, many of the basic requirements needed for effective
securitization were not met. Today's panel will discuss how in recent
years the securitization process created incentives throughout the
chain of participants to emphasize loan volume over loan quality,
contributing to the build-up and collapse of the subprime mortgage
market and the broader economy.
Today we find ourselves in the opposite position from a few years
back, with hardly any issuances in key markets that could help return
lending to responsible levels. So this afternoon's hearing is about how
to strengthen the securitization markets and enact any needed changes
to ensure that securitization can be used in ways that expand credit
without harming consumers and the capital markets.
I have asked today's witnesses to address a number of key issues,
including the role securitization played in the financial crisis, the
current conditions of these markets, and what changes may be needed to
Federal oversight of the securitization process.
Unfortunately, a number of the banks who issue these securities
could not find anyone in their workforce who was willing to testify
today.
I welcome you all and look forward to your testimony.
______
PREPARED STATEMENT OF PATRICIA A. McCOY
George J. and Helen M. England Professor of Law, and Director,
Insurance Law Center, University of Connecticut School of Law
October 7, 2009
During the housing bubble, private-label securitization financed
the majority of subprime and nontraditional mortgages. \1\ This system
proceeded on the assumption that housing prices would keep going up.
When housing prices fell and people could not refinance out of
unaffordable loans, investors lost confidence in private-label mortgage
securitization and the system collapsed in August 2007.
---------------------------------------------------------------------------
\1\ I use the term ``nonprime'' to refer to subprime loans plus
other nontraditional mortgages. Subprime mortgages carry higher
interest rates and fees and are designed for borrowers with impaired
credit. Nontraditional mortgages encompass a variety of risky mortgage
products, including option payment ARMs, interest-only mortgages, and
reduced documentation loans. Originally, these nontraditional products
were offered primarily in the ``Alt-A'' market to people with near-
prime credit scores but intermittent or undocumented income sources.
Eventually, interest-only ARMs and reduced documentation loans
penetrated the subprime market as well.
---------------------------------------------------------------------------
This statement begins with a thumbnail sketch of securitization.
Then I describe the role played by securitization in the financial
crisis. Following that, I analyze the inherent flaws in private-label
mortgage securitization. The statement goes on to describe current
conditions in that market. I close by describing needed reforms.
I. An Introduction to Securitization
Back in the 1970s, banks had to hold home mortgages in portfolio
until those loans were paid off. This destabilized banks that made
mortgages because they got their financing from demand deposits, but
invested those deposits in illiquid mortgages. This ``term mismatch''
between assets and liabilities was a direct cause of the 1980s savings
and loan crisis.
Starting in the late 1970s, securitization burst on the scene and
eliminated the need for lenders to hold their mortgages in portfolio.
The idea behind securitization is ingenious: bundle a lender's loans,
sell them to a bankruptcy-remote trust, repackage the monthly loan
payments into bonds rated by rating agencies, back the bonds with the
underlying mortgages as collateral, and sell those bonds to investors.
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Investment banks ``structured'' these securitization deals by
dividing the bonds into ``tranches'' (French for ``slice''). The best
tranche, with the lowest expected default rate, carried an AAA rating,
was paid off first, and offered the lowest rate of return. The lower
tranches were rated AA, A, etc., on down to the junior-most tranche,
known as the equity tranche. The equity tranche was paid off last and
was the first to absorb any losses from the loans.
Securitization was prized for accomplishing four things. First,
lenders were able to get their mortgages off their books. Second,
securitization appeared to manage the risks of mortgages by slicing and
dicing those risks and spreading them among millions of investors with
assorted tolerances for risk. Third, securitization opened up huge new
pools of capital to finance home mortgages. Finally, securitization
freed lenders from relying principally on insured deposits in order to
make loans. Instead, in a continuous cycle, lenders could make loans,
sell those loans through securitization, and then plow the proceeds
into a new batch of loans, which in turn would be securitized. This
paved the way for a new breed of nonbank subprime lenders, who had
little in the way of capital reserves, were free from Federal banking
regulation, and were inured to the reputational constraints of banks
and thrifts.
At first, securitization was limited to prime loans, which were
mostly securitized through the two Government-sponsored entities (GSEs)
Fannie Mae and Freddie Mac. Once the market gained confidence about its
ability to price subprime mortgages, securitization expanded to the
subprime market in the early 1990s. Although the GSEs made limited
forays into the subprime market and later expanded those forays around
2005, most subprime securitizations did not take place through the
GSEs, but rather through the ``private-label'' securitization market.
The private-label market lacked the same degree of public
accountability that was expected of Fannie Mae and Freddie Mac as GSEs.
By 2006, two-thirds or more of subprime mortgages were being
securitized through the private-label market.
II. The Role of Securitization in the Financial Crisis
A. How Private-Label Securitization Increased the Risk of Mortgage
Lending
Before securitization, lenders usually did it all: they solicited
loan applicants, underwrote and funded the loans, serviced the loans,
and held the loans in portfolio. Lenders earned profits on loans from
interest payments as well as from upfront fees. If the loans went into
default, the lenders bore the losses. Default was such a serious
financial event that lenders took care when underwriting loans.
All that changed with private-label securitization. Securitization
allowed lenders to offload most of the default risk associated with
nonprime loans. Under the ``originate-to-distribute'' model, lenders
could make loans intending to sell them to investors, knowing that
investors would bear the financial brunt if the loans went belly-up.
Similarly, securitization altered the compensation structure of
nonprime lenders. Lenders made their money on upfront fees collected
from borrowers and the cash proceeds from securitization offerings, not
on the interest payments on loans.
Lenders liked the security of being paid in advance, instead of
having to wait for uncertain monthly payments over the life of loans.
And, because they could pass the lion's share of the default risk onto
faceless investors, lenders had less reason to care about how well
their loans performed. In my examinations of internal records of major
nonprime lenders, including Federal thrift institutions and national
banks, too often I found two sets of underwriting standards: high
standards for the loans they kept on their books and lax standards for
the loans that they securitized.
At their peak, investment grade, \2\ nonprime residential mortgage-
backed securities (RMBS) were considered excellent investments because
they supposedly posed minimal default risk while offering high returns.
Investors clamored for these bonds, creating demand for ever-riskier
loans.
---------------------------------------------------------------------------
\2\ The top four ratings issued by a rating agency are
``investment grade'' ratings. For Standard & Poor's, these are ratings
of AAA, AA, A, and BBB; for Moody's, Aaa, Aa, A, and Baa. Any rating
below investment grade is considered junk bond status.
---------------------------------------------------------------------------
Lenders were not the only players in the chain between borrowers
and investors. Investment banks played significant roles as
underwriters of nonprime securitizations. Lehman Brothers, Bear
Stearns, Merrill Lynch, JPMorgan, Morgan Stanley, Citigroup, and
Goldman Sachs underwrote numerous private-label nonprime
securitizations. From 2000 through 2002, when IPO offerings dried up
during the 3-year bear market, RMBS and CDO deals stepped into the
breach and became one of the hottest profit centers for investment
banks.
Investment banks profited from nonprime underwriting by collecting
a percentage of the sales proceeds, either in the form of discounts,
concessions, or commissions. Once an offering was fully distributed,
the underwriter collected its fee in full. This compensation system for
the underwriters of subprime offerings caused Donna Tanoue, the former
Chairman of the Federal Deposit Insurance Corporation, to warn: ``[T]he
underwriter's motivation appears to be to receive the highest price . .
. on behalf of the issuer--not to help curb predatory loans.''
Tanoue's warning proved prophetic. In February 2008, Fitch Ratings
projected that fully 48 percent of the subprime loans securitized by
Wall Street in 2006 would go into default. Despite that dismal
performance, 2006 produced record net earnings for Goldman Sachs,
Morgan Stanley, Merrill Lynch, Lehman Brothers, and Bear Stearns. That
year, manager pay reflected the bottom-line importance that investment
banks placed on private-label RMBS, with managing directors in the
mortgage divisions of investment banks earning more on average in 2006
than their counterparts in other divisions.
B. How Securitization Fueled Contagion
Ultimately, private-label mortgage securitization turned out to be
an edifice built on a rotting foundation. Once that foundation gave
way, rising nonprime delinquencies mushroomed into international
contagion for a number of reasons. For example, the same loan often
served as collateral for multiple bonds, including an RMBS, a CDO, and
a CDO of CDOs. If the loan went into default, it would jeopardize
repayment for all three bonds. In addition, if defaults led to
downgrades on those bonds, those assets were highly correlated. If
rating agencies downgraded one issue, other issues came into question
as well.
Collateral is another reason why nonprime loans infected other
markets. Many large institutional investors bought nonprime bonds that
they later pledged as security for other types of loans. Banks, for
instance, pledged their nonprime bonds as security for short-term loans
from other banks on the market for interbank credit. Major corporations
borrowed money from other corporations on the short-term commercial
paper market by issuing paper backed by nonprime bonds. As the value of
nonprime bonds fell, lenders began calling loans and ultimately the
interbank lending and asset-backed commercial paper markets slowed to a
crawl.
Banks also reinfected themselves with subprime risks by buying
private-label RMBS and CDOs and effectively taking those risks back on
their books. When they sustained major losses on those bonds, they
reined in their lending, adding fuel to the recession.
General investor panic is the final reason for contagion. Even in
transactions involving no nonprime collateral, concerns about the
nonprime crisis had a ripple effect, making it hard for companies and
cities across-the-board to secure financing. Banks did not want to lend
to other banks out of fear that undisclosed nonprime losses might be
lurking on their books. Investors did not want to buy other types of
securitized bonds, such as those backed by student loans or car loans,
because they lost faith in ratings and could not assess the quality of
the underlying collateral. Stocks in commercial banks, insurance
companies, and Wall Street firms took a beating because investors did
not know where nonprime assets were hidden and feared more nonprime
write-downs. Because they did not know exactly who was tainted by
nonprime, investors stopped trusting practically everyone.
III. Inherent Flaws in Private-Label Mortgage Securitization
A. The Lemons Problem
In hindsight, private-label mortgage securitization turned out to
resemble the used car business in one respect. Both businesses have
motivations to sell ``lemons.'' In other words, they have structural
incentives to sell products carrying hidden defects and a heightened
risk of failure.
There are two main reasons for this lemons problem. First,
securitization resulted in a misalignment of compensation and risk.
Each company in the securitization process was able to collect upfront
fees, while shifting default risk to downstream purchasers. Although
investors tried to protect themselves through recourse clauses and
structures making lenders retain the equity tranches, those contractual
safeguards often broke down. Lenders were able to hedge their equity
tranches or shed them by resecuritizing them as CDOs. Similarly, too
many originators lacked the capital to honor their recourse obligations
in full.
Second, securitization fueled a relentless demand for volume and
volume-based commissions. In the process, the quest for volume pushed
lending standards steadily downward in order to maintain market share.
This became a challenge in 2003, when interest rates began rising
again, ending the refinancing boom. Securitizers needed another source
of mortgages in order to increase the rate of securitization and the
fees it generated. The ``solution'' was to expand the market through
nontraditional mortgages, especially interest-only loans and option
payment ARMs offering negative amortization. Lenders also relaxed their
underwriting standards on traditional products to qualify more
borrowers. This expansion of credit swept a larger portion of the
population into the potential homeowner pool, driving up housing demand
and prices, and consumer indebtedness. Many big investment banks,
including Lehman Brothers and Bear Stearns, went so far as to buy
subprime lenders in order to have an assured pipeline of mortgages to
securitize.
In short, the incentive structure of securitization caused the
lemons problem to grow worse over time. Not only did private-label
securitization sell lemons, those lemons grew more rotten as the
housing bubble grew. In the process, securitization actors played the
ends against the middle, injuring borrowers and investors alike.
B. Harm to Borrowers
Private-label securitization hurt numerous borrowers. First,
investor appetite for high-yield RMBS caused originators to peddle
risky mortgages, to the exclusion of safer loans. Second, compensation
methods such as yield spread premiums saddled many borrowers with
costlier mortgages than they qualified for. Third, borrowers whose
loans were securitized lost important legal rights without their
consent.
On the first point: As mentioned above, in order to maintain volume
while satisfying investor demand for high-yield bonds, investment banks
and lenders had to continually tap new groups of borrowers with lower
credit scores and less disposable income. For many of these cash-
strapped borrowers, low monthly payments were a primary consideration.
In order to offer the lure of lower initial payments, lenders concocted
bafflingly complex loans combining a host of risky features, including
adjustable-rate terms, teaser rates, high margins, stiff prepayment
penalties, and no amortization or even negative amortization. Evidence
is now coming to light that investment banks or large investors in many
cases dictated those underwriting guidelines to originators.
The front-end payments of these hazardous mortgages were attractive
to unsuspecting borrowers and usually lower than the payments on a
plain vanilla fixed-rate mortgage. But the back-end risks of those
mortgages were daunting, yet difficult or impossible for borrowers to
discern. Worse yet, to qualify individual borrowers, lenders often
threw full income verification out the window.
There was a second way in which investor demand for higher yield
hurt many borrowers. Because investors paid more for higher yields,
lenders offered mortgage brokers higher compensation in the form of
yield spread premiums to convince borrowers who probably qualified for
cheaper loans to unwittingly pay higher interest rates. The Wall Street
Journal estimated that by year-end 2006, 61 percent of subprime
mortgages went to borrowers with high enough credit scores to qualify
for cheaper prime loans. \3\ Yield spread premiums artificially
inflated the interest rates that borrowers had to pay, substantially
increasing the likelihood that nonprime loans would default and go into
foreclosure. Economists have estimated the size of this risk. For every
1 percent increase in the initial interest rate of a home mortgage, the
chance that a household will lose its home rises by 16 percent a year.
---------------------------------------------------------------------------
\3\ Rick Brooks and Ruth Simon, ``Subprime Debacle Traps Even Very
Creditworthy'', Wall St. J., Dec. 3, 2007, at A1.
---------------------------------------------------------------------------
Finally, under the Uniform Commercial Code in many States,
borrowers whose loans are securitized lose valuable legal rights
without their consent or financial compensation. This doctrine, known
as the ``holder-in-due-course rule,'' prohibits borrowers whose loans
are securitized from raising common types of fraud or other misconduct
in the making of their loans against all subsequent purchasers of their
loan notes. In many case, this shields investment banks, rating
agencies, and investors from borrower suits for fraud. Although
borrowers can still raise fraud as a claim or defense against their
mortgage brokers and lenders, many of those entities are bankrupt today
and thus judgment-proof. More importantly, once a loan is securitized,
any suit for foreclosure will be brought by the investor or securitized
trust, not the mortgage broker or lender. In those cases, the holder-
in-due course rule prevents borrowers who were defrauded from even
raising the fraud as a defense to foreclosure.
C. Harm to Investors
The lack of transparency in securitization also hurt investors. The
securities disclosures for private-label RMBS lacked crucial
information to investors. In addition, product complexity made it
difficult or impossible for investors to grasp the risks associated
with many offerings. Finally, both problems caused investors to place
undue reliance on credit ratings, which proved to be badly inflated.
1. Inadequate Securities Disclosures--For most of the housing
bubble, the Securities and Exchange Commission (SEC) had no rule
requiring disclosures specifically tailored to RMBS or CDOs. The SEC
adopted Regulation AB in an attempt to redress that gap, but the rule
did not go into effect until January 1, 2006, too late to cover earlier
private-label offerings.
Once the rule went into effect, it was riddled with holes. First,
Reg AB only applies to public offerings of asset-backed securities. An
investment bank could simply bypass Reg AB by structuring the offering
as a private offering limited to big institutional investors. In
private offerings, SEC disclosures are lighter or left to private
negotiation, based on the idea that institutional investors have clout
to demand the information they need. Wall Street took full advantage of
this loophole, meaning that CDOs were almost always sold through
private offerings with seriously deficient disclosures.
Even when Reg AB did apply--i.e., in public offerings of asset-
backed securities--the disclosures were too skimpy to be of use. The
SEC modeled many of Reg AB's disclosures on the reporting requirements
for corporate issuers. Corporations usually have track records to speak
of, so securities disclosures for those issuers focus on recent past
performance. But past performance was irrelevant for most offerings of
RMBS and CDOs, which involved relatively new mortgages. In essence, Reg
AB puts the wrong information under the microscope.
Instead, investors in nonprime bonds needed standardized
information on the risk characteristics of the individual loans in the
loan pool. But Reg AB does not require that level of detail. While the
rule encouraged investment banks to make tapes with loan level data
available to investors online, it did not force them to do so. Instead,
Reg AB simply mandates a summary of the aggregate characteristics of
the loan pool. That made it difficult to discern whether the riskiest
loans were going to the strongest borrowers or to the worst borrowers
in the loan pool.
Similarly, too many prospectuses and offering memoranda for
private-label offerings stated that the lenders reserved the right to
make exceptions to their underwriting standards in individual cases. In
2006 and 2007, there were offerings in which the exceptions--in other
words, loans that flunked the lender's underwriting standards--
outweighed the number of loans that conformed to the lender's stated
standards. The exact (and often high) percentage of exceptions was not
disclosed to investors.
Nor does Reg AB make investment banks disclose the due diligence
reports they commissioned from outside firms, even when those reports
contained evidence of deteriorating lending standards. Too often,
investment banks withheld those reports from investors and ratings
agencies.
Reg AB is also deficient regarding the performance of individual
loans. While Reg AB requires some reporting on loan performance, it is
only for the first year following the offering, not for the life of the
loans.
All told, there was a dearth of useful publicly available
information on the loan pools underlying private-label RMBS and CDOs.
The SEC disclosure scheme for nonprime RMBS and CDOs was so misbegotten
and riddled with exceptions that those securities operated in a fact-
free zone. Investors and analysts who wanted to do serious due
diligence could not get the facts they needed to figure out the true
risk presented by the loans. Without those facts, investors often
overpaid for those securities. Furthermore, the dearth of key public
information also impeded the development of a healthy resale market in
those bonds, which became a big problem later on when banks tried to
unload toxic subprime assets off their books.
2. Complex Products--Many private-label RMBS and CDOs were so
complex that due diligence was too costly or impossible for investors.
CDOs are a good example. Typically, a CDO consisted of junior tranches
of RMBS from different offerings, sometimes paired with other types of
asset-backed securities involving receivables from things like credit
cards or auto loans. At best, the investor received data on the quality
of the underlying bonds. But it was impossible for the investor to x-
ray the offering in order to analyze the underlying home mortgages,
credit card borrowers, or auto loans themselves. That was even more
impossible when the CDO was a ``synthetic CDO'' made up of credit
default swaps on RMBS and asset-backed securities.
Even in regular RMBS, complexity was a big problem. One issue was
the sheer number of tranches. Another was the fact that many private-
label RMBS offerings featured complex credit enhancement rules about
who would receive cash flows from the mortgages in what amounts,
depending on changes in the amount of subordination or
overcollateralization. This meant that investors could not just stop
with estimating expected losses from the mortgages. They also had to
analyze who would get what cash flows when, based on a changing
kaleidoscope of scenarios. \4\ In addition, too many offerings were
made on a ``to be announced'' or ``TBA'' basis, which meant that
investors could not scrutinize the underlying loans because the loans
had not yet been put in the loan pool. Finally, many securitization
deals involved custom features that undermined standardization.
---------------------------------------------------------------------------
\4\ Ingo Fender and Janet Mitchell, ``The Future of
Securitization: How To Align Incentives?'', Bis. Quarterly Review 27,
30, 32 (Sept. 2009).
---------------------------------------------------------------------------
Of course, this discussion begs the question whether investors
would have done adequate investigation in any case when the housing
bubble was at its height and euphoria prevailed. But back then, even
investors who wanted to do serious due diligence would have met
insuperable obstacles. More recently, lack of transparency and
complexity have blocked the formation of an active, liquid resale
market that would enable banks to remove impaired RMBS and CDOs from
their books.
3. Overreliance on Credit Ratings--Poor disclosures and overly
complex deals caused investors to over rely on credit ratings.
Meanwhile, the rating agencies had financial incentives to understate
the risks of nonprime RMBS and CDOs. The investment banks that
underwrote nonprime securitizations paid the rating agencies to provide
them with investment-grade ratings. The rating agencies touted the top-
rated nonprime bonds--ranging from AAA down to A--as hardly ever
defaulting.
Under banking and insurance laws, banks and insurance companies can
only invest in types of bonds permitted by law. Private-label RMBS and
CDOs carrying investment grade ratings are on the permissible list, so
long as those ratings are rendered by rating agencies designated
Nationally Recognized Statistical Rating Organizations (NRSROs) by the
SEC. These regulatory rules encouraged institutional investors in
search of higher yields to buy the top-rated nonprime RMBS and CDOs.
During the housing bubble, rating fees on private-label RMBS and
CDOs were the fastest-growing sector of the rating agency business.
Issuers paid the rating agencies handsome fees from these deals,
spurring the rating agencies to rate offerings for which there was
scant historical default data. Similarly, the rating agencies used
flawed models which assumed never-ending housing price appreciation and
were not updated with new default data. Nor did most investors realize
that an AAA rating for an RMBS offering was different than, and
inferior to, an AAA rating for a corporate bond. \5\
---------------------------------------------------------------------------
\5\ In large part, and in contrast with corporate bonds, this is
because downgrades of a tranched RMBS tend to make downgrades of other
RMBS tranches more likely. Fender and Mitchell, supra note 4, at 33.
---------------------------------------------------------------------------
D. Impediments to Loan Modifications
Deal provisions in private-label securitizations have also
paralyzed constructive workouts of many distressed home loans. Today,
securitized trusts, not lenders, hold the vast majority of those loans.
The complexity of the securitized deals often pits servicers against
investors and investors against each other. Too often, the servicers
opt for foreclosing on property, instead of arranging workouts that
would allow homeowners to stay in their homes. The irony of this
approach is that, in many cases, workouts in the form of loan
forbearance or loan modifications would result in a higher recovery.
There are several explanations for this seemingly irrational
behavior, including inadequate staffing levels and compensation clauses
that cause servicers to earn more money from foreclosures than
workouts. But the main reason why more workouts do not occur is that
many pooling and servicing agreements place constraints on servicers'
ability to negotiate loan workouts. Some limit the percent of the loan
pool that can be modified. Others have vague prohibitions allowing
modifications only to the extent they are in the best interests of the
investors. Even when those agreements give servicers latitude to modify
loans, servicers are reluctant to modify loans because they fear
lawsuits by warring trancheholders for breach of fiduciary duty.
This hold-up problem has stymied Federal regulators' attempts to
speed up loan modifications and halt the vicious cycle of falling home
prices. With no Federal legislation to force modifications, regulators
have only had limited success. Meanwhile, loan workouts are crawling at
a snail's pace, leading foreclosed homes to be dumped on the market in
record numbers and pushing home prices further down in the process.
IV. Current Conditions in the Private-Label Securitization Markets
Due to the problems just described, the markets for private-label
RMBS and CDOs are essentially dead. The securitization markets for auto
loans, credit cards, and student loans are open, but their volume has
dropped sharply due to general concerns about the soundness of the
securitization process.
For all intents and purposes, the Federal Government has become the
financier of first resort for residential mortgages. In 2008, agency
mortgage-backed securities--in other words, RMBS issued by Fannie Mae,
Freddie Mac, and Ginnie Mae (FHA loans)--accounted for over 96 percent
of the U.S. RMBS market. Private-label mortgage-backed securitization
accounted for less than 4 percent of the market that year.
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
This disparity widened in the first 6 months of 2009, when the
relative market shares of agency and private-label mortgage-backed
securitization were 99 percent and 1 percent. \6\ In second quarter
2009, moreover, 38.4 percent of private-label RMBS transactions were
re-REMICs of old loans that were repackaged into tranches of good and
bad loans. According to the Securities Industry and Financial Markets
Association (SIFMA), the ``private label market remains dormant due to
reduced lending, lack of investor demand, low liquidity,'' and rising
delinquencies and foreclosures. \7\
---------------------------------------------------------------------------
\6\ I use the term ``agency'' to refer to GNMA, Fannie Mae and
Freddie Mac mortgage-backed securities and collateralized mortgage
obligations. The term ``private-label'' includes RMBS and CMOs.
\7\ Securities Industry and Financial Markets Association,
Research Report 2009 Q2 (August 2009), at 2, 9.
---------------------------------------------------------------------------
As these numbers suggest, private investors are largely shunning
the private-label mortgage securitization market in favor of other
investments, including agency RMBS. In the meantime, the Federal
Reserve has become a major investor in agency RMBS, having begun
purchases in this market in December 2008. The Fed has pledged to
buying up to $1.25 trillion in agency RMBS before the end of this year,
in an effort to help lower home mortgage interest rates.
Other securitization markets associated by investors with mortgages
are also dormant. SIFMA reports that the private-label commercial MBS
primary market ``remains closed.'' \8\ Similarly, global issuance of
CDOs has essentially come to a halt.
---------------------------------------------------------------------------
\8\ Id. at 9.
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Outside of the mortgage sector, auto loan, credit card, and student
loan securitizations have fallen by over half since 2007. All three
sectors became paralyzed in mid-2008, prompting the Federal Reserve to
revive these markets with the Term Asset-Backed Securities Lending
Facility (TALF). Spreads soared in 2008 and have since fallen, although
have not completely recovered. This suggests that investor concerns
about the general integrity of the securitization process spilled over
to other sectors.
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Although TALF has helped to revive these markets, particularly in
the auto and credit card areas, delinquencies and charge-offs continue
to climb.
V. Needed Reforms
Private-label mortgage securitization will undoubtedly return in
one form or another. And just as certainly, investors will eventually
forget the lessons from this crisis. To avoid repeating the mistakes of
the past, it is essential to put private-label mortgage securitization
on sound footing going forward.
A. Proposals To Realign Incentives
Discussions about reforming private-label securitization often
revolve around proposals to realign the incentives of originators and
investment banks. The idea is to give them sufficient ``skin in the
game'' to care about soundly underwritten loans. Thus, the Obama
Administration has proposed \9\ requiring securitizers to retain at
least 5 percent of the credit risk on each asset in the asset-backed
securities that they issue. \10\ Securitizers would also be barred from
resecuritizing or hedging that retained risk. Section 213 of the
Mortgage Reform and Anti-Predatory Lending Act, H.R. 1728, passed by
the House of Representatives on May 7, 2009, contains a similar
proposal.
---------------------------------------------------------------------------
\9\ Financial Regulatory Reform Proposal, Title IX, 951,
www.treas.gov/initiatives/regulatoryreform/.
\10\ The implementing agencies would also have to adopt provisions
allocating the risk retention obligation between the securitizer and
the originator.
---------------------------------------------------------------------------
There are other incentive-based proposals to improve loan
underwriting. One involves increased capital: in other words, requiring
commercial and investment banks --especially too-big-to-fail banks--to
hold more capital, both against the tranches they retain and against
other aspects of securitization that could come back to haunt them,
such as recourse clauses and structured investment vehicles.
Another proposal is to realign originators' compensation with loan
performance. Accounting standards could be changed to eliminate
immediate recognition of gain on sale by originators at the time of
securitization. And there are two promising proposals to curb reckless
originations by independent mortgage brokers. One would prohibit pay
incentives such as yield spread premiums for steering customers to
costlier or riskier loans. H.R. 1728, 103. Another proposal would make
full payout of compensation to mortgage brokers contingent on good
performance of the loan.
A final idea along these lines is to require lenders and
securitizers to make stronger representations and warranties to
investors, accompanied by stiffer recourse provisions for loans that
violate those reps and warranties. The American Securitization Forum
has advanced this reform.
All of these proposals are good ideas. However, they are not
enough, together or alone, to ensure sound underwriting. Take the risk
retention requirement, for example. It is doubtful whether the ban on
hedging is even enforceable, since ``sometimes firms pool their risk
and set hedges against several positions at once.'' \11\ More
importantly, requiring risk retention does not solve the fact that
banks, once they got loans off of their books through securitization,
assumed that risk again by investing in toxic subprime RMBS and CDOs.
---------------------------------------------------------------------------
\11\ Fender and Mitchell, supra note 4, at 41.
---------------------------------------------------------------------------
As for capital requirements, more capital is essential for
depository institutions and investment banks. But capital is no
panacea. Banks have proven adept at evading minimum capital
requirements. Furthermore, the credit crisis raised serious concerns
about the newly adopted Basel II capital standards, which were designed
to lower capital and allow large internationally active banks--i.e.,
too-big-to-fail banks--to set their own minimum capital.
Stronger reps and warranties, backed by stiffer recourse, are
likewise advisable. But the crisis has shown that recourse provisions
are only as good as a lender's solvency. Since the credit crisis began,
most nonbank subprime lenders have gone out of business. In addition,
126 banks and thrifts have failed since 2007. Some institutions failed
precisely due to their inability to meet investor demands for recourse.
\12\
---------------------------------------------------------------------------
\12\ See, e.g., Office of Inspector General, Department of the
Treasury, ``Safety and Soundness: Material Loss Review of NetBank,
FSB'' (OIG-08-032, April 23, 2008), www.ustreas.gov/inspector-general/
audit-reports/2008/OIG08032.pdf.
---------------------------------------------------------------------------
Even when recourse can be had, negotiations can be long and drawn-
out. Moreover, if a recourse provision is not ironclad, a solvent
lender may be able to escape it. For example, any provisions that would
condition recourse on the lender's knowledge that the reps and
warranties were violated--creating a Sergeant Schultz ``I know
nothing'' defense--usually would be meaningless if the misconduct in
question was committed by an independent mortgage broker. That would
include situations where the lender failed to adequately supervise the
broker, which often was the case.
For all of these reasons, having ``skin in the game'' is not enough
to ensure sound loan underwriting. As discussed below, more is needed
in the form of minimum underwriting standards.
B. Improved Due Diligence by Investors
Meanwhile, investors need the ability to do better due diligence.
Three major reforms are needed to provide investors with the
information that they need to make sound investment decisions about
private-label mortgage-related bonds. First is improved transparency,
second is product simplification and standardization, and third is
rating agency reform.
Transparency--The SEC should require securitizers to provide
investors with all of the loan-level data they need to assess the risks
involved. See Obama Administration Proposal, Title IX, 952. In
addition, the SEC should require securitizers and servicers to provide
loan-level information on a monthly basis on the performance of each
loan and the incidence of loan modifications and recourse. These
disclosures should be made in public offerings and private placements
alike. In addition, TBA offerings should be prohibited because it is
impossible for investors to do due diligence on those loan pools.
Product Simplification and Standardization--The Government should
encourage simpler, standardized securitization products, whether
through the REMIC tax rules or rules governing permissible investments
by insured banks and thrifts. Similarly, the Government should explore
ways to build a liquid secondary trading market in private-label RMBS
and other bonds.
Rating Agency Reform--The most critical rating agency reform is
banning the ``issuer pays'' system, in which issuers pay for ratings.
That would help ensure that rating agencies serve the interests of
investors, not issuers. In addition, it is necessary to require the
rating agencies to create a new, different ratings scale for mortgage
structured finance to distinguish it from the ratings for corporate
bonds. Finally, NRSRO designations need to be abolished.
The Obama Administration's proposal takes a different approach. The
proposal would subject NRSROs to enhanced SEC oversight, including
expanded public disclosures. In addition, the Administration would
require rating agencies to have systems to ``manage, and disclose''
their conflicts of interest. Title IX, subtitle C.
While better investor due diligence is necessary to improve
private-label mortgage securitization, it is not enough. At the height
of every business cycle, memories grow dim and euphoria takes hold.
During bubbles, when default rates are low, investors are apt to cast
aside basic due diligence precautions to grab the chance of a high-
yield investment. This temptation is particularly great for
institutional money managers, who have cash they need to put to work
and face pressure to report the same high returns as their competitors.
For all of these reasons, minimum Federal underwriting standards are a
needed supplement to investor due diligence.
C. Protecting Borrowers and the Financial System
We cannot assume that investors will monitor adequately or that
standardization will be achieved. Furthermore, none of the measures
outlined above addresses the obstacles to loan modifications. Two
additional measures are needed to protect borrowers and the larger
economic system from reckless loans and unnecessary foreclosures.
1. Uniform Minimum Underwriting Standards Enforceable by
Borrowers--The downward spiral in underwriting standards drove home the
need for uniform consumer protection standards that apply to all
financial services providers. In fact, a new study by the Center for
Community Capital at the University of North Carolina (Chapel Hill)
finds that States that mandated strong loan underwriting standards had
lower foreclosure rates than States without those laws. \13\
---------------------------------------------------------------------------
\13\ Center for Community Capital, State Anti-Predatory Lending
Laws (October 5, 2009), http://www.ccc.unc.edu/news/
AG_study_release_5[2].10.2009.pdf.
---------------------------------------------------------------------------
The Federal Reserve's 2008 rule for higher-cost loans accomplished
part of this goal, \14\ but all loans need protection, not just
subprime loans. The Obama Administration proposal, H.R. 1728, and H.R.
3126 would solve this problem by creating one set of uniform Federal
laws that apply to all financial services providers across the country,
regardless of entity, charter, or geographic location. To prevent a
race to the bottom in which regulators compete to relax lending
standards, the Administration proposal and H.R. 3126 would consolidate
the authority to administer those laws in a new Consumer Financial
Protection Agency. Under both, the standards would constitute a floor,
in which weaker State laws are federally preempted. States would remain
free to enact stricter consumer protections so long as those
protections were consistent with Federal law.
---------------------------------------------------------------------------
\14\ Federal Reserve System, Truth in Lending: Final rule;
official staff commentary, 73 FED. REG. 44522, 44536 (July 30, 2008).
The Board intended to cover the subprime market, but not the prime
market. See, id. at 44536-37.
---------------------------------------------------------------------------
These Federal standards do three things. First, the standards would
ensure proper loan underwriting based on the consumer's ability to
repay. Second, the standards would prohibit unfair or deceptive
practices in consumer credit products and transactions. Finally, the
standards would promote transparency through improved consumer
disclosures. Bottom-line, the proposed standards would help make it
possible for consumers to engage in meaningful comparison shopping,
with no hidden surprises.
In the event these standards are violated, injured borrowers need
an affirmative claim for relief as well as a defense to foreclosure.
Both the claim and the defense should be available against loan
originators. Limiting relief to loan originators does not help
borrowers with securitized loans, however, if their loans later go into
foreclosure or their originators become judgment-proof. When a
securitized loan is foreclosed on, for example, the lender is not the
plaintiff; rather, foreclosure is instituted by the servicer, the owner
of the loan, or its designee (generally the Mortgage Electronic
Registration Systems or MERS). Consequently, fairness requires allowing
injured borrowers to raise violations as a defense to foreclosure
against those entities. Similarly, giving borrowers an affirmative
claim against assignees for violations of Federal lending standards by
originators will spur investors and investment banks to insist on
proper underwriting of loans and afford injured borrowers relief when
their originators are judgment-proof or a securitized trust sues for
foreclosure. The Administration's proposal and H.R. 1728, 204, both
contain assignee liability provisions designed to accomplish these
objectives.
Some fear that a borrower right of action against securitized
trusts and investment banks would reduce access to credit. A 2008 study
by Dr. Raphael Bostic et al. examined that question by looking at the
effect of assignee liability provisions in nine State antipredatory
lending laws on the availability of subprime credit. The study found
``no definitive effect of assignee liability on the likelihood of
subprime originations, even when the [assignee] liability provisions
are in their strongest form.'' Subprime originations rose in six of the
nine States studied that had assignee liability, relative to the
control State. Results were mixed in the other three States, depending
on how subprime lending was defined. No State reported a consistent
drop in subprime originations. \15\
---------------------------------------------------------------------------
\15\ Raphael Bostic, Kathleen C. Engel, Patricia A. McCoy, Anthony
Pennington-Cross, and Susan Wachter, ``The Impact of Predatory Lending
Laws: Policy Implications and Insights'', In Borrowing To Live:
Consumer and Mortgage Credit Revisited 138 (Nicolas P. Retsinas and
Eric S. Belsky eds., Joint Center for Housing Studies of Harvard
University and Brookings Institution Press, 2008), working paper
version at http://www.jchs.harvard.edu/publications/finance/
understanding_consumer_credit/papers/ucc08-9_bostic_et_al.pdf.
---------------------------------------------------------------------------
In short, assignee liability is not likely to impede access to
credit. To the contrary, borrower relief will provide needed incentives
for originators, Wall Street, and investors to only securitize loans
that borrowers can repay. Providing that relief would go a long way
toward avoiding the biggest threat to access to credit, which is a
repeat collapse of private-label securitization.
2. Remove Artificial Barriers to Cost-Effective Loan
Modifications--Right now, too many distressed loans are needlessly
going to foreclosure despite the availability of cost-effective loan
modifications. Not only do these foreclosures oust homeowners from
their homes, they needlessly depress home values for everyone else. It
is time to cut this Gordian knot.
Most securitized loan pools are created as ``Real Estate Mortgage
Investment Conduits,'' or REMICs, under the Federal tax code. Any
securitization vehicle that qualifies for REMIC treatment is exempt
from Federal income taxes. Congress or the Internal Revenue Service
should amend the REMIC rules to disqualify future mortgage pools from
favored REMIC tax treatment unless pooling and servicing agreements and
related deal documents are drafted to give servicers ironclad
incentives to participate in large-scale loan modifications when
specific triggers are hit. \16\
---------------------------------------------------------------------------
\16\ See, Michael S. Barr and James A. Feldman, Issue Brief:
Overcoming Legal Barriers to the Bulk Sale of At-Risk Mortgages (Center
for American Progress April 2008).
---------------------------------------------------------------------------
______
PREPARED STATEMENT OF GEORGE P. MILLER
Executive Director, American Securitization Forum
October 7, 2009
On behalf of the American Securitization Forum, I appreciate the
opportunity to testify before this Subcommittee as it explores problems
and solutions associated with the securitization process.
The American Securitization Forum (ASF) is a broad-based
professional forum through which participants in the U.S.
securitization market advocate their common interests on important
legal, regulatory and market practice issues. ASF members include over
350 firms, including investors, mortgage and consumer credit lenders
and securitization issuers, financial intermediaries, legal and
accounting firms, and other professional organizations involved in the
securitization markets. The ASF also provides information, education,
and training on a range of securitization market issues and topics
through industry conferences, seminars and similar initiatives. ASF is
an affiliate of the Securities Industry and Financial Markets
Association. \1\
---------------------------------------------------------------------------
\1\ For more information on ASF, please visit our Web site: http:/
/www.americansecuritization.com. For more information on the Securities
Industry and Financial Markets Association, please see: http://
www.sifma.org.
---------------------------------------------------------------------------
My testimony today will address the following topics:
1. The role and importance of securitization to the financial system
and U.S. economy;
2. Current conditions in the securitization market;
3. Limitations and deficiencies in securitization revealed by the
recent financial market crisis; and
4. Views on certain securitization policy and market reform
initiatives now underway or under consideration.
I. The Role and Importance of Securitization to the Financial System
and U.S. Economy
Securitization--generally speaking, the process of pooling and
financing consumer and business assets in the capital markets by
issuing securities, the payment on which depends primarily on the
performance of those underlying assets--plays an essential role in the
financial system and the broader U.S. economy. Over the past 25 years,
securitization has grown from a relatively small and unknown segment of
the financial markets to a mainstream source of credit and financing
for individuals and businesses alike.
In recent years, the role that securitization has assumed in
providing both consumers and businesses with credit is striking:
currently, there is over $12 trillion of outstanding securitized
assets, \2\ including mortgage-backed securities (MBS), asset-backed
securities (ABS), and asset-backed commercial paper. This represents a
market nearly double the size of all outstanding marketable U.S.
Treasury securities--bonds, bills, notes, and TIPS combined. \3\
Between 1990 and 2006, issuance of mortgage-backed securities grew at
an annually compounded rate of 13 percent, from $259 billion to $2
trillion a year. \4\ In the same time period, issuance of asset-backed
securities secured by auto loans, credit cards, home equity loans,
equipment loans, student loans and other assets, grew from $43 billion
to $753 billion. \5\ In 2006, just before the downturn, nearly $2.9
trillion in mortgage- and asset-backed securities were issued. As these
data demonstrate, securitization is clearly an important sector of
today's financial markets.
---------------------------------------------------------------------------
\2\ SIFMA, ``Asset-Backed Securities Outstanding'', http://
www.sifma.org/uploadedFiles/Research/Statistics/
SIFMA_USABSOutstanding.pdf .
\3\ U.S. Department of the Treasury, ``Monthly Statement of the
Public Debt of the United States: August 31, 2009'', (August 2009).
http://www.treasurydirect.gov/govt/reports/pd/mspd/2009/opds082009.pdf.
\4\ National Economic Research Associates, Inc. (NERA), ``Study of
the Impact of Securitization on Consumers, Investors, Financial
Institutions and the Capital Markets'', p. 16 (June 2009). http://
www.americansecuritization.com/uploadedFiles/ASF_NERA_Report.pdf .
\5\ SIFMA, ``U.S. ABS Issuance'', http://www.sifma.org/
uploadedFiles/Research/Statistics/SIFMA_USABSIssuance.pdf.
---------------------------------------------------------------------------
The importance of securitization becomes more evident by observing
the significant proportion of consumer credit it has financed in the
U.S. It is estimated that securitization has funded between 30 and 75
percent of lending in various markets, including an estimated 59
percent of outstanding home mortgages. \6\ Securitization plays a
critical role in nonmortgage consumer credit as well. Historically,
most banks have securitized 50-60 percent of their credit card assets.
\7\ Meanwhile, in the auto industry, a substantial portion of
automobile sales are financed through auto ABS. \8\ Overall, recent
data collected by the Federal Reserve Board show that securitization
has provided over 25 percent of outstanding U.S. consumer credit. \9\
In the first half of 2009 alone, securitization financed over $9.5
billion in student loans. \10\ Securitization also provides an
important source of commercial mortgage loan financing throughout the
U.S., through the issuance of commercial mortgage-backed securities.
---------------------------------------------------------------------------
\6\ Citigroup, ``Does the World Need Securitization?'' pp. 10-11
(Dec. 2008).http://www.americansecuritization.com/uploadedFiles/
Citi121208_restart_securitization.pdf.
\7\ Ibid., p. 10.
\8\ Ibid., p. 10.
\9\ Federal Reserve Board of Governors, ``G19: Consumer Credit'',
(September 2009). http://www.federalreserve.gov/releases/g19/current/
g19.htm.
\10\ SIFMA, ``U.S. ABS Issuance'', http://www.sifma.org/
uploadedFiles/Research/Statistics/SIFMA_USABSIssuance.pdf.
---------------------------------------------------------------------------
Over the years, securitization has grown in large measure because
of the benefits and value it delivers to transaction participants and
to the financial system. Among these benefits and value are the
following:
1. Efficiency and Cost of Financing. By linking financing terms to
the performance of a discrete asset or pool of assets, rather
than to the future profitability or claims-paying potential of
an operating company, securitization often provides a cheaper
and more efficient form of financing than other types of equity
or debt financing.
2. Incremental Credit Creation. By enabling capital to be recycled
via securitization, lenders can obtain additional funding from
the capital markets that can be used to support incremental
credit creation. In contrast, loans that are made and held in a
financial institution's portfolio occupy that capital until the
loans are repaid.
3. Credit Cost Reduction. The economic efficiencies and increased
liquidity available from securitization can serve to lower the
cost of credit to consumers. Several academic studies have
demonstrated this result. A recent study by National Economic
Research Associates, Inc., concluded that securitization lowers
the cost of consumer credit, reducing yield spreads across a
range of products including residential mortgages, credit card
receivables and automobile loans. \11\
\11\ National Economic Research Associates, Inc. (NERA), ``Study
of the Impact of Securitization on Consumers, Investors, Financial
Institutions and the Capital Markets'', (June 2009), p. 16. http://
www.americansecuritization.com/uploadedFiles/ASF_NERA_Report.pdf.
---------------------------------------------------------------------------
4. Liquidity Creation. Securitization often offers issuers an
alternative and cheaper form of financing than is available
from traditional bank lending, or debt or equity financing. As
a result, securitization serves as an alternative and
complementary form of liquidity creation within the capital
markets and primary lending markets.
5. Risk Transfer. Securitization allows entities that originate
credit risk to transfer that risk to other parties throughout
the financial markets, thereby allocating that risk to parties
willing to assume it.
6. Customized Financing and Investment Products. Securitization
technology allows for precise and customized creation of
financing and investment products tailored to the specific
needs of issuers and investors. For example, issuers can tailor
securitization structures to meet their capital needs and
preferences and diversify their sources of financing and
liquidity. Investors can tailor securitized products to meet
their specific credit, duration, diversification and other
investment objectives. \12\
---------------------------------------------------------------------------
\12\ The vast majority of investors in the securitization market
are institutional investors, including banks, insurance companies,
mutual funds, money market funds, pension funds, hedge funds and other
large pools of capital. Although these direct market participants are
institutions, many of them--pension funds, mutual funds and insurance
companies, in particular--invest on behalf of individuals, in addition
to other account holders.
Recognizing these and other benefits, policymakers globally have
taken steps to help encourage and facilitate the recovery of
securitization activity. The G-7 finance ministers, representing the
world's largest economies, declared that ``the current situation calls
for urgent and exceptional action . . . to restart the secondary
markets for mortgages and other securitized assets.'' \13\ The
Department of the Treasury stated in March that ``while the intricacies
of secondary markets and securitization . . . may be complex, these
loans account for almost half of the credit going to Main Street,''
\14\ underscoring the critical nature of securitization in today's
economy. The Chairman of the Federal Reserve Board recently noted that
securitization ``provides originators much wider sources of funding
than they could obtain through conventional sources, such as retail
deposits'' and also that ``it substantially reduces the originator's
exposure to interest rate, credit, prepayment, and other risks.'' \15\
Echoing that statement, Federal Reserve Board Governor Elizabeth Duke
recently stated that the ``financial system has become dependent upon
securitization as an important intermediation tool,'' \16\ and last
week the International Monetary Fund (IMF) noted in its Global
Financial Stability Report that ``restarting private-label
securitization markets, especially in the United States, is critical to
limiting the fallout from the credit crisis and to the withdrawal of
central bank and Government interventions.'' \17\ There is clear
recognition in the official sector of the importance of the
securitization process and the access to financing that it provides
lenders, and of its importance to the availability of credit that
ultimately flows to consumers, businesses and the real economy.
---------------------------------------------------------------------------
\13\ G-7 Finance Ministers and Central Bank Governors Plan of
Action (Oct. 10, 2008). http://www.treas.gov/press/releases/hp1195.htm.
\14\ U.S. Department of the Treasury, ``Road to Stability:
Consumer & Business Lending Initiative'', (March 2009). http://
www.financialstability.gov/roadtostability/lendinginitiative.html.
\15\ Bernanke, Ben S., ``Speech at the UC Berkeley/UCLA Symposium:
The Mortgage Meltdown, the Economy, and Public Policy, Berkeley,
California'', Board of Governors of the Federal Reserve System (Oct.
2008). http://www.federalreserve.gov/newsevents/speech/
bernanke20081031a.htm.
\16\ Duke, Elizabeth A., ``Speech at the AICPA National Conference
on Banks and Savings Institutions, Washington, DC'', Board of Governors
of the Federal Reserve System (Sept. 2009). http://
www.federalreserve.gov/newsevents/speech/duke20090914a.htm.
\17\ International Monetary Fund, ``Restarting Securitization
Markets: Policy Proposals and Pitfalls'', Global Financial Stability
Report: Navigating the Financial Challenges Ahead (Oct. 2009), p. 33.
http://www.imf.org/external/pubs/ft/gfsr/2009/02/pdf/text.pdf.
---------------------------------------------------------------------------
Restoration of function and confidence to the securitization
markets is a particularly urgent need, in light of capital and
liquidity constraints currently confronting financial institutions and
markets globally. As mentioned above, at present nearly $12 trillion in
U.S. assets are funded via securitization. With the process of bank de-
leveraging and balance sheet reduction still underway, and with
increased bank capital requirements on the horizon, the funding
capacity provided by securitization cannot be replaced with deposit-
based financing alone in the current or foreseeable economic
environment. Just last week, the IMF estimated that a financing ``gap''
of $440 billion will exist between total U.S. credit capacity available
for the nonfinancial sector and U.S. total credit demand from that
sector for the year 2009. \18\ Moreover, nonbank finance companies, who
have played an important role in providing financing to consumers and
small businesses, are particularly reliant on securitization to fund
their lending activities, since they do not have access to deposit-
based funding. Small businesses, who employ approximately 50 percent of
the Nation's workforce, depend on securitization to supply credit that
is used to pay employees, finance inventory and investment, and other
business purposes. Furthermore, many jobs are made possible by
securitization. For example, a lack of financing for mortgages hampers
the housing industry; likewise, constriction of trade receivable
financing can adversely affect employment opportunities in the
manufacturing sector. To jump start the engine of growth and jobs,
securitization is needed to help restore credit availability.
---------------------------------------------------------------------------
\18\ International Monetary Fund, ``The Road to Recovery'', Global
Financial Stability Report: Navigating the Financial Challenges Ahead
(Oct. 2009), p. 29. http://www.imf.org/external/pubs/ft/gfsr/2009/02/
pdf/text.pdf.
---------------------------------------------------------------------------
Simply put, the absence of a properly functioning securitization
market, and the funding and liquidity this market has historically
provided, adversely impacts consumers, businesses, financial markets,
and the broader economy. The recovery and restoration of confidence in
securitization is therefore a necessary ingredient for economic growth
to resume, and for that growth to continue on a sustained basis into
the future.
II. Current Conditions in the Securitization Market
The U.S. securitization markets experienced substantial dislocation
during the recent financial market turmoil, with a virtual collapse of
both supply and demand in the new-issue market, very substantial
reductions in liquidity, widespread declines in securities prices and
valuations, and increases in risk premiums throughout the secondary
market. While there have been signs of recovery in certain parts of the
securitization market throughout the first three calendar quarters of
2009, some market segments--most notably, private-label residential
mortgage backed securities--remain dormant, with other securitization
asset classes and market sectors remaining significantly challenged.
In the asset-backed securities market, total issuance volume
remains at a relatively low level, with 2009 issuance projected to
reach $130 billion, roughly in line with the $140 billion issued in
2008 but sharply down from the $750 billion issued in 2006. \19\
Although issuance rates in nearly all major asset classes, including
credit cards, auto and equipment loans, and student loans, picked up in
the second quarter of 2009, a recent ASF survey showed that market
participants expect securitization issuance rates to return to only
half of their predownturn levels over the next 2 to 3 years. For
residential mortgage-backed securities, 2009 to date has seen over $1.2
trillion in issuance, compared with a yearlong total of $1.3 trillion
in 2008 and $2.1 trillion in 2006. However, in 2009, less than 1
percent of this has been issued without a Government or GSE guarantee
(i.e., private-label MBS); this is compared with private-label MBS
comprising over 23 percent of all issuance during the time period from
1996 to 2006. \20\ Furthermore, private-label MBS transactions that
have occurred in 2009 involved pools of seasoned, conforming loans--no
major private-label residential mortgage-backed securities deal of
which we are aware has directly financed new mortgage loan origination
this year.
---------------------------------------------------------------------------
\19\ SIFMA, ``U.S. ABS Issuance'', http://www.sifma.org/
uploadedFiles/Research/Statistics/SIFMA_USABSIssuance.pdf.
\20\ SIFMA, ``U.S. Mortgage-Related Issuance'', http://
www.sifma.org/uploadedFiles/Research/Statistics/
SIFMA_USMortgageRelatedIssuance.pdf.
---------------------------------------------------------------------------
Part of the reason for this involves a broad retreat from risk by
many investors. The events of 2007 and 2008, especially in the RMBS
markets, resulted in significant losses for many investors. While it
seems unlikely that some types of investors, such as those who
purchased securitized instruments issued by structured investment
vehicles (SIVs) or certain types of collateralized debt obligations
(CDOs), will play a significant role in the future MBS and ABS markets,
the number of traditional securitization investors has also diminished,
and along with it, the liquidity they have provided to both senior and
subordinate parts of the market. Replacing at least a portion of this
investor base is a significant challenge faced by participants in
today's market.
Certain programs sponsored by the Federal Government--in
particular, the TALF program--have been successful in stimulating parts
of the new-issue securitization market. President Obama described TALF
as the Government's ``largest effort ever to help provide auto loans,
college loans, and small business loans to the consumers and
entrepreneurs who keep this economy running,'' \21\ and in many ways,
TALF is among the most successful of the Government's efforts to
bolster the consumer economy. As of September 2009, TALF has directly
financed $46 billion \22\ of ABS issuances out of the approximately $80
billion of ABS eligible for TALF that has been issued since March. \23\
Due in significant measure to TALF, credit costs on consumer ABS have,
across the board, returned to levels more in line with their historical
trends than the extremely high levels that were seen in late 2008 and
early 2009. For example, 3-year AAA credit card spreads to benchmark
rates had ballooned to more than 500 basis points, or 5 percent, above
LIBOR by January 2009, but have retracted to a level less than 1
percent above LIBOR. \24\ While this is not quite back to the spread
levels seen over the years leading up to the crisis, it represents a
more stable and economical level for issuers that translates into more
affordable rates for borrowers. In recent months a number of issuers
have been able to sell, at economical levels, transactions without the
support of TALF. \25\ Clearly there are other factors at play in this
recovery, including a generally more benign credit market, but one
cannot dismiss the considerable and positive impact of TALF.
---------------------------------------------------------------------------
\21\ Obama, Barack, ``Remarks of President Barack Obama--Address
to Joint Session of Congress'', (Feb. 24, 2009). http://
www.whitehouse.gov/the_press_office/remarks-of-President-Barack-Obama-
address-to-joint-session-of-congress.
\22\ SIFMA, ``TALF'', http://www.sifma.org/research/
research.aspx?ID=10256#TALF.
\23\ Allison, Herbert M., ``Written Testimony: Senate Committee on
Banking, Housing and Urban Affairs'', (Sept. 2009). http://
www.ustreas.gov/press/releases/tg298.htm.
\24\ JPMorgan Securitized Products Weekly, September 18, 2009, pp.
22-23.
\25\ See, for example: ``AmeriCredit's $725 Million Auto ABS
transaction'', (July 2009) http://www.reuters.com/article/pressRelease/
idUS140529+31-Aug-2009+BW20090831; JPMorgan's $2.53 billion credit card
ABS deal, (Sept. 2009) http://online.wsj.com/article/
SB125311472402316179.html.
---------------------------------------------------------------------------
TALF has helped somewhat to bring investors back to the parts of
consumer ABS markets that are not directly eligible for the program,
although the markets for debt rated lower than AAA are still
struggling. For example, 5-year single-A rated credit card ABS, which
are not TALF eligible, saw an even more severe spread widening than
that of AAA during the height of the disruption in late 2008. By
January 2009 spreads had ballooned to more than 15 percent above LIBOR,
but have since come back in to lower levels. \26\ The subordinate ABS
markets are still relatively dormant, and unless banks are able to
finance a greater portion of the capital structure, credit origination
via securitization cannot be fully restored.
---------------------------------------------------------------------------
\26\ ``JPMorgan Securitized Products Weekly'', September 18, 2009,
pp. 22-23.
---------------------------------------------------------------------------
Notwithstanding the success of the TALF program and the restoration
of a modest degree of securitization financing and liquidity in some
market segments, significant challenges remain, including establishing
a stable, sustainable, and broad-based platform for future
securitization market issuance and investment activity that is less
reliant on direct Government support.
III. Limitations and Deficiencies in Securitization Revealed by the
Recent Financial Market Crisis
The recent financial market crisis revealed several limitations and
weaknesses in securitization market activity. Among the multiple (and,
in many cases, interrelated) deficiencies revealed were the following:
1. Risk management failures, including the excessive or imprudent
use of leverage and mismanagement of liquidity risk. Many
market participants--including financial intermediaries,
investors, and others--established large, leveraged risk
positions in securitized instruments. A significant number of
these market positions were, in effect, highly levered triggers
which, when tripped by an adverse rating action or downward
price movement, caused widespread deleveraging and further
price reductions. At the same time, large parts of the
securitization market became reliant on cheap, short term
liquidity to finance long-term assets. When this liquidity
disappeared and financing was either repriced or withdrawn
completely, a more systematic deleveraging and unwinding
process ensued.
2. Credit ratings methodologies and assessments that proved to be
overly optimistic, and excessive reliance on credit ratings.
Especially in parts of the residential mortgage market, a
favorable economic environment and persistent increase in
housing prices masked gaps in credit rating agency models and
methodologies that did not sufficiently factor in the risk of
nationwide housing price declines and a high correlation in the
performance of the assets underlying certain mortgage and
asset-backed securities. At the same time, market participants
became overly reliant on credit ratings, and many failed to
perform or to act upon their own assessment of the risks
created by certain securitized transaction structures.
3. Deteriorating underwriting standards and loan quality.
Underwriting standards declined precipitously throughout
various segments of the credit markets, including but not
limited to subprime mortgages, with housing prices rising
steeply and credit and liquidity in plentiful supply. As loan
demand and competition among lending institutions intensified,
asset quality declined, leaving securitized instruments
vulnerable to credit-related performance impairments.
4. Gaps in data integrity, reliability and standardization.
Especially in parts of the residential mortgage market, a
combination of explosive lending growth, operational
weaknesses, the absence of standardized and comparable loan-
level data, an increasing prevalence of fraud and other factors
caused investors broadly to question the accuracy and integrity
of performance data relating to the assets underlying
securitizations. This led to a massive loss of confidence and
widespread aversion to securitized risk, including asset
classes and transaction structures that were far removed from
the direct source of these concerns.
5. A breakdown in checks and balances and lack of shared
responsibility for the system as a whole. While many within the
securitization industry were aware of the general deterioration
in credit underwriting standards and the other factors outlined
above, no single party or group of market participants enforced
sufficient discipline across all parts of the interdependent
securitization value chain. Weaknesses and deficiencies in one
part of the chain thus impaired the function of the chain in
its entirety.
It is important to note that the weaknesses outlined above are not
inherent in securitization per se. Instead, they relate to the manner
in which securitization was used in some settings by some market
participants. In general, the amount of risk inherent in a
securitization is equal to the risk that is embedded in the securitized
assets themselves. However, in retrospect it is clear that
securitization technology can be used in ways that can reduce and
distribute risk (i.e., can be beneficial to the financial system), or
that increase and concentrate that risk (i.e., can be detrimental to
the financial system). Ancillary practices and strategies employed in
some securitization transactions by some market participants--for
example, the use of additional leverage; reliance on short-term funding
for long-term liabilities; or the absence of effective risk management
controls--can amplify and concentrate those risks. This is especially
true when such practices and strategies relate to large dollar volumes
of transactions and risk positions held by multiple participants
throughout the financial system.
It is also important to recognize that many of the deficiencies
outlined above were prevalent, or at least more heavily concentrated,
in certain securitization market products and sectors, rather than
characterizing conditions or practices in the securitization market as
a whole. In fact, the most consequential deficiencies were concentrated
in portions of the residential mortgage market--and the subprime
mortgage market, in particular--and in certain types of CDOs, SIVs and
similar securities arbitrage structures. These transactions--many of
which relied on high degrees of leverage--generated significant
incremental demand for underlying securitization products. However,
much of that demand was ``artificial,'' in the sense that production of
underlying securitization products (e.g., subordinated risk tranches of
subprime RMBS) was driven by demand from CDOs and SIVs, rather than by
the financing needs of lenders or borrowing needs of consumers. In
other parts of the securitization market, including prime RMBS, credit
card, auto and student loan ABS, and asset-backed commercial paper
conduits, among others, securitization activity largely remained
focused on its historical role of financing the credit extension
activities of lenders, and the credit needs of their consumer and
business customers.
IV. Views on Securitization Policy and Market Reform Initiatives
Numerous policy and market reforms aimed at the securitization
market have been advanced in response to the broader financial market
crisis. Global policymaking bodies have proposed a series of
securitization reforms as part of their broader response to financial
market turmoil, and in the United States, both legislative and
regulatory responses are under active consideration. At the same time,
industry participants and their representative organizations are moving
forward with important reforms to securitization market practices and
to retool key parts of the market's operational infrastructure.
Overall, we believe that a targeted combination of thoughtful
policy reforms, coupled with industry initiatives to improve the
securitization market infrastructure, will help to establish a more
stable and lasting platform for future securitization market activity.
In general, we believe that these policy and industry reform measures
should facilitate the ability to originate and fund of a wide range of
consumer and business credit via securitization. However, this activity
must be supported by improved data and transparency that enables
securitized risk to be evaluated and priced efficiently by market
participants, and by enhanced operational controls (including but not
limited to asset origination practices, due diligence and quality
review practices, standardized and more effective representations and
warranties, standardization of key documentation provisions and rating
agency methodologies, among others) that provide necessary assurances
to investors and other market participants regarding the accuracy,
integrity and reliability of securitization data and transaction
structures. At the same time, we believe that it is important, as a
recent IMF report noted, to consider the individual and combined
effects of various reform measures under consideration, to ensure that
they do not inadvertently stifle otherwise sound and desirable
securitization activity. \27\
---------------------------------------------------------------------------
\27\ The exact language used by the IMF in its Global Financial
Stability Report states: ``While most of the current proposals are
unambiguously positive for securitization markets and financial
stability, some proposals--such as those designed to improve the
alignment of securitizer and investor interests and accounting changes
that will result in more securitized assets remaining on balance
sheets--may be combined in ways that could halt, not restart,
securitization, by inadvertently making it too costly for
securitizers.'' See, ``The Road to Recovery'', (Oct. 2009), p. 29.
http://www.imf.org/external/pubs/ft/gfsr/2009/02/pdf/text.pdf.
---------------------------------------------------------------------------
In the United States, a primary policy focus is on legislative
proposals advanced by the Obama Administration, which in turn reflect
many of the reform themes and initiatives under consideration globally.
Together with other reforms being pursued by Federal regulatory
agencies and accounting standards setters, these securitization reform
initiatives may be broadly categorized as follows:
1. Increased Data Transparency, Disclosure, and Standardization; and
Improvements to the Securitization Infrastructure. Initiatives
designed to increase the type and amount of information and
data (including loan-level data) that is captured and disclosed
with respect to securitized instruments, and to improve and
standardize that information and data as well as key
documentation provisions, market practices and procedures
employed in securitization transactions.
2. Required Risk Retention and Other Incentive Alignment Mechanisms.
Mandated requirements for asset originators and/or securitizers
to retain an economic interest in securitization transactions,
and other mechanisms designed to produce a closer alignment of
economic risks and incentives of originators, securitizers, and
end investors.
3. Increased Regulatory Capital Requirements and Limitations on Off-
Balance Sheet Accounting. Increases in regulatory capital
required to be held against securitized exposures by regulated
financial institutions, as a means of creating an additional
safety and soundness buffer against potential losses associated
with those exposures, and revisions to generally accepted
accounting standards that restrict off-balance sheet accounting
for securitized transactions and produce more widespread
accounting consolidation of the assets and liabilities of
securitization special purpose entities.
4. Credit Rating Agency Reforms. Various reforms intended to
eliminate or minimize conflicts of interest, and to promote the
accuracy, integrity and transparency of methodologies and
processes that credit rating agencies apply to securitization
transactions.
A summary of ASF's views on each of these reform directions and
initiatives are set forth below.
A. Increased Transparency, Disclosure, Standardization; and
Improvements to the Securitization Market Infrastructure
ASF supports increased transparency and standardization in the
securitization markets, and related improvements to the securitization
market infrastructure. We believe that such efforts should be focused
on those areas and products where preexisting practices have been
determined to be deficient, and where improvements can help to restore
confidence and function to the related market segment(s).
Our principal focus in this area is ASF's Project on Residential
Securitization Transparency and Reporting (Project RESTART), which is
initially directed at addressing transparency and standardization
deficiencies in the residential mortgage-backed securities (RMBS)
market. Prior studies and market surveys conducted by ASF have clearly
identified the RMBS market as most in need of these types of reform.
Overall, Project RESTART seeks to address transparency and
standardization needs in the RMBS market via the substantial injection
of new disclosures and reporting by issuers and servicers on new
transactions as well as on the trillions of dollars of outstanding
private-label RMBS. Project RESTART would create a uniform set of data
standards for such disclosure and reporting, including at the loan
level. This will create a more level playing field where issuers
provide the same information at the initiation of a securitization
transaction and on an ongoing basis throughout the life of that
transaction. With these standards in place, information provided by
different issuers will be more comparable and capable of meaningful
evaluation by investors and other market participants. In addition to
supporting investment analysis, these data and standardization
improvements will also support more robust and reliable rating agency,
due diligence, quality review and valuation processes, and other
downstream applications that will benefit from more robust, reliable
and comparable underlying data.
Project RESTART for RMBS transactions consists of the following
phases: (i) the Disclosure Package, which will provide substantially
more loan-level data than is currently available to investors, rating
agencies and other parties, and standardize the presentation of
transaction-level and loan-level data to allow for a more ready
comparison of transactions and loans across issuers; (ii) the Reporting
Package, which will provide for monthly updating of critical loan-level
information that will enable improvements in the ability of investors,
rating agencies and other market participants to analyze the
performance of outstanding securities; (iii) Model RMBS Representations
and Warranties, which will provide assurances to investors in RMBS
transactions regarding the allocation and assumption of risk associated
with loan origination and underwriting practices; (iv) Model Repurchase
Procedures, which will be used to enforce the Model Representations and
Warranties and to clearly delineate the roles and responsibilities of
transaction parties in the repurchase process; (v) Model Pre-
Securitization Due Diligence Standards, which will buttress due
diligence and quality review practices relating to mortgage
underwriting and origination practices and the data supplied to market
participants through the Disclosure Package; and (vi) Model Servicing
Provisions for Pooling and Servicing Agreements, which will create more
standardized documentation provisions and work rules in key areas, such
as loss mitigation procedures that servicers may employ in dealing with
delinquent or defaulting loans.
Final versions of the Disclosure and Reporting Packages were
released by ASF in July 2009, with industry implementation beginning in
2010. Work continues on the other Project RESTART workstreams
identified above, with an immediate focus on the development of Model
RMBS Representations and Warranties, which are used to act as a
``return policy'' to guard against the risk of defective mortgage loans
being sold into a securitization trust. Much like a defective product
is returned to the store from which it was sold, a defective mortgage
loan will be ``returned'' to the issuer through its removal from a
securitization trust for cash. A mortgage loan is ``defective'' if it
materially breaches one of the representations and warranties. Examples
of defects range from a general fraud in a loan's origination to a
failure to properly verify a borrower's income or employ an independent
appraiser. The ASF supports 100 percent risk retention for defective
loans that result from an originator's failure to meet specified
underwriting criteria.
Although Project RESTART has initially been focused on the RMBS
market, members of the ASF have begun development of the ASF Credit
Card ABS Disclosure Package, which seeks to provide increased
transparency and standardization to the Credit Card ABS market.
Finally, ASF believes that every mortgage loan should be assigned a
unique identification number at origination, which would facilitate the
identification and tracking of individual loans as they are sold or
financed in the secondary market, including via RMBS securitization.
ASF recently selected a vendor who will work with us to provide this
unique Loan ID, which is called the ASF LINCTM.
Implementation of the ASF LINC will enable market participants to
access Project RESTART's valuable loan-level information without
violating privacy laws by removing personal nonpublic information and
other protected information from the process.
B. Required Risk Retention
ASF supports initiatives to align the economic interests of asset
originators and securitization sponsors with investors. As suggested
above, we believe that the principal goal of these efforts should be to
establish and reinforce commercial incentives for originators and
sponsors to create and fund assets that conform to stated underwriting
standards and securitization eligibility criteria, thereby making those
parties economically responsible for the stated attributes and
underwriting quality of securitized loans. The creation and maintenance
of effective mechanisms of this type will facilitate responsible
lending, as well as a more disciplined and efficient funding of
consumer assets via securitization (i.e., where the varying credit and
performance risks presented by different types of securitized assets
can be properly evaluated and priced in the capital markets).
Securitization risk retention proposals currently under
consideration, including legislation advanced by the Obama
Administration, call for securitization sponsors and/or asset
originators to retain an economic interest in a material portion of the
credit risk that the sponsor and/or asset originator conveys to a third
party via a securitization transaction.
As noted above, we support the concept of requiring retention of a
meaningful economic interest in securitized loans as a means of
creating a better alignment of incentives among transaction
participants. Many securitizations already embed this concept through
various structuring mechanisms, including via the retention of
subordinated or equity risk in the securitization, holding portfolio
assets bearing credit exposure that is similar or identical to that of
securitized assets, and representations and warranties that require
originators or sponsors to repurchase assets that fail to meet stated
securitization eligibility requirements, among others. However, we do
not believe that mandated retention of specific portions of credit
risk--one such form of economic interest--necessarily constitutes the
sole or most effective means of achieving this alignment in all cases.
There are numerous valid and competing policy goals that stand in
opposition to requiring the retention of credit risk in securitized
assets and exposures. Among others, these include the proper isolation
of transferred assets (i.e., meeting legal criteria necessary to effect
a ``true sale''); reduction and management of risk on financial
institutions' balance sheets; balance sheet management; and the
redeployment of capital to enable financial institutions to originate
more credit than their limited capital resources would otherwise allow.
Balancing these competing and worthwhile policy goals suggests that
retention and incentive alignment mechanisms other than universal
credit risk retention requirements should be considered. This viewpoint
was echoed by the IMF last week in its Global Financial Stability
Report, which expressed strong concerns about the potential unintended
negative consequences of implementing suggested credit risk retention
requirements and instead indicated that regulatory authorities ``should
consider other mechanisms that incentivize due diligence and may be
able to produce results comparable to a retention requirement,
including, perhaps, representations and warranties.'' \28\
---------------------------------------------------------------------------
\28\ International Monetary Fund, ``Restarting Securitization
Markets: Policy Proposals and Pitfalls.'' Global Financial Stability
Report: Navigating the Financial Challenges Ahead (Oct. 2009), p. 31.
http://www.imf.org/external/pubs/ft/gfsr/2009/02/pdf/text.pdf.
---------------------------------------------------------------------------
We believe that the risk or ``skin in the game'' traditionally
retained by originators of RMBS is embodied in the representations and
warranties that issuers provide with respect to the mortgage loans sold
into the securitization trust. These representations and warranties are
designed to ensure that the loans are free from undisclosed origination
risks, leaving the investor primarily with normal risks of loan
ownership, such as the deterioration of the borrower's credit due to
loss of employment, disability or other ``life events.'' However, many
market participants have indicated that the traditional representations
and warranties and their related remedy provisions have not
sufficiently provided a means to return defective loans to the
originator. Because of this, the ASF has sought to enhance and
standardize these items through the previously discussed Project
RESTART Model RMBS Representations and Warranties and Model Repurchase
Provisions.
We therefore believe that to the extent legislation is adopted to
require risk retention, regulators should have flexibility to develop
and apply alternative retention mechanisms. This flexibility should
include the ability for regulators to specify permissible forms and
amounts of retention, how retention requirements may be calculated and
measured, the duration of retention requirements, whether and to what
extent hedging or risk management of retained positions is permissible,
and other implementation details.
Finally, we believe that it is imperative to achieve global
harmonization and consistency of policy approaches to securitization
risk retention. Different approaches are being considered and/or have
been adopted in different jurisdictions. \29\ Given the global nature
of securitization activity and the mobility of global capital among
jurisdictions, significant competitive disparities and inefficiencies
may be produced by introducing substantively different retention
standards throughout the world's financial markets. We believe that is
essential for policymakers to coordinate their approaches in this area.
---------------------------------------------------------------------------
\29\ One such approach was adopted by the European Parliament in
May 2009. Article 122a to the Capital Requirements Directive prohibits
EU banks from investing in securitizations unless the originator
retains on an ongoing basis at least 5 percent of the material net
economic interest of the securities securitized. The article proposes
four ways the 5 percent retention requirement may be applied. The
article's requirement is scheduled to go into effect on December 31,
2010, for new issues, and December 31, 2014, for existing
securitizations where new underlying exposures are added or subtracted
after that date. For more information, see: http://
www.europarl.europa.eu/sides/getDoc.do?type=TA&reference=P6-TA-2009-
0367&language=EN&ring=A6-2009-0139#BKMD-35.
---------------------------------------------------------------------------
C. Increased Regulatory Capital and Limitations on Off-Balance Sheet
Financing
The Obama Administration has advocated that risk-based regulatory
capital requirements should appropriately reflect the risk of
structured credit products, including the concentrated risk of senior
tranches and resecuritizations and the risk of exposures held in highly
leveraged off-balance sheet vehicles. Global policymakers have also
advocated for minimizing opportunities for financial institutions to
use securitization to reduce their regulatory capital requirements
without a commensurate reduction in risk.
Consistent with the above views, the Basel Committee on Banking
Supervision has amended the Basel II risk-based capital framework to
require additional regulatory capital to be held against certain
resecuritizations (such as CDOs), on the basis that previous rules
underestimated the risks inherent in such structures. In the U.S., the
combined bank regulatory agencies recently issued proposals that would
continue to link risk-based capital requirements to whether an
accounting sale has occurred under U.S. GAAP. Given that recent
accounting changes (which will generally take effect in January 2010)
will make it very difficult to achieve GAAP sales in many
securitizations, including both term asset-backed securities and asset-
backed commercial paper vehicles, these proposed rules will likely
materially increase the capital that financial institutions will be
required to hold in against securitizations, since many securitized
assets will remain on or return to those institutions' balance sheets.
ASF supports efforts to addresses weaknesses in the risk-based
capital framework that have been revealed in certain securitization
products by the recent financial market dislocation, and agrees that
regulatory capital levels should adequately reflect the risks of
different types of securitization transactions. Furthermore, ASF
supports efforts to reduce or eliminate opportunities for regulatory
capital arbitrage that are unrelated to differences in the risk
profiles of securitization instruments.
We therefore believe that increases in regulatory capital
requirements for certain securitizations may be appropriate, based on
the conclusion that they present more risk than had been previously
understood (for example, because of their use of leverage or where
underlying risk positions are more highly correlated than they were
assumed to be, as in the case of certain CDOs and SIVs). However, a
broader increase in capital requirements for securitization across the
board, that is not tied to the differing risk profiles of different
transactions, may produce very negative consequences for the economic
viability of securitization. In turn, this outcome could unduly
constrain the ability of financial institutions to originate and fund
consumer and business credit demand, particularly as the broader
economy begins to recover.
ASF is particularly concerned that linking risk-based capital
requirements to accounting outcomes--particularly when those outcomes
are produced by the application of accounting standards that are not
themselves risk-based--is an inappropriate policy response. We believe
that the resulting increase in regulatory capital required to be held
against securitized assets held on financial institutions' balance
sheets will grossly misrepresent the actual, incremental risk inherent
in those assets. We believe that a more targeted approach to revising
the securitization risk-based capital framework is warranted. Last week
ASF asked the U.S. bank regulatory agencies for a 6-month moratorium
relating to any changes in bank regulatory capital requirements
resulting from the implementation of FASB's Statements 166 and 167. We
believe that this action is necessary to avoid a potentially severe
capital and credit shock to the financial system as of January 1st,
when the new accounting rules generally take effect. We will be
providing detailed input and recommendations to bank regulatory
agencies and other policymakers on this important topic by the October
15th deadline.
D. Rating Agency Reforms
ASF supports credit rating reform in the securitization markets,
focusing on steps designed to increase the quality, accuracy and
integrity of credit ratings and the transparency of the ratings
process. Credit ratings have occupied a central role in the
securitization markets, providing investors and other market
participants with expert views on the credit performance and risks
associated with a wide range of securitization products. As an
outgrowth of the financial market crisis, confidence in rating agencies
and the ratings process for securitization have been significantly
impaired. We believe that a restoration of such confidence is a
necessary step in restoring broader confidence and function to the
securitization markets.
Various credit rating reform measures targeting the securitization
markets have been advanced by policymakers, and a number of proposals
have been adopted or remain under consideration by the Securities
Exchange Commission. Our views on some of the more significant
proposals affecting the securitization market are summarized below:
\30\
---------------------------------------------------------------------------
\30\ For more detail on ASF's views on these and other credit
rating agency reform proposals, see the series of letters submitted to
the SEC by ASF between May and September of 2008. These letters may be
found at: http://www.americansecuritization.com/uploadedFiles/
ASFpercent20CRA percent20-percent20ratings percent20scale.pdf (May
2008); http://www.americansecuritization.com/uploadedFiles/Release_34-
57967_ASF_Comment_Letter_.pdf (July 2008); and http://
www.americansecuritization.com/uploadedFiles/ASF_Final_SEC_CRA--
Letter_9_5_08.pdf (Sept. 2008).
1. Conflicts of Interest. We support measures aimed at developing
and enhancing strong conflict of interest policies and rules
governing the operations of credit rating agencies. We believe
that effective management and disclosure of actual and
potential conflicts is a necessary component for ensuring
---------------------------------------------------------------------------
transparency and integrity in the rating process.
2. Differentiation of Structured Finance Ratings. ASF supports full
and transparent disclosure of the basis for structured finance
ratings, so that the risk of securitizations can be understood
and differentiated from risks presented by other types of
credit instruments. However, we strongly oppose proposals
advocating that a special ratings designation or modifier be
required for structured finance ratings. We believe that such a
designation or modifier would not convey any meaningful
information about the rating, and would require significant
revisions to private investment guidelines that incorporate
ratings requirements.
3. Ratings Performance Disclosure. We support the publication in a
format reasonably accessible to investors of a record of all
ratings actions for securitization instruments for which
ratings are published. We believe that publication of these
data will enable investors and other market participants to
evaluate and compare the performance, stability, and quality of
ratings judgments over time.
4. Disclosure of Ratings Methods and Processes. ASF strongly
supports enhanced disclosure of securitization ratings methods
and processes, including information relating to the use of
ratings models and key assumptions utilized by those models.
5. Reliance on Ratings. We believe that investors and other market
participants, including regulators, should not place an undue
reliance on credit ratings, and should employ other mechanisms
for performing an independent credit analysis. However, ASF
believes that credit ratings are an important part of existing
regulatory regimes, and that steps aimed at reducing or
eliminating the use of ratings in regulation should be
considered carefully, to avoid undue disruption to market
function and efficiency.
Conclusion
The securitization market is an essential mechanism for supporting
credit creation and capital formation throughout the consumer and
business economy. Its role is even more important today, when other
sources of credit and financing are limited, due to balance sheet,
capital, and liquidity constraints facing financial institutions.
Securitization activity was significantly impaired as a consequence of
the financial market crisis. While portions of the securitization
market have recovered to some extent throughout 2009, other market
segments remain significantly challenged.
The financial market crisis revealed weaknesses in several key
areas of securitization market activity. Targeted reforms are needed,
and a number are being pursued through both public- and private-sector
responses. In pursuing market reforms and redressing these weaknesses,
care should be taken to avoid imposing undue impediments to the
restoration of securitization activity that could adversely impact
credit availability and retard economic recovery and growth.
Thank you for the opportunity to share these views, and I look
forward to answering any questions that Members of the Subcommittee may
have.
______
PREPARED STATEMENT OF ANDREW DAVIDSON
President, Andrew Davidson and Company
October 7, 2009
Mr. Chairman and Members of the Subcommittee, I appreciate the
opportunity to testify before you today about securitization. My
expertise is primarily in the securitization of residential mortgages
and my comments will be primarily directed toward those markets.
Securitization has been a force for both good and bad in our
economy. A well functioning securitization market expands the
availability of credit for economic activity and home ownership. It
allows banks and other financial institutions to access capital and
reduces risk. On the other hand a poorly functioning securitization
market may lead to misallocation of capital and exacerbate risk. \1\
---------------------------------------------------------------------------
\1\ Portions of this statement are derived from ``Securitization:
After the Fall'', Anthony Sanders and Andrew Davidson, forthcoming.
---------------------------------------------------------------------------
Before delving into a discussion of the current crisis, I would
like to distinguish three types of capital markets activities that are
often discussed together: Securitization, Structuring, and Derivatives.
\2\
---------------------------------------------------------------------------
\2\ See, Andrew Davidson, Anthony Sanders, Lan-Ling Wolff, and
Anne Ching, ``Securitization'', 2003, for a detailed discussion of
securitization and valuation of securitized products.
---------------------------------------------------------------------------
Securitization is the process of converting individual loans into
securities that can be freely transferred. Securitization serves to
separate origination and investment functions.
Without securitization investors would need to go through a very
complex process of transferring the ownership of individual loans. The
agency mortgage-backed securities (MBS) from Ginnie Mae, Fannie Mae,
and Freddie Mac are one of the most successful financial innovations.
However, as the last years have taught us, the so-called, ``originate
to sell'' model, especially as reflected in private-label (nonagency)
MBS, has serious shortcomings.
Structuring is the process of segmenting the cash flows of one set
of financial instruments into several bonds which are often called
tranches. The collateralized mortgage obligation or CMO is a classic
example of structuring. The CMO transforms mortgage cash flows into a
variety of bonds that appeal to investors from short-term stable bonds,
to long-term investments. Private label MBS use a second form of
structuring to allocate credit risk. A typical structure uses
subordination, or over-collateralization, to create bonds with
different degrees of credit risk. The collateralized debt obligation or
CDO is a third form of structuring. In this case, bonds, rather than
loans, are the underlying collateral for the CDO bonds which are
segmented by credit risk. Structuring allows for the expansion of the
investor base for mortgage cash flows, by tailoring the bonds
characteristics to investor requirements. Unfortunately, structuring
has also been used to design bonds that obfuscate risk and return.
Derivatives, or indexed contracts, are used to transfer risk from
one party to another. Derivatives are a zero sum game in that one
investor's gain is another's loss. While typically people think of
swaps markets and futures markets when they mention derivatives, the
TBA (to be announced) market for agency pass-through mortgages is a
large successful derivative market. The TBA market allows for trading
in pass-through MBS without the need to specify which pool of mortgages
will be delivered. More recently a large market in mortgage credit risk
has developed. The instruments in this market are credit default swaps
(CDS) and ABX, an over-the-counter index based on subprime mortgage
CDS. Derivatives allow for risk transfer and can be powerful vehicles
for risk management. On the other hand, derivatives may lead to the
creation of more risk in the economy as derivate volume may exceed the
underlying asset by substantial orders of magnitude.
For any of these products to be economically useful they should
address one or more of the underlying investment risks of mortgages:
funding, interest rate risk, prepayment risk, credit risk, and
liquidity. More than anything else mortgages represent the funding of
home purchases. The twelve trillion of mortgages represents funding for
the residential real estate of the country. Interest rate risk arises
due to the fixed coupon on mortgages. For adjustable rate mortgages it
arises from the caps, floors and other coupon limitations present in
residential mortgage products. Interest rate risk is compounded by
prepayment risk. Prepayment risk reflects both a systematic component
that arises from the option to refinance (creating the option features
of MBS) as well as the additional uncertainty created by the difficulty
in accurately forecasting the behavior of borrowers. Credit risk
represents the possibility that borrowers will be unable or unwilling
to make their contractual payments. Credit risk reflects the borrower's
financial situation, the terms of the loan and the value of the home.
Credit risk has systematic components related to the performance of the
economy, idiosyncratic risks related to individual borrowers and
operational risks related to underwriting and monitoring. Finally,
liquidity represents the ability to transfer the funding obligation
and/or the risks of the mortgages.
In addition to the financial characteristics of these financial
tools, they all have tax, regulatory and accounting features that
affect their viability. In some cases tax, regulatory and accounting
outcomes rather than financial benefit are the primary purpose of a
transaction. In developing policy alternatives each of these
activities: securitization, structuring and derivatives, pose distinct
but interrelated challenges.
Role of Securitization in the Current Financial Crisis
The current economic crisis represents a combination of many
factors and blame can be laid far and wide. Additional analysis may be
required to truly assess the causes of the crisis. Nevertheless I
believe that securitization contributed to the crisis in two important
ways. It contributed to the excessive rise in home prices and it
created instability once the crisis began.
First, the process of securitization as implemented during the
period leading up to the crisis allowed a decline in underwriting
standards and excessive leverage in home ownership. The excess lending
likely contributed to the rapid rise in home prices leading up to the
crisis. In addition to the well documented growth in subprime and Alt-A
lending, we find that the quality of loans declined during the period
from 2003 to 2005, even after adjusting for loan to value ratios, FICO
scores, documentation type, home prices and other factors reflected in
data available to investors. The results of our analysis are shown in
Figure 1. It shows that the rate of delinquency for loans originated in
2006 is more than 50 percent higher than loans originated in 2003. The
implication is that the quality of underwriting declined significantly
during this period, and this decline was not reflected in the data
provided to investors. As such it could reflect fraud,
misrepresentations and lower standard for verifying borrower and
collateral data. The net impact of this is that borrowers were granted
credit at greater leverage and at lower cost than in prior years.
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
In concrete terms, the securitization market during 2005 and 2006
was pricing mortgage loans to an expected lifetime loss of about 5
percent. Our view is that even if home prices had remained stable,
these losses would have been 10 percent or more. Given the structure of
many of these loans, with a 2-year initial coupon and an expected
payoff by the borrower at reset, the rate on the loans should have been
200 or 300 basis points higher. That is, initial coupons should have
been over 10 percent rather than near 8 percent.
Our analysis further indicates that this lower cost of credit
inflated home prices. The combination of relaxed underwriting standards
and affordability products, such as option-arms, effectively lowered
the required payment on mortgages. The lower payment served to increase
the price of homes that borrowers could afford. Figure 2 shows the
rapid rise in the perceived price that borrowers could afford in the
Los Angeles area due to these reduced payment requirements. Actual home
prices then followed this pattern. Generally we find that
securitization of subprime loans and other affordability products such
as option arms were more prevalent in the areas with high amounts of
home price appreciation during 2003 to 2006. To be clear, not all of
the affordability loans were driven by securitization, as many of the
option arms remained on the balance sheet of lending institutions.
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Figure 3 provides an indication of the magnitude of home price
increases that may have resulted from these products on a national
basis. Based on our home price model, we estimate that home prices may
have risen by 15 percent at the national level due to lower effective
interest rates. In the chart, the gap between the solid blue line and
the dashed blue line reflects the impact of easy credit on home prices.
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
On the flip side, we believe that the shutting down of these
markets and the reduced availability of mortgage credit contributed to
the sharp decline in home prices we have seen since 2006 as shown in
Figure 4. Without an increase in effective mortgage rates, home prices
might have sustained their inflated values as shown by the dashed blue
line. \3\
---------------------------------------------------------------------------
\3\ See, http://www.ad-co.com/newsletters/2009/Jun2009/
Valuation_Jun09.pdf for more details.
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Thus the reduced focus on underwriting quality lead to an
unsustainable level of excess leverage and reduced borrowing costs
which helped to inflate home prices. When these ``affordability''
products were no longer sustainable in the market, they contributed to
the deflation of the housing bubble.
The way securitization was implemented during this period fostered
high home prices through poor underwriting, and the end of that era may
have led to the sharp decline in home prices and the sharp decline in
home prices helped to spread the financial crisis beyond the subprime
market.
The second way that securitization contributed to the current
economic crisis is through the obfuscation of risk. For many structures
in the securitization market: especially collateralized debt
obligations, structured investment vehicles and other
resecuritizations, there is and was insufficient information for
investors to formulate an independent judgment of the risks and value
of the investment. As markets began to decline in late 2007, investors
in all of these instruments and investors in the institutions that held
or issued these instruments were unable to assess the level of risk
they bore.
This lack of information quickly became a lack of confidence and
led to a massive deleveraging of our financial system. This
deleveraging further depressed the value of these complex securities
and led to real declines in economic value as the economy entered a
severe recession. In addition, regulators lacked the ability to assess
the level of risk in regulated entities, perhaps delaying corrective
action or other steps that could have reduced risk levels earlier.
Limitations of Securitization Revealed
To understand how the current market structure could lead to
undisciplined lending and obfuscation of risk it is useful to look at a
simplified schematic of the market. \4\
---------------------------------------------------------------------------
\4\ Adapted from ``Six Degrees of Separation'', August 2007, by
Andrew Davidson http://www.securitization.net/pdf/content/
ADC_SixDegrees_1Aug07.pdf.
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
In the simplest terms, what went wrong in the subprime mortgage in
particular and the securitization market in general is that the people
responsible for making loans had too little financial interest in the
performance of those loans and the people with financial interest in
the loans had too little involvement in the how the loans were made.
The secondary market for nonagency mortgages, including subprime
mortgages, has many participants and a great separation of the
origination process from the investment process. Each participant has a
specialized role. Specialization serves the market well, as it allows
each function to be performed efficiently. Specialization, however also
means that risk creation and risk taking are separated.
In simplified form the process can be described as involving:
A borrower--who wants a loan for home purchase or refinance
A broker--who works with the borrower and lenders to
arrange a loan
A mortgage banker--who funds the loan and then sells the
loan
An aggregator--(often a broker-dealer) who buys loans and
then packages the loans into a securitization, whose bonds are
sold to investors.
A CDO manager--who buys a portfolio of mortgage-backed
securities and issues debt
An investor--who buys the CDO debt
Two additional participants are also involved:
A servicer--who keeps the loan documents and collects the
payments from the borrower
A rating agency--that places a rating on the mortgage
securities and on the CDO debt
This chart is obviously a simplification of a more complex process.
For example, CDOs were not the only purchasers of risk in the subprime
market. They were however a dominant player, with some estimating that
they bought about 70 percent of the lower rated classes of subprime
mortgage securitizations. What is clear even from this simplified
process is that contact between the provider of risk capital and the
borrower was very attenuated.
A central problem with the securitization market, especially for
subprime loans was that no one was the gate keeper, shutting the door
on uneconomic loans. The ultimate CDO bond investor placed his trust in
the first loss investor, the rating agencies, and the CDO manager, and
in each case that trust was misplaced.
Ideally mortgage transactions are generally structured so that
someone close to the origination process would take the first slice of
credit risk and thus insure that loans were originated properly. In the
subprime market, however it was possible to originate loans and sell
them at such a high price, that even if the mortgage banker or
aggregator retained a first loss piece (or residual) the transaction
could be profitable even if the loans did not perform well.
Furthermore, the terms of the residuals were set so that the owner of
the residual might receive a substantial portion of their cash flows
before the full extent of losses were known.
Rating agencies set criteria to establish credit enhancement levels
that ultimately led to ratings on bonds. The rating agencies generally
rely on historical statistical analysis to set ratings. The rating
agencies also depend on numeric descriptions of loans like loan-to-
value ratios and debt-to-income ratios to make their determinations.
Rating agencies usually do not review loans files or ``re-underwrite''
loans. Rating agencies also do not share in the economic costs of loan
defaults. The rating agencies methodology allowed for the inclusion of
loans of dubious quality into subprime and Alt-A mortgage pools,
including low documentation loans for borrowers with poor payment
histories, without the offsetting requirement of high down payments.
To help assure investors of the reliability of information about
the risks of purchased loans, the mortgage market has developed the
practice of requiring ``representations and warranties'' on purchased
loans. These reps and warrants as they are called, are designed to
insure that the loans sold meet the guidelines of the purchasers. This
is because mortgage market participants have long recognized that there
is substantial risk in acquiring loans originated by someone else. An
essential component in having valuable reps and warrants is that the
provider of those promises has sufficient capital to back up their
obligations to repurchase loans subsequently determined to be
inconsistent with the reps and warrants. A financial guarantee from an
insolvent provider has no value.
Representations and warranties are the glue that holds the process
together; if the glue is weak the system can collapse.
The rating agencies also established criteria for Collateralized
Debt Obligations that allowed CDO managers to produce very highly
leveraged portfolios of subprime mortgage securities. The basic
mechanism for this was a model that predicted the performance of
subprime mortgage pools were not likely to be highly correlated. That
is defaults in one pool were not likely to occur at the same time as
defaults in another pool. This assumption was at best optimistic and
most likely just wrong.
In the CDO market the rating agencies have a unique position. In
most of their other ratings business, a company or a transaction exists
or is likely to occur and the rating agency reviews that company or
transaction and establishes ratings. In the CDO market, the criteria of
the rating agency determine whether or not the transaction will occur.
A CDO is like a financial institution. It buys assets and issues debt.
If the rating agency establishes criteria that allow the institution to
borrow money at a low enough rate or at high enough leverage, then the
CDO can purchase assets more competitively than other financial
institutions. If the CDO has a higher cost of debt or lower leverage,
then it will be at a disadvantage to other buyers and will not be
brought into existence. If the CDO is created, the rating agency is
compensated for its ratings. If the CDO is not created, there is no
compensation. My view is that there are very few institutions that can
remain objective given such a compensation scheme.
CDO bond investors also relied upon the CDO manager to guide them
in the dangerous waters of mortgage investing. Here again investors
were not well served by the compensation scheme. In many cases CDO
managers receive fees that are independent of the performance of the
deals they manage. While CDO managers sometimes keep an equity interest
in the transactions they manage, the deals are often structured in such
a way that that the deal can return the initial equity investment even
if some of the bonds have losses. Moreover, many of the CDOs were
managed by start-up firms with little or no capital.
Nevertheless, much of the responsibility should rest with the
investors. CDO bond investors were not blind to the additional risks
posed by CDO investing. CDOs generally provided higher yields than
similarly rated bonds, and it is an extremely naive, and to my mind,
rare, investor who thinks they get higher returns without incremental
risk. It is not unusual, however, for investors not to realize the
magnitude of additional risk they bear for a modest incremental return.
Ultimately it is investors who will bear the losses, and investors must
bear the bulk of the burden in evaluating their investments. There were
clear warning signs for several years as to the problems and risk of
investing in subprime mortgages. Nevertheless, investors continued to
participate in this sector as the risks grew and reward decreased.
As expressed herein, the primary problem facing securitization is a
failure of industrial organization. The key risk allocators in the
market, the CDO managers, were too far from the origination process
and, at best, they believed the originators and the rating agencies
were responsible for limiting risk. At the origination end, without the
discipline of a skeptical buyer, abuses grew. The buyer was not
sufficiently concerned with the process of loan origination and the
broker was not subject to sufficient constraints.
Current Conditions of the Mortgage-backed Securities Market
More than 2 years after the announcement of the collapse of the
Bear Stearns High Grade Structured Credit Enhanced Leverage Fund the
mortgage market remains in a distressed state. Little of the mortgage
market is functioning without the direct involvement of the U.S.
Government, and access to financing for mortgage originators and
investors is still limited.
Fortunately there are the beginning signs of stabilization of home
prices, but rising unemployment threatens the recovery. In the
secondary market for mortgage-backed securities there has been
considerable recovery in price in some sectors, but overall demand is
being propped up by large purchases of MBS by the Federal Reserve Bank.
In addition, we find that many of our clients are primarily focused
on accounting and regulatory concerns related to legacy positions, and
less effort is focused on the economic analysis of current and future
opportunities. That situation may be changing as over the past few
months we have seen some firms begin to focus on longer term goals.
The Effectiveness of Government Action
I have not performed an independent analysis of the effectiveness
of Government actions, so by comments are limited to my impressions.
Government involvement has been beneficial in a number of
significant respects. Without Government involvement in Fannie Mae,
Freddie Mac, and FHA lending programs, virtually all mortgage lending
could have stalled. What lending would have existed would have been for
only the absolute highest quality borrowers and at restrictive rates.
In addition Government programs to provide liquidity have also been
beneficial to the market as private lending was reduced to extremely
low levels. Government and Federal Reserve purchases of MBS have kept
mortgage rates low. This has probably helped to bolster home prices.
On the other hand the start/stop nature of the buying programs
under TARP and PPIP has probably been a net negative for the market.
Market participants have held back on investments in anticipation of
Government programs that either did not materialize or were
substantially smaller in scope than expected.
Furthermore Government efforts to influence loan modifications,
while beneficial for some home owners, and possibly even investors,
have created confusion and distrust. Investors are more reluctant to
commit capital when the rules are uncertain. In my opinion there has
been excessive focus on loan modifications as a solution to the current
crisis. Loan modifications make sense for a certain portion of
borrowers whose income has been temporarily disrupted or have
sufficient income to support a modestly reduce loan amount and the
willingness to make those payments. However for many borrowers, loan
modifications cannot produce sustainable outcomes. In addition, loan
modifications must deal with the complexities of multiple liens and
complex ownership structures of mortgage loans. Short sales, short
payoffs, and relocation assistance for borrowers are other alternatives
that should be given greater weight in policy development.
The extensive Government involvement in the mortgage market has
likely produced significant positive benefits to the economy. However
unwinding the Government role will be quite complex and could be
disruptive to the recovery. Government programs need to be reduced and
legislative and regulatory uncertainties need to be addressed to
attract private capital back into these markets.
Legislative and Regulatory Recommendations
I believe that the problems in the securitization market were
essentially due to a failure of industrial organization. Solutions
should address these industrial organization failures. While some may
seek to limit the risks in the economy, I believe a better solution is
to make sure the risks are borne by parties who have the capacity to
manage the risks or the capital to bear those risks. In practical
terms, this means that ultimately bond investors, as the creators of
leverage, must be responsible for limiting leverage to economically
sustainable levels that do not create excessive risk to their
stakeholders. Moreover, lenders should not allow equity investors to
have tremendous upside with little exposure to downside risk. Equity
investors who have sufficient capital at risk are more likely to act
prudently. Consequently, all the information needed to assess and
manage risks must be adequately disclosed and investors should have
assurances that the information they rely upon is accurate and timely.
Likewise when the Government acts as a guarantor, whether explicitly or
implicitly, it must insure that it is not encouraging excessive risk
taking and must have access to critical information on the risks borne
by regulated entities.
In this light, I would like to comment on the Administration
proposals on Securitization in the white paper: ``Financial Regulatory
Reform: A New Foundation.'' \5\ Recommendations 1 and 2 cover similar
ground:
---------------------------------------------------------------------------
\5\ http://www.financialstability.gov/docs/regs/
FinalReport_web.pdf pp. 44-46.
1. Federal banking agencies should promulgate regulations that
require originators or sponsors to retain an economic interest
in a material portion of the credit risk of securitized credit
---------------------------------------------------------------------------
exposures.
The Federal banking agencies should promulgate regulations that
require loan originators or sponsors to retain 5 percent of the
credit risk of securitized exposures.
2. Regulators should promulgate additional regulations to align
compensation of market participants with longer term
performance of the underlying loans.
Sponsors of securitizations should be required to provide
assurances to investors, in the form of strong, standardized
representations and warranties, regarding the risk associated
with the origination and underwriting practices for the
securitized loans underlying ABS.
Clearly excessive leverage and lack of economic discipline was at
the heart of the problems with securitization. As described above the
market failed to adequately protect investors from weakened
underwriting standards. Additional capital requirements certainly
should be part of the solution. However, such requirements need to be
constructed carefully. Too little capital and it will not have any
effect; too much and it will inhibit lending and lead to higher
mortgage costs. The current recommendation for retention of 5 percent
of the credit risk does not seem to strike that balance appropriately.
When a loan is originated there are several kinds of credit related
risks that are created. In addition to systematic risks related to
future events such as changes in home prices and idiosyncratic risks
such as changes in the income of the borrower, there are also
operational risks related to the quality of the underwriting and
servicing. An example of an underwriting risk is whether or not the
borrower's income and current value of their home were verified
appropriately. Originators are well positioned to reduce the
operational risks associated with underwriting and fight fraud, but
they may be less well positioned to bear the long term systematic and
idiosyncratic risks associated with mortgage lending. Investors are
well positioned to bear systemic risks and diversify idiosyncratic
risks, but are not able to assess the risks of poor underwriting and
servicing. The securitization process should ensure that there is
sufficient motivation and capital for originators to manage and bear
the risks of underwriting and sufficient information made available to
investors to assess the risks they take on.
The current form of representations and warranties is flawed in
that it does not provide a direct obligation from the originator to the
investor. Instead representations and warranties pass through a chain
of ownership and are often limited by ``knowledge'' and capital. In
addition current remedies are tied to damages and in a rising home
price market calculated damages may be limited. Thus a period of rising
home prices can mask declining credit quality and rising violations of
representations and warranties.
Therefore, incentives and penalties should be established to limit
unacceptable behavior such as fraud, misrepresentations, predatory
lending. If the goal is to prevent fraud, abuse and misrepresentations
rather than to limit risk transfer then there needs to be a better
system to enforce the rights of borrowers and investors than simply
requiring a originators to retain a set percentage of credit risk.
I have proposed \6\ a ``securitization certificate'' which would
travel with the loan and would be accompanied by appropriate assurances
of financial responsibility. The certificate would replace
representations and warranties, which travel through the chain of
buyers and sellers and are often unenforced or weakened by the
successive loan transfers. The certificate could also serve to protect
borrowers from fraudulent origination practices in the place of
assignee liability. Furthermore the certificate should be structured so
that there are penalties for violations regardless of whether or not
the investor or the borrower has experienced financial loss. The record
of violations of these origination responsibilities should publically
available.
---------------------------------------------------------------------------
\6\ http://www.ad-co.com/newsletters/2008/Feb2008/Credit_Feb08.pdf
and ``Securitization: After the Fall'', Anthony Sanders and Andrew
Davidson, forthcoming.
---------------------------------------------------------------------------
I have constructed a simple model of monitoring fraudulent loans.
\7\ Some preliminary results are shown in Table 1. These simulations
show the impact of increasing the required capital for a seller and of
instituting a fine for fraudulent loans beyond the losses incurred.
These results show that under the model assumptions, without a fine for
fraud, sellers benefit from originating fraudulent loans. The best
results are obtained when the seller faces fines for fraud and has
sufficient capital to pay those fines. The table below shows the
profitability of the seller and buyer for various levels of fraudulent
loans. In the example below, the profits of the seller increase from
.75 with no fraudulent loans to .77 with 10 percent fraudulent loans,
even when the originator retains 5 percent capital against 5 percent of
the credit risk. On the other hand, the sellers profit falls from .75
to .44 with 10 percent fraudulent loans even though the retained
capital is only 1 percent, but there is a penalty for fraudulent loans.
Thus the use of appropriate incentives can reduce capital costs, while
increasing loan quality.
---------------------------------------------------------------------------
\7\ The IMF has produced a similar analysis and reached similar
conclusions. http://www.imf.org/external/pubs/ft/gfsr/2009/02/pdf/
chap2.pdf.
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
Under this analysis the Treasury proposals would not have a direct
effect on fraud. In fact, there is substantial risk the recommended
approach of requiring minimum capital requirements for originators to
bear credit risk would lead to either higher mortgage rates or
increased risk taking. A better solution is to create new mechanisms to
monitor and enforce the representations and warranties of originators.
With adequate disclosure of risks and a workable mechanism for
enforcing quality controls the securitization market can more
effectively price and manage risk.
Recommendation 3 addresses the information available to investors:
3. The SEC should continue its efforts to increase the
transparency and standardization of securitization markets and
be given clear authority to require robust reporting by issuers
of asset backed securities (ABS).
Increased transparency and standardization of securitization
markets would likely to better functioning markets. In this area,
Treasury charges the SEC and ``industry'' with these goals. I believe
there needs to be consideration of a variety of institutional
structures to achieve these goals. Standardization of the market can
come from many sources. Possible candidates include the SEC, the
American Securitization Forum, the Rating Agencies and the GSEs, Fannie
Mae and Freddie Mac.
I believe the best institutions to standardize a market are those
which have an economic interest in standardization and disclosure. Of
all of these entities the GSEs have the best record of standardizing
the market; this was especially true before their retained portfolios
grew to dominate their income. (As I will discuss below, reform of the
GSEs is essential for restoring securitization.) I believe a revived
Fannie Mae and Freddie Mac, limited primarily to securitization,
structured as member-owned cooperatives, could be an important force
for standardization and disclosure.
While the other candidates could achieve this goal they each face
significant obstacles.
The SEC operates primarily through regulation and therefore may not
be able to adapt to changing markets. While the ASF has made
substantial strides in this direction, the ASF lacks enforcement power
for its recommendations and has conflicting constituencies. The rating
agencies have not shown the will or the power to force standardization,
and such a role may be incompatible with their stated independence.
Recommendations 4 and 5 address the role of rating agencies in
securitization.
4. The SEC should continue its efforts to strengthen the
regulation of credit rating agencies, including measures to
require that firms have robust policies and procedures that
manage and disclose conflicts of interest, differentiate
between structured and other products, and otherwise promote
the integrity of the ratings process.
5. Regulators should reduce their use of credit ratings in
regulations and supervisory practices, wherever possible.
In general I believe that the conflicts of interest facing rating
agencies and their rating criteria were well known and easily
discovered prior to the financial crisis. Thus I do not believe that
greater regulatory authority over rating agencies will offer
substantial benefits. In fact, increasing competition in ratings or
altering the compensation structure of rating agencies may not serve to
increase the accuracy of ratings, since most users of ratings issuers
as well as investors are generally motivated to seek higher ratings.
(Only if the regulatory reliance on rating agencies is reduced will
these structural changes be effective.) To the extent there is reliance
on rating agencies in the determination of the capital requirement for
financial institutions, a safety and soundness regulators for financial
institutions, such the FFIEC or its successor, should have regulatory
authority over the rating agencies.
Rather than focus on better regulation, I support the second aspect
of Treasury's recommendations on rating agencies (recommendation 5) and
believe it would be better for safety and soundness regulators to
reduce their reliance on ratings and allow the rating agencies to
continue their role of providing credit opinions that can be used to
supplement credit analysis performed by investors. To reduce reliance
on ratings, regulators, and others will need alternative measures of
credit and other risks. I believe that the appropriate alternative to
ratings is analytical measures of risk. Analytical measures can be
adopted, refined, and reviewed by regulators. In addition regulators
should insist that regulated entities have sufficient internal capacity
to assess the credit and other risks of their investments. In this way
regulators would have greater focus on model assumptions and model
validation and reduced dependence on the judgment of rating agencies.
The use of quantitative risk measures also requires that investors and
regulators have access to sufficient information about investments to
perform the necessary computations. Opaque investments that depend
entirely upon rating agency opinions would be clearly identified.
Quantitative measures can also be used to address the concerns raised
in the report about concentrations of risk and differentiate structured
products and direct corporate obligations.
I recently filed a letter with the National Association of
Insurance Commissioners on the American Council of Life Insurers'
proposal to use an expected loss measure as an alternative to ratings
for nonagency MBS in determining risk based capital. Here I would like
to present some of the key points in that letter:
An analytical measure may be defined as a number, or a value,
that is computed based on characteristics of a specific bond,
its collateral and a variety of economic factors both
historical and prospective. One such analytical measure is the
probability of default and another measure is the expected loss
of that bond. While an analytical measure is a numeric value
that is the result of computations, it should be noted that
there may still be some judgmental factors that go into its
production. In contrast, a rating is a letter grade, or other
scale, assigned to a bond by a rating agency. While ratings
have various attributes, generally having both objective and
subjective inputs, there is not a particular mathematical
definition of a rating.
Analytical measures may be useful for use by regulators because
they have several characteristics not present in ratings.
1. An analytical measure can be designed for a specific purpose.
Specific analytical measures can be designed with particular
policy or risk management goals in mind. Ratings may reflect a
variety of considerations. For example, there is some
uncertainty as to whether ratings represent the first dollar of
loss or the expected loss, or how expected loss is reflected in
ratings.
2. Analytical measures can be updated at any frequency. Ratings are
updated only when the rating agencies believe there has been
sufficient change to justify an upgrade, downgrade or watch.
Analytical measures can be computed any time new information is
available and will show the drift in credit quality even if a
bond remains within the same rating range.
3. Analytical measures can take into account price or other investor
specific information. Ratings are computed for a bond and
generally reflect the risk of nonpayment of contractual cash
flows. However, the risk to a particular investor of owning a
bond will at least partially depend on the price that the bond
is carried in the portfolio or the composition of the
portfolio.
4. Regulators may contract directly with vendors to produce
analytical results and may choose the timing of the
calculations. On the other hand, ratings are generally
purchased by the issuer at the time of issuance. Not only may
this introduce conflicts of interest, but it also creates a
greater focus on initial ratings than on surveillance and
updating of ratings. In addition, once a regulator allows the
use of a particular rating agency it has no further involvement
in the ratings process.
5. Analytical measures based on fundamental data may also be
advantageous over purely market-based measures. As market
conditions evolve values of bonds may change. These changes
reflect economic fundamentals, but may also reflect supply/
demand dynamics, liquidity and risk preferences. Measures fully
dependent on market prices may create excessive volatility in
regulatory measures, especially for companies with the ability
to hold bonds to maturity.
Even if regulators use analytical measures of risk, ratings from
rating agencies as independent opinions would still be valuable to
investors and regulators due to the multifaceted nature of ratings and
rating agency analysis can be used to validate the approaches and
assumptions used to compute particular analytical measures.
Additional measures beyond the credit risk of individual securities
such as stress tests, market value sensitivity and measures of
illiquidity may also be appropriate in the regulatory structure. The
use of analytical measures rather than ratings does not eliminate the
potential for mistakes. In general, any rigid system can be gamed as
financial innovation can often stay ahead of regulation. To reduce this
problem regulation should be based on principles and evolve with the
market. Regulators should always seek to build an a margin of safety as
there is always a risk that the theory underlying the regulatory regime
falls short and that some participants will find mechanisms to take
advantage of the regulatory structure.
Finally, as discussed by the Administration in the white paper, the
future of securitization for mortgages requires the resolution of the
status of Fannie/Freddie and role of FHA/GNMA. As stated above, I
believe that continuation of Fannie Mae and Freddie Mac as member owned
cooperatives would serve to establish standards, and provide a vehicle
for the delivery of Government guarantees if so desired. The TBA, or to
be announced, market has been an important component in the success of
the fixed rate mortgage market in the United States. Careful
consideration should be given to the desirability of fixed rate
mortgages and the mechanisms for maintaining that market in discussions
of the future of the GSEs.
______
PREPARED STATEMENT OF J. CHRISTOPHER HOEFFEL
Executive Committee Member, Commercial Mortgage Securities Association
October 7, 2009
The Commercial Mortgage Securities Association (CMSA) is grateful
to Chairman Reed, Ranking Member Bunning, and the Members of the
Subcommittee for giving CMSA the opportunity to share its perspective
concerning the securitized credit markets for commercial real estate.
In responding to the specific questions the Subcommittee has asked
witnesses to address, we will focus on securitization in the commercial
real estate (CRE) mortgage context and address the following issues:
(1) the challenges facing the $3.5 trillion market for commercial real
estate finance; (2) the unique structure of the commercial market and
the need to customize regulatory reforms accordingly to support, and
not undermine, our Nation's economic recovery; and, (3) efforts to
restore the availability of credit by promoting and enhancing the
viability of commercial mortgage-backed securities (CMBS).
CMSA and the Current State of the Market
CMSA represents the full range of CMBS market participants,
including investment and commercial banks; rating agencies; accounting
firms, servicers; other service providers; and investors such as
insurance companies, pension funds, and money managers. CMSA is a
leader in the development of standardized practices and in ensuring
transparency in the commercial real estate capital market finance
industry.
Because our membership consists of all constituencies across the
entire market, CMSA has been able to develop comprehensive responses to
policy questions to promote increased market efficiency and investor
confidence. For example, our members continue to work closely with
policymakers in Congress, the Administration, and financial regulators,
providing practical advice on measures designed to restore liquidity
and facilitate lending in the commercial mortgage market (such as the
Term Asset-Backed Securities Loan Facility (TALF) and the Public-
Private Investment Program (PPIP)). CMSA also actively participates in
the public policy debates that impact the commercial real estate
capital markets.
The CMBS market is a responsible and key contributor to the overall
economy that historically has provided a tremendous source of capital
and liquidity to meet the needs of commercial real estate borrowers.
CMBS helps support the commercial real estate markets that fuel our
country's economic growth. The loans that are financed through those
markets help provide jobs and services to local communities, as well as
housing for millions of Americans in multifamily dwellings.
Unfortunately, the recent turmoil in the financial markets coupled
with the overall downturn in the U.S. economy have brought the CMBS
market to a standstill and created many pressing challenges,
specifically:
No liquidity or lending--While the CMBS market provided
approximately $240 billion in commercial real estate financing
in 2007 (nearly 50 percent of all commercial lending), CMBS
issuance fell to $12 billion in 2008, despite strong credit
performance and high borrower demand. There has been no new
private label CMBS issuance year-to-date in 2009, as the
lending markets remain frozen;
Significant loan maturities through 2010--At the same time,
there are significant commercial real estate loan maturities
this year and next--amounting to hundreds of billions of
dollars--but the capital necessary to refinance these loans
remains largely unavailable and loan extensions are difficult
to achieve; and
The U.S. economic downturn persists--The U.S. recession
continues to negatively affect both consumer and business
confidence, which impacts commercial and multifamily occupancy
rates and rental income, as well as business performance and
property values.
Significantly, it is important to note that the difficulties faced
by the overall CRE market are not attributable solely to the current
trouble in the CMBS market, but also stem from problems with unsecured
CRE debt, such as construction loans. As described by Richard Parkus,
an independent research analyst with Deutsche Bank who has testified
before both the Joint Economic Committee and the TARP Oversight Panel,
while the overall CRE market will experience serious strain (driven by
poor consumer confidence and business performance, high unemployment
and property depreciation), it is the nonsecuritized debt on the books
of small and regional banks that will be most problematic, as the
projected default rates for such unsecuritized commercial debt have
been, and are expected to continue to be, significantly higher than
CMBS loan default rates.
As recently as early this year, default rates in the CMBS market,
which have historically been low (less than .50 percent for several
years) still hovered around a mere 1.25 percent. Unfortunately, the
economic recession that began as a crisis of liquidity in some sectors
transformed into a crisis in confidence that affected all sectors, and
it was only a matter of time before CMBS was affected. No matter the
strength of our fundamentals and loan performance, once investors lost
confidence and began to shy away from mortgaged-backed securities, CMBS
could not avoid the contagion.
This unfortunate combination of circumstances leaves the broader
CRE sector and the CMBS market with several overarching problems: (1) a
liquidity gap, i.e., the difference between borrowers' demand for
credit and the nearly nonexistent supply of credit; (2) an equity gap
(the difference between the current market value of commercial
properties and what is owed on them, which will be extremely difficult
to refinance as current loans mature); and (3) the fact that potential
CMBS sponsors are very reluctant to take the risk of trying to
aggregate loans for securitization, since there is no assurance that
private sector investors will buy the securities, all of which serves
to simply perpetuate the cycle of frozen credit markets.
Unique Characteristics of the CMBS Market
There are a number of important distinctions between CMBS and other
asset-backed securities (ABS) markets, and those distinctions should be
considered in fashioning any broad securitization-related regulatory
reforms. These differences relate not only to the structure of
securities, but also to the underlying collateral, the type and
sophistication of the borrowers, as well as to the level of
transparency in CMBS deals.
Commercial Borrowers
Commercial borrowers are highly sophisticated businesses with cash
flows based on business operations and/or tenants under leases. This
characteristic stands in stark contrast to the residential market
where, for example, loans were underwritten in the subprime category
for borrowers who may not have been able to document their income, or
who may not have understood the effects of factors like floating
interest rates and balloon payments on their mortgage's affordability.
Additionally, securitized commercial mortgages have different terms
(generally 5-10 year ``balloon'' loans), and they are, in the vast
majority of cases, nonrecourse loans. This means that if the borrower
defaults, the lender can seize the collateral, although it may not
pursue a claim against the borrower for any deficiency in recovery.
This dramatically decreases the cost of default because the loan work-
out recoveries in the CMBS context tend to be significantly more
efficient than, for example, the residential loan foreclosure process.
Structure of CMBS
There are multiple levels of review and diligence concerning the
collateral underlying CMBS, which help ensure that investors have a
well informed, thorough understanding of the risks involved.
Specifically, in-depth property-level disclosure and review are done by
credit rating agencies as part of the process of rating CMBS bonds.
Moreover, nonstatistical analysis is performed on CMBS pools. This
review is possible given that there are only 100-300 commercial loans
in a pool that support a bond, as opposed, for example, to tens of
thousands of loans in residential mortgage-backed securities pools.
This limited number of loans allows market participants (investors,
rating agencies, etc.) to gather detailed information about income
producing properties and the integrity of their cash flows, the credit
quality of tenants, and the experience and integrity of the borrower
and its sponsors, and thus conduct independent and extensive due
diligence on the underlying collateral supporting their CMBS
investments.
First-Loss Investor (``B-Piece Buyer'') Re-Underwrites Risk
CMBS bond issuances include a first-loss, noninvestment grade bond
component. The third-party investors that purchase these lowest-rated
securities (referred to as ``B-piece'' or ``first-loss'' investors)
conduct their own extensive due diligence (usually including, for
example, site visits to every property that collateralizes a loan in
the loan pool) and essentially re-underwrite all of the loans in the
proposed pool. Because of this, the B-piece buyers often negotiate the
removal of any loans they consider to be unsatisfactory from a credit
perspective, and specifically negotiate with bond sponsors or
originators to purchase this noninvestment-grade risk component of the
bond offering. This third-party investor due diligence and negotiation
occurs on every deal before the investment-grade bonds are issued.
Greater Transparency
A wealth of transparency currently is provided to CMBS market
participants via the CMSA Investor Reporting Package (CMSA IRP). The
CMSA IRP provides access to loan, property and bond-level information
at issuance and while securities are outstanding, including updated
bond balances, amount of interest and principal received, and bond
ratings, as well as loan-level and property-level information on an
ongoing basis. The ``CMSA IRP'' has been so successful in the
commercial space that it is now serving as a model for the residential
mortgage-backed securities market.
Current Efforts To Restore Liquidity
Private investors are absolutely critical to restoring credit
availability in the capital finance markets. Accordingly, Government
initiatives and reforms must work to encourage private investors--who
bring their own capital to the table--to come back to the capital
markets.
Treasury Secretary Geithner emphasized this need when he stressed
during the introduction of the Administration's Financial Stability
Plan that ``[b]ecause this vital source of lending has frozen up, no
financial recovery plan will be successful unless it helps restart
securitization markets for sound loans made to consumers and
businesses--large and small.'' The importance of restoring the
securitization markets is recognized globally as well, with the
International Monetary Fund noting in its most recent Global Financial
Stability Report that ``restarting private-label securitization
markets, especially in the United States, is critical to limiting the
fallout from the credit crisis and to the withdrawal of central bank
and Government interventions.'' \1\
---------------------------------------------------------------------------
\1\ International Monetary Fund, ``Restarting Securitization
Markets: Policy Proposals and Pitfalls'', Ch. 2, Global Financial
Stability Report: Navigating the Financial Challenges Ahead (October
2009), at 33 (``Conclusions and Policy Recommendations'' section)
available at http://www.imf.org/external/pubs/ft/gfsr/2009/02/pdf/
text.pdf.
---------------------------------------------------------------------------
As a centerpiece of the Financial Stability Plan, policymakers hope
to restart the CMBS and other securitization markets through innovative
initiatives (such TALF and the PPIP), and CMSA welcomes efforts to
utilize private investors to help fuel private lending. In this regard,
the TALF program for new CMBS issuance has been particularly helpful in
our space, as evidenced in triple-A CMBS cash spreads tightening almost
immediately after the program was announced, as one example.
To this end, CMSA continues to engage in an ongoing dialogue with
many members of the relevant Congressional committees, as well as with
key policymakers at the Treasury Department, Federal Reserve and other
agencies, and with participants in various sectors of the commercial
real estate market. The focus of our efforts has been on creative
solutions to help bring liquidity back to the commercial real estate
finance markets. We appreciate policymakers' recognition, as evidenced
by programs like TALF and PPIP, that a major part of the solution will
be to bring private investors back to the market through
securitization. We also appreciate the willingness of Congress and
other policymakers to listen to our recommendations on how to make
these programs as effective as possible.
However, there is still a long way to go toward recovery in the CRE
market, despite the early success of the TALF program. The market faces
the overarching problems of the liquidity and equity gaps. This is
driven in part by the absence of any aggregation mechanism--
securitizers are unwilling to bear all of the noncredit risks (like
interest rate changes) they must currently take on between the time a
loan is made and when it can be securitized (a process that takes
months across a pool of loans). This is especially true now when there
still is uncertainty as to whether there will be willing investors at
the end of the process.
CMSA also is committed to working on additional long-term solutions
to ensure the market is able to meet ongoing commercial borrowing
demands. For example, CMSA supports efforts to facilitate a U.S.
commercial covered bond market in order to provide an additional source
of liquidity through new and diverse funding sources. We will continue
to work with Congress on the introduction of comprehensive legislation
that would include high quality commercial mortgage loans and CMBS as
eligible collateral in the emerging covered bond marketplace.
Financial Regulatory Reform and Commercial Real Estate
The Administration has proposed new and unprecedented financial
regulatory reform proposals that will change the nature of the
securitized credit markets which are at the heart of recovery efforts.
The securitization reform proposals appear to be prompted by some of
the practices that were typical in the subprime and residential
securitization markets. At the outset, we must note that CMSA does not
oppose efforts to address such issues, as we have long been an advocate
within the industry for enhanced transparency and sound practices.
As a general matter, however, policymakers must ensure that any
regulatory reforms are tailored to address the specific needs of each
securitization asset class. As discussed above, the structure of the
CMBS market has incorporated safeguards that minimize the risky
securitization practices that policymakers hope to address. Thus, the
securitization reform initiatives should be tailored to take these
differences into account. In doing so, policymakers can protect the
viability of the markets that already are functioning in a way that
does not pose a threat to overall economic stability, and ensure that
such markets can continue to be a vital component of the economic
recovery solution.
CMSA and its members are concerned that certain aspects of the
Administration's securitization reform proposals could undermine rather
than support the Administration's many innovative efforts to restart
the securitization markets, effectively stalling recovery efforts by
making lenders less willing or able to extend loans and investors less
willing or able to buy CMBS bonds--two critical components to the flow
of credit in the commercial market.
The two aspects of the securitization reform proposal that are of
utmost concern to CMSA are a plan to require bond issuers or
underwriters (referred to as ``securitizers'' in the Administration's
draft securitization reform bill) to retain at least 5 percent of the
credit risk in any securitized asset they sell, and an associated
restriction on the ability of issuers to hedge the 5 percent retained
risk. Again, CMSA does not oppose these measures per se, but emphasizes
that they should be tailored to reflect key differences between the
different asset-backed securities markets.
Significantly, we are not alone in advocating a tailored approach.
The IMF, which recently expressed concern that U.S. and European
retained risk proposals may be too simplistic, warned that
``[p]roposals for retention requirements should not be imposed
uniformly across the board, but tailored to the type of securitization
and underlying assets to ensure that those forms of securitization that
already benefit from skin in the game and operate well are not
weakened. The effects induced by interaction with other regulations
will require careful consideration.''
Five Percent Risk Retention for Securitizers
The retention of risk is an important component regardless of who
ultimately retains it: the originator, the issuer, or the first-loss
buyer. As explained above, the CMBS structure has always had a third-
party in the first-loss position that specifically negotiates to
purchase this risk. Most significantly, these third-party investors are
able to, and do, protect their own interests in the long-term
performance of the bonds rather than relying merely on the underwriting
and representations of securitizers or originators. First-loss buyers
conduct their own extensive credit analysis on the loans, examining
detailed information concerning every property--before buying the
highest risk bonds in a CMBS securitization. In many cases, the holder
of the first-loss bonds is also related to the special servicer who is
responsible on behalf of all bondholders as a collective group for
managing and resolving defaulted loans through workouts or foreclosure.
Thus, the policy rationale for imposing a risk retention
requirement on issuers or underwriters as ``securitizers'' that could
preclude them from transferring the first-loss position to third
parties is unnecessary in this context, because, although the risk is
transferred, it is transferred to a party that is acting as a
``securitizer'' and that is fully cognizant, through its own diligence,
of the scope and magnitude of the risk it is taking on. In effect, when
it comes to risk, the first-loss buyer is aware of everything the
issuer or underwriter is aware of.
Because the CMBS market is structured differently than other
securitization markets, policymakers' focus in this market should be on
the proper transfer of risk (e.g., sufficient collateral disclosure,
adequate due diligence and/or risk assessment procedures on the part of
the risk purchaser), analogous to what takes place in CMBS
transactions. Therefore, CMBS securitizers should be permitted to
transfer risk to B-piece buyers who--in the CMBS context at least--act
as ``securitizers.'' To require otherwise would hamper the ability of
CMBS lenders to originate new bond issuances, by needlessly tying up
their capital and resources in the retained risk, which in turn, would
squelch the flow of credit at a time when our economy desperately needs
it.
CMSA therefore suggests that securitization legislation include a
broader definition of ``securitizer'' than is presently in the
Administration's draft bill, to include third parties akin to the CMBS
first-loss investors. Such an approach will provide explicit
recognition of the ability to transfer retained risk to third parties
under circumstances in which the third party agrees to retain the risk
and is capable of adequately protecting its own interests.
Prohibition on Hedging of Retained Risk
In conjunction with the retained risk requirement, there also is a
proposal to prohibit ``securitizers'' from hedging that risk. Rather
than adopting an outright ban on hedging the retained risk, however,
the measure needs to be designed to strike a balance between fulfilling
the legislation's objective of ensuring that securitizers maintain an
appropriate stake in the risks they underwrite. Such tailoring is
necessary to avoid imposing undue constraints on ``protective''
mechanisms that are legitimately used by securitizers to maintain their
financial stability.
Several risks inherent in any mortgage or security exposure arise
not from imprudent loan origination and underwriting practices, but
from outside factors such as changes in interest rates, a sharp
downturn in economic activity, or regional/geographic events such as a
terrorist attack or weather-related disaster. Securitizers attempt to
hedge against these market-oriented factors in keeping with current
safety and soundness practices, and some examples in this category of
hedges are interest rate hedges using Treasury securities, relative
spread hedges (using generic interest-rate swaps), and macroeconomic
hedges (that, for example, are correlated with changes in GDP or other
macroeconomic factors). The hallmark of this category is that these
hedges seek protection from factors the securitizer does not control,
and the hedging has neither the purpose nor the effect of shielding the
originators or sponsors from credit exposures on individual loans.
As such, hedges relate to generally uncontrollable market forces
that cannot be controlled independently. There is no way to ensure that
any such hedge protects 100 percent of an investment from loss--
particularly as it pertains to a CMBS transaction that, for example, is
secured by a diverse pool of loans with exposure to different
geographic locations, industries and property types. Therefore, loan
securitizers that must satisfy a retention requirement continue to
carry significant credit risk exposure that reinforces the economic tie
between the securitizer and the issued CMBS even in the absence of any
hedging constraints.
For these reasons, securitization reform legislation should not
seek to prohibit securitizers from using market-oriented hedging
vehicles. Instead, if a limitation is to be placed on the ability to
hedge, it should be targeted to prohibit hedging any individual credit
risks within the pool of risks underlying the securitization. Because
these types of vehicles effectively allow the originator or issuer to
completely shift the risk of default with respect to a particular loan
or security, their use could provide a disincentive to engage in
prudent underwriting practices--the specific type of disincentive
policymakers want to address.
Retroactive Changes to Securitization Accounting
Beyond the specific securitization reform proposals that have been
circulated by the Administration in draft legislation, there are two
other policy initiatives that greatly concern CMSA because of the
adverse effect these initiatives can have on the securitization market:
retroactive changes to the rules for securitization accounting, and
differentiated credit rating symbols for structured finance products.
Retroactive changes to securitization accounting rules known as FAS
166 and 167, which were recently adopted by the Financial Accounting
Standards Board (FASB), throw into question the future of securitized
credit markets. \2\ The new rules eliminate Qualified Special Purposes
Entities (QSPEs), which are the primary securitization accounting
vehicle for all asset-backed securities including CMBS, as well as
change the criteria for the sales treatment and consolidation of
financial assets. These accounting standards are important to issuers
and investors, and for the liquidity of capital markets as a whole,
because they free up balance sheet capacity to enable issuers to make
more loans and do more securitizations, and they enable investors to
invest more of their capital into the market. Under the new rules,
however, issuers may not receive sales accounting treatment, while
investors may be forced to consolidate an entire pool of loans on their
balance sheet, despite owning only a small fraction of the loans pool.
---------------------------------------------------------------------------
\2\ More specifically, these two standards provide accounting
guidance on when a sale of a financial instrument has occurred and how
to account for the sale, and guidance on when a securities issuer, B-
piece buyer or servicer needs to consolidate the securities and
liabilities on its balance sheet. The current rules facilitate
securitization by allowing issuers to receive ``sales treatment'' for
the assets they securitize, such that these assets are reflected on the
balance sheet of the investors that purchase the bonds, rather than the
issuers' balance sheet. Moreover, under present rules, investors
reflect only the fraction of the securitization deal that they actually
own on their balance sheet.
---------------------------------------------------------------------------
The implementation date of FAS 166 and 167 is January 1, 2010, and
it will be applied retroactively. The elimination of QSPEs therefore
will impact trillions of dollars of outstanding asset-backed
securities, including investors in these assets. These significant and
retroactive changes will pose a serious threat to unlocking the frozen
credit markets and another impediment to the Administration's wide-
ranging efforts to restart the securitized credit markets. CMSA and a
diverse coalition of 15 trade groups have raised concerns about the
timing and scope of FAS 166 and 167 given the impact these rule changes
could have on credit availability. These concerns have been echoed by
the Federal Reserve and other banking regulators, which wrote to FASB
in December 2008 to highlight the adverse impact these rule changes
could have on the credit markets.
More recently, Federal Reserve Board Member Elizabeth Duke
capsulized the concerns shared by the industry when she cautioned that:
[i]f the risk retention requirements, combined with accounting
standards governing the treatment of off-balance-sheet
entities, make it impossible for firms to reduce the balance
sheet through securitization and if, at the same time, leverage
ratios limit balance sheet growth, we could be faced with
substantially less credit availability. I'm not arguing with
the accounting standards or the regulatory direction. I am just
saying they must be coordinated to avoid potentially limiting
the free flow of credit . . . . As policymakers and others work
to create a new framework for securitization, we need to be
mindful of falling into the trap of letting either the
accounting or regulatory capital drive us to the wrong model.
This may mean we have to revisit the accounting or regulatory
capital in order to achieve our objectives for a viable
securitization market. \3\
---------------------------------------------------------------------------
\3\ ``Regulatory Perspectives on the Changing Accounting
Landscape'', Speech by Governor Elizabeth A. Duke at the AICPA National
Conference on Banks and Savings Institutions, Washington, DC, September
14, 2009, available at http://www.federalreserve.gov/newsevents/speech/
duke20090914a.htm.
Policymakers and standard setters, including FASB and the SEC, need
to proceed cautiously and deliberately in this regard, so that
accounting rule changes do not hamper the recovery of the
securitization markets.
Credit Rating Agency Reform
One aspect of the reforms currently being considered for credit
rating agencies is a previously rejected proposal to require credit
ratings to be differentiated for certain types of structured financial
products (requiring the use of ``symbology,'' such as ``AAA.SF'').
Generally speaking, ``differentiation'' is an overly simplistic and
broad proposal that provides little value or information about credit
ratings. Thus, CMSA's members, and specifically the investors the
symbology is geared to inform, continue to have serious concerns about
differentiation, although we are strong supporters of more effective
means of strengthening the credit ratings system in order to provide
investors with the information they need to make sound investment
decisions.
In fact, a broad coalition of market participants--including
issuers, investors, and borrowers seeking access to credit--remain
overwhelming opposed to differentiation because it will serve only to
increase confusion and implementation costs, while decreasing
confidence and certainty regarding ratings. Such effects would, in
turn, create market volatility and undermine investor confidence and
liquidity, which could exacerbate the current constraints on borrowers'
access to capital, at a time when other policymakers are employing
every reasonable means to get credit flowing again.
In this regard, it is worth noting that the concept of
differentiation has been examined extensively and rejected in recent
years by the House Committee on Financial Services, as well as by the
SEC and the ratings agencies themselves, \4\ for most (if not all) of
the foregoing reasons. Nothing has changed in the interim.
---------------------------------------------------------------------------
\4\ In early 2008, the CRAs sought feedback on various
differentiation proposals, which elicited overwhelming opposition from
investors. For example, see the results of Moody's Request for Comment:
``Should Moody's Consider Differentiating Structured Finance and
Corporate Ratings?'' (May 2008). Moody's received more than 200
responses, including ones from investors that together held in excess
of $9 trillion in fixed income securities.
---------------------------------------------------------------------------
Accordingly, Congress should not include a differentiation
requirement as part of any credit rating agency reform bill, but
instead should include language consistent with that already passed
last year by the House Committee on Financial Services in the Municipal
Bond Fairness Act. That legislation would require CRAs to use ratings
symbols that are consistent for all types of securities, recognizing
the fact that a single and consistent ratings structure is critical to
bond investors who want the ability to compare a multitude of
investment options across asset classes. Ultimately, investors (who are
critical to the Nation's economic recovery) expect and demand a common
rating structure to provide a meaningful foundation for our markets and
ratings system. Such consistency will promote certainty and confidence
among investors and all market participants.
In terms of credit ratings performance CMSA devoted significant
resources over the last few years to affirmatively enhance transparency
in credit ratings. Such enhancements will be far more effective in
providing investors with the information they need to make the most
informed decisions than a differentiated ratings structure. Instead of
differentiated ratings, what CMBS investors have consistently sought is
new, targeted transparency and disclosures about the ratings of
structured products, to build on the already robust information CRAs
provide in their published methodology, presale reports, and
surveillance press releases.
In comments filed with the SEC in July 2008, CMSA listed a number
of recommendations for enhancements that would serve the investor
community, such as publication of more specific information regarding
NRSRO policies and procedures related to CMBS valuations; adoption of a
standard presale report template with specified information regarding
methodology and underwriting assumptions; and adoption of a standard
surveillance press release with specified information regarding the
ratings. Such information would allow investors to better understand
the rating methodology and make their own investment determinations.
Fundamentally, CMSA believes that one of the keys to long term
viability is market transparency. As previously mentioned transparency
is one of the hallmarks of our market, as exemplified by the
unqualified success of our Investor Reporting Package. As we endeavor
to continually update our reporting package and provide additional
standardized information to market participants, one of our most
important proactive initiatives is the ongoing process of creating
model offering documents and providing additional disclosure fields
with regard to additional subordinate debt that may exist outside the
CMBS trust. To this end, CMSA is working with the Federal Reserve Board
to ensure the expanded disclosure meets their information needs under
TALF.
Conclusion
There are enormous challenges facing the commercial real estate
sector. While regulatory reforms are important and warranted, these
proposals should not detract from or undermine efforts to get credit
flowing, which is critical to economic recovery. Moreover, any policies
that make debt or equity interests in commercial real estate less
liquid will have a further negative effect on property values and the
cost of capital. Accordingly, we urge Congress to ensure that
regulatory reform measures are tailored to account for key differences
in the various securitization markets.
______
PREPARED STATEMENT OF WILLIAM W. IRVING
Portfolio Manager, Fidelity Investments
October 7, 2009
Good afternoon Chairman Reed, Ranking Member Bunning, and Members
of the Subcommittee. I am Bill Irving, an employee of Fidelity
Investments, \1\ where I manage a number of fixed-income portfolios and
play a leading role in our investment process in residential mortgage-
backed securities (RMBS). This experience has certainly shaped my
perspective on the role of securitization in the financial crisis, the
condition of the securitization markets today, and policy changes
needed going forward. I thank you for the opportunity to share that
perspective with you in this hearing. At the outset, I want to
emphasize that the views I will be expressing are my own, and do not
necessarily represent the views of my employer, Fidelity Investments.
---------------------------------------------------------------------------
\1\ Fidelity Investments is one of the world's largest providers
of financial services, with assets under Administration of $3.0
trillion, including assets under management of more than $1.4 trillion
as of August 31, 2009. Fidelity offers investment management,
retirement planning, brokerage, and human resources and benefits
outsourcing services to over 20 million individuals and institutions as
well as through 5,000 financial intermediary firms. The firm is the
largest mutual fund company in the United States, the number one
provider of workplace retirement savings plans, the largest mutual fund
supermarket and a leading online brokerage firm. For more information
about Fidelity Investments, visit Fidelity.com.
---------------------------------------------------------------------------
Summary
I will make three main points. First, the securitized markets
provide an important mechanism for bringing together investors and
borrowers to provide credit to the American people for the financing of
residential property, automobiles, and retail purchases. Securitization
also provides a major source of funding for American businesses for
commercial property, agricultural equipment, and small-business
investment. My second point is that the rapid growth of the markets led
to some poor securitization practices. For example, loan underwriting
standards got too loose as the interests of issuers and investors
became misaligned. Furthermore, liquidity was hindered by a
proliferation of securities that were excessively complex and
customized. My third and final point is that in spite of these
demonstrated problems, the concept of asset securitization is not
inherently flawed; with proper reforms to prevent weak practices, we
can harness the full potential of the securitization markets to benefit
the U.S. economy.
Brief Review of the Financial Crisis
To set context, I will begin with a brief review of the financial
crisis. This view is necessarily retrospective; I do not mean to imply
that investors, financial institutions or regulators understood all
these dynamics at the time. In the middle of 2007, the end of the U.S.
housing boom revealed serious deficiencies in the underwriting of many
recently originated mortgages, including subprime loans, limited-
documentation loans, and loans with exotic features like negative
amortization. Many of these loans had been packaged into complex and
opaque mortgage-backed securities (MBS) that were distributed around
the world to investors, some of whom relied heavily on the opinion of
the rating agencies and did not sufficiently appreciate the risks to
which they were exposed. \2\
---------------------------------------------------------------------------
\2\ At Fidelity, we consider the opinions of the rating agencies,
but we also do independent credit research on each issuer or security
we purchase.
---------------------------------------------------------------------------
The problems of poorly understood risks in these complex securities
were amplified by the leverage in the financial system. For example, in
2007, large U.S. investment banks had about $16 of net assets for each
dollar of capital. \3\ Thus, a seemingly innocuous hiccup in the
mortgage market in August 2007 had ripple effects that quickly led to a
radical reassessment of what is an acceptable amount of leverage. What
investors once deemed safe levels of capital and liquidity were
suddenly considered far too thin. As a result, assets had to be sold to
reduce leverage. This selling shrank the supply of new credit and
raised borrowing costs. In fact, the selling of complex securities was
more than the market could bear, resulting in joint problems of
liquidity and solvency. Suddenly, a problem that had started on Wall
Street spread to Main Street. Companies that were shut off from credit
had to cancel investments, lay off employees and/or hoard cash. Many
individuals who were delinquent on their mortgage could no longer sell
their property at a gain or refinance; instead, they had to seek loan
modifications or default.
---------------------------------------------------------------------------
\3\ Source: SNL Financial, and company financials.
---------------------------------------------------------------------------
This de-leveraging process created a vicious cycle. Inability to
borrow created more defaults, which led to lower asset values, which
caused more insolvency, which caused more de-leveraging, and so forth.
Home foreclosures and credit-card delinquencies rose, and job layoffs
increased, helping to create the worst recession since the Great
Depression.
Role Played by Asset Securitization in the Crisis
Without a doubt, securitization played a role in this crisis. Most
importantly, the ``originate-to-distribute'' model of credit provision
seemed to spiral out of control. Under this model, intermediaries found
a way to lend money profitably without worrying if the loans were paid
back. The loan originator, the warehouse facilitator, the security
designer, the credit rater, and the marketing and product-placement
professionals all received a fee for their part in helping to create
and distribute the securities. These fees were generally linked to the
size of the transaction and most of them were paid up front. So long as
there were willing buyers, this situation created enormous incentive to
originate mortgage loans solely for the purpose of realizing that up-
front intermediation profit.
Common sense would suggest that securitized assets will perform
better when originators, such as mortgage brokers and bankers, have an
incentive to undertake careful underwriting. A recent study by the
Federal Reserve Bank of Philadelphia supports this conjecture. \4\ The
study found evidence that for prime mortgages, private-label
securitized loans have worse credit performance than loans retained in
bank portfolios. Specifically, the study found that for loans
originated in 2006, the 2-year default rate on the securitized loans
was on average 15 percent higher than on loans retained in bank
portfolios. This observation does not necessarily mean that issuers
should be required to retain a portion of their securities, but in some
fashion, the interests of the issuers and the investors have to be kept
aligned.
---------------------------------------------------------------------------
\4\ Elul, Ronel, ``Working Paper No. 09-21 Securitization and
Mortgage Default: Reputation vs. Adverse Selection'', Federal Reserve
Bank of Philadelphia. September 22, 2009.
---------------------------------------------------------------------------
Flawed security design also played a role in the crisis. In its
simplest form, securitization involves two basic steps. First, many
individual loans are bundled together into a reference pool. Second,
the pool is cut up into a collection of securities, each having a
distinct bundle of risks, including interest-rate risk, prepayment
risk, and credit risk. For example, in a simple sequential structure,
the most senior bond receives all available principal payments until it
is retired; only then does the second most senior bond begin to receive
principal; and so on. In the early days of securitization, the process
was kept simple, and there were fewer problems. But over time, cash-
flow rules grew increasingly complex and additional structuring was
employed. For example, the securities from many simple structures were
rebundled into a new reference pool, which could then be cut into a new
set of securities. In theory, there is no limit to the amount of
customization that is possible. The result was excessive complexity and
customization. The complexity increased the challenge of determining
relative value among securities, and the nonuniformity hurt liquidity
when the financial system was stressed.
One example of poor RMBS design is the proliferation of securities
with complex rules on the allocation of principal between the senior
and subordinate bonds. Such rules can lead to counter-intuitive
outcomes in which senior bonds take write-downs while certain
subordinate bonds are paid off in full. A second example of poor design
is borrower ability to take out a second-lien mortgage without
notifying the first-lien holder. This ability leads to a variety of
thorny issues, one of which is simply the credit analysis of the
borrower. If a corporation levered further, the senior unsecured debt
holder would surely be notified, but that is not so in RMBS.
Other Factors Contributing to the Crisis
Securitization of assets played a role in the crisis, but there
were several additional drivers. Low interest rates and a bubble
mentality in the real estate market also contributed to the problem.
Furthermore, in the case of securitized assets, there were plenty of
willing buyers, many of them highly levered. In hindsight, this high
demand put investors in the position of competing with each other,
making it difficult for any of them to demand better underwriting, more
disclosure, simpler product structures, or other favorable terms.
Under-estimation of risk is always a possibility in capital markets, as
the history of the stock market amply demonstrates. That possibility
does not mean that capital markets, or asset securitization, should be
discarded.
Benefits of Asset Securitization
When executed properly, there are many potential benefits of
allowing financial intermediaries to sell the loans they originate into
the broader capital markets via the securitization process. For one,
this process provides loan originators much wider sources of funding
than they could obtain through conventional sources like retail
deposits. For example, I manage the Fidelity Ginnie Mae Fund, which has
doubled in size in the past year to over $7 billion in assets; the MBS
market effectively brings together shareholders in this Ginnie Mae Fund
with individuals all over the country who want to purchase a home or
refinance a mortgage. In this manner, securitization breaks down
geographic barriers between lenders and borrowers, thereby improving
the availability and cost of credit across regions.
A second benefit of securitization is it generally provides term
financing which matches assets against liabilities; this stands in
contrast to the bank model, a substantial mismatch can exist between
short-term retail deposits and long-term loans. Third, it expands the
availability of credit across the country's socio-economic spectrum,
and provides a mechanism through which higher credit risks can be
mitigated with structural enhancements. Finally, it fosters competition
among capital providers to ensure more efficient pricing of credit to
borrowers.
Current Conditions of Consumer ABS and Residential MBS Markets
At present, the RMBS and ABS markets are sharply bifurcated. On one
side are the sectors that have received Government support, including
consumer ABS and Agency MBS (i.e., MBS guaranteed by Fannie Mae,
Freddie Mac, and Ginnie Mae); these sectors are, for the most part,
functioning well. On the other side are the sectors that have received
little or no such support, such as the new-issue private-label RMBS
market, which remains stressed, resulting in a lack of fresh mortgage
capital for a large segment of the housing market.
Consumer ABS
The overall size of the consumer debt market is approximately $2.5
trillion; \5\ this total includes both revolving debt (i.e., credit-
card loans) and nonrevolving debt (e.g., auto and student loans).
Approximately 75 percent takes the form of loans on balance sheets of
financial institutions, while the other 25 percent has been
securitized. \6\
---------------------------------------------------------------------------
\5\ Source: Federal Reserve, www.federalreserve.gov/releases/g19/
current/g19.htm.
\6\ Source: Federal Reserve, www.federalreserve.gov/releases/g19/
current/g19.htm.
---------------------------------------------------------------------------
From 2005 through the third quarter of 2008, auto and credit card
ABS issuance ranged between $160 billion and $180 billion per year. \7\
However, after the collapse of Lehman Brothers in September 2008, new
issuance came to a virtual halt. With the ABS market effectively shut
down, lenders tightened credit standards to where only the most credit
worthy borrowers had access to credit. As a result, the average
interest rate on new-car loans provided by finance companies increased
from 3.28 percent at end of July 2008 to 8.42 percent by the end of
2008. \8\
---------------------------------------------------------------------------
\7\ Source: Bloomberg.
\8\ Federal Reserve, www.federalreserve.gov/releases/g19/hist/
cc_hist_tc.html.
---------------------------------------------------------------------------
Issuance did not resume until March 2009 when the Term Asset-Backed
Securities Loan Facility (TALF) program began. Thanks to TALF, between
March and September of this year, there has been $91 billion of card
and auto ABS issuance. \9\ Coincident with the resumption of a
functioning auto ABS market, the new-car financing rate fell back into
the 3 percent range and consumer access to auto credit has improved,
although credit conditions are still more restrictive than prior to the
crisis. While TALF successfully encouraged the funding of substantial
volumes of credit card receivables in the ABS market, it is worth
noting that credit card ABS issuance has recently been suspended due to
market uncertainty regarding the future regulatory treatment of the
sector.
---------------------------------------------------------------------------
\9\ Source: Bloomberg.
---------------------------------------------------------------------------
While interest rates on top tier New Issue ABS are no longer
attractive for investors to utilize the TALF program, TALF is still
serving a constructive role by allowing more difficult asset types to
be financed through securitization. Examples include auto dealer
floorplans, equipment loans to small businesses, retail credit cards,
nonprime auto loans, and so forth.
Residential MBS
The overall size of the residential mortgage market is
approximately $10.5 trillion, which can be decomposed into three main
categories:
1. Loans on bank balance sheets: \10\ $3.5 trillion.
---------------------------------------------------------------------------
\10\ Source: Federal Reserve, www.federalreserve.gov/econresdata/
releases/mortoutstand/current.htm.
2. Agency MBS: \11\ $5.2 trillion.
---------------------------------------------------------------------------
\11\ Source: eMBS, www.embs.com.
---------------------------------------------------------------------------
a. Fannie Mae: $2.7 trillion.
b. Freddie Mac: $1.8 trillion.
c. Ginnie Mae: $0.7 trillion.
3. Private-Label MBS: \12\ $1.9 trillion.
---------------------------------------------------------------------------
\12\ Source: Loan Performance.
---------------------------------------------------------------------------
a. Prime: $0.6 trillion.
b. Alt-A: $0.8 trillion.
c. Subprime: $0.5 trillion.
Thanks to the extraordinary Government intervention over the past
year, the Agency MBS market is performing very well. This intervention
had two crucial components. First, on September 7, 2008, the director
of the Federal Housing Finance Agency (FHFA) placed Fannie Mae and
Freddie Mac into conservatorship. This action helped reassure tens of
thousands of investors in Agency unsecured debt and mortgage-backed
securities that their investments were supported by the Federal
Government, in spite of the sharp declines in home prices across the
country. The second component of the Government intervention was the
Federal Reserve's pledge to purchase $1.25 trillion of Agency MBS by
the end of 2009.
Year to date, as of the end of September 2009, the Fed had
purchased $905 billion Agency MBS, while net supply was only $448
billion. \13\ Thus, the Fed has purchased roughly 200 percent of the
year-to-date net supply. Naturally, this purchase program has reduced
the spread between the yields on Agency MBS and Treasuries; we estimate
the reduction to be roughly 50 basis points. As of this week, the
conforming-balance \14\ 30-year fixed mortgage rate is approximately
4.85 percent, which is very close to a generational low. \15\
---------------------------------------------------------------------------
\13\ Source: JPMorgan, ``Fact Sheet: Federal Reserve Agency
Mortgage-Backed Securities Purchase Program''.
\14\ As of 2009, for the contiguous States, the District of
Columbia and Puerto Rico, the general conforming limit is $417,000; for
high-cost areas, it can be as high as $729,500.
\15\ Source: HSH Associates, Financial Publishers.
---------------------------------------------------------------------------
In contrast, the new-issue private-label MBS market has received no
Government support and is effectively shut down. From 2001 to 2006,
issuance in this market had increased almost four-fold from $269
billion to $1,206 billion. \16\ But when the financial crisis hit, the
issuance quickly fell to zero. Issuance in 2007, 2008 and 2009 has been
$759 billion, $44 billion and $0, respectively. \17\ Virtually the only
source of financing for mortgage above the conforming-loan limit (so-
called ``Jumbo loans'') is a bank loan. As a result, for borrowers with
high-credit quality, the Jumbo mortgage rate is about 1 percentage
point higher than its conforming counterpart. \18\
---------------------------------------------------------------------------
\16\ Source: Loan Performance.
\17\ Source: Loan Performance.
\18\ Source: HSH Associates, Financial Publishers.
---------------------------------------------------------------------------
At first glance, the higher cost of Jumbo financing may not seem to
be an issue that should concern policymakers, but what is bad for this
part of the mortgage market may have implications for other sectors. If
the cost of Jumbo financing puts downward pressure on the price of
homes costing (say) $800,000, then quite likely there will be downward
pressure on the price of homes costing $700,000, and so forth. Pretty
soon, there is downward pressure on homes priced below the conforming
limit. In my opinion, at the same time that policymakers deliberate the
future of the Fannie Mae and Freddie Mac, they should consider the
future of the mortgage financing in all price and credit-quality tiers.
Recommended Legislative and Regulatory Changes
The breakdown in the securitization process can be traced to four
root causes: aggressive underwriting, overly complex securities,
excessive leverage, and an over-reliance on the rating agencies by some
investors. Such flaws in the process have contributed to the current
financial crisis. However, when executed properly, securitization can
be a very effective mechanism to channel capital into our economy to
benefit the consumer and commercial sectors. Keep in mind that
securitization began with the agency mortgage market, which has
successfully provided affordable mortgage financing to millions of U.S.
citizens for over 35 years. \19\ To ensure that the lapses of the
recent past are not repeated, I recommend that regulatory and
legislative efforts be concentrated in four key areas.
---------------------------------------------------------------------------
\19\ Fannie Mae, Freddie Mac, and Ginnie Mae issued their first
MBS in 1981, 1971, and 1970, respectively. Source: ``Fannie Mae and
Freddie Mac: Analysis of Options for Revising the Housing Enterprises
Long-term Structures'', GAO Report to Congressional Committees,
September, 2009.
---------------------------------------------------------------------------
First, promote improved disclosure to investors at the initial
marketing of transactions as well as during the life of the deal. For
example, originators should provide detailed disclosure on the
collateral characteristics and on exceptions to stated underwriting
procedures. Furthermore, there should be ample time before a deal is
priced for investors to review and analyze a full prospectus, not just
a term sheet.
Second, strong credit underwriting standards are needed in the
origination process. One way to support this goal is to discourage the
up-front realization of issuers' profits. Instead, issuers'
compensation should be aligned with the performance of the security
over its full life. This issue is complex, and will likely require
specialized rules, tailored to each market sector.
Third, facilitate greater transparency of the methodology and
assumptions used by the rating agencies to determine credit ratings. In
particular, there should be public disclosure of the main assumptions
behind rating methodologies and models. Furthermore, when those models
change or errors are discovered, the market should be notified.
Fourth, support simpler, more uniform capital structures in
securitization deals. This goal may not readily be amenable to
legislative action, but should be a focus of industry best practices.
Taking such steps to correct the defects of recent securitization
practices will restore much-needed confidence to this critical part of
our capital markets, thereby providing improved liquidity and capital
to foster continued growth in the U.S. economy.
Additional Material Supplied for the Record
Prepared Statement of the Mortgage Bankers Association
The Mortgage Bankers Association (MBA) \1\ appreciates the
opportunity to provide this statement for the record of the Senate
Banking Securities, Insurance, and Investment Subcommittee hearing on
the securitization of assets.
---------------------------------------------------------------------------
\1\ The Mortgage Bankers Association (MBA) is the national
association representing the real estate finance industry, an industry
that employs more than 280,000 people in virtually every community in
the country. Headquartered in Washington, DC, the association works to
ensure the continued strength of the Nation's residential and
commercial real estate markets; to expand homeownership and extend
access to affordable housing to all Americans. MBA promotes fair and
ethical lending practices and fosters professional excellence among
real estate finance employees through a wide range of educational
programs and a variety of publications. Its membership of over 2,400
companies includes all elements of real estate finance: mortgage
companies, mortgage brokers, commercial banks, thrifts, Wall Street
conduits, life insurance companies and others in the mortgage lending
field. For additional information, visit MBA's Web site:
www.mortgagebankers.org.
---------------------------------------------------------------------------
Asset-backed securities are a fundamental component of the
financial services system because they enable consumers and businesses
to access funding, organize capital for new investment opportunities,
and protect and hedge against risks. As policymakers evaluate
securitization's role in the recent housing finance system's
disruptions, MBA believes it is important to keep in mind the benefits
associated with securitization when it is used prudently by market
participants.
Securitization describes the process in which relatively illiquid
assets are packaged in a way that removes them from the institution's
balance sheet and sold as more liquid securities. Securities backed by
residential or commercial mortgages are an example of asset
securitization.
Securitization is an effective means of risk management for many
institutions. For example, the accumulation of many loans in a single
asset sector creates concentration risk on a financial institution's
balance sheet. If that sector becomes distressed, these large
concentrations could place the solvency of the financial institution at
risk. However, securitization provides a remedy to avoid concentration
risk by disbursing the exposure more widely across the portfolios of
many investors. In this way, the exposure of any one investor is
minimized. As demonstrated by the current business cycle however, if
the entire system is hit by a significant systemic shock, all investors
will face losses from these exposures, as diversification does not
protect investors from systemic events.
Securitization also enables various market sectors to create
synergies by combining their particular areas of expertise. For
example, community-based financial institutions are known for their
proficiency in originating loans because of their relationships with
local businesses and consumers, and their knowledge of local economic
conditions. Securitization links these financial institutions to others
that may be more adept at matching asset risks with investor appetites.
As the last 2 or 3 years have demonstrated, when it is not
understood, or poorly underwritten, securitization can cause meaningful
harm to investors, lenders, borrowers and other segments of the
financial services system. Since the economic and housing finance
crisis began, investors have shunned securitization products, including
mortgage-backed securities (MBS), particularly those issued by private
entities. As a result, central banks and governments have taken up the
slack with various programs to support securitization markets. MBA
believes this has been an important, yet ultimately unsustainable,
course of action.
One key to the process is to create an environment where investors
can accurately evaluate the risks in the various investment
opportunities available to them, and have confidence that their
analysis of the risk is consistent with what the underlying risk will
turn out to be. No investments are risk-free. But reliable instruments
allow responsible investors to evaluate whether the instrument's risk
profile is within the boundaries of an investor's risk tolerance.
When considering how to reestablish a safe and sound environment
for securitization of real estate-related assets, MBA believes the
following components must be addressed:
Risk Assessment: Risk assessment is an imperfect science,
but it is crucial for securitization to enable accurate,
effective, and stable risk assessment. Equally important,
third-party assessments of risk must be highly credible to be
widely used or adopted.
Aligning Risks, Rewards, and Penalties: A key consideration
for the market going forward will be ensuring the alignment of
risks with rewards and penalties. Loan attributes, such as
whether a loan is adjustable-rate or fixed rate, or does or
does not have a prepayment restriction, shift risks between the
borrower and the investors. If investors or other market
participants are not accountable for the risks they take on,
they are prone to act irresponsibly by taking on greater risks
than they otherwise would.
Aligning Rewards With Long-Term Performance: Given the
long-term nature of a mortgage contract, as well as the
imperfect state of risk assessment, some risks inherent in a
mortgage asset may not appear for some time after the asset has
changed hands. It is important to consider the degree to which
participants in the mortgage process can be held accountable
for the long-term performance of an asset.
Ensuring Capital Adequacy of Participants: Participants
throughout the market need adequate levels of capital to
protect against losses. Capital adequacy is keenly dependent on
the assessment of risks outlined above. The greater the risks,
as assessed, the greater the capital needed. In times of rapid
market deterioration, when model and risk assumptions change
dramatically, capital needs may change dramatically as well. If
market participants that have taken on certain risks become
undercapitalized, they may not be able to absorb those risks
when necessary--forcing others to take on unanticipated risks
and losses.
Controlling Fraud Between Parties in the System: A key
consideration for effective securitization is the degree to
which fraud can be minimized. Key considerations include the
ability to identify and prosecute fraud, and the degree to
which fraud is deterred.
Transparency: In order to attract investors, another key
consideration for securitization is transparency. The less
transparent a market is, the more poorly understood it will be
by investors, and the higher will be the yield those investors
demand to compensate for the uncertainty.
The task of improving transparency and accountability involves both
policy and operational issues. Public debate typically focuses on the
policy issues--what general types of information should be disclosed,
and who should share and receive this information. However, the
operational issues are equally important to establishing and
implementing a functional system that promotes and supports the goals
of transparency and accountability. We are submitting testimony today
to stress the importance to market transparency and investor confidence
of better loan tracking and more accessible, complete, and reliable
loan and security data across the primary and secondary mortgage
markets.
Loan and Security Tracking
Improving transparency in the real estate finance system is
considered essential to restoring investor confidence in the
securitization market. Because the real estate finance system embraces
multiple parties--loan originators, loan aggregators (servicers) and
securitizers--we need transparency solutions that flow from and span
the complete mortgage value chain.
The goal, we think, is relatively easy to state: key information
about mortgages, the securities built upon those mortgages, and the
people and companies that create them, should all be linked and tracked
over time, so our financial system is more transparent and the
strengths and risks of various products can be properly assessed and
appreciated. Loans need to be tracked, for example, to help identify
fraud and distinguish the performance of various mortgage products and
securities types.
Just as the vehicle identification number, or ``VIN,'' has evolved
from a simple serial number into a valuable tool for consumers,
enabling a potential purchaser to research the history of any car or
truck, a comprehensive mortgage/security numbering system would be the
key to tracking MBS history and performance.
Achieving such a goal is very doable because the essential
components are already in place. With relatively minor modifications
these existing systems can evolve into the tools necessary to meet the
challenge of transparency and accountability.
On the mortgage end of the value chain there is MERS. \2\ This
national loan registry is already used by virtually all mortgage
originators, aggregators, and securitizers to track individual
mortgages by means of a unique, 18-digit Mortgage Identification
Number, or ``MIN.'' For each registered mortgage, the MIN and the MERS
database tracks information regarding the originator, the borrower, the
property, the loan servicer, the investors, and any changes of
ownership for the life of the loan. MERS currently tracks more than 60
million loans and is embedded in every major loan origination system,
servicing system, and delivery system in the United States, so total
adoption would be swift and inexpensive.
---------------------------------------------------------------------------
\2\ ``MERS'' is formally known as MERSCORP, Inc., and is the owner
and operator of the MERS System. MBA, along with Fannie Mae, Freddie
Mac, and other industry participants, is a shareholder in MERS.
---------------------------------------------------------------------------
On the securitization end of the value chain, the American Bankers
Association has a product called CUSIP that generates a 9-digit
identification number for most types of securities, including MBS. The
CUSIP number uniquely identifies the company or issuer and the type of
security instrument.
Together, these two identifiers solve the loan and security
tracking problem, with the MIN tracking millions of individual mortgage
loans and the CUSIP tracking thousands of unique financial instruments
created each year in the United States. Loan-level information for
every mortgage and mortgage-backed security would be available at the
touch of a button, for example, the credit rating agencies would have
needed information to assess more accurately the risk of a given
security and track its performance relative to other securities over
time.
As the Congress looks to reform the capital markets, it should
require that these two complementary identification systems be linked
and that they be expanded in scope to track the decisions of all market
participants--originators, aggregators and securitizers. In this way,
throughout the value chain, participants that contributed to the
creation of high-risk mortgages and selling of high-risk securities may
be identified and held accountable.
With a system like this in place, the Congress, regulators and the
market as a whole would have a means of distinguishing with much more
precision the quality of financial products and could enforce the
discipline that has not been previously possible.
Data Standards
The Mortgage Industry Standards Maintenance Organization, Inc.
(``MISMO'') \3\ has been engaged for the past 8 years in developing
electronic data standards for the commercial and residential real
estate finance industries. These standards, which have been developed
through a structured consensus-building process, are grounded in the
following principles that we believe characterize a robust, transparent
system of data reporting:
---------------------------------------------------------------------------
\3\ MISMO is a wholly owned subsidiary of the Mortgage Bankers
Association.
First, there must be concrete definitions of the data
elements that are going to be collected, and these definitions
must be common across all the related products in the market.
Different products (such as conforming and nonconforming loans)
may require different data elements, but any data elements that
are required for both products should have the same
---------------------------------------------------------------------------
definitions.
Second, there should be a standardized electronic reporting
format by which these data elements are shared across the
mortgage and security value chain and with investors. The
standards should be designed so that information can freely
flow across operating systems and programs with a minimum of
reformatting or rekeying of data to facilitate desired
analytics. Rekeying results in errors, undermining the
reliability of data. MISMO's standards are written in the XML
(Web based) computer language. This is the language used in the
relaunch earlier this week of the Federal Register's Web site.
As reported in The Washington Post on October 5, 2009, this Web
site has been received with great praise for allowing
researchers and other users to extract information readily from
the Register for further analysis and reuse without rekeying.
Mortgage and securities data transmitted using MISMO's data
standards can similarly be extracted and used by investors and
regulators for customized analytics. XML is also related to and
compatible with the XBRL web language that the Securities and
Exchange Commission (SEC) is implementing for financial
reporting.
Third, the definitions and the standards should be
nonproprietary and available on a royalty-free basis, so that
third-parties can easily access and incorporate those standards
into their work, whether it be in the form of a new loan
origination software package or an improved analytical tool for
assessing loan and security performance or fraud detection.
Fourth, to the extent that the data includes nonpublic
personal information, the system must maintain the highest
degree of confidentiality and protect the privacy of that
information.
True transparency requires that information is not only available,
but also understandable and usable. The incorporation of these four
principles into any new data reporting regime will help ensure that the
goal of transparency and accountability is realized.
We believe that the standards of MISMO and MERS satisfy these
elements for the conforming mortgage market. Their relative positions
in the real estate finance process provide them with unique insight and
an objective perspective that we believe could be very useful to
improving transparency and accountability in the nonconforming market.
Increasing the quality and transparency of loan-level mortgage and
MBS-related data is an essential step so that investor confidence may
be restored and the risk of a similar securitization crisis of the kind
we are experiencing in the future can be minimized. This objective is
paramount to all market participants, and as such all participants have
an interest in achieving a solution. However, because it is so
critical, the ultimate solution must also be able to withstand the
scrutiny of investors, Government regulators, and academics. It must be
widely perceived as a fair, appropriate, and comprehensive response to
the challenges at hand.
In conclusion, MBA reiterates its request for Congress and other
policymakers to be mindful of the important role of securitization to
housing finance and the entire financial services system. As the
Congress looks to reform the capital markets, we look forward to
working with you to developing a framework with a solid foundation
based on the key considerations outlined above.