[JPRT, 111th Congress]
[From the U.S. Government Publishing Office]
CONGRESSIONAL OVERSIGHT PANEL
DECEMBER OVERSIGHT REPORT *
----------
A REVIEW OF TREASURY'S FORECLOSURE PREVENTION PROGRAMS
[GRAPHIC] [TIFF OMITTED] TONGRESS.#13
December 14, 2010.--Ordered to be printed
* Submitted under Section 125(b)(1) of Title 1 of the Emergency
Economic Stabilization Act of 2008, Pub. L. No. 110-343
CONGRESSIONAL OVERSIGHT PANEL DECEMBER OVERSIGHT REPORT
U.S. GOVERNMENT PRINTING OFFICE
62-622 WASHINGTON : 2010
-----------------------------------------------------------------------
For sale by the Superintendent of Documents, U.S. Government Printing Office,
http://bookstore.gpo.gov. For more information, contact the GPO Customer Contact Center, U.S. Government Printing Office. Phone 202�09512�091800, or 866�09512�091800 (toll-free). E-mail, [email protected].
CONGRESSIONAL OVERSIGHT PANEL
DECEMBER OVERSIGHT REPORT *
__________
A REVIEW OF TREASURY'S FORECLOSURE PREVENTION PROGRAMS
[GRAPHIC] [TIFF OMITTED] TONGRESS.#13
December 14, 2010.--Ordered to be printed
* Submitted under Section 125(b)(1) of Title 1 of the Emergency
Economic Stabilization Act of 2008, Pub. L. No. 110-343
CONGRESSIONAL OVERSIGHT PANEL
Panel Members
Sen. Ted Kaufman, Chairman
Richard H. Neiman
Damon Silvers
J. Mark McWatters
Kenneth Troske
C O N T E N T S
----------
Page
Executive Summary................................................ 1
Section One: Introduction........................................ 4
Part One: Where HAMP Stands Today................................ 5
A. Background................................................ 5
B. Pre-HAMP Foreclosure Mitigation Efforts................... 6
C. HAMP's Structure.......................................... 9
D. HAMP's Performance........................................ 12
1. Trial Modifications................................... 13
2. Conversion to Permanent Modification.................. 15
3. Permanent Modifications............................... 17
4. Borrowers Dropped from the Program.................... 21
5. Redefault............................................. 25
E. Performance of Non-HAMP Modifications..................... 26
F. Treasury's Other Foreclosure Mitigation Programs.......... 28
G. Barriers to Success....................................... 33
1. What are HAMP's Goals?................................ 33
2. Factors Affecting HAMP Success........................ 39
Part Two: The Future of HAMP..................................... 59
H. Treasury's Implementation of HAMP and Possible
Improvements............................................... 60
1. Role of Fannie Mae and Freddie Mac.................... 61
2. Current Oversight Mechanisms.......................... 69
3. Implementation Failures and Ways to Improve........... 70
I. Redefaults of Modified Mortgages.......................... 73
1. The Great Depression-Era Home Owners' Loan Corporation 75
2. Redefaults in Other Loan Modification Efforts......... 77
3. Current Trends in Loan Modifications and Redefaults... 78
4. HAMP Redefaults....................................... 80
Conclusions and Recommendations.................................. 90
Annex I: Lessons from the Home Owners' Loan Corporation of the
1930s and 1940s................................................ 95
A. Background................................................ 95
B. The HOLC's Operations..................................... 96
C. How the HOLC Compares to HAMP............................. 102
D. Lessons from the HOLC..................................... 104
Section Two: Additional Views.................................... 106
A. J. Mark McWatters and Professor Kenneth R. Troske......... 106
Section Three: Correspondence with Treasury Update............... 110
Section Four: TARP Updates Since Last Report..................... 111
Section Five: Oversight Activities............................... 138
Section Six: About the Congressional Oversight Panel............. 139
Appendices:
APPENDIX I: LETTER FROM SPECIAL MASTER PATRICIA GEOGHEGAN TO
CHAIRMAN TED KAUFMAN RE: FOLLOW UP TO EXECUTIVE
COMPENSATION HEARING, DATED NOVEMBER 18, 2010.............. 140
======================================================================
DECEMBER OVERSIGHT REPORT
_______
December 14, 2010.--Ordered to be printed
_______
EXECUTIVE SUMMARY *
In April 2010, in its most recent report on Treasury's
foreclosure prevention programs, the Panel raised serious
concerns about the timeliness, accountability, and
sustainability of Treasury's efforts. As the Panel noted at the
time, ``It now seems clear that Treasury's programs, even when
they are fully operational, will not reach the overwhelming
majority of homeowners in trouble . . . . Treasury is still
struggling to get its foreclosure programs off the ground as
the crisis continues unabated.''
---------------------------------------------------------------------------
* The Panel adopted this report with a 5-0 vote on December 13,
2010.
---------------------------------------------------------------------------
In the intervening eight months, Treasury has tweaked its
main foreclosure prevention effort, the Home Affordable
Modification Program (HAMP), but the changes have not resolved
the Panel's core concerns. The Panel now estimates that, if
current trends hold, HAMP will prevent only 700,000 to 800,000
foreclosures--far fewer than the 3 to 4 million foreclosures
that Treasury initially aimed to stop, and vastly fewer than
the 8 to 13 million foreclosures expected by 2012. Because
Treasury's authority to restructure HAMP ended on October 3,
2010, the program's prospects are unlikely to improve
substantially in the future.
In some regards, the program's failure to make a dent in
the foreclosure crisis may seem surprising. HAMP's premise was
straightforward: Because the foreclosure process allows lenders
to recover only a small fraction of the value of a mortgage
loan, lenders should generally prefer to avoid foreclosure by
voluntarily reducing a borrower's monthly payments to
affordable levels. Through HAMP, Treasury attempted to sweeten
this deal by offering incentive payments to all parties to a
mortgage loan modification. Yet despite the apparent strength
of HAMP's economic logic, the program has failed to help the
vast majority of homeowners facing foreclosure.
A major reason is that mortgages are, in practice, far more
complicated than a one-to-one relationship between borrower and
lender. In particular, banks typically hire loan servicers to
handle the day-to-day management of a mortgage loan, and the
servicer's interests may at times sharply conflict with those
of lenders and borrowers. For example, although lenders suffer
significant losses in foreclosures, servicers can turn a
substantial profit from foreclosure-related fees. As such, it
may be in the servicer's interest to move a delinquent loan to
foreclosure as soon as possible. HAMP attempted to correct this
market distortion by offering incentive payments to loan
servicers, but the effort appears to have fallen short, in part
because servicers were not required to participate. Another
major obstacle is that many borrowers have second mortgages
from lenders who may stand to profit by blocking the
modification of a first mortgage. For these reasons among many
others, HAMP's straightforward plan to encourage modifications
has proven ineffective in practice.
While HAMP's most dramatic shortcoming has been its poor
results in preventing foreclosures, the program has other
significant flaws. For example, despite repeated urgings from
the Panel, Treasury has failed to collect and analyze data that
would explain HAMP's shortcomings, and it does not even have a
way to collect data for many of HAMP's add-on programs.
Further, Treasury has refused to specify meaningful goals by
which to measure HAMP's progress, while the program's sole
initial goal--to prevent 3 to 4 million foreclosures--has been
repeatedly redefined and watered down. Treasury has also failed
to hold loan servicers accountable when they have repeatedly
lost borrower paperwork or refused to perform loan
modifications. Treasury has essentially outsourced the
responsibility for overseeing servicers to Fannie Mae and
Freddie Mac, but both companies have critical business
relationships with the very same servicers, calling into
question their willingness to conduct stringent oversight.
Freddie Mac in particular has hesitated to enforce some of its
contractual rights related to the foreclosure process, arguing
that doing so ``may negatively impact our relationships with
these seller/servicers, some of which are among our largest
sources of mortgage loans.'' Treasury bears the ultimate
responsibility for preventing such conflicts of interest, and
it should ensure that loan servicers are penalized when they
fail to complete loan modifications appropriately.
Many of the problems now plaguing HAMP are inherent in its
design and cannot be resolved at this late date. Other
problems, however, can still be mitigated. For instance,
Treasury should enable borrowers to apply for loan
modifications more easily--for example, by allowing online
applications. Treasury should also carefully examine HAMP's
track record to pin down the factors that define successful
loan modifications so that similar modifications can be
encouraged in the future.
Perhaps most critically, Treasury should carefully monitor
and, where appropriate, intervene in cases in which borrowers
are falling behind on their HAMP-modified mortgages. Preventing
redefaults is an extremely powerful way of magnifying HAMP's
impact, as each redefault prevented translates directly into a
borrower keeping his home. Delinquencies that are flagged in
their early stages can potentially be brought current through a
repayment plan, but delinquencies that are left unchecked have
the potential to undermine even the modest progress made by
HAMP. Worse still, each redefault represents thousands of
taxpayer dollars that have been spent merely to delay rather
than prevent a foreclosure.
Finally, Treasury should accept that HAMP will not reach
its original goals and provide a meaningful framework for
evaluating the program in the future. Treasury continues to
state that HAMP will expend $30 billion in Troubled Asset
Relief Program funding, yet the Congressional Budget Office
recently estimated that all of Treasury's foreclosure programs
combined will spend only $12 billion. Given the Panel's cost
estimates for Treasury's other foreclosure-related efforts,
HAMP thus appears likely to spend only around $4 billion. Had
Treasury acknowledged this reality before its crisis authority
expired, it could have made material changes to HAMP or
reallocated the money to a more effective program. Now, that
option is gone.
For this reason, Treasury's reluctance to acknowledge
HAMP's shortcomings has had real consequences. Absent a
dramatic and unexpected increase in HAMP enrollment, many
billions of dollars set aside for foreclosure mitigation may
well be left unused. As a result, an untold number of borrowers
may go without help all because Treasury failed to acknowledge
HAMP's shortcomings in time.
SECTION ONE
Introduction
The Emergency Economic Stabilization Act (EESA), the
October 2008 legislation that granted Treasury the authority to
create the Troubled Asset Relief Program (TARP), included a
mandate that TARP funds be used in a manner that ``protects
home values'' and ``preserves homeownership.'' \1\ To fulfill
that mandate, Treasury in 2009 allocated $50 billion in TARP
funds for a new mortgage modification program called the Home
Affordable Modification Program (HAMP).
---------------------------------------------------------------------------
\1\ 12 U.S.C. Sec. 5201(2)(A), (B). For a discussion of the
authority of the Secretary of the Treasury to use TARP funds to create
a program such as HAMP, see Congressional Oversight Panel, April
Oversight Report: Evaluating Progress on TARP Foreclosure Mitigation
Programs, at 147-171 (Apr. 14, 2010) (online at cop.senate.gov/
documents/cop-041410-report.pdf) (hereinafter ``April 2010 Oversight
Report'').
---------------------------------------------------------------------------
The same legislation established the Congressional
Oversight Panel, along with a specific charge to issue periodic
reports on TARP foreclosure mitigation efforts. The Panel's
first foreclosure mitigation oversight report was issued in
March 2009, concurrent with the announcement of HAMP. The
report established numerous standards for evaluating the
Administration's foreclosure mitigation program, including: (1)
whether it resulted in affordable monthly payments; (2) whether
it dealt with negative equity; (3) whether it addressed second
liens; and (4) whether it counteracted incentives for mortgage
servicers not to modify troubled loans. Seven months later, in
October 2009, the Panel examined the Administration's
implementation of HAMP. This report identified three main
concerns with the program: (1) that it lacked sufficient scope
to prevent many foreclosures, including those caused by
unemployment and negative equity; (2) that it was not achieving
scale quickly enough; and (3) that it was not providing a
permanent solution to homeowners who needed help.
The Panel again assessed HAMP in April 2010. Foreclosures
were continuing at a rapid pace, and the report found that
Treasury's response continued to lag the crisis. The Panel
articulated three major concerns with HAMP: (1) the failure of
the program to deal with the foreclosure crisis in a timely
way; (2) the unsustainable nature of many HAMP modifications,
given the large debt load and negative equity that many
participating homeowners continued to carry; and (3) the need
for greater accountability in HAMP, particularly with regard to
the activities of participating servicers.
Treasury's foreclosure mitigation programs have grown and
evolved since the initial announcement in March 2009. More
TARP-funded foreclosure prevention initiatives were announced
shortly before the release of the Panel's most recent
foreclosure report; those programs are discussed in Section B.4
of this report.\2\ However, this month's report focuses
primarily on HAMP, since it is Treasury's marquee foreclosure
prevention initiative, and because many of the other
initiatives remain in early stages, with no record of results
on which to be assessed.\3\
---------------------------------------------------------------------------
\2\ Treasury has since created two additional foreclosure
prevention programs under the TARP: the Hardest Hit Fund (HHF), which
provides foreclosure prevention funding to 19 states and the District
of Columbia, and the Federal Housing Administration (FHA) Short
Refinance Program, which will allow for the refinancing of certain
mortgages by the FHA.
\3\ For an analysis of the programs' structures, see April 2010
Oversight Report, supra note 1, at 8-29.
---------------------------------------------------------------------------
Part One: Where HAMP Stands Today
A. Background
Despite government and private sector efforts to modify
troubled loans and thus stop ``preventable'' foreclosures, the
number of foreclosures remains extremely high, with
approximately 250,000 foreclosure starts and over 100,000
foreclosure completions per month.
Figure 1 below shows the number of foreclosure starts and
completions each month.
FIGURE 1: FORECLOSURE STARTS AND COMPLETIONS BY MONTH (JULY 2007-
SEPTEMBER 2010) \4\
[GRAPHIC] [TIFF OMITTED] 62622A.001
---------------------------------------------------------------------------
\4\ HOPE NOW Alliance, Appendix--Mortgage Loss Mitigation
Statistics: Industry Extrapolations (Monthly for Dec. 2008 to Nov.
2009) (online at www.hopenow.com/industry-data/
HOPE%20NOW%20National%20Data%20July07%20to%20Nov09%20v2%20(2).pdf);
HOPE NOW Alliance, Industry Extrapolations and Metrics (May 2010)
(online at www.hopenow.com/industry-data/
HOPE%20NOW%20Data%20Report%20(May)%2006-21-2010.pdf); HOPE NOW
Alliance, Industry Extrapolations and Metrics (Sept. 2010) (online at
www.hopenow.com/industry-data/HOPE%20NOW%20Data%20
Report%20(September)%20101010%20v2.pdf) (hereinafter ``HOPE NOW
Alliance Industry Extrapolations and Metrics'').
Figure 2 below shows the percentage of all mortgages that
are in delinquency and the percentage of all mortgages that are
counted as foreclosure inventory, meaning they are somewhere in
the foreclosure process. Please note that the foreclosure
inventory is stacked on top of delinquency, that is,
delinquency is currently around 10 percent, and foreclosure
inventory is roughly 4 percent, not 14 percent. Although both
factors are at historically high levels, they have been
relatively steady for the past two years.
FIGURE 2: DELINQUENCY AND FORECLOSURE RATES (Q1 2006-Q3 2010) \5\
[GRAPHIC] [TIFF OMITTED] 62622A.002
---------------------------------------------------------------------------
\5\ Mortgage Bankers Association, National Delinquency Survey Q3
2010 (Nov. 18, 2010).
The Federal Reserve has recently estimated foreclosures
over the next two years: ``All told, we expect about two and
one-quarter million foreclosure filings [in 2010] and again
next year, and about two million more in 2012.'' \6\ The Center
for Responsible Lending has also released a foreclosure
forecast of 9 million foreclosures between 2009 and 2012.\7\
Since approximately 5 million foreclosures have been completed
since the beginning of 2009, this seems to be generally in line
with the Federal Reserve's prediction.
---------------------------------------------------------------------------
\6\ Sarah Bloom Raskin, member, Board of Governors of the Federal
Reserve System, Remarks at the National Consumer Law Center's Consumer
Rights Litigation Conference, Boston, Massachusetts, Problems in the
Mortgage Servicing Industry, at 2 (Nov. 12, 2010) (online at
www.federalreserve.gov/newsevents/speech/bloomraskin20101112a.pdf). See
also House Committee on Financial Services, Written Testimony of
Elizabeth A. Duke, member, Board of Governors of the Federal Reserve
System, Robo-Signing, Chain of Title, Loss Mitigation, and Other Issues
in Mortgage Servicing, at 5 (Nov. 18, 2010) (online at
financialservices.house.gov/Media/file/hearings/111/Duke111810.pdf)
(``Over the first half of this year, we have seen a further 1.2 million
foreclosure filings, and an additional 2.4 million homes were somewhere
in the foreclosure pipeline at the end of June. All told, we expect
about 2.25 million foreclosure filings this year and again next year,
and about 2 million more in 2012.'').
\7\ Center for Responsible Lending, Soaring Spillover (May 2009)
(online at www.responsiblelending.org/mortgage-lending/research-
analysis/soaring-spillover-3
9.pdf).
---------------------------------------------------------------------------
B. Pre-HAMP Foreclosure Mitigation Efforts
In the wake of the financial crisis of late 2008, Treasury
developed HAMP as the latest in a series of federal government
initiatives to stem the growing foreclosure problem. At the
time, foreclosures had been rising for several years already,
leading to an increase in the number of empty homes owned by
banks and weakening the banking system at a time when
policymakers felt that restoring economic and bank stability
was crucial. The history of prior efforts, which met with
limited success, provides a useful context for examining HAMP
and its performance to date.
As the housing boom peaked and began its long downward
slide in 2006, policymakers appeared to take for granted that
foreclosures would not require government intervention. In
previous housing recessions of the post-WWII era, foreclosures
and mortgage modifications had been left to the discretion of
private lenders and loan servicers.\8\ After all, modifications
generally maximize value for the lender or investor as compared
to foreclosure, so it was logical to assume that such
modifications would occur. Yet, by mid-2007, it appeared that
many foreclosures were proceeding even in instances where loan
modifications would appear to be economically preferable to the
lender or mortgage investors.
---------------------------------------------------------------------------
\8\ The Home Owners' Loan Corporation, a depression-era federal
government mortgage modification program, is discussed in Annex I.
---------------------------------------------------------------------------
Policymakers continued to look to the private sector for a
more aggressive response to the situation, but began to nudge
them toward a more organized response in hopes of achieving
greater results. As a result, in October 2007, the HOPE NOW
Alliance was formed as a voluntary coalition of mortgage
companies and industry organizations designed to centralize and
coordinate foreclosure mitigation efforts. Although both
Treasury and the Department of Housing and Urban Development
(HUD) were consulted and strongly promoted the effort, the
federal government is not an official sponsor.\9\ Initially,
HOPE NOW met with limited success. For instance, a 2009 study
found that only 49 percent of HOPE NOW workouts had reduced the
borrower's monthly payment, and 34 percent had actually
resulted in a higher monthly payment.\10\ However, HOPE NOW
recently reported that 91 percent of the nearly 150,000
modifications completed in August 2010 involved payment
reductions.\11\ As of August 2010, HOPE NOW participants report
a total of 10.7 million mortgage ``solutions'' since the
inception of the alliance, including 3.2 million proprietary
modifications. However, the Mortgage Metrics Report compiled by
the Office of Comptroller of the Currency and the Office of
Thrift Supervision reports only 1.3 million such
modifications.\12\ Proprietary modifications are discussed
further in Section E below.
---------------------------------------------------------------------------
\9\ See U.S. Department of the Treasury, Statement by Secretary
Henry M. Paulson, Jr. on Announcement of New Private Sector Alliance--
HOPE NOW (Oct. 10, 2007) (online at 205.168.45.71/press/releases/
hp599.htm).
\10\ Congressional Oversight Panel, March Oversight Report:
Foreclosure Crisis: Working Toward a Solution, at 31 (Mar. 6, 2009)
(online at cop.senate.gov/documents/cop-030609-report.pdf) (hereinafter
``March 2009 Oversight Report''). See also Sonia Garrison et al.,
Continued Decay and Shaky Repairs: The State of Subprime Loans Today,
Center for Responsible Lending Study, at 7 (Jan. 2009) (online at
www.responsiblelending.org/mortgage-lending/research-analysis/
continued_decay_and_shaky_repairs.pdf). This study reported only 20
percent of modifications resulted in lower payments.
\11\ HOPE NOW Alliance, HOPE NOW: Nine out of Ten Proprietary Loan
Mods in August Included Principal ` Interest Payment Reduction (Oct. 7,
2010) (online at www.hopenow.com/press_release/files/
August%202010%20Data%20Release_FINAL.pdf).
\12\ See Congressional Oversight Panel, Written Testimony of Joseph
H. Evers, deputy comptroller for large bank supervision, Office of the
Comptroller of the Currency, COP Hearing on TARP Foreclosure Mitigation
Programs, at 4, 7-8 (Oct. 27, 2010) (online at cop.senate.gov/
documents/testimony-102710-evers.pdf) (stating that loan servicers
modified 1,239,896 loans between the start of 2008 and the end of the
first quarter of 2010, including 121,731 HAMP modifications, and that
an additional 164,473 non-HAMP modifications were done in the second
quarter of 2010).
---------------------------------------------------------------------------
The first official federal government foreclosure
mitigation program was FHA Secure, announced in August 2007,
which refinanced adjustable-rate mortgages into fixed-rate
mortgages insured by the Federal Housing Administration (FHA).
FHA Secure permitted the refinancing of delinquent and
underwater borrowers, which was rare in the private sector.
However, delinquencies had to be attributable to the loan
resetting, and borrowers could not generally show any
delinquencies in the six-month period prior to the rate reset.
Borrowers participating in this program were therefore able to
refinance their existing underwater mortgages into safer loans
at a time when lenders were tightening underwriting standards
and underwater borrowers were unable to refinance in the
private market. As the Panel has noted previously, however,
this was accomplished at the cost of having the taxpayer insure
a large number of negative equity mortgages. FHA Secure was
closed down at the end of 2008. Although the program refinanced
nearly half a million loans, only 4,128 of these were
delinquent at the time of refinancing. The Panel has previously
attributed FHA Secures failure to its restrictive borrower
criteria.\13\
---------------------------------------------------------------------------
\13\ March 2009 Oversight Report, supra note 10, at 35. See also
Kate Berry, HUD Mulling How to Widen FHA Refi Net, American Banker
(Feb. 15, 2008) (online at www.americanbanker.com/issues/173_33/-
344173-1.html); Michael Corkery, Mortgage `Cram-Downs' Loom as
Foreclosures Mount, Wall Street Journal (Dec. 31, 2008) (online at
online.wsj.com/article/SB123068005350543971.html).
---------------------------------------------------------------------------
Following the lackluster results stemming from the private
sector initiatives, policymakers determined that a new
government program was the next appropriate step. Accordingly,
HOPE for Homeowners was established by Congress in July 2008 to
permit FHA insurance of refinanced distressed mortgages. While
less restrictive in some areas than FHA Secure, the program did
not guarantee negative equity loans. Since the goal of the
program was specifically to encourage principal reduction
modifications of negative equity loans, guaranteeing them as is
would have defeated the purpose, as well as likely been
impossible under FHA`s 97 percent LTV statutory limit.
Nonetheless, although HOPE for Homeowners was predicted to help
400,000 homeowners, it managed to refinance only a handful of
loans. This was likely due to the program`s poor initial
design, lack of flexibility, and its reliance on voluntary
principal write-downs, which lenders were very reluctant to do,
a pattern also seen in HAMP.\14\
---------------------------------------------------------------------------
\14\ See March 2009 Oversight Report, supra note 10, at 36. See
also Dina ElBoghdady, HUD Chief Calls Aid on Mortgages A Failure,
Washington Post (Dec. 17, 2008) (online at www.washingtonpost.com/wp-
dyn/content/article/2008/12/16/AR2008121603177.html).
---------------------------------------------------------------------------
In the same month HOPE for Homeowners was created, the
Federal Deposit Insurance Corporation (FDIC) took over IndyMac,
one of the largest subprime lenders. Soon afterwards, the FDIC
announced a loan modification program to assist the 65,000
delinquent borrowers with loans in IndyMac`s non-securitized
portfolio. Although no FDIC funds were allocated specifically
for these modifications, loss-sharing agreements were signed
with the purchasers of IndyMac`s assets. A number of other,
similar efforts were instituted with smaller failed lenders
taken over by the FDIC. The IndyMac program and other FDIC
foreclosure mitigation efforts had limited reach, but may have
influenced the structure of HAMP.\15\ This and other aspects of
the FDIC IndyMac program are discussed in more detail in
Section I.2.
---------------------------------------------------------------------------
\15\ See March 2009 Oversight Report, supra note 10, at 32-33;
Congressional Oversight Panel, October Oversight Report: An Assessment
of Foreclosure Mitigation Efforts After Six Months, at 83-84 (Oct. 9,
2009) (online at cop.senate.gov/documents/cop-100909-report.pdf)
(hereinafter ``October 2009 Oversight Report''). See also Charles
Duhigg, Fighting Foreclosures, F.D.I.C. Chief Draws Fire, New York
Times (Dec. 11, 2008) (online at www.nytimes.com/2008/12/11/business/
11bair.html).
---------------------------------------------------------------------------
Even with increasing government intervention throughout
this timeframe, foreclosures continued to surge. It became
clear that the private sector was either unable or unwilling to
conduct mortgage modifications on its own of a scope and scale
necessary to stem the tide. Most likely, this was due to
rational behavior on the part of servicers. As discussed below,
there are incentives built into the mortgage servicing system
that encourage servicers to prefer foreclosure in many cases,
and discourage certain types of modifications. HAMP, and
specifically its servicer incentive payments, were created to
overcome the additional costs that servicers incur in modifying
loans, and to compensate them in part for the income they may
forgo by choosing modification over foreclosure.
C. HAMP's Structure
HAMP is designed to provide a path to modification in those
cases in which modification is the economically preferable
outcome, from the perspective of the lender or investor who
owns the loan, to foreclosure. Because such modifications are
in the interest of both the borrower and the lender, it would
seem to follow that mortgage servicers should be providing
modifications in those cases without any payments from the
government. As observed by David Stevens, commissioner of the
Federal Housing Administration, ``To be frank, too often during
this crisis, that private sector engagement hasn't happened. In
some instances, we've seen market actors refuse to participate.
In others, we've seen them participate half-heartedly.'' \16\
The failure of the private sector to reduce debt service
payments on a substantial number of mortgages in part led to
Treasury's decision to create HAMP.
---------------------------------------------------------------------------
\16\ David H. Stevens, commissioner, Federal Housing
Administration, Remarks at the Mortgage Bankers Association Annual
Convention, at 7 (Oct. 26, 2010).
---------------------------------------------------------------------------
HAMP provides financial incentives to mortgage servicers to
modify mortgages for homeowners at risk of default, and
incentives for the beneficiaries of these modifications to stay
current on their mortgage payments going forward.\17\
Participation in the program by servicers is voluntary, but
once a servicer elects to participate, adherence to the program
standards is mandatory for all the servicer's loans. If a
participating servicer has a borrower who qualifies for HAMP,
the lender must first reduce monthly payments until they are no
more than 38 percent of the borrower's gross monthly income.
Treasury will then match, dollar for dollar, further reductions
required to bring the monthly payments down to 31 percent of
the borrower's income.
---------------------------------------------------------------------------
\17\ Servicers of government-sponsored enterprise (GSE) mortgages
are required to participate in HAMP for their GSE portfolio. Servicers
of non-GSE mortgages may elect to sign a servicer participation
agreement (SPA) in order to participate in the program. Once an
agreement has been signed, the participating servicer must evaluate all
mortgages under HAMP unless the participation contract is terminated.
---------------------------------------------------------------------------
Only borrowers whose mortgage servicers have opted into the
program may apply for assistance. Pooling and servicing
agreements (PSAs) for securitized mortgages may also limit the
latitude that servicers have to modify loans.\18\ Furthermore,
borrowers must meet the following criteria to be eligible:
---------------------------------------------------------------------------
\18\ The decision to modify securitized mortgages rests with the
servicer, and servicers are instructed to manage loans as if for their
own account and maximize the net present value of the loan.
Nevertheless, some PSAs contain additional restrictions that can hamper
servicers' ability to modify mortgages. Sometimes the modification is
forbidden outright, sometimes only interest rates can be adjusted, not
principal, and sometimes there are limitations on the amount by which
interest rates can be adjusted. Other times the total number of loans
that can be modified is capped (typically at 5 percent of the pool),
the number of times a loan may be modified will be capped, or the
number of modifications in a year will be capped. Generally, the term
of a loan cannot typically be extended beyond the last maturity date of
any loan in the securitized pool. Additionally, servicers are sometimes
required to purchase any loans they modify at the face value
outstanding (or even with a premium). This functions as an
antimodification provision. See March 2009 Oversight Report, supra note
10, at 42-44.
---------------------------------------------------------------------------
The home must be owner-occupied, not vacant,
and not condemned;
The remaining balance on a single unit home
must be no more than $729,750, with higher limits for
properties containing up to four units;
The borrower must be delinquent, or default
must be reasonably foreseeable, and the borrower must
demonstrate financial hardship, including the fact that
he or she has insufficient liquid assets to make the
required monthly payments; and
The borrower must have a monthly ``front-
end'' debt-to-income (DTI) ratio of more than 31
percent, meaning that the monthly mortgage payment must
be greater than 31 percent of the borrower's gross
monthly income.\19\
---------------------------------------------------------------------------
\19\ U.S. Department of the Treasury, Introduction of the Home
Affordable Modification Program, Supplemental Directive 09-01, at 6
(Apr. 6, 2009) (online at www.hmpadmin.com/portal/programs/docs/
hamp_servicer/sd0901.pdf) (hereinafter ``Introduction of the Home
Affordable Modification Program,''). ``Monthly payment'' means monthly
mortgage payment before modification, including both principal and
interest, plus applicable taxes, hazard insurance, flood insurance,
condominium association fees and homeowners' association fees, as
applicable. ``Front-end'' DTI refers to a ratio of the borrower's
monthly mortgage payment to their monthly income, as opposed to a
``back-end'' DTI which compares all of a borrowers debt service
payments (including credit cards, car payments, etc.) to their income.
---------------------------------------------------------------------------
If a borrower meets these criteria, the servicer must then
use Treasury's Net Present Value (NPV) model to determine
whether or not a HAMP modification makes economic sense from
the lender's perspective. The NPV model calculates net present
values for the expected income from the mortgage under a HAMP
modification, and the expected income with no modification
(generally a foreclosure and home sale scenario). The two
figures are then compared. If the mortgage has a greater value
under the HAMP modification, it is said to be ``NPV positive,''
in which case a participating servicer must offer the borrower
a HAMP modification.
HAMP prescribes a ``waterfall'' to determine what type of
modification should be offered. First, the servicer should
consider whether lowering the loan's interest rate, to as low
as 2 percent, would result in a monthly front-end DTI ratio of
less than 31 percent. If the ratio would still be too high, the
next step should be extending the loan period out to as long as
40 years. If the DTI ratio would still be greater than 31
percent, the final step is principal forbearance.\20\
---------------------------------------------------------------------------
\20\ As described in Section F, an alternate waterfall moves the
principal reduction option to earlier in the process.
---------------------------------------------------------------------------
Once approved for assistance through HAMP, a borrower must
successfully complete a trial period, typically three months,
during which the borrower makes payments on the modified
mortgage. A borrower who remains current through the trial
period becomes eligible for a permanent modification, under
which the terms of the trial modification remain in effect for
a period of five years.\21\ For each year that the borrower
remains current under the modified mortgage, he or she receives
a $1,000 incentive payment from HAMP, for up to five years.
After the five year term is up, the interest rate on the loan
can increase by a maximum of 1 percent per year until it
reaches the prevailing Freddie Mac average interest rate at the
time the HAMP modification was made. As of December 2, 2010,
Freddie Mac's average interest rate on a 30-year fixed rate
conforming mortgage is 4.46 percent. Incentive payments also
flow to the mortgage servicer and investor.\22\ For mortgages
that are not backed by government-sponsored entities Fannie Mae
or Freddie Mac (the GSEs), the funding comes from $29.9 billion
currently set aside from the TARP for foreclosure
mitigation.\23\ For GSE mortgage modifications, $25 billion has
been set aside from the Housing and Economic Relief Act of
2008.
---------------------------------------------------------------------------
\21\ ``Current'' means that no payment is more than 30 days
overdue. See Introduction of the Home Affordable Modification Program,
supra note 19, at 17-18.
\22\ The incentive payments for investors and borrowers are
included when calculating the net present value of a loan modification
under Treasury's NPV analysis. See Introduction of the Home Affordable
Modification Program, supra note 19, at 22-25.
\23\ The original funding amount allotted for HAMP was $50 billion.
In May 2009, the $1.2 billion reduction in TARP due to the passage of
the Helping Families Save Their Homes Act was officially allocated to
HAMP. See Helping Families Save Their Homes Act of 2009, Pub. L. No.
111-22 Sec. 402(f) (2009) (online at financialservices.house.gov/
FinancialSvcsDemMedia/file/public%20laws/111-22.pdf). The $50 billion
HAMP funding was later reduced to a ceiling of $30.6 billion by the
Dodd-Frank Wall Street Reform and Consumer Protection Act. To date
Treasury has expended less than $800 million on HAMP.
---------------------------------------------------------------------------
Under HAMP's Principal Reduction Alternative (PRA)
initiative, which became effective on October 1, 2010,
servicers must conduct an additional evaluation of borrowers
who meet the HAMP criteria and whose mortgages are
significantly underwater, resulting in a loan-to-value (LTV)
ratio, at current market prices, of more than 115 percent. If
the mortgage meets these criteria, the servicer must evaluate
it to determine, as with the original HAMP NPV test, whether
the NPV of a principal reduction under the program is greater
than the NPV under no modification. If so, the servicer has the
option to offer a principal reduction under the PRA. These
principal reductions are voluntary for servicers, unlike
interest payment reductions which are a mandatory part of HAMP.
A servicer who elects to offer a principal reduction first
reduces the principal until the LTV is 115 percent or the DTI
is no more than 31 percent, whichever happens first. Then the
servicer follows the HAMP guidelines for completing the
modification. The amount of principal reduced is treated
initially only as forbearance. Each year for three years,
however, a third of that amount is forgiven if the borrower
remains current on the modified loan.\24\
---------------------------------------------------------------------------
\24\ Principal forbearance is not the same as principal reduction.
In the first case, the borrower's unpaid principal balance remains
unchanged but is restructured to reduce monthly payments. In the latter
case, a certain amount of the principal is forgiven by the lender.
Principal forbearance means that the loan's unpaid principal balance is
not fully amortized over the remaining life of the loan. The balance,
less the amount subject to forbearance, is then amortized or ``spread
out'' over the remaining period of the loan, lowering the amount due
each month. The amount by which the principal was reduced is then due
as a balloon payment at the end of the loan, or when the property is
sold. In the case of principal reduction, a certain amount of the
principal is forgiven and is no longer owed by the borrower. See U.S.
Department of the Treasury, Making Home Affordable: Borrower Frequently
Asked Questions (Oct. 12, 2010) (online at makinghomeaffordable.gov/
borrower-faqs.html).
---------------------------------------------------------------------------
The NPV test is key to HAMP's strategy. The program is not
intended to prevent all foreclosures, but rather to encourage
modification in those cases in which the value of a
modification is greater than the value of a foreclosure. One
might very reasonably ask why a government program involving
servicer payments is necessary when servicers should already be
modifying these loans out of self-interest. However, there are
other factors at work that may affect a servicer's decision,
particularly the many incentives in securitized mortgages for
servicers to prefer foreclosure.\25\
---------------------------------------------------------------------------
\25\ For a detailed explanation of these factors, see March 2009
Oversight Report, supra note 10. See also Office of the Special
Inspector General for the Troubled Asset Relief Program, Quarterly
Report to Congress, at 163-176 (Oct. 26, 2010) (online at sigtarp.gov/
reports/congress/2010/October2010_Quarterly_Report_to_Congress.pdf)
(hereinafter ``SIGTARP Quarterly Report to Congress'').
---------------------------------------------------------------------------
During the period that the mortgage is in default, the
servicer typically must continue to make payments to the
investor out of its own pocket. It may also incur other
expenses related to holding the troubled mortgage in its
portfolio. If the mortgage goes into foreclosure, the servicer
is reimbursed for these expenses before any of the money passes
to the investors. Moreover, if a mortgage goes into
foreclosure, the servicer holding the loan at that time
typically handles the foreclosure, earning various fees for
this service. Additionally, servicers earn income from float on
the payments they collect--interest earned from the short-term
investment of mortgage payments made between the time the
servicer receives it from the borrower and the time it must be
remitted to the investors.\26\
---------------------------------------------------------------------------
\26\ Privately modified loans generally follow a similar pattern of
reimbursement of the servicer for costs incurred, such as advancing
coupon payments to investors while loans in the pool are in default.
Individual PSAs can differ in their terms, however.
---------------------------------------------------------------------------
Therefore, it is in the servicer's interest to keep a
mortgage for as long as it is producing an income stream and,
once it goes into default, to ensure that the mortgage goes
through foreclosure. HAMP's servicer incentive payments are
designed, at least in part, to overcome these incentives that
distort servicer decision making and lead to unnecessary
foreclosures.
HAMP also offers incentive payments to borrowers over the
course of the modification. These payments are designed to
encourage borrowers to remain in the program and continue to
pay their mortgages as well as to encourage participation in
the first place. While it would seem unnecessary to pay
borrowers to do what should already be in their interest,
Treasury obviously felt otherwise, probably due to the poor
track record of prior programs at obtaining and keeping
borrower participation. Additionally, the added income from
these borrower incentives is considered in the NPV model and
may help some marginal HAMP applicants to achieve an NPV
positive result, thus obtaining a HAMP modification.\27\
---------------------------------------------------------------------------
\27\ U.S. Department of the Treasury, Home Affordable Modification
Program: Base Net Present Value (NPV) Model v4.0: Model Documentation,
at 33 (Oct. 1, 2010) (online at www.hmpadmin.com//portal/programs/docs/
hamp_servicer/npvmodeldocumentationv4.pdf) (hereinafter ``HAMP Base NPV
Model v4.0: Model Documentation'').
---------------------------------------------------------------------------
D. HAMP's Performance
After an influx of new trial modifications in 2009 and
early 2010, the pace of entry into the program has fallen off
considerably, according to the most recent data on the
program's performance. Moreover, although some headway has been
made in reducing the enormous number of borrowers in trial
modifications awaiting conversion to permanent status, a
sizeable backlog remains. Finally, while nearly 1.4 million
trial modifications have been initiated since the start of the
program, the number of borrowers who have dropped out of the
program remains high. To date, HAMP has processed 519,648
permanent modifications.\28\
---------------------------------------------------------------------------
\28\ Data provided by Treasury.
---------------------------------------------------------------------------
1. Trial Modifications
As of October 31, 2010, approximately 1.4 million trial
modifications had been initiated under HAMP. Of these, 20,998
were initiated in October 2010. Between May and October 2010,
each month posted, on average, approximately 23,000 new trial
modifications, down from a high of almost 160,000 in October
2009.\29\
---------------------------------------------------------------------------
\29\ Data provided by Treasury. These figures represent new trial
modification first payments reported in October 2010.
---------------------------------------------------------------------------
Figure 3 below shows the total number of trial
modifications granted by month (e.g. October 2010:
approximately 21,000 modifications) as well as the disposition
of loans in each monthly cohort of HAMP modifications. Note
that some trial modifications from the earliest months of the
program remain active as trials.
FIGURE 3: DISPOSITION OF HAMP TRIAL MODIFICATIONS BY VINTAGE (MARCH
2009-SEPTEMBER 2010) \30\
[GRAPHIC] [TIFF OMITTED] 62622A.003
---------------------------------------------------------------------------
\30\ Data provided by Treasury.
Home mortgages have been traditionally divided into three
categories of borrower credit quality, although these
categories are not clearly defined. In the traditional usage,
prime mortgages are loans to borrowers with good credit
(typically above FICO 620) and adequate income. Alt-A mortgages
are also loans to borrowers with prime (A) credit. However,
Alt-As usually do not require income documentation (they are
``stated income'' loans), which is useful for small business
owners and independent contractors who have variable income,
but making the loans susceptible to fraud. Subprime mortgages
refer to loans to borrowers with poor credit (below 620).
Although in the past the Prime, Alt-A, and Subprime categories
formerly did not indicate anything about mortgage type (e.g.,
fixed or floating rate, interest only or fully amortizing), the
terms have come to be associated with specific loan types
without regard to credit score in recent usage. For example,
``subprime'' is often used to refer to more exotic and risky
mortgage types such as option ARMs.
Despite the fact that much of the analysis of the housing
crisis has focused on subprime loans, the majority of loans
past due as of June 2010 were prime mortgages.\31\ Prime
mortgages received more than half of all HAMP modifications in
the second quarter of 2010, while subprime and alt-A mortgages
each received less than a quarter.\32\
---------------------------------------------------------------------------
\31\ Office of the Comptroller of the Currency and Office of Thrift
Supervision, Mortgage Metrics Report, Second Quarter 2010, at 22 (Sept.
2010) (online at www.ots.treas.gov/_files/490019.pdf) (hereinafter
``OCC/OTS Mortgage Metrics Report, Second Quarter 2010'').
\32\ Id. at 22.
---------------------------------------------------------------------------
Treasury has attributed the dramatic decrease in new trial
modifications to several factors. According to officials, most
of the decline is likely due to the institution of a verified
income requirement in June 2010. Information provided by
borrowers (including income), now must be documented before a
trial modification can be initiated. ``Stated income'' (i.e.
non-verified) was previously allowed for trial modifications.
Additionally, Treasury has said that servicers increasingly
shifted their attention from initiating new trial modifications
to converting the existing backlog to permanent modifications.
Finally, Treasury has indicated that the declining overall
mortgage delinquency rate may play a role as well. There is
reason to doubt that this latter factor has been a major cause
of the decline, however. Figure 4 below shows new trial
modifications and mortgage delinquency from the third quarter
of 2009 to the third quarter of 2010. Clearly, the decline in
delinquency has been relatively slight, while the decline in
new trials has been much more severe.
FIGURE 4: TRIAL MODIFICATIONS VS. MORTGAGE DELINQUENCY (Q3 2009-Q3
2010) \33\
[GRAPHIC] [TIFF OMITTED] 62622A.004
---------------------------------------------------------------------------
\33\ Mortgage Bankers Association, National Delinquency Survey Q3
2010 (Nov. 18, 2010); Data provided by Treasury.
The change to verified income cannot completely explain the
decrease either, since the number of new trial modifications
began dropping off long before the up-front verified
documentation standard was implemented in July 2010. It is
possible that the program has already reached the majority of
borrowers who can be helped. In the early months of the
program, there was a large pool of borrowers awaiting help.
Once many of these homeowners entered HAMP or other programs,
there were simply fewer potential applicants who met HAMP
criteria.\34\
---------------------------------------------------------------------------
\34\ OCC/OTS Mortgage Metrics Report, Second Quarter 2010, supra
note 31, at 22. See also Fitch Ratings, U.S. RMBS: Still Under a
Shadow, at 5 (Nov. 1, 2010) (noting that ``the number of remaining
borrowers eligible for a loan modification appears to be declining, as
new loan modification activity has declined from its peak in 2009.'').
---------------------------------------------------------------------------
Among borrowers starting trial modifications, ``curtailment
of income'' remains the most common reason provided to explain
the borrowers' economic hardship: around 47 percent of
borrowers gave this reason for their hardship, reflecting in
part the effect of underemployment. Unemployment accounts for
only around 6 percent.\35\ This is not surprising since the
program originally was structured in such a way that made it
almost impossible for an unemployed borrower to receive help
under HAMP without another source of income. Furthermore, an
applicant from a household in which one spouse has lost a job,
while the other remains employed may select ``curtailment of
income'' instead of ``unemployment.'' In July 2010, Treasury
rolled out an add-on program, the HAMP Unemployment Program,
which offers forbearance for those who are unemployed and are
receiving unemployment benefits. Additionally, several states,
through the Hardest Hit Fund (HHF), have taken steps to address
the effects of unemployment on homeowners' ability to keep
their homes. For further discussion of these programs, see
Section F. Figure 5 below details the top economic hardship
reasons for trial modification starts.
---------------------------------------------------------------------------
\35\ Data provided by Treasury.
---------------------------------------------------------------------------
FIGURE 5: ECONOMIC HARDSHIP REASONS FOR TRIAL MODIFICATION STARTS \36\
---------------------------------------------------------------------------
\36\ Data provided by Treasury.
[GRAPHIC] [TIFF OMITTED] 62622A.005
2. Conversion to Permanent Modification
In its previous reports, the Panel expressed concern
regarding the rate at which trial modifications were converting
to permanent modifications. As noted in the October 2009
report, only 1.26 percent of modifications had converted to
permanent modifications, or approximately 2,000 borrowers. The
April 2010 report found that the conversion rate had improved
to 23.1 percent, or around 4,000 borrowers, although only 9.7
percent converted within the standard three months. Through
September 2010, 38.4 percent of HAMP conversions happened
within three months.\37\ While, as noted in the Panel's prior
reports, the earlier conversion rates were based on a fairly
small loan pool, HAMP is demonstrating improvement in
conversion of modifications from trial to permanent. As
mentioned above, however, the number of new trial modifications
has declined substantially in recent months.
---------------------------------------------------------------------------
\37\ Data provided by Treasury.
---------------------------------------------------------------------------
In this area, some servicers have performed markedly better
than others. Among the top servicers, for example, Wachovia
Mortgage and HomeEq Servicing have conversion rates of 89
percent and 95 percent respectively. In contrast, Bank of
America's rate is closer to 30 percent.\38\ Conversations with
Treasury officials indicate that much of the difference in
conversion rates between servicers is due to the recent switch
from stated income trials to verified income, and because
stated income was used primarily by larger servicers, such as
Bank of America. As a result, these servicers still have large
pools of difficult to convert, stated income modifications.
---------------------------------------------------------------------------
\38\ U.S. Department of the Treasury, Making Home Affordable
Program: Servicer Performance Report Through October 2010, at 4 (Nov.
19, 2010) (online at www.financialstability.gov/docs/
Oct%202010%20MHA%20Public%20Final.pdf) (hereinafter ``MHA Servicer
Performance Report'').
---------------------------------------------------------------------------
During the early days of HAMP, Treasury focused on
providing payment relief to as many borrowers as possible.
Servicers were able to grant trial modifications to borrowers
with no documentation required. However, the documentation
required to convert the trial to a permanent modification
proved to be a challenge as demonstrated by the earlier anemic
conversion rate. Accordingly, mortgage servicers utilizing
stated income began to develop a backlog of trial modifications
awaiting a decision. In its last report on foreclosure
mitigation, the Panel noted steps that Treasury had taken to
ensure that servicers would clear these cases in a timely
manner. As of October 31, 2010, the backlog of modifications
that had been in a trial period for six months or longer had
fallen to 69,400. Three servicers--Bank of America, JPMorgan
Chase, and CitiMortgage--comprise more than two-thirds of the
backlog, and not surprisingly, all three utilized stated income
trials prior to the new standard. Bank of America alone
services approximately half of the backlog of aged trials.\39\
---------------------------------------------------------------------------
\39\ Id. at 4.
---------------------------------------------------------------------------
While it is good that the servicers appear to be clearing
out the backlog of trial modifications eligible for conversion,
much of the increase in the conversion rate is linked to the
decline in number of people entering new trial modifications,
and therefore a decline in the total number of borrowers in
trial modifications, and not from an actual increase in the
number of borrowers moving to permanent modifications. That is,
while the conversion rate is improving, it is primarily a
result of the smaller pool of borrowers eligible for
conversion. Treasury expects the HAMP conversion rate to
increase significantly going forward as trial modifications are
now started only after the borrower has provided all
documentation, previously a major stumbling block, and
servicers shift their focus more to processing the existing
backlog of trial modifications.\40\ As the Panel has noted
before, different pieces of the HAMP modification process, such
as initiating trial modifications and converting trials to
permanent modifications are linked, and a narrow focus by
servicers on one aspect to the exclusion of others may lead to
tradeoffs that diminish overall modification success.\41\
Currently, as Treasury has acknowledged, a focus by servicers
on converting the backlog has caused the number of new trials
to decline.\42\
---------------------------------------------------------------------------
\40\ Treasury conversations with Panel staff (Nov. 16, 2010).
\41\ April 2010 Oversight Report, supra note 1, at 69-70.
\42\ Treasury conversations with Panel staff (Nov. 16, 2010).
---------------------------------------------------------------------------
3. Permanent Modifications
Through October 2010, 519,648 homeowners have been able to
obtain a conversion to a permanent modification through HAMP.
Of these, 483,342 are currently active modifications. The
remaining 36,306 represent 491 loans that have been paid off
and 35,815 that have been cancelled due to redefault.\43\
---------------------------------------------------------------------------
\43\ Data provided by Treasury; MHA Servicer Performance Report,
supra note 38, at 4.
---------------------------------------------------------------------------
There were a total of 23,750 new permanent modifications in
October 2010. The number of new permanent modifications peaked
in April 2010 at 68,291 and has declined steadily since then.
The number of new permanent conversions is now averaging less
than half the number of new monthly permanent modifications at
the peak.\44\ This trend is closely tied to the significant
decrease in trial modifications coming into the pipeline rather
than a failure to convert modifications once they enter the
pipeline. This development raises the question of whether HAMP
has already surpassed its maximum effectiveness and will
continue the trend of diminishing results.
---------------------------------------------------------------------------
\44\ Data provided by Treasury; MHA Servicer Performance Report,
supra note 38, at 2.
---------------------------------------------------------------------------
All HAMP permanent modifications have used interest rate
reductions in order to reach the program's affordability
target. In addition, more than 57 percent of the modifications
include a term extension, and 30 percent feature principal
forbearance.\45\ Principal reduction remains relatively rare,
with only around 3 percent of the permanent modifications
offering a principal write-down as of early October 2010.\46\
Prior to modification, the median interest rate of HAMP
participants is 6.63 percent. This drops to 2 percent after the
permanent modification. Similarly, monthly payments decline
from a median value of $1,434 before modification to $838 after
modification, a difference of $596.\47\
---------------------------------------------------------------------------
\45\ MHA Servicer Performance Report, supra note 38, at 3.
\46\ Data provided by Treasury; MHA Servicer Performance Report,
supra note 38, at 3.
\47\ Data provided by Treasury.
---------------------------------------------------------------------------
Figure 6 below shows the number of all active permanent
modifications and redefaults by month.
FIGURE 6: MONTHLY HAMP PERMANENT MODIFICATIONS AND REDEFAULTS (FEBRUARY
2010-SEPTEMBER 2010) \48\
---------------------------------------------------------------------------
\48\ ``Monthly Active Permanent Modifications'' and ``Monthly
Permanent Modification Redefaults'' are derived from cumulative
``Active Permanent Modifications'' and ``Permanent Modifications
Canceled'' (excluding loans paid off) levels from March 2010 to October
2010 recorded in the Making Home Affordable Program's monthly Servicer
Performance Reports. For these monthly reports, see U.S. Department of
the Treasury, Reports and Documents (online at financialstability.gov/
latest/reportsanddocs.html) (accessed Dec. 10, 2010) (hereinafter
``Treasury Reports and Documents'').
[GRAPHIC] [TIFF OMITTED] 62622A.006
Figure 7 below summarizes the characteristics of all
permanent modifications granted under HAMP. Forgiveness and
forebearance are explained in footnote 24 above. UPB refers to
the unpaid principal balance. PI refers to debt service
payments--principal and interest.
FIGURE 7: SUMMARY DATA ABOUT PERMANENT MODIFICATIONS \49\
---------------------------------------------------------------------------
\49\ Data provided by Treasury.
[GRAPHIC] [TIFF OMITTED] 62622A.007
a. Borrower Debt
Permanent HAMP modifications offer much more affordable
mortgage payments over unmodified mortgages, but they
nonetheless leave borrowers with very high overall debt levels.
Prior to modification, the median borrower receiving a HAMP
modification was paying 45 percent of his or her pre-tax income
towards the mortgage (front-end DTI), and nearly 80 percent of
pre-tax income toward all debts (back-end DTI). After receiving
a HAMP permanent modification, the median borrower's front-end
DTI had been reduced to 31 percent, and their back-end DTI fell
to 63 percent.\50\
---------------------------------------------------------------------------
\50\ Since the purpose of HAMP is to lower high-DTI borrowers to 31
percent or below, and since servicers have no incentive to lower
payments below this, the average borrower DTI is the same as the 31
percent limit. Data provided by Treasury.
---------------------------------------------------------------------------
However, even these post-modification DTIs are higher than
those allowed by most mortgage underwriting standards. To
qualify for a new FHA loan, for example, a borrower typically
needs to show a back-end DTI ratio of no more than 41 percent
of gross income, while a borrower receiving a mortgage through
Freddie Mac must not have a back-end DTI ratio of more than 45
percent.\51\ The situation worsens once state and federal taxes
are subtracted from income. Pre-modification borrowers appear
to be spending nearly all of their after-tax income on debt
service, with other expenses presumably paid for with credit.
Even at the improved 63 percent back-end DTI of borrowers
following a HAMP modification, debt service payments will still
consume approximately 80 percent of after-tax income, which is
of course the income borrowers actually have to spend.\52\
Furthermore, the Panel is particularly concerned that the post-
modification back-end DTI ratio appears to be rising--up 4
percent since the Panel's April 2010 foreclosure report, where
post-modification back-end median DTI was 59 percent.\53\ These
high and rising DTIs do not bode well for the long term success
of the program.
---------------------------------------------------------------------------
\51\ Federal Housing Administration, FHA Requirements: Debt Ratios
(online at www.fha.com/fha_requirements_debt.cfm) (accessed Dec. 10,
2010); Federal Home Loan Mortgage Corporation, Underwriting Reminders
for Loan Prospector Caution Risk Class Mortgages, at 4 (Oct. 2010)
(online at www.freddiemac.com/learn/pdfs/uw/caution_remind.pdf).
\52\ A rough calculation of after-tax income can give a more
realistic picture of the financial situation of HAMP participants. The
median HAMP participant earning $32,000 a year, filing as the head of
household, and taking the standard deduction, would pay approximately
$2,954 in federal income taxes, $2,448 in Social Security and Medicare
taxes and, in this example, $1,000 in state taxes, although this latter
figure will vary greatly depending on the state. The borrower in this
case would then have an after-tax income of $25,598.
The median HAMP participant's pre-tax, pre-mod, back-end DTI of 80
percent equates to $25,555 in total debt service owed annually, divided
into $14,422 in mortgage payments and $11,133 of service on other
debts. If accurate, total debt service is nearly as much as the after-
tax income above (a 100 percent after-tax back-end DTI). After a HAMP
permanent modification, the borrower would still be paying $20,272
annually on all debts, including $9,930 in mortgage payments (a savings
of $4,493 annually). Interestingly, service on other debts apparently
falls as well after modification, down $790 to $10,342. After taxes and
debt service, the median HAMP participant has just $5,326 per year, or
$444 per month, for all other expenses, including food, clothing,
health care, education, etc.
\53\ See April 2010 Oversight Report, supra note 1, at 43; Data
provided by Treasury. Sustainability of HAMP modifications is discussed
further in Section G.2.g, infra.
---------------------------------------------------------------------------
Figure 8 below shows the distribution of (pre-tax) back-end
DTIs for borrowers receiving permanent modifications from HAMP.
Since these borrowers have already been in trials for at least
three months, these DTIs are post-modification. The bulk of
borrowers receiving permanent modifications are clearly either
in the more moderate 31 percent-40 percent DTI category or the
very high over 80 percent DTI. In fact, nearly one-third of
post-modification borrowers have back-end DTI in excess of 80
percent. This is troubling, as it indicates that the overall 63
percent post-modification DTI statistic contains a very large
sub-segment of heavily indebted borrowers.
FIGURE 8: NUMBER OF PERMANENT MODIFICATIONS BY BACK-END DTI (AS OF
SEPTEMBER 30, 2010) \54\
---------------------------------------------------------------------------
\54\ Data provided by Treasury.
[GRAPHIC] [TIFF OMITTED] 62622A.008
The question of what DTI level is sustainable remains open,
and complicated by HAMP's focus on front-end DTIs. Other debts
such as second liens, credit card debt, and car debt are not
factored into the front-end DTI used to determine HAMP
eligibility, despite the fact that most HAMP applicants have
substantial amounts of such debt. It would appear that back-end
DTI may be the more important metric for gauging a homeowner's
overall financial picture and their ultimate ability to remain
current on a mortgage. Indeed some borrowers, although heavily
indebted overall, do not meet HAMP's front-end DTI eligibility
threshold. Faith Schwartz, senior adviser for the HOPE NOW
Alliance, testified that HAMP's 31 percent minimum front-end
DTI for eligibility was considered ``aggressive'' when HAMP was
first rolled out, but that even this level is too high for many
homeowners who wind up in foreclosure because their front-end
DTI is too low to make them HAMP-eligible.\55\ Any additional
assistance necessary to bring down the back-end DTI ratio could
theoretically come from a number of sources, including federal,
state, or local agencies, or from private sources.\56\
Obviously, however, in any effort of this sort to reduce back-
end DTIs, policymakers will have to weigh the costs and
benefits of any additional assistance, as well as how the
burden of these costs would be distributed between servicers,
investors, and taxpayers.\57\
---------------------------------------------------------------------------
\55\ Congressional Oversight Panel, Written Testimony of Faith
Schwartz, senior advisor, HOPE NOW Alliance, COP Hearing on TARP
Foreclosure Mitigation Programs, at 7 (Oct. 27, 2010) (online at
cop.senate.gov/documents/testimony-102710-schwartz.pdf).
\56\ For instance, Treasury's 2MP program is designed to decrease
second lien payments which would reduce back-end DTI. See Section F,
infra.
\57\ Continuing with the median HAMP participant example discussed
in footnote 52, supra, an additional subsidy sufficient to put the
borrower at a 50 percent pre-tax back-end DTI would raise the
borrower's after-tax, after debt service, disposable income by $4,272
($356 per month) from $5,326 ($444 per month) to $9,598 ($800 per
month). Although an extra $356 a month could greatly help many
distressed borrowers living at the edge of their means, it is difficult
to determine how much such a policy would actually reduce foreclosures
in the long run. The cost of such a program is also difficult to
estimate, as it would depend on how the subsidy was structured (direct
payment of debt service vs. principal reduction) and which debts are
affected (the low-interest first mortgage or higher interest debts such
as credit cards).
---------------------------------------------------------------------------
b. Negative Equity
Permanent modifications have not, however, made much of an
impact on the depth of borrowers' negative equity. Of all
active permanent modifications, nearly 95 percent have an
unpaid principal balance that is higher than it was before
modification.\58\ Negative equity remains high, with more than
76 percent of mortgages in permanent modification still
carrying a negative LTV ratio. In fact, LTV ratios have
increased, on average, after modifications. The median LTV
ratio is approximately 120 percent before a modification and
about 125 percent after the modification.\59\ Although HAMP may
provide a more affordable monthly payment for homeowners, it
does not address the problems caused by mortgages that are
deeply underwater. Negative equity can restrict the ability of
homeowners to move, whether for family reasons or to pursue
greater job opportunities, since home sale proceeds will not be
sufficient to repay their loan. It also provides an incentive
for borrowers who can afford to pay their mortgages to stop
paying intentionally and walk away from their homes if they
believe that they will remain deeply underwater for a long
time, a decision known as a ``strategic default.''
---------------------------------------------------------------------------
\58\ The increase does not represent a true increase in the
obligations of the borrower because it represents mostly capitalization
of arrearages and escrow requirements. Data provided by Treasury.
\59\ Data provided by Treasury. Much of this increase in LTV is due
to adding missed mortgage payments to the principal balance.
---------------------------------------------------------------------------
The value of futures contracts based on the Case-Shiller
Housing Price Index can be a useful indicator of market
expectations for future home prices. A chart showing the
current prices for these contracts for the 10 largest
Metropolitan Statistical Areas (MSAs) and the composite of
these MSAs is shown as Figure 45 in the Metrics section. The
futures market expects that over the next five years housing
prices will remain relatively flat in all MSAs as well as the
10-MSA composite. Although the predictive accuracy of these
prices is limited, especially for the more illiquid contracts,
and as they go further out in time, the prices seem to indicate
that few of the educated observers of the housing market who
trade these contracts expect a rapid rise in home prices. It
would therefore appear that borrowers at 120 percent LTV or
higher, such as the average HAMP participant, have a slim
chance of returning to positive equity in the foreseeable
future.\60\
---------------------------------------------------------------------------
\60\ Further discussion of Case-Shiller futures can be found below
in Section Three: TARP Updates Since Last Report (see Section D.3 on
housing prices).
---------------------------------------------------------------------------
4. Borrowers Dropped from the Program
As servicers began to work through the significant backlog
in the program, many borrowers in trial modifications were
denied permanent modifications and dropped from the program. In
total, servicers have cancelled 541,907 trial modifications
through October 2010.\61\ According to Treasury, the most
common reasons for trial modification cancellations are
insufficient documentation, trial plan payment default, and
ineligibility because the first mortgage payment is already at
or below the program DTI standard of 31 percent. Specifically,
among the HAMP servicers, just under 30 percent of trial
modifications have been cancelled because of incomplete
requests. Surprisingly, the percentage did not change
significantly between the period preceding June 1, 2010--the
date the verified income requirement was implemented--and the
period following that date. Trial plan defaults, however, have
crept up slightly. Prior to June 1, 2010, approximately 21
percent of cancellations were due to a default during the trial
period, and 24 percent for the period following June 1. Among
those with the first payment due on or before June 1, 2010,
nearly 8 percent did not go forward because of a negative NPV
test result, meaning that foreclosure was found to be more cost
effective than modification. Among those with a first payment
due after June 1, 2010, only 4 percent were cancelled for this
reason. Figure 9 below shows reasons for permanent conversion
denials for the period prior to June 1, 2010, and post June 1,
2010, respectively.\62\
---------------------------------------------------------------------------
\61\ In addition to these canceled modifications that failed to
convert from trial to permanent status, an additional 177,580 trial
modifications were disqualified through October 2010. Data provided by
Treasury.
\62\ Data provided by Treasury.
---------------------------------------------------------------------------
FIGURE 9: CONVERSION DENIAL REASONS \63\
---------------------------------------------------------------------------
\63\ Data provided by Treasury.
[GRAPHIC] [TIFF OMITTED] 62622A.009
Figure 10 below shows the status of cancelled trial
modifications for the eight largest servicers. Only 3.9 percent
experienced a foreclosure sale, while 13 percent are in the
foreclosure process. Another 41.3 percent of the borrowers
received an alternative modification, although the alternative
modification terms are not necessarily comparable to a HAMP
modification.\64\ See Section E for a discussion of alternative
modifications.
---------------------------------------------------------------------------
\64\ MHA Servicer Performance Report, supra note 38, at 5. These
eight are the servicers with the largest allocated HAMP cap amounts.
---------------------------------------------------------------------------
FIGURE 10: STATUS OF CANCELLED TRIAL MODIFICATIONS (AS OF SEPTEMBER
2010) \65\
---------------------------------------------------------------------------
\65\ MHA Servicer Performance Report, supra note 38, at 5.
[GRAPHIC] [TIFF OMITTED] 62622A.010
There is a tremendous variation in the reasons given by
different servicers for trial plan failures. Some notable
outliers are identified with circles in Figure 11 below. For
example, a significantly greater percentage of Wells Fargo's
trial modifications failed because of low front-end DTI (<31
percent) than was the case for other servicers. Of the trial
modifications terminated by Citibank, nearly half were due to
``ineligible mortgages.'' JPMorgan Chase's borrowers had NPV
negative modifications at a much higher rate than other
servicers. This might be a function of other servicers ruling
out more modifications before even getting to NPV, or it could
be a function of the particular inputs JPMorgan Chase uses for
its NPV test.\66\ Of the trial modifications terminated by GMAC
and Litton Loan Servicing, over half were caused by a failure
to receive complete paperwork. Nearly 60 percent of failed
trials at Saxon Mortgage Services, Inc., were due to trial plan
default. At present, the Panel is unsure why these wide
discrepancies exist. It might be explained by heterogeneity in
the loan pools or by servicer choices in coding denials, but it
could also be explained by particular servicer behavior or
decisions. Treasury should be closely monitoring this issue and
provide an explanation for these wide discrepancies.
---------------------------------------------------------------------------
\66\ Treasury permits larger servicers such as JPMorgan Chase to
adjust certain inputs to the NPV model to suit their particular
circumstances. This is discussed further in Section G.2.f, infra.
---------------------------------------------------------------------------
FIGURE 11: REASONS FOR TRIAL PLAN FAILURE FOR LARGE HAMP SERVICERS
(>20,000 TRIAL MODIFICATIONS), MARCH 2009-JUNE 2010 \67\
---------------------------------------------------------------------------
\67\ Data provided by Treasury.
[GRAPHIC] [TIFF OMITTED] 62622A.011
5. Redefault
There are also a number of borrowers who are unable to
remain current even after receiving a permanent modification
through HAMP and ultimately redefault on their mortgages. As of
October 2010, 35,815 borrowers with permanent modifications had
redefaulted. Although the trend has been toward lower rates of
redefault, October posted an increase. Actual redefaults to
date have been lower than the redefault rate assumed in the NPV
model. However, the NPV model considers the likelihood of
redefault over the entire 5-year term of a modification, while
the current redefault rate only looks at the experience of the
program since its inception. Many of these loans are only a few
months into their modifications, so it is too soon to tell if
the current redefault rates will continue.
It is important to note that Treasury does not have
complete data on HAMP permanent modification performance.
Treasury lacks complete or valid performance information for
63,169 permanent HAMP modifications or 13 percent of all
permanent modifications. The inclusion of full information for
these permanent modifications could potentially raise or lower
redefault rates. The Panel urges Treasury to ensure that going
forward it has complete, valid performance data on all HAMP
permanent modifications. For further discussion of HAMP
redefaults, see Section D.5.
E. Performance of Non-HAMP Modifications
Taken as a whole, 54 percent of the nearly 1.4 million
temporary modifications that have been initiated under HAMP
have ultimately failed.\68\ What became of the borrowers who
are no longer in the program? According to Treasury, the
majority of the borrowers go on to receive modifications
through the servicers' own proprietary modification programs.
Under program guidelines, participating servicers must consider
borrowers who were unable to receive a HAMP modification for a
proprietary modification, but there are no specific eligibility
or modifications standards, only the requirement that borrowers
not receiving a HAMP modification must be considered for a
proprietary modification according to the servicers' own
standards. Thus, whereas HAMP modifications are standardized
and the terms are publicly transparent, non-HAMP modifications
are not, creating concern about their sustainability.
---------------------------------------------------------------------------
\68\ Data provided by Treasury. These failed modifications include
disqualified and cancelled trials as well as disqualified permanent
modifications.
---------------------------------------------------------------------------
In addition, proprietary modifications serve borrowers who
are not eligible for HAMP. Of the 5.1 million first lien
mortgages that were more than 60 days delinquent as of October
31, 2010, Treasury estimates that only around 1.5 million are
HAMP-eligible. The most common reasons for ineligibility, are:
(1) having a servicer that is not participating in HAMP; (2)
having an ineligible loan, such as a loan backed by FHA or
Department of Veterans Affairs (VA); (3) having a front-end DTI
ratio of less than 31 percent, and (4) having a loan for a home
reported as non-owner occupied at the time of origination.\69\
---------------------------------------------------------------------------
\69\ Data provided by Treasury. Having a non-participating servicer
excludes 600,000 borrowers, while the other most common reasons for
ineligibility affect approximately 700,000 borrowers each. HAMP
applicants are evaluated via a waterfall method. Therefore, a borrower
who is disqualified because his or her servicer is not participating
may also be ineligible for other reasons, but is only counted among
those disqualified at the top of waterfall.
---------------------------------------------------------------------------
Banks report that they are increasing their number of non-
HAMP modifications to help those who are not eligible or do not
qualify for a HAMP modification. Statistics gathered by HOPE
NOW, an alliance of mortgage servicers, show that the number of
loan modifications implemented through private channels is
outpacing the number of HAMP modifications.\70\ Year-to-date,
there have been more than twice as many modifications performed
outside of Treasury programs as HAMP modifications,\71\ and in
October 2010 alone, 81 percent of all completed modifications
were done outside of HAMP.\72\ In addition to various types of
modifications, HAMP and proprietary, some homeowners have
sought assistance through other payment plans. More than 18
percent of the home retention actions initiated in the second
quarter of 2010 were payment plans.\73\
---------------------------------------------------------------------------
\70\ HOPE NOW Alliance, Industry Extrapolations and Metrics
(October 2010), at 3-4 (Dec. 6, 2010) (online at www.hopenow.com/
industry-data/HOPE%20NOW%20Data%20Report%20(October)%2012-05-
2010%20v2.pdf) (hereinafter ``HOPE NOW Industry Extrapolations and
Metrics'').
\71\ HOPE NOW Industry Extrapolations and Metrics, supra note 70,
at 3.
\72\ Year-to-date, nearly 87 percent of all completed modifications
have been performed outside of HAMP. HOPE NOW Industry Extrapolations
and Metrics, supra note 70, at 4.
\73\ OCC/OTS Mortgage Metrics Report, Second Quarter 2010, supra
note 31, at 21.
---------------------------------------------------------------------------
While the number of proprietary modifications currently
outpaces HAMP modifications, HAMP still produces, on average, a
modification offering more relief to the borrower and having a
lower likelihood of redefault. For those modifications made
during the second quarter of 2010, non-HAMP modifications
reduced monthly payments, on average, half as much as HAMP
payment reductions, and of the nearly 10 percent of loan
modifications that result in unchanged or increased monthly
payments, nearly all are non-HAMP modifications.\74\ The lower
average payment reductions for non-HAMP modifications seems to
impact directly their potential to redefault, as non-HAMP
modifications have a redefault rate six months after
modification of 22.4 percent, compared to only 10.8 percent for
HAMP modifications.\75\ While the redefault rate six months
after modification is a more preliminary data point that does
not encompass the full scope of likely redefaults, it is the
longest time span currently available for rates that break out
HAMP and non-HAMP modifications. HAMP modifications have a
redefault rate of 21 percent twelve months after modification
for loans 90 or more days past due,\76\ while the same rates
for all modifications (HAMP and non-HAMP combined) are 48.6
percent and 36.6 percent for loans modified in the first
quarter and second quarter of 2009, respectively.\77\
---------------------------------------------------------------------------
\74\ OCC/OTS Mortgage Metrics Report, Second Quarter 2010, supra
note 31, at 30-32.
\75\ The redefault rate is for fourth quarter 2009 loan
modifications that have aged six months and are 60 or more days
delinquent. OCC/OTS Mortgage Metrics Report, Second Quarter 2010, supra
note 31, at 35.
\76\ Data provided by Treasury.
\77\ OCC/OTS Mortgage Metrics Report, Second Quarter 2010, supra
note 31, at 37.
---------------------------------------------------------------------------
Beyond these data points, HAMP continues to provide a
roadmap (i.e., the ``waterfall'') for servicers to use when
considering various tools to use in mortgage modifications. As
Ms. Schwartz noted at the Panel's October foreclosure
mitigation hearing, ``The first most important contribution of
HAMP is that all servicers who signed up for HAMP must review
all homeowners for eligibility. The HAMP process offers
homeowners a first line of defense to avoid foreclosure. Second
is the importance of the HAMP waterfall. Investors, servicers,
lenders, and nonprofits, and homeowners have a uniform map of
activity that is necessary to ensure delinquent homeowners who
seek help are being considered for a solution prior to
foreclosure.'' \78\ HAMP has created a uniform approach to
mortgage modifications: participating servicers must evaluate
borrowers for and (assuming criteria are met) provide them with
a HAMP modification first, ensuring a set affordability
standard, and only consider a proprietary modification, the
terms of which are determined completely by the servicer, after
it has been determined that HAMP is not an option.
---------------------------------------------------------------------------
\78\ Congressional Oversight Panel, Testimony of Faith Schwartz,
senior advisor, HOPE NOW Alliance, Transcript: COP Hearing on TARP
Foreclosure Mitigation Programs (Oct. 27, 2010) (publication
forthcoming) (online at cop.senate.gov/hearings/library/hearing-102710-
foreclosure.cfm) (hereinafter ``Testimony of Faith Schwartz'').
---------------------------------------------------------------------------
Unfortunately, while some limited data do exist on non-HAMP
programs, the data collection efforts have not been
sufficiently robust. The lack of comprehensive, reliable data
makes it difficult to make an apples-to-apples comparison of
HAMP and non-HAMP programs. While the average monthly payment
reduction and redefault rates are available, the specifics of
the structure of the modifications are unknown. Since its
initial report on foreclosure in March 2009, the Panel has
stressed the importance of collecting data on loan performance,
loss mitigation efforts, and foreclosure. While data collection
has improved, including the data collected through HAMP, non-
HAMP modifications do not have the same level of transparency.
The Panel questions why a system of record was not established
to track non-HAMP modifications at the outset and why one still
does not exist. The HOPE NOW Alliance has increased its
collection of data on this type of modification but, as of
publication, the data are not available. Mortgage servicers
representing the bulk of the mortgage servicing industry are
owned or controlled by federally regulated entities.\79\
Regulators of those entities, typically the Federal Reserve
System and the Office of the Comptroller of the Currency, could
and should be collecting and reporting this information from
the mortgage servicers under their jurisdiction. Given the need
for specific and robust data on non-HAMP modifications to
analyze properly their effectiveness and viability and further
inform the continued analysis of HAMP, the Panel reiterates its
call for Treasury to increase or aid in data collection.
---------------------------------------------------------------------------
\79\ For instance the top four servicers, Bank of America, Wells
Fargo, JPMorgan and Citigroup, who service over 50 percent of all
mortgages (by outstanding principal balance), are all bank holding
companies. Congressional Oversight Panel, November Oversight Report:
Examining the Consequences of Mortgage Irregularities for Financial
Stability and Foreclosure Mitigation, at 59 (Nov. 16, 2010) (online at
cop.senate.gov/documents/cop-111610-report.pdf) (hereinafter ``November
2010 Oversight Report''); Federal Financial Institutions Examination
Council, Top 50 BHCs (Sept. 30, 2010) (online at www.ffiec.gov/
nicpubweb/nicweb/top50form.aspx); House Financial Services,
Subcommittee on Housing and Community Opportunity, Written Testimony of
John Walsh, acting comptroller of the currency, Office of the
Comptroller of the Currency, Robo-Signing, Chain of Title, Loss
Mitigation, and Other Issues in Mortgage Servicing, at 2 (Nov. 18,
2010) (online at financialservices.house.gov/Media/file/hearings/111/
Walsh111810.pdf) (``The OCC supervises all national banks and their
operating subsidiaries, including their mortgage servicing operations.
The servicing portfolios of the eight largest national bank mortgage
servicers account for approximately 63 percent of all mortgages
outstanding in the United States--nearly $33.3 million loans totaling
almost $5.8 trillion in principal balances as of June 30, 2010.'').
---------------------------------------------------------------------------
F. Treasury's Other Foreclosure Mitigation Programs
Establishing HAMP was part of Treasury's initial response
to the housing crisis, and it has remained Treasury's largest
foreclosure mitigation program, both in terms of funding and
borrowers. However, as the crisis evolved and HAMP's reach
began to prove too limited, Treasury announced and rolled out
the following additional programs: Home Price Decline
Protection (HPDP), PRA, Home Affordable Unemployment Program
(UP), Home Affordable Foreclosure Alternatives (HAFA), Second
Lien Modification Program (2MP), HHF, FHA Refinance Program,
and various agency-specific programs to encourage
modifications. Although many of these programs have been in
existence for many months to over a year, at this time there
are either no data available or such a limited data pool that
meaningful analysis of these programs is not possible.
Therefore, this report will confine its primary analysis to
HAMP. However, it is important to understand that HAMP
currently operates in conjunction with these other programs.
Home Price Decline Protection (HPDP)
The HPDP, announced on July 31, 2009, and effective the
following day,\80\ was designed to address the issue of
investor objections to modifications due to fear of a potential
future decline in home values. For an investor who anticipates
a future decline in home values, it is preferable to foreclose
today rather than consent to a modification that might fail and
end in a future foreclosure. Under this program, investors
receive incentive payments that accrue over a 24-month period
to mitigate potential losses and encourage their consent to
proposed modifications.
---------------------------------------------------------------------------
\80\ U.S. Department of the Treasury, Treasury Announces Home Price
Decline Protection Incentives (July 31, 2009) (online at
www.financialstability.gov/latest/tg_07312009.html).
---------------------------------------------------------------------------
Principal Reduction Alternative (PRA)
Treasury's PRA program was announced on June 3, 2010, and
went into effect on October 1, 2010. This program was a
response to the large number of underwater mortgages, as
principal reductions can be a significant means of preventing
foreclosures and redefaults. It operates much like HAMP, except
that instead of postponing payments on a portion of the
mortgage, the PRA program forgives that portion altogether.\81\
Servicers are required to evaluate a loan that is HAMP eligible
and has a mark-to-market loan-to-value ratio greater than 115
percent with both the standard HAMP waterfall and an
alternative waterfall that includes principal reduction as the
required second step, and then must use the NPV model to
evaluate the modifications proposed by both waterfalls. If the
NPV result generated by the standard waterfall is positive,
servicers must modify the loan; if the NPV result generated by
the alternative waterfall is positive, servicers are encouraged
but not required to perform a HAMP modification including
principal forgiveness; if the NPV result for both waterfalls is
negative, loan modification is not required.\82\ Thus, the
final decision on whether to grant a principal reduction is
ultimately up to the servicer. Investors receive standard
incentive payments as well as a percentage of each dollar
forgiven.
---------------------------------------------------------------------------
\81\ U.S. Department of the Treasury, Modification of Loans with
Principal Reduction Alternative, Supplemental Directive 10-05 (June 3,
2010) (online at www.hmpadmin.com/portal/programs/docs/hamp_servicer/
sd1005.pdf).
\82\ Id. The alternative waterfall inserts principal reduction in
between the standard waterfall's step one (capitalization) and step two
(interest rate reduction).
---------------------------------------------------------------------------
Home Affordable Unemployment Program (UP)
Treasury's unemployment program, announced on March 26,
2010,\83\ and effective July 1, 2010,\84\ was designed to
assist unemployed homeowners by granting a temporary
forbearance of a portion of their monthly mortgage payment for,
at a minimum, the lesser of three months or until employment is
regained.\85\ During the forbearance period, payments are
reduced to no more than 31 percent of the borrower's gross
monthly income, including unemployment benefits.\86\ In order
to be eligible, a borrower must hold a mortgage that: is
secured by the borrower's principal residence, is first-lien
and originated on or before January 1, 2009, has an unpaid
principal amount equal to or less than $729,750, is delinquent
or default is reasonably foreseeable, has not yet been modified
under the HAMP, and has not yet received UP forbearance.
Additionally, the borrower must: be eligible for HAMP, make the
request before becoming seriously delinquent (three months
overdue on monthly payments), and be unemployed. An unemployed
borrower who requests HAMP assistance must be evaluated for and
receive UP forbearance before being considered for a HAMP
modification, if all criteria are met. Servicers may require as
a pre-condition to approval that borrowers be in receipt of
unemployment benefits for up to three months before the
forbearance period begins. Once in the program, if a borrower
regains employment, the forbearance ends. If the borrower still
meets HAMP eligibility, the servicer must consider him for a
permanent HAMP modification, and any arrearages are capitalized
as part of the standard HAMP process. If the borrower is still
unemployed at the end of the forbearance period, the borrower
will be considered for a HAMP foreclosure alternative, like
HAFA or PRA.\87\ A unique feature of this program is that no
TARP funds are obligated to it.
---------------------------------------------------------------------------
\83\ U.S. Department of the Treasury, Housing Program Enhancements
Offer Additional Options for Struggling Homeowners (Mar. 26, 2010)
(online at www.financialstability.gov/latest/pr_03262010.html).
\84\ U.S. Department of the Treasury, Home Affordable Unemployment
Program, Supplemental Directive 10-04 (May 11, 2010) (online at
www.hmpadmin.com//portal/programs/docs/hamp_servicer/sd1004.pdf)
(hereinafter ``Supplemental Directive 10-04'').
\85\ Servicers have the discretion to extend the minimum
forbearance period in increments in accordance with investor and
regulatory guidelines.
\86\ See Supplemental Directive 10-04, supra note 84.
\87\ See Supplemental Directive 10-04, supra note 84.
---------------------------------------------------------------------------
Recent research has suggested that the loss of a job is a
significant factor in a homeowner's determination of whether or
not to default on a loan and that foreclosure prevention policy
that addresses this finding would be more effective than
traditional modifications. Researchers at the Federal Reserve
Banks of Boston and Atlanta conclude that unaffordable loans,
defined as those with high monthly payments relative to income
at the time of origination, are unlikely to be the main reason
borrowers default. Their analysis concludes a better served
policy would be one that cushions the immediate effects of job
loss or other adverse life events instead of conducting
mortgage modifications aimed at long-term affordability.\88\ As
nearly 60 percent of borrowers report loss of income through
reduction in hours or lost jobs as the primary reason for
permanent modification, Treasury's unemployment program has the
potential to reduce significantly foreclosures prompted by job
loss.\89\
---------------------------------------------------------------------------
\88\ Christopher Foote et al., Reducing Foreclosures: No Easy
Answers, Federal Reserve Bank of Atlanta Working Paper No. 2009-15 (May
2009) (online at www.frbatlanta.org/filelegacydocs/wp0915.pdf)
(hereinafter ``Federal Reserve Bank of Atlanta Working Paper'').
\89\ Data provided by Treasury.
---------------------------------------------------------------------------
Home Affordable Foreclosure Alternatives (HAFA)
Announced on November 30, 2009,\90\ but not effective until
April 5, 2010,\91\ Treasury's HAFA program was created to
encourage the use of short sales and deeds-in-lieu of
foreclosure for HAMP-eligible borrowers unable to qualify for
modifications of currently underwater mortgages. Servicers
agree to forfeit the ability to seek a deficiency judgment in
exchange for borrowers engaging in short sales or issuing a
deed-in-lieu of foreclosure. Essentially, a servicer agrees to
accept the property itself in satisfaction of a borrower's
mortgage obligation. All parties receive financial incentives
in the form of relocation assistance, one-time completion, and
reimbursement to release subordinate liens.
---------------------------------------------------------------------------
\90\ U.S. Department of the Treasury, Making Home Affordable: HAMP
Update--New Program Offers Borrowers Foreclosure Alternatives (Nov. 30,
2009) (online at www.hmpadmin.com/portal/news/docs/2009/
hampupdate113009.pdf).
\91\ U.S. Department of the Treasury, Making Home Affordable: HAMP
Updates--New Supplemental Directives Issued (Mar. 26, 2010) (online at
www.hmpadmin.com/portal/news/docs/2010/hampupdate032610.pdf).
---------------------------------------------------------------------------
Second Lien Modification Program (2MP)
The 2MP was announced on April 28, 2009, and went into
effect on August 14, 2009.\92\ This program was created to
address the issue of homeowners remaining distressed even after
their first liens were modified because there was also a second
lien on the property. Borrowers are eligible to apply for the
2MP after their corresponding first liens have been modified
under HAMP. Although servicer participation is voluntary, once
on board, servicers must modify or extinguish the second liens
of all eligible borrowers.\93\ All 2MP modifications must
consist of: an interest rate reduction, an extension of term
years matching that of the first lien modification, and
principal forbearance or principal reduction matching the
percentage of any principal forbearance or reduction of the
first lien.\94\ All parties receive incentive payments for
their participation.
---------------------------------------------------------------------------
\92\ U.S. Department of the Treasury, Making Home Affordable:
Second Lien Modification Program Details Announced (Aug. 14, 2009)
(online at www.hmpadmin.com/portal/news/docs/2009/
hampupdate081409.pdf).
\93\ U.S. Department of the Treasury, Making Home Affordable
Program, Policy Update, Supplemental Directive 10-16 (Nov. 23, 2010)
(online at www.hmpadmin.com/portal/programs/docs/hamp_servicer/
sd1016.pdf).
\94\ Id.
---------------------------------------------------------------------------
Hardest Hit Fund (HHF)
The HHF provides TARP money to state-run foreclosure
mitigation programs in specific states hit hardest by home
value decreases and high unemployment rates. In a series of
announcements, Treasury has stated that 18 states and the
District of Columbia are eligible for HHF funding. Before
receiving the funds, eligible states must submit and receive
approval for their plans to use the money. To date, there have
been four rounds of funds approved and expended under the HHF.
The first round of HHF funding was announced on February 19,
2010, and the plan to distribute the first round of funds was
approved on June 23, 2010. The first-round money went to
qualifying states where the average home price declined by more
than 20 percent from its peak: Arizona, California, Florida,
Michigan, and Nevada.\95\ The second-round funding went to the
top five states (excluding those included in the first round)
with the highest shares of their state populations living in
counties in which the unemployment rate exceeded 12 percent, on
average, over the twelve calendar months in 2009. The states
receiving the second round of funding are North Carolina, Ohio,
Oregon, Rhode Island, and South Carolina.\96\ The third-round
funding went to qualifying states, including those that
received funding in prior rounds, with an unemployment rate at
or above the national average during the previous 12 months:
Alabama, California, Florida, Georgia, Illinois, Indiana,
Kentucky, Michigan, Mississippi, Nevada, New Jersey, North
Carolina, Ohio, Oregon, Rhode Island, South Carolina,
Tennessee, and the District of Columbia.\97\ The fourth-round
dollars were provided to existing HHF participants to be used
in approved programs.\98\ A total of $7.6 billion in TARP funds
has been allocated for the HHF.
---------------------------------------------------------------------------
\95\ U.S. Department of the Treasury, Update on HFA Hardest-Hit
Fund (Mar. 5, 2010) (online at www.makinghomeaffordable.gov/
pr_03052010.html).
\96\ U.S. Department of the Treasury, Update to the HFA Hardest Hit
Fund Frequently Asked Questions, at 4 (Mar. 29, 2010) (online at
financialstability.gov/docs/
Hardest%20Hit%20public%20QA%200%2029%2010.pdf).
\97\ U.S. Department of the Treasury, Troubled Asset Relief Program
Section 105(a) Report--August 2010 (Sept. 10, 2010) (online at
www.financialstability.gov/docs/105CongressionalReports/
August%202010%20105(a)%20Report_final_9%2010%2010.pdf).
\98\ U.S. Department of the Treasury, Troubled Asset Relief
Program--Two Year Retrospective (Oct. 2010) (online at
www.financialstability.gov/docs/
TARP%20Two%20Year%20Retrospective_10%2005%2010_transmittal%20letter.pdf)
(hereinafter ``TARP Two Year Retrospective'').
---------------------------------------------------------------------------
All state recipients of HHF funds are using at least a
portion of those funds to aid unemployed homeowners.
California, for example, has developed a program targeting
unemployed homeowners. Under that program, homeowners could
receive a mortgage subsidy for up to 6 months, with monthly
benefit of up to $1,500 or 50 percent of existing total monthly
mortgage payment. Other states, such as Arizona, have developed
programs intended to help any struggling homeowner with a
demonstrated hardship and who meets certain qualifications.
Among the recognized hardships are unemployment as well as
medical condition, divorce, and death. Many of the proposed HHF
programs are aimed at low- to moderate-income families,
requiring those eligible to have a household income of no more
than 120 or 140 percent of median household income. These
programs are all in the early stages of implementation, and
there are not yet data available on the programs' results.
In August 2010, HUD announced the $1 billion Emergency
Homeowners Loan Program (EHLP) to complement Treasury's Hardest
Hit Fund and continue to target unemployed borrowers at risk of
foreclosure. The program will provide assistance to homeowners
in Puerto Rico and 32 states not funded by Treasury's
Innovation Fund for Hardest Hit Housing Markets program through
a declining balance, deferred payment bridge loan for up to
$50,000 to assist borrowers with arrearages and mortgage
payments for up to 24 months. HUD intends to start taking
applications by the end of this year.\99\
---------------------------------------------------------------------------
\99\ The HUD EHLP loans will be non-recourse, zero-interest,
subordinate loans. U.S. Department of Housing and Urban Development,
Emergency Homeowner Loan Program--Summary (Oct. 8, 2010) (online at
www.hud.gov/offices/hsg/sfh/hcc/msgs/EHLP100810.pdf). The Pennsylvania
Housing Finance Agency implemented a similar program, the Homeowners'
Emergency Mortgage Assistance Loan Program (HEMAP), which targets
borrowers facing foreclosure who are dealing with financial hardship
due to circumstances beyond their control (i.e., unemployment, wage
loss, illness). HEMAP provides a loan of up to $60,000 that cannot
exceed a period of 24 months (or 36 months in the case of unemployment)
to cover mortgage payments. The loan is considered a mortgage lien
against the homeowner's property. Non-continuing loan recipients must
begin repayment immediately following loan closing at a monthly amount
based on income but not less than $25 per month. Payment increases are
based on 40 percent of a homeowner's net monthly income less total
monthly housing expense. Continuing loan recipients begin repayment
immediately following termination of continuing loan disbursements.
Pennsylvania Housing Finance Agency, Frequently Asked Questions (FAQ)
(online at www.phfa.org/hsgresources/faq.aspx) (accessed Dec. 1, 2010).
---------------------------------------------------------------------------
FHA Short Refinance Program
The FHA Short Refinance Program was announced on March 26,
2010,\100\ and went into effect on September 7, 2010.\101\ This
program was created to refinance non-FHA-insured underwater
mortgages into above-water, FHA-insured mortgages. Eligible
borrowers are not guaranteed a refinance, and program
participation is voluntary for servicers on a case-by-case
basis. Under the terms of the refinance, the existing lien
holder is given a cash payment equal to 97.75 percent of the
current home value and issued a subordinate second lien for up
to 17.25 percent of the current home value, if applicable.
Thus, the existing lien holder can retain an interest in up to
115 percent of the current home value, but any interest above
115 percent of the current home value is deemed forgiven. The
new FHA-insured mortgage payments can be no more than 31
percent of the borrower's gross monthly income.
---------------------------------------------------------------------------
\100\ U.S. Department of Housing and Urban Development, Housing
Program Enhancements Offer Additional Options for Struggling Homeowners
(Mar. 26, 2010) (online at portal.hud.gov/portal/page/portal/HUD/press/
press_releases_media_advisories/2010/HUDNo.10-058).
\101\ U.S. Department of Housing and Urban Development, FHA
Launches Short Refi Opportunity for Underwater Homeowners (Aug. 6,
2010) (online at portal.hud.gov/portal/page/portal/HUD/press/
press_releases_media_advisories/2010/HUDNo.10-173).
---------------------------------------------------------------------------
Other Agency Programs
Lastly, Treasury has coordinated with the FHA, the United
States Department of Agriculture, and the VA to encourage
modification of the mortgages they insure. These agencies have
each developed their own versions of HAMP, in support of which
Treasury provides incentive payments to servicers and
borrowers. No incentive payments are made to investors, because
they already have the benefit and protection of a government
loan guarantee.
As stated earlier, there is not yet sufficient data
available to analyze whether these programs are achieving their
goals or operating effectively. While the Panel appreciates the
length of time involved in moving a program from inception to
data validation, the timely availability of data is a key
component of transparency and oversight, and the Panel urges
Treasury to continue working towards this goal. For an analysis
of the structural aspects of these programs, and the Panel's
recommendations as to addressing new issues surrounding
foreclosure mitigation programs overall, see the Panel's April
2010 report.
G. Barriers to Success
1. What Are HAMP's Goals?
In considering whether a program has been a success, the
goals and metrics outlined for that program offer an important
yardstick. When the stated objectives are limited or not
meaningful, the scope of oversight and analysis is narrowed.
Treasury's only explicitly stated goal for HAMP is to offer
three to four million homeowners lower mortgage payments
through a modification.\102\ As noted in the Panel's April
report, the meaning of even that one goal has shifted over
time. Treasury and the administration initially stated a goal
of ensuring that three to four million homeowners avoid
foreclosure and remain in their homes.\103\ This nebulous
target gave rise to uncertainty about Treasury's goal: did
Treasury intend HAMP's three to four million target to refer to
permanent modifications, as was widely assumed? Or did the goal
refer to trial modifications started, or merely trial
modifications offered?
---------------------------------------------------------------------------
\102\ U.S. Department of the Treasury, Making Home Affordable
Program: Servicer Performance Report Through January 2010, at 2 (Feb.
18, 2010) (online at www.financialstability.gov/docs/press/
January%20Report%20FINAL%2002%2016%2010.pdf).
\103\ April 2010 Oversight Report, supra note 1, at 63; Office of
the Special Inspector General for the Troubled Asset Relief Program,
Factors Affecting Implementation of the Home Affordable Modification
Program (Mar. 25, 2010) (online at sigtarp.gov/reports/audit/2010/
Factors_Affecting_Implementation_of_the_Home_Affordable_Modification_Pro
gram.pdf) (hereinafter ``SIGTARP Report--Factors Affecting
Implementation of HAMP'').
---------------------------------------------------------------------------
A year into the program's life, Treasury clarified its goal
by stating its intended objective is to offer three to four
million modifications, with one Treasury official estimating
that the number of permanent modifications would be only
between 1.5 and 2 million.\104\ This not only cut what was
assumed to be the initial modification goal in half but also
established a relatively meaningless measurement, trial
modification offers, as the touchstone for program success.
---------------------------------------------------------------------------
\104\ SIGTARP Report--Factors Affecting Implementation of HAMP,
supra note 103, at 10.
---------------------------------------------------------------------------
Recently, Treasury has claimed that the existence of a
government foreclosure program--HAMP--accelerated the number of
proprietary modifications.\105\ While counting these non-HAMP
modifications as successful ``HAMP'' modifications would allow
Treasury to come closer to its three to four million target,
the lack of public data by servicers on the terms of
proprietary modifications hinders the ability to assess
properly whether these modifications are actually helping
homeowners. While Treasury believes that HAMP accelerated the
pace of proprietary modifications, when pressed, Treasury
acknowledges that there is no clear causal link between HAMP
and proprietary modifications.\106\
---------------------------------------------------------------------------
\105\ Treasury conversations with Panel staff (Oct. 21, 2010). See
TARP Two Year Retrospective, supra note 98, at 67 (``A cancelled trial
modification does not mean that the program has completely failed a
homeowner or that the borrower will inevitably face foreclosure: HAMP
explicitly requires servicers to consider these borrowers for other
foreclosure prevention options including proprietary modifications or
other options like a short sale or deed-in-lieu of foreclosure that
also prevent a foreclosure sale. The broader HAMP program provides
borrowers with a range of assistance; success can only be measured on
an aggregate basis, taking account of homeowners' individual situations
and outcomes.'').
\106\ Treasury conversations with Panel staff (Oct. 21, 2010).
---------------------------------------------------------------------------
Treasury has not explicitly stated its own definition of
program success, noting only that it plans to evaluate success
in the context of the state of the economy over time,\107\ and
remains elusive on this issue when questioned by the Panel and
others.\108\ Using its singular goal, however, one can
reasonably infer that success, from Treasury's point of view,
is measured by reaching three to four million trial
modification offers. Under this definition, the program has not
been ``successful'' to date, as nearly 1.65 million trial plan
offers have been extended through October 2010. As noted in
Section D above, trial modification offers have significantly
declined since September 2009 and seem to have leveled off
recently. Although Treasury is halfway to its goal with two
years remaining in the program's life, and the long-term trend
is uncertain, at the current rate of trial offers, HAMP is not
on target to meet its one goal.\109\ Assuming a continued pace
of 20,000 to 30,000 offers per month going forward,\110\ HAMP
will provide a cumulative total of only 2.17 to 2.43 million
trial modification offers.\111\ And if the current pace of new
permanent modifications continues, the program will result in
only about 1.26 million permanent modifications, many of which
will likely end in redefaults.\112\
---------------------------------------------------------------------------
\107\ Treasury conversations with Panel staff (Oct. 21, 2010).
\108\ See Congressional Oversight Panel, Testimony of Phyllis
Caldwell, chief of the Homeownership Preservation Office, U.S.
Department of the Treasury, Transcript: COP Hearing on TARP Foreclosure
Mitigation Programs (Oct. 27, 2010) (publication forthcoming) (online
at cop.senate.gov/hearings/library/hearing-102710-foreclosure.cfm)
(hereinafter ``Transcript Testimony of Phyllis Caldwell''). See also
Congressional Oversight Panel, Written Responses of Timothy F.
Geithner, secretary, U.S. Department of the Treasury, COP Hearing with
Treasury Secretary Timothy Geithner (June 22, 2010) (online at
cop.senate.gov/documents/testimony-062210-geithner-qfr.pdf); This Week
with Jake Tapper, Interview with Treasury Secretary Timothy Geithner,
ABC News Television Broadcast (Feb. 7, 2010) (online at abcnews.go.com/
ThisWeek/week-transcript-treasury-secretary-timothy-geithner/
story?id=9758951).
\109\ MHA Servicer Performance Report, supra note 38, at 2.
\110\ The Panel previously noted in its October 2009 report that
Treasury officials stated a goal of modifying 25,000 to 30,000 loans
per week; however, the actual pace is currently only 20,000 to 30,000
trial modifications per month. October 2009 Oversight Report, supra
note 15, at 4; Data provided by Treasury.
\111\ Panel staff calculation using assumed range of 20,000 to
30,000 trial modification offers per month over the remaining 26 months
of the program's life, added to the actual cumulative total of 1.65
million offers through October 2010.
\112\ Panel staff calculation that assumes 28,311 new permanent
modification each month through December 2012, based on the average
number of new permanent modifications between August-October 2010.
---------------------------------------------------------------------------
The Congressional Budget Office (CBO) last month projected
that Treasury will spend only $12 billion on all TARP housing
programs, including HAMP and the Hardest Hit Fund, out of the
$45.6 billion in TARP funds allocated for those programs.\113\
While the CBO did not publish a breakdown of how it expects the
projected $12 billion to be divided between the various TARP
housing programs, if one assumes that the Hardest Hit Fund-
recipient states will spend the $7.6 billion allocated to them
in grants, that would leave only $4.4 billion to be spent on
HAMP and the other TARP housing programs.\114\
---------------------------------------------------------------------------
\113\ Congressional Budget Office, Report on the Troubled Asset
Relief Program--November 2010, at 7 (Nov. 29, 2010) (online at
www.cbo.gov/ftpdocs/119xx/doc11980/11-29-TARP.pdf) (hereinafter ``CBO
Report on the TARP--November 2010'').
\114\ Through November 2010, Treasury has spent $652.4 million on
the first-lien portion of HAMP, $71.3 million on HPDP, $4.3 million on
HAFA, $959,1335 on 2MP, and $8,990 on FHA-HAMP. Data provided by
Treasury.
---------------------------------------------------------------------------
Treasury has declined to state publicly any metrics or
benchmarks by which HAMP should be judged, a fact that has
frustrated Congress and TARP oversight bodies, and has made
clear to the Panel that it has no other unarticulated goals for
HAMP.\115\ As stated earlier, the absence of additional, more
meaningful goals hinders the Panel's ability to perform
oversight and adequately assess HAMP's level of success or
failure. The Panel has expressed concerns about Treasury's lack
of transparency and accountability in regards to its goals for
HAMP, noting in its April report that ``. . . Treasury needs to
take care to communicate its goals, its strategies, and its
measures of success for its programs. Its stated goal of
modifying three to four million mortgages has proven too vague.
. . .'' \116\ Further, the Panel has continued to urge Treasury
to develop specific objectives for other program measures.\117\
In its most recent Quarterly Report to Congress, the Special
Inspector General for the Troubled Asset Relief Program
(SIGTARP) clearly pointed out Treasury's lack of meaningful
program benchmarks or measures of success:
---------------------------------------------------------------------------
\115\ Treasury conversations with Panel staff (Oct. 21, 2010).
\116\ April 2010 Oversight Report, supra note 1, at 95.
\117\ Congressional Oversight Panel, Transcript: COP Hearing on
TARP Foreclosure Mitigation Programs (Oct. 27, 2010) (publication
forthcoming) (online at cop.senate.gov/hearings/library/hearing-102710-
foreclosure.cfm).
SIGTARP, along with the other TARP oversight bodies
(GAO and the Congressional Oversight Panel), has long
argued that Treasury should adopt meaningful benchmarks
and goals for HAMP, including setting forth its
expectations and goals for the most meaningful aspect
of HAMP--permanent modifications that offer secure,
sustainable relief to the program's intended
beneficiaries. Remarkably, Treasury has steadfastly
rejected these recommendations, and now finds itself
defending a program that is failing to meet TARP's goal
of ``preserv[ing] homeownership.'' . . . and it has
steadfastly and explicitly declined to articulate well-
considered, consistent, and meaningful success
standards for HAMP. . . . Instead it continues to cite
the number of HAMP trial modifications, as opposed to
permanent modifications, as an indication of success. .
. . While it may be true that many homeowners may
benefit from temporarily reduced payments even though
the modification ultimately fails, Treasury's claim
that ``every single person'' who participates in HAMP
gets ``a significant benefit'' is either hopelessly out
of touch with the real harm that has been inflicted on
many families or a cynical attempt to define failure as
success. Worse, Treasury's apparent belief that all
failed trial modifications are successes may preclude
it from seeking to make the meaningful changes
necessary to provide the ``sustainable'' mortgage
relief for struggling families it first promised.\118\
---------------------------------------------------------------------------
\118\ Senate Committee on Finance, Written Testimony of Neil
Barofsky, Special Inspector General for the Troubled Asset Relief
Program, An Update on the TARP Program, at 3 (July 21, 2010) (online at
finance.senate.gov/imo/media/doc/072110nbtest.pdf).
During a July Senate Finance Committee hearing with TARP
oversight bodies, Senator Charles Grassley (R-Iowa) similarly
noted, ``Moreover, Treasury still has not established
performance goals or benchmarks for HAMP, meaning that there is
no effective way for us to know whether this 50 billion dollar
program is accomplishing its intended purpose. That's not
accountability, that's not transparency--that's just more
taxpayer money flying out the window.'' \119\ Richard Hillman,
managing director of financial markets and community investment
team at the Government Accountability Office (GAO), emphasized
that even as Treasury created new programs and modified
existing programs under the MHA initiative, it failed to
redefine the program's reach and goals, undercutting program
transparency.\120\ Despite Treasury's frequent changes to the
program, despite HAMP's inadequate performance compared to
Treasury's initially stated goal, and despite the urgings of
Congress and oversight bodies to articulate a meaningful
benchmark for success beyond the initial, opaque goal of
``reaching'' three to four million homeowners, Treasury has
failed to do so.
---------------------------------------------------------------------------
\119\ Senate Committee on Finance, Opening Statement of Senator
Chuck Grassley, An Update on the TARP Program, at 2 (July 21, 2010)
(online at finance.senate.gov/imo/media/doc/072110CG.pdf).
\120\ Senate Committee on Finance, Written Testimony of Richard
Hillman, managing director, Financial Markets and Community Investment
Team, U.S. Government Accountability Office, An Update on the TARP
Program, at 11 (July 21, 2010) (online at finance.senate.gov/imo/media/
doc/072110rhtest.pdf) (``Treasury announced several potentially
substantial new HAMP-funded efforts in March 2010, but did not say how
many borrowers these programs were intended to reach. . . .We noted
that Treasury needed to ensure that future public reporting on this
program [principal reduction initiative] provided program transparency
and address the potential question of whether borrowers were being
treated fairly.'').
---------------------------------------------------------------------------
As SIGTARP noted in its March 2010 report on HAMP, ``the
anticipated benefits of a program . . . and how a program's
success or failure is defined are important to provide a
reference point for discussions about whether a program is
worth the resources devoted to it and whether the program is
functioning as it should.'' \121\ While Treasury has clarified
its target goal of trial modification offers, it has not
provided a clear definition of program success. Meeting a goal
that has no basis in whether borrowers have actually received
help does not make a program successful, and, as noted in
Section D.2, intense focus on one goal over another can limit
broader success. In order to demonstrate the legitimacy and
effectiveness of its chief foreclosure mitigation program, the
Panel believes Treasury should clearly articulate its
definition of success for HAMP, allowing the defined benchmarks
to be the true measure of HAMP's worth.
---------------------------------------------------------------------------
\121\ SIGTARP Report--Factors Affecting Implementation of HAMP,
supra note 103, at 8.
---------------------------------------------------------------------------
The Panel stated in its April 2010 report that Treasury's
success with HAMP will be measured by the number of homeowners
who avoid foreclosure, not the number of mortgages
modified.\122\ It is difficult to track precisely the number of
homeowners who have avoided foreclosure because of HAMP, as
there are no public data for foreclosure avoidance due to trial
modification only, so the most recent modification results are
the closest proxy. As of October 31, 2010, HAMP has 156,408
active trial modifications and 483,342 active permanent
modifications.\123\ Thus, 639,750 homeowners are currently
avoiding foreclosure, on either a temporary basis or for five
years, through HAMP.\124\ Foreclosure starts since HAMP's
inception, on the other hand, reached a total of approximately
4.4 million in October 2010.\125\ These results mean that since
HAMP was unveiled, there have been just over nine foreclosure
starts for every one permanent, or five-year, HAMP
modification.\126\
---------------------------------------------------------------------------
\122\ April 2010 Oversight Report, supra note 1, at 95.
\123\ Data provided by Treasury.
\124\ Data provided by Treasury. (The trial period typically lasts
for three months to provide immediate relief and ensure that the new
payment plan will work for the borrower.)
\125\ The total of 4.4 million starts includes foreclosure starts
from March 2009, the time of HAMP's inception, through October 2010.
The number of foreclosure starts is an imperfect data point, as it is
incomplete, potentially redundant, and is extrapolated out to determine
national totals; however, it is helpful in providing the ability to
track trends in foreclosures. These flaws emphasize the need for the
government to collect data on foreclosure starts and completions on a
nationwide basis. Over this same time period, the number of foreclosure
completions, or actual foreclosure sales, reached nearly 1.8 million.
HOPE NOW Alliance Foreclosure Data.
\126\ Panel staff calculation using total foreclosure starts
through October 2010 of 4.4 million and total permanent HAMP
modifications that are active or paid off (excluding those that were
disqualified or cancelled) of 483,833 as of October 31, 2010. Data
provided by Treasury.
---------------------------------------------------------------------------
While the number of foreclosure starts that were HAMP
eligible is unknown, as of October 31, 2010, only 29 percent of
loans 60 or more days delinquent were estimated to be eligible
for HAMP modification, as noted in Section D.\127\ As the Panel
noted in its April 2010 report, certain HAMP exclusions prevent
aid from flowing to speculators, wealthy borrowers, those
covered by other modification programs, or those unlikely to
benefit from HAMP. Other HAMP exclusions, however, such as
borrowers with nonparticipating servicers and those with DTI
less than 31 percent, may be preventing a significant number of
modifications and more positive program results. Foreclosure
rates remain at historically very elevated levels, and housing
prices have not recovered significantly compared to the highs
reached in the middle of the decade, as indicated in Figure 12
below. Therefore, it is unclear whether HAMP has made
appreciable improvement in the unstated but implicit goals of
reducing foreclosures and stabilizing the housing markets.
---------------------------------------------------------------------------
\127\ Data provided by Treasury.
---------------------------------------------------------------------------
FIGURE 12: FORECLOSURE RATE AND HOME VALUES \128\
---------------------------------------------------------------------------
\128\ Bloomberg data (graph created from FORLTOTL and SPCSUSS
indices) (accessed Dec. 9, 2010). Data for the Case-Shiller Price Index
are normalized to 100 at March 2000.
[GRAPHIC] [TIFF OMITTED] 62622A.012
Despite HAMP's lack of clearly articulated goals and
metrics, the program has had a positive impact in another way,
even if inadvertently so. It has provided standardization of
the mortgage modification process, an issue noted as a
significant industry problem in the past,\129\ by encouraging
the use of a common model, reporting platform, and
affordability measurement.\130\ Phyllis Caldwell, chief of
Treasury's Homeownership Preservation Office, noted at the
Panel's most recent hearing on foreclosure mitigation, ``. . .
it [HAMP] has helped transform the way the entire mortgage
servicing industry operates. HAMP established a universal
affordability standard, a 31 percent debt to income ratio.''
\131\ Ms. Schwartz reiterated these sentiments, stating, ``. .
. it's very integral and important that the government stepped
forward to put a protocol in place for modifications and that
this protocol would have been very difficult to get into place
otherwise . . . . And HAMP offers uniformity of approach, which
is fair and systematic in its approach for all homeowners at
risk.'' \132\ While Treasury was not the first entity to
propose or employ a 31 percent standard, through the parameters
of HAMP and substantial servicer participation, Treasury
solidified the wide acceptance of 31 percent as an industry
affordability standard.\133\
---------------------------------------------------------------------------
\129\ See Joseph R. Mason, Mortgage Loan Modification: Promises and
Pitfalls, at 17 (Oct. 3, 2007) (online at www.lebow.drexel.edu/PDF/
Docs/20071003LoanModificationPaper.pdf).
\130\ Transcript Testimony of Phyllis Caldwell, supra note 108.
\131\ Transcript Testimony of Phyllis Caldwell, supra note 108.
\132\ Testimony of Faith Schwartz, supra note 78.
\133\ In testimony before the Senate Banking, Housing, and Urban
Affairs Committee in November 2008, Martin D. Eakes, CEO of Self-Help
and Center for Responsible Lending, noted the industry standard of 38
percent DTI but urged Congress to push Treasury to utilize TARP to
create a modification program that required 31 percent DTI, referencing
the FDIC proposal at the time to use TARP funds for a modification
program with a 31 percent HTI (housing-to-income). Eakes noted that the
standard had moved over time from a lending standard of 25 percent DTI
to the current 38 percent. See Senate Committee on Banking, Housing,
and Urban Affairs, Written Testimony of Martin D. Eakes, chief
executive officer, Self-Help Credit Union and the Center for
Responsible Lending, Oversight of the Emergency Economic Stabilization
Act: Examining Financial Institution Use of Funding Under the Capital
Purchase Program, at 8 (Nov. 13, 2008) (online at banking.senate.gov/
public/ index.cfm? FuseAction=Files. View&FileStore_id=26c73c7a-bafc-
4ff1-88f9-e984d672c9f6). Gregory Palm, Executive Vice President and
General Counsel of The Goldman Sachs Group, also testified at the
hearing that Litton Loan Servicing, Goldman Sachs' affiliate, was
already employing a 31 percent DTI standard. See Senate Committee on
Banking, Housing, and Urban Affairs, Testimony of Gregory K. Palm,
executive vice president and general counsel, The Goldman Sachs Group,
Inc., Transcript: Oversight of the Emergency Economic Stabilization
Act: Examining Financial Institution Use of Funding Under the Capital
Purchase Program, at 8 (Nov. 13, 2008) (publication forthcoming)
(online at banking.senate.gov/public/ index.cfm? FuseAction= Hearings.
Hearing& Hearing_ ID=1d38de7d- 67db- 4614-965b-edf5749f1fa3). The
IndyMac Federal program implemented by the FDIC utilized a DTI standard
of 38 percent. Federal Deposit Insurance Corporation, FDIC Implements
Loan Modification for Distressed IndyMac Mortgage Loans (Aug. 20, 2008)
(online at www.fdic.gov/news/news/press/2008/pr08067.html).
---------------------------------------------------------------------------
Nonetheless, when viewed in light of the millions of
foreclosure completions since 2007 and the large number waiting
in the pipeline due to continued hardships from high
unemployment rates and lower home values, HAMP has failed to
make a significant dent in the number of foreclosures and does
not appear likely to do so in the future.
2. Factors Affecting HAMP Success
In many cases, mortgage modifications are economically
rational; the investor loses less money on a modification than
the losses that it would incur under a foreclosure. In a well-
functioning market, modifications, principal forgiveness, and
refinancings will occur as long as they are in the best
interests of all parties. If a mortgage modification allows a
lender or investor to sustain a smaller loss than a foreclosure
and allows a borrower to remain in his home at a sustainably
affordable monthly payment, then incentives to modify the loan
should be unnecessary. However, as discussed in Section B, the
housing market has been in a state of disequilibrium, and the
private sector failed to conduct modifications on its own,
which triggered government intervention in the form of HAMP and
the use of incentives to enhance the likelihood that the
decision scale tipped in favor of modification.
HAMP is premised upon the idea of economically rational
mortgage modifications--beneficial to both borrower and
investor--as embodied in its NPV model. The model is explicitly
designed to determine whether a modification or a foreclosure
makes more sense economically for the mortgage holder. If the
NPV model shows that it is economically better to modify the
mortgage, participating servicers are compelled to make the
modification. HAMP also added financial incentive payments for
investors, servicers, and borrowers, some of which are included
in the NPV model calculations, to promote participation
further. These incentives were designed to encourage mortgage
modifications by lessening the financial burden on all parties,
especially given the dearth of private-sector modifications at
the outset of HAMP and to offset any disincentives.\134\
---------------------------------------------------------------------------
\134\ Transcript Testimony of Phyllis Caldwell, supra note 108
(``No, I think it's that when you look back at the beginning of the
program--again, HAMP was a--is a voluntary program--getting the
servicers, the investors and the homeowners to the table and to change
their business model to--to do that required some incentives.'').
---------------------------------------------------------------------------
As of the October 3, 2010 deadline for enrolling in HAMP,
105 servicers,\135\ covering nearly 90 percent of all non-GSE
mortgage loans, had signed Servicer Participation
Agreements.\136\ Servicers of GSE mortgages are required to
evaluate mortgages for HAMP. Thus, the program clearly has
broad coverage of the mortgage market.
---------------------------------------------------------------------------
\135\ Treasury conversations with Panel staff (Oct. 21, 2010).
\136\ TARP Two Year Retrospective, supra note 98, at 65.
---------------------------------------------------------------------------
Yet, despite this broad commitment by servicers to
participate in a program that offers additional payments on top
of an often economically rational decision, HAMP's results
remain underwhelming. Why has HAMP failed to deliver? It is
unclear what the additional cost would be to provide sufficient
incentives to induce greater numbers of modifications, and
there does not appear to be one single answer as to why HAMP
has not resulted in more modifications. Rather, it appears that
multiple factors have inhibited otherwise sensible mortgage
modifications.
a. Voluntary Nature of the Program
A key characteristic that has limited Treasury's influence
over servicers is the voluntary nature of the program for
servicers of non-GSE loans.
Although Treasury has been able to pressure servicers to
sign participation agreements, Treasury has little ability to
pressure servicers when it comes to actually making
modifications. This structure has created an imbalance of power
between Treasury and the servicers, and has hindered Treasury's
ability to influence, oversee, and compel servicers into more
aggressive action. According to Treasury officials, ``Because
it is a voluntary program, our abilities to enforce specific
performance are extremely limited,'' \137\ and it ``makes
aggressive enforcement difficult.'' \138\ The California
Reinvestment Coalition has noted that the voluntary nature of
the program is a hindrance to success. Its recent survey of
mortgage counselors indicated a growing frustration with
servicers not following HAMP requirements and Treasury not
enforcing servicer compliance.\139\ Treasury has focused on
establishing a tone of program compliance, directing servicers
to fix issues instead of doling out penalties.\140\ Although
Treasury oversees servicers and encourages compliance, there is
little real accountability for servicers that fail to adhere to
program standards, lose borrower submitted paperwork,
unnecessarily delay the process, or otherwise don't make
modifications. In describing this system of incentives with no
real sticks, Professor Katherine Porter, a professor of law who
testified at the Panel's October hearing, said, ``. . .
[servicers] have gorged themselves at a buffet of carrots, and
they're still not doing what we want them to do.'' \141\
---------------------------------------------------------------------------
\137\ Treasury conversations with Panel staff (Sept. 10, 2010).
\138\ Treasury conversations with Panel staff (Oct. 21, 2010).
\139\ California Reinvestment Coalition, Chasm Between Words and
Deeds IV: HAMP Is Not Working, at 2, 9-11 (July 2010) (online at
www.calreinvest.org/system/assets/234.pdf) (hereinafter ``Chasm Between
Words and Deeds IV: HAMP Is Not Working'').
\140\ Treasury conversations with Panel staff (Oct. 21, 2010).
\141\ Congressional Oversight Panel, Testimony of Katherine Porter,
professor of law, University of Iowa College of Law, Transcript: COP
Hearing on TARP Foreclosure Mitigation Programs (Oct. 27, 2010)
(publication forthcoming) (online at cop.senate.gov/ hearings/ library/
hearing-102710-foreclosure.cfm).
---------------------------------------------------------------------------
The Panel has previously noted that servicers need to face
``meaningful monetary penalties'' for noncompliance with SPAs
and denial of modification for an unexplained reason, a breach
of their contractual obligations under HAMP SPAs.\142\ However,
Treasury has seemed reluctant to do more than vaguely threaten
the potential for clawbacks of HAMP payments.\143\ Despite
rampant anecdotal stories of servicer errors, to date, no
servicer has experienced a clawback or other financial
repercussion. The steepest penalty Treasury has levied to date
has been withholding payments to servicers due to data issues.
But after remedying the problem, these servicers will receive
any missed incentive payments, despite a prolonged remediation
process.\144\ By signing HAMP participation agreements,
servicers indicated that they agreed with the NPV model and
would provide modifications based on the model's results. They
knew they were legally obligated to do so when they signed.
---------------------------------------------------------------------------
\142\ April 2010 Oversight Report, supra note 1, at 55.
\143\ See House Committee on Financial Services, Written Testimony
of Herbert M. Allison, Jr., assistant secretary for financial
stability, U.S. Department of the Treasury, The Private Sector and
Government Response to the Mortgage Foreclosure Crisis, at 3 (Dec. 8,
2009) (online at financial services. house. gov/ media/ file/ hearings/
111/ herb_allison.pdf); Rob Chrisman, Cinco de Mayo: HAMP Incentives
Denied to NonCompliant Servicers? High Balance Loans; PennyMac
Earnings; Flood Zone Alert, Mortgage News Alert (May 5, 2010) (online
at www.mortgagenewsdaily.com/ channels/ pipelinepress/ 05052010-hamp-
greece.aspx).
\144\ Treasury noted that it is currently considering no longer
accruing missed incentive payments to the five servicers that have
failed to remedy their data issues. Treasury conversations with Panel
staff (Oct. 21, 2010).
---------------------------------------------------------------------------
An additional challenge of voluntary participation is the
issue of potential servicer withdrawal. Despite a formal
contract for those who voluntarily participate, the contract
does not explicitly address servicer withdrawal. It is unclear
whether or not servicers can withdraw without Treasury's
consent, creating uncertainty regarding the ramifications
should a servicer decide unilaterally to exit the program. The
Servicer Participation Agreements (SPAs) require participating
servicers to offer HAMP modifications to all eligible
borrowers, but the contracts do not explicitly compel servicers
to remain in the program. Some servicers have exited HAMP for
various reasons,\145\ and Treasury has indicated that in those
instances it worked with the servicer to ensure that the HAMP-
modified loans remained in the program or were transferred to
another HAMP servicer. Further, Treasury noted that its concern
with servicer exit is most applicable to the larger servicers.
Should a servicer choose to exit the program without Treasury's
consent, Treasury has indicated to the Panel that it could
potentially sue the servicer for specific performance.\146\
Treasury encouraged servicer participation by adding incentives
to the program and by sharing some of the financial burden of
modifying a loan.\147\ Servicers that do not participate also
face the public stigma of being seen as unwilling to help
homeowners avoid foreclosure.
---------------------------------------------------------------------------
\145\ Treasury conversations with Panel staff (Oct. 21, 2010).
Treasury noted that servicers might exit due to having better
proprietary modification options or because of the costs of
participating in HAMP (e.g., labor and time needed to complete
additional paperwork). Treasury indicated that the latter has occurred.
\146\ Treasury conversations with Panel staff (Oct. 21, 2010).
\147\ Servicers that make modifications to get a homeowner down to
a 38 percent mortgage DTI ratio receive aid from Treasury in the form
of half of the remaining cost of getting the borrower down to a final
31 percent DTI ratio. This amount is not paid by Treasury, though,
until the modification moves to permanent status. Servicers also
receive an up-front payment for each eligible modification meeting
program guidelines and yearly Pay for Success payments for up to three
years as long as the borrower remains in the program. Initially, these
payments were $1,000, but Treasury increased them to $1,500 to
encourage servicers to increase their use of foreclosure alternatives
and engage in additional outreach to homeowners unable to complete a
modification. U.S. Department of the Treasury, Making Home Affordable
Program Enhancements to Offer More Help for Homeowners (Mar. 25, 2010)
(online at making home affordable.gov/ docs/ HAMP%20
Improvements_Fact_% 20Sheet_ 032510%20FINAL2.pdf) (hereinafter ``MHA
Program Enhancements to Offer More Help for Homeowners'').
---------------------------------------------------------------------------
A program structure in which participation is voluntary and
the repercussions of exit are uncertain has presented a key
problem within HAMP: namely a lack of accountability for
performance. Without more useful, publicly available
performance data, transparency is compromised, and without
stringent enforcement of contractual obligations, servicer
underperformance is unpunished. Of course, such an inability to
enforce standards would be an Achilles heel in virtually any
voluntary foreclosure mitigation program design.
b. Accounting Issues
Because of the latitude participating servicers have in
ultimately granting a modification with little accountability
for performance, many external incentives or disincentives will
remain crucial in shaping the decision to modify a loan. As
discussed in the Panel's April 2010 Report, the current
accounting rules provide investors a disincentive to modify
loans under HAMP.\148\ Since HAMP loans are contractually
modified, under the current accounting rules there is immediate
loss recognition for financial institutions.\149\ For other
loans that are not modified, the current accounting rules
provide a financial institution more discretion to determine
when a loss should be recognized.\150\ In addition, write-downs
on loans would require financial institutions to boost their
regulatory capital ratios. As a result, financial institutions
are reluctant to write down mortgages since their capital
structures have already been weakened by a variety of factors,
including write-downs already taken on residential and
commercial real estate loans, losses taken on other loans due
to the recession, and recent actions by Fannie Mae and Freddie
Mac to require banks to buy back mortgages that the banks had
previously sold to them.\151\
---------------------------------------------------------------------------
\148\ See April 2010 Oversight Report, supra note 1, at 74. Under
generally accepted accounting principles (GAAP), once the terms of a
loan are contractually modified, the modified loan is accounted for as
a ``troubled debt restructuring.'' A troubled debt restructuring occurs
when the terms of a loan have been modified due to the borrower's
financial difficulties, and a long-term concession has been granted to
the borrower. Examples of such concessions include interest rate
reductions, principal forbearance, principal forgiveness, and term
extensions, all of which may be used to modify loans in HAMP.
\149\ For all restructured loans, GAAP requires that a loss be
recognized if the difference in cash flows to be received under the
modified loan is less than the cash flows of the original loan. By
nature of the modified terms of the loan under HAMP (i.e., reduction of
interest to be received and/or principal forbearance or forgiveness)
the entity's future cash flows to be received will be less than the
current loan payoff amount. See April 2010 Oversight Report, supra note
1, at 74 n. 243.
\150\ Current accounting rules require that a loss contingency is
required to be recognized only if it is probable that an asset has been
impaired and the amount of the loss can be reasonably estimated. See
April 2010 Oversight Report, supra note 1, at 74 n. 244.
\151\ Financial institutions are faced with write-downs on both
first and second lien mortgages. See April 2010 Oversight Report, supra
note 1, at 75.
---------------------------------------------------------------------------
There continues to be tension between Treasury's goal of
mitigating foreclosures and its goal of maintaining adequate
capital levels at large banks. A new proposed accounting rule
that would eliminate the disincentives to modify loans under
HAMP is currently under consideration. This rule, which is to
be adopted at the earliest in 2013, would require a financial
institution to account for all loans similar to the accounting
for HAMP loans.\152\ Unfortunately, the rule will come too late
to make a difference for HAMP, which will stop offering
modifications in 2012.
---------------------------------------------------------------------------
\152\ On May 26, 2010, the Financial Accounting Standards Board
(FASB) issued proposed Accounting Standard Update (ASU), Accounting for
Financial Instruments and Revisions to the Accounting for Derivative
Instruments and Hedging Activities--Financial Instruments (Topic 825)
and Derivatives and Hedging (Topic 815). The rule applies to all
entities and there is a phase in period for non-public entities that
have less than $1 billion in total assets. The rule would eliminate an
entity's discretion to determine when a loan impairment loss should be
recognized and would also require that loans and other financial
instruments be classified on the balance sheet at fair value. Changes
in fair value would be reported at each reporting period and would
either flow through an entity's equity or net income. An adjustment to
equity (other comprehensive income) would be made if the entity does
not plan to sell the loan. Otherwise, the fair value adjustment would
be required to flow through the income statement. See Financial
Accounting Standards Board, Accounting for Financial Instruments and
Revisions to the Accounting for Derivative Instruments and Hedging
Activities--Financial Instruments (Topic 825) and Derivatives and
Hedging (Topic 815), Financial Accounting Series FASB Exposure Draft,
at paragraph 38 at 40 (May 26, 2010) (online at www. fasb. org/cs/
BlobServer? blobcol=urldata&blob table= Mungo Blobs&blobkey=
id&blobwhere= 1175820761372&blobheader= application%2Fpdf).
---------------------------------------------------------------------------
c. Second Liens
Since the Panel's March 2009 oversight report, the Panel
has been highlighting the significant financial and legal
barriers that second liens impose on the successful
implementation of HAMP.\153\ Because more than 40 percent of
homes and approximately 50 percent of HAMP participants have
second liens,\154\ it is important to reemphasize the
importance of several of those impediments.\155\ Though
Treasury has recently implemented several initiatives designed
to address the problems caused by second liens, these programs
have not yet produced data sufficient to evaluate their
success.\156\
---------------------------------------------------------------------------
\153\ April 2010 Oversight Report, supra note 1, at 15.
\154\ Amherst Securities Group LP, Amherst Mortgage Insight: 2nd
Liens--How Important, at 3 (Jan. 29, 2010) (hereinafter ``Amherst
Mortgage Insight: Second Liens''); House Committee on Financial
Services, Testimony of Phyllis Caldwell, chief of the Homeownership
Preservation Office, U.S. Department of the Treasury, Transcript: Robo-
Signing, Chain of Title, Loss Mitigation, and Other Issues in Mortgage
Servicing (Nov. 18, 2010) (publication forthcoming) (online at
financialservices.house.gov/Hearings/hearingDetails.aspx?NewsID=1376.)
\155\ Based on data representing approximately a third of all
second liens outstanding and over 75 percent of all first liens
outstanding, as of September 2010, the average second lien is
approximately $55,000, while the average first lien is roughly
$170,000. In general, second liens that are home equity loans tend to
be larger in dollar value than close-end second liens, with approximate
average sizes of $57,000 and $46,000, respectively. Data provided to
Panel staff by CoreLogic (Dec. 8, 2010).
\156\ The second lien program was announced in April 2009, four
months later, in August 2009, Treasury released the first guidance
regarding the program. SIGTARP Report--Factors Affecting Implementation
of HAMP, supra note 103, at 20. During the summer, banks commented on
the interim rule and requested that second liens modified in tandem
with first liens be given the same favorable treatment as the first
liens, meaning that as long as certain criteria are met, loans modified
under the program will retain the risk weight assigned to that loan
prior to the modification. Letter from Gregory A. Baer, deputy general
counsel for corporate law, Bank of America, to Jennifer J. Johnson,
secretary, Board of Governors of the Federal Reserve System, et al.
(July 30, 2009) (online at www.federalreserve.gov/SECRS/2009/August/
20090810/R-1361/R-1361_073009_21279_428725101156_1.pdf); Letter from
David Pommerehn, counsel, Legislative and Regulatory Affairs, Consumer
Bankers Association, to Office of the Comptroller of the Currency, et
al. (July 30, 2009) (online at 192.147.69.84/regulations/laws/federal/
2009/09c04AD42.PDF). In December 2009, the rule on risk based capital
guidelines for liens modified under the program became effective not
reflecting the banks' request on the treatment of second liens. U.S.
Department of the Treasury, Risk-Based Capital Guidelines; Capital
Adequacy Guidelines; Capital Maintenance; Capital--Residential Mortgage
Loans Modified Pursuant to the Home Affordable Mortgage Program, 74
Fed. Reg. 60138-60140 (Nov. 20, 2009) (online at www.occ.gov/news-
issuances/federal-register/74fr60137.pdf) (``the agencies have
determined that allowing a banking organization to risk weight junior-
lien mortgage loans at less than 100 percent is not appropriate other
than in those circumstances already permitted by the agencies general
risk-based capital rules''). The first servicer signed up for the
program in January 2010. April 2010 Oversight Report, supra note 1, at
14-15. In March 2010, Treasury issued new guidelines for the second
lien program. U.S. Department of the Treasury, Making Home Affordable
Program--Update to the Second Lien Modification Program (2MP),
Supplemental Directive 09-05 Revised (Mar. 26, 2010) (online at
www.hmpadmin.com/portal/programs/docs/second_lien/sd0905r.pdf). To
date, there is limited data to report for the second lien programs as
Ms. Caldwell indicated, ``we don't have data to report yet, as the
program really got started at the beginning of October [2010].''
Transcript Testimony of Phyllis Caldwell, supra note 108. Treasury
expects to have more complete data and the programs in early March
2011. Treasury conversations with Panel staff (Nov. 23, 2010). As of
September 30, 2010, 19 first lien servicers representing 60 percent of
the second liens are participating in the program, and information
supplied to SIGTARP on October 15, 2010 indicates that only 21
permanent modifications have been performed under the program. SIGTARP
Quarterly Report to Congress, supra note 25, at 67, 69.
---------------------------------------------------------------------------
Second liens are loans where the collateral interest for
that loan is second in ranking to the primary or first
lien.\157\ In general terms this means that in the foreclosure
process, the first lien would need to be compensated in full
before the second lien would be paid from the proceeds of the
sale. In other words, the first liens have lien priority.
Therefore, in cases where the first lien is underwater and the
home is in foreclosure, the second lien holder is generally not
entitled to any of the proceeds from the sale.
---------------------------------------------------------------------------
\157\ It is possible that a home could have more than two liens;
however, as most homes only have two liens, in this section second
liens are often used as a proxy for all subordinated liens and the
analysis with respect to second liens holders can often by extrapolated
to apply to other subsequent lien holders.
---------------------------------------------------------------------------
Loan modifications, including some of those performed under
HAMP, could affect lien priority.\158\ The law in this area,
however, is not always clear.\159\ In general, in the absence
of a clause in the first lien reserving the right to modify, a
modification performed between the first lien holder and the
borrower could jeopardize the lien priority of the first lien
holder.\160\ The usual standard to determine whether lien
priority remains intact is if a modification materially
prejudices the second lien holder.\161\ For instance,
modifications that lower the interest rate or reduce principal
are generally considered favorable to the second lien holder
and should not affect lien priority, while modifications that
increase principal can prejudice the second lien holder and
might affect lien priority.\162\ Furthermore, it is possible
that a court would only subordinate (rank lower in lien
priority) the additional amounts due to the modification.\163\
However, though a more drastic remedy, courts have on occasion
elevated the second lien before the entire first lien in
priority.\164\
---------------------------------------------------------------------------
\158\ ``HAMP does not require extinguishment of subordinate lien
instruments as a condition of modification. However, servicers must
follow investor guidance to ensure first lien priority.'' U.S.
Department of the Treasury, Making Home Affordable Program: Handbook
for Servicers of Non-GSE Mortgages-Version 3.0, at 42 (Dec. 2, 2010)
(online at www.hmpadmin.com/portal/programs/docs/hamp_servicer/
mhahandbook_30.pdf) (hereinafter ``MHA Handbook for Servicers of Non-
GSE Mortgages''); House Oversight and Government Reform, Subcommittee
on Domestic Policy, Written Testimony of David Berenbaum, chief program
officer, National Community Reinvestment Coalition, Foreclosures
Continue: What Needs to Change in the Administration's Response?, at 21
(Feb. 25, 2010) (online at oversight.house.gov/images/stories/Hearings/
Domestic_Policy/2010/022510_Foreclosure/
022310_DP_David_Berenbaum_022510.pdf).
\159\ Furthermore, real estate law is state specific and states
differ in their approaches to determinations of lien priority.
\160\ In general, the first lien holder retains the priority to any
subsequent modification if the modification is not materially
prejudicial to the second lien holder or there is a clause in the
contract that allows for a modification. However, there is a limited
case law and scholarship is mixed on the enforceability of this clause.
An example of such a provision would read, ``This mortgage shall also
secure all extensions, amendments, modifications, or alterations of the
secured obligation including amendments, modifications, or alterations
that increase the amount of the secured obligation or the interest rate
on the secured obligation.'' Restatement (Third) of Prop. (Mortgages)
Sec. 7.3 (1997).
\161\ The determination of what is considered materially
prejudicial is generally a question of fact to be determined by the
court and court decisions in this area vary. Restatement (Third) of
Prop. (Mortgages) Sec. 7.3 (1997).
\162\ There is mixed case law on modifications that extend the
length of the first lien. In general, this type of modification aids in
affordability and can be considered favorable to the second lien
holder; however, there is legal precedent where loan priority is
shifted due to this type of modification. Restatement (Third) of Prop.
(Mortgages) Sec. 7.3 (1997) (``Absent an increase in the interest rate
or principal amount of the mortgage obligation, courts routinely hold
that such modifications do not defeat the mortgagee's priority as
against intervening lienors. The assumption is that such transactions
reduce the likelihood of foreclosure of the senior mortgage and that
they are therefore beneficial to the interests of junior lienors. See,
e.g., Crutchfield v. Johnson & Latimer, 8 So.2d 412 (Ala. 1942); Lennar
Northeast Partners v. Buice, 57 Cal.Rptr.2d 435 (Cal.Ct. App.1996);
Eurovest Ltd. v. 13290 Biscayne Island Terrace Corp., 559 So.2d 1198
(Fla.Dist.Ct.App.1990); State Life Insurance Co. v. Freeman, 31 N.E.2d
375 (Ill.Ct.App.1941); Guleserian v. Fields, 218 N.E.2d 397
(Mass.1966); Shultis v. Woodstock Land Dev. Assoc., 594 N.Y.S.2d 890
(N.Y.App.Div.1993); Skaneateles Savings Bank v. Herold, 376 N.Y.S.2d
286 (N.Y.App.Div.1975), aff'd, 359 N.E.2d 701 (N.Y. 1976); Resolution
Trust Corp. v. BVS Development, Inc., 42 F.3d 1206 (9th Cir.1994); In
re Fowler, 83 B.R. 39 (Bankr.Mont. 1987); In re Earl, 147 B.R. 60
(Bankr. N.D.N.Y.1992); Kratovil & Werner, Mortgage Extensions and
Modification, 8 Creighton L. Rev. 595, 607 (1975); 1 G. Nelson & D.
Whitman, Real Estate Finance Law, Sec. 9.4 (3d ed. 1993). Contra
Citizens and Southern National Bank v. Smith, 284 S.E.2d 770 (S.C.1981)
(extension of senior mortgage results in loss of its priority as
against intervening lienor).
\163\ For instance, a homeowner borrows $100,000 from Lender A to
buy a property, Blackacre, and uses Blackacre as collateral against the
note. Then the homeowner borrows $20,000 from Lender B, again using
Blackacre as collateral. Both liens are properly recorded in a timely
fashion and in accordance with local laws and there is no clause
reserving the right to modify in the agreement with Lender A.
Subsequently, the homeowner and Lender A modify their agreement in a
way which has the effect of increasing the total principal outstanding
to $110,000. If a court decides on an equitable solution, it might
decide that from the proceeds of a foreclosure sale the ranking would
be such that Lender A is entitled to $100,000, then Lender B $20,000
and thirdly, Lender A would be entitled to the additional amount due to
the modification, $10,000. Restatement (Third) of Prop. (Mortgages)
Sec. 7.3 (1997).
\164\ Using the example describe above in footnote 163, if a court
decides to elevate the second lien over the first lien, then Lender B
would be entitled to his entire $20,000 before Lender A would get paid
back any of the $110,000 that homeowner still owes him. Restatement
(Third) of Prop. (Mortgages) Sec. 7.3 (1997).
---------------------------------------------------------------------------
While the HAMP modifications are geared to making the first
liens more affordable and do utilize tools such as interest and
principal reductions (which can be viewed as favorable to the
second lien holder), the vast majority of active HAMP
modifications have a higher principal balance after the
modification than before the modification.\165\ Though this
increase in principal balance is mostly due to capitalization
of arrearages and escrow requirements, it is unclear how courts
would treat this principal increase.\166\ Even if courts were
to decide that HAMP modifications do not change or even
partially affect the lien priority, this legal murkiness could
complicate or at least extend the foreclosure process.
Therefore, in order to assure the first lien holder their
position of seniority, it is more prudent to have that lien
holder reach an agreement with the subsequent lien
holders.\167\
---------------------------------------------------------------------------
\165\ See Section D.3.b, supra, for a discussion on the increase of
principal balance.
\166\ Restatement (Third) of Prop. (Mortgages) Sec. 7.3 (1997).
\167\ April 2010 Oversight Report, supra note 1, at 13-16;
Restatement (Third) of Prop. (Mortgages) Sec. 7.3 (1997).
---------------------------------------------------------------------------
In addition, there are reasons of equity why having an
agreement with the second lien holder, especially an agreement
that also modifies or extinguishes the second lien, is
important to the modification process. Unless the second lien
is modified or extinguished, many of the concessions the first
lien holder makes to help the borrower could disproportionately
accrue to the benefit of the second lien holder.\168\
Therefore, for these legal and financial reasons, second lien
holders can act as a hold out and make modification more
difficult.
---------------------------------------------------------------------------
\168\ April 2010 Oversight Report, supra note 1, at 13-14.
---------------------------------------------------------------------------
Similar to first liens, losses on modified second liens
generally need to be recognized sooner than losses on loans
that have not been modified. In general, if a second lien is
modified, the loss requires immediate recognition.\169\
However, despite the fact that many such liens may not have any
recoverable value at current home prices, many second lien
holders currently carry these liens at improbably high
accounting values for various reasons, including that they are
optimistic that home prices will recover shortly, they believe
they can extract some value from the first lien holder, many of
the second liens are actually current, or because not doing so
would damage their capital position.\170\ Therefore,
recognizing these losses could have significant economic costs,
including that it might require the second lien holder to raise
additional capital or increase other reserves.\171\ Treasury
has indicated that it does not require second lien write-downs
and is ``indifferent to it in the [. . .] first lien program.''
\172\
---------------------------------------------------------------------------
\169\ See Section G.2.b, supra, for a more in-depth discussion of
accounting issues.
\170\ Second liens are current in many cases where first liens are
in default. This could be for a variety of reasons, including that
second liens tend to have lower payments, and that since many of them
are home equity lines of credit, the borrowers want to keep those lines
of credit open for future needs. In addition, second liens tend not to
be securitized. Therefore, the servicing bank is also the investor and
might be more aggressive in collecting payments. Furthermore, many of
the second liens are home equity lines of credit and could be
technically current even if the lien holder is not being paid, as the
unpaid interest can simply be rolled into the balance due. April 2010
Oversight Report, supra note 1, at 119-120; Amherst Mortgage Insight:
Second Liens, supra note 154, at 11-12. See also Section G.2.b, supra.
\171\ April 2010 Oversight Report, supra note 1, at 15-16.
\172\ Transcript Testimony of Phyllis Caldwell, supra note 108.
According to Treasury, borrower qualification for HAMP is based on
first lien affordability. Treasury conversations with Panel staff (Oct.
21, 2010).
---------------------------------------------------------------------------
As the vast majority of second liens are not securitized
and are held directly by banks and other financial
institutions, a write-down of second lien portfolios under a
modification program could cause substantial accounting losses
and thereby impact the solvency of some banks. In fact, the
largest four banks--JPMorgan Chase, Bank of America, Citigroup,
and Wells Fargo--hold 43 percent or $420.0 billion of all U.S.
revolving and second liens ($975.3 billion).\173\ To give a
sense of the magnitude, the value of the second liens held by
these banks is roughly comparable to just under 80 percent of
the amount of their Tier 1 capital, the principal measure used
by regulators to determine the adequate capitalization of a
bank. Specifically, at the end of the third quarter 2010, the
four largest banks reported $420.0 billion in second lien
mortgages while having total Tier 1 capital of $535.2
billion.\174\ In the case of Wells Fargo, their second lien
exposure actually exceeds the amount of their Tier 1 capital.
---------------------------------------------------------------------------
\173\ In terms of market size, the first lien market is roughly 10
times the size of the second lien market. However, since almost all
second liens are not securitized, the exposure that banks have with
respect to second liens tends to be disproportionately larger than
their exposure from first liens. For instance, at the end of the third
quarter, the largest four banks hold $420.0 billion in second liens,
while holding $793.1 billion in first liens. Board of Governors of the
Federal Reserve System, Flow of Funds Accounts of the United States, at
96 (Sept. 30, 2010) (online at www.federalreserve.gov/RELEASES/z1/
Current/z1.pdf) (hereinafter ``Federal Reserve--Flow of Funds Accounts
of the United States''); Data provided by Federal Reserve staff (Dec.
1, 2010); Data on first and second lien holdings of the four largest
banks accessed through SNL Financial data service (Dec. 9, 2010).
\174\ The Tier 1 Capital amount is based on reporting by the banks,
not their holding companies, and therefore may not include all second
liens held by affiliates. Data accessed through SNL Financial data
service (Dec. 9, 2010).
---------------------------------------------------------------------------
Figure 13 shows that these banks are some of the largest
investors in second lien mortgages, and Figure 14 details their
second lien exposure as compared to their Tier 1 capital.
FIGURE 13: OWNERSHIP OF SECOND-LIEN MORTGAGES BY INSTITUTION TYPE, AS
OF Q3 2010 \175\
[Dollars in billions]
---------------------------------------------------------------------------
\175\ Federal Reserve--Flow of Funds Accounts of the United States,
supra note 173 at 96; Data on second lien holdings of the four largest
banks accessed through SNL Financial data service (Dec. 9, 2010).
[GRAPHIC] [TIFF OMITTED] 62622A.013
FIGURE 14: SECOND LIENS COMPARED TO TIER 1 CAPITAL AS OF THE THIRD QUARTER OF 2010 \176\
[Dollars in billions]
----------------------------------------------------------------------------------------------------------------
Second Liens as
Company Second Liens Tier 1 Capital a Percent of
Tier 1 Capital
----------------------------------------------------------------------------------------------------------------
Bank of America Corporation............................ $137.4 $164.8 83
JPMorgan Chase Co..................................... 108.6 139.4 78
Citigroup Inc.......................................... 52.0 125.4 41
Wells Fargo Company................................... 122.0 105.6 116
Total Top 4........................................ $420.0 $535.2 78
----------------------------------------------------------------------------------------------------------------
\176\ Data accessed through SNL Financial data service (Dec. 9, 2010).
In addition, the four largest commercial banks, Bank of
America, Citigroup, JPMorgan Chase, and Wells Fargo, are both
servicers of first and second liens as well as owners of large
second lien portfolios. Therefore, this potential for a
significant write-down of second lien mortgages creates a
profound conflict of interest for these entities, putting their
financial well-being at odds with their duties as first lien
servicers.\177\
---------------------------------------------------------------------------
\177\ For instance, Bank of America and Chase have indicated that
for first liens that they service, they own 15 percent and 10 percent,
respectively, of the related second liens. House Committee on Financial
Services, Written Testimony of Barbara Desoer, president, Bank of
America Home Loans, Second Liens and Other Barriers to Principal
Reduction as an Effective Foreclosure Mitigation Program, at 6 (Apr.
13, 2010) (online at financialservices.house.gov/media/file/hearings/
111/desoer.pdf); House Committee on Financial Services, Written
Testimony of David Lowman, chief executive officer, JPMorgan Chase Home
Lending, Second Liens and Other Barriers to Principal Reduction as an
Effective Foreclosure Mitigation Program, at 5 (Apr. 13, 2010) (online
at financialservices.house.gov/media/file/hearings/111/
jpmc_lowman_4.13.10.pdf).
---------------------------------------------------------------------------
Finally, the salient objective of HAMP is to make homes
more affordable. Statistically, having a second lien decreases
the likelihood that the property is affordable.\178\ If a
homeowner has two liens, even if that homeowner is granted a
modification on the first lien, the homeowner might still not
have the funds to make his or her mortgage payments, as the
modified first lien payment combined with the cost of the
second lien might still be beyond the homeowner's reach.\179\
The existence of multiple liens also increases the likelihood
that there is negative equity in the home and makes it more
likely that the homeowner will default.\180\
---------------------------------------------------------------------------
\178\ The likelihood that properties financed through multiple
liens have aggregate mortgage payments greater than 30 percent of
income is nearly 22 percent, as compared to an only 2.2 percent
likelihood for properties with only one lien against the property.
David Bernstein, A Presentation: Seconds First: The Role of Second
Liens in the Mortgage Crisis and Rescue, at 7 (Nov. 10, 2008) (online
at ssrn.com/abstract=1299144).
\179\ Second liens, as additional borrower debt, contribute to
back-end DTI. See Section D.3.a., discussing how borrower debt affects
the affordability and sustainability of a modification.
\180\ Amherst Mortgage Insight: Second Liens, supra note 154, at 5.
---------------------------------------------------------------------------
Though the administration's two announced second lien
programs--2MP and the second-lien portion of the FHA Short
Refinance Program--are structured to better align the interests
of the first and second lien holders in order to increase the
likelihood that a borrower's modification will be successful,
it is unclear that all of the impediments caused by second
liens have been adequately addressed.
d. Misaligned Incentives
Mortgage servicers, often divisions of large banks, are
responsible for collecting payments from borrowers on behalf of
the investors who own the loans. These servicers play a key
decision-making role on HAMP modifications--they are the firms
that have entered into contracts with Treasury to make
modifications under HAMP--but they have different financial
incentives than the investors who own the loans. Hence, as the
Panel described in its April 2010 report, HAMP may be faltering
because of a principal-agent problem. The servicers are
interested in maximizing their revenue from each loan while
minimizing their expenses.\181\ Servicers typically incur most
of their costs on the front-end when ``boarding'' the loans
into their systems. Since they make money from servicing on an
on-going basis it is in their interest to keep servicing loans
that are current. However, once a loan is on the verge of
becoming delinquent or defaults, servicers' incentives change:
because the costs associated with servicing a delinquent loan
often exceed the revenues that a servicer can generate from the
same loan, it may be in the servicer's interest to move to
foreclosure as soon as possible.\182\ Consequently, the
servicer's financial interests may be at odds with the
interests of the loan investors, who will generally receive far
less from a foreclosure than they will from a modification. The
picture is even more complicated when the mortgages have been
securitized, because the holders of the senior tranches of
mortgage-backed securities tend to benefit from more rapid
foreclosures, while the holders of the junior tranches benefit
from a more drawn-out foreclosure process.\183\ This
misalignment of incentives and the disputes that can result
from it are sometimes called ``tranche warfare.'' \184\ The
complexity and opacity of the situation combined with the lack
of accountability--where it is not clear whose interests the
servicers are serving--appear to be undermining HAMP's
effectiveness.
---------------------------------------------------------------------------
\181\ April 2010 Oversight Report, supra note 1, at 71-74.
\182\ See October 2009 Oversight Report, supra note 15, at 71-74.
See also SIGTARP Quarterly Report to Congress, supra note 25, at 162
(``Typically, if a borrower's payments are not made promptly and in
full, servicers must advance to the investor the required amount of the
monthly payment owed by the borrower, although in some circumstances
servicers are required to advance only the unpaid interest. . . . As
more loans in a servicer's portfolio become delinquent, these advance
payments can strain servicers' cash supplies or their lines of credit.
Servicers thus have incentives to pursue aggressive collection
techniques to ``cure'' a loan and return it to a performing status or
to place the property into foreclosure, which may enable them to
recover their advances more quickly. . . . Servicers often borrow to
fund these advances, and they incur interest expenses on that
borrowing.'').
\183\ See Amherst Securities Group LP, Amherst Mortgage Insight:
The Affidavit Fiasco--Implications for Investors in Private Label
Securities (Oct. 12, 2010) (``In general, when liquidations are
postponed, servicers generally continue to advance (``as long as it is
deemed recoverable''). These advances are repaid to a servicer at
liquidation, which increases the loss severity at liquidation. That
helps the more junior bonds, which continue to receive interest
payments for a longer period (they never expected to receive principal,
anyway). But postponement hurts the more senior bonds, as (1) they
experience higher losses due to the higher severities and (2) the
average life lengthens for their bonds, a negative for bonds purchased/
carried at a discount.'').
\184\ Larry Cordell et al., The Incentives of Mortgage Servicers
and Designing Loan Modifications to Address the Mortgage Crisis, at
233-234, in Lessons from the Financial Crisis: Causes, Consequences,
and Our Economic Future (Robert W. Kolb ed., 2010).
---------------------------------------------------------------------------
e. Pooling and Servicing Agreements
The contracts between the investors in a mortgage and the
mortgage's servicers are known as PSAs. In some cases, as the
Panel has noted in its previous reports, these agreements may
be serving as contractual barriers to HAMP modifications.\185\
While PSAs generally do not bar loan modifications, they
typically restrict the ability of servicers to extend the term
of the loan by more than one year, and they often make it
difficult for servicers to reduce the loan principal as part of
a modification.\186\ These contractual restrictions may be
hindering the ability of servicers to qualify borrowers for
HAMP, since term extensions are a key tool for reducing monthly
payments to the level that the program requires.
---------------------------------------------------------------------------
\185\ March 2009 Oversight Report, supra note 10, at 42-44; October
2009 Oversight Report, supra note 15, at 28-29; April 2010 Oversight
Report, supra note 1, at 72. But see John P. Hunt, What Do Subprime
Securitization Contracts Actually Say About Loan Modification?
Preliminary Results and Implications, at 8 (Mar. 25, 2009) (online at
www.law.berkeley.edu/files/
Subprime_Securitization_Contracts_3.25.09.pdf) (``The most common rules
are that the servicer must follow generally applicable servicing
standards, service the loans in the interest of the certificateholders
and/or the trust, and service the loans as it would service loans held
for its own portfolio. Notably, these conditions taken together can be
read as attempting to cause the loan to be serviced as they would have
been if they had not been securitized.'').
\186\ April 2010 Oversight Report, supra note 1, at 72.
---------------------------------------------------------------------------
f. Treasury's Base NPV Model
Treasury's base NPV model is in many ways the linchpin for
HAMP. Treasury requires that participating servicers use the
NPV test for all HAMP eligible loans that are in imminent
default or are at least 60 days delinquent.\187\ The NPV test
is run before a homeowner is offered a trial modification.\188\
The base NPV model computes the difference between the expected
values of those loans under unmodified and modified scenarios.
This is the gateway into HAMP. As long as agreed to by the
investor,\189\ a servicer must offer a HAMP modification when
an NPV evaluation yields a positive value, meaning the total
discounted value of expected cash flows is greater for the
modified loan than the non-modified loan.\190\ A loan
modification is not required when the NPV is negative.\191\
---------------------------------------------------------------------------
\187\ The determination of delinquency is based on the Mortgage
Bankers Association delinquency calculation; however, the determination
for imminent default for non-GSE mortgages is left to the servicer with
very little guidance from Treasury. MHA Handbook for Servicers of Non-
GSE Mortgages, supra note 158, at 58, 73 (``All loans that meet HAMP
eligibility criteria and are either deemed to be in imminent default or
delinquent as to two or more payments must be evaluated using a
standardized NPV test.''); U.S. Department of the Treasury, Home
Affordable Modification Program: Base Net Present Value (NPV) Model
Specifications, at 3 (June 11, 2009) (online at www.hmpadmin.com/
portal/programs/docs/hamp_servicer/npvoverview.pdf) (hereinafter ``HAMP
Base NPV Model Specifications''). Large servicers with at least $40
billion on their servicing books have the option of using the base NPV
model or creating their own proprietary model. To date, no servicer has
elected to customize the model and Treasury indicated that it would
remove the new model as an option, if no servicer makes such an
election by December. HAMP Base NPV Model v4.0: Model Documentation,
supra note 27, at 41; Treasury conversations with Panel staff (Oct. 20,
2010).
\188\ HAMP Base NPV Model v4.0: Model Documentation, supra note 27,
at 3; MHA Handbook for Servicers of Non-GSE Mortgages, supra note 158,
at 73.
\189\ See Section G.2.e, supra.
\190\ MHA Handbook for Servicers of Non-GSE Mortgages, supra note
158, at 73-74; HAMP Base NPV Model Specifications, supra note 187.
\191\ MHA Handbook for Servicers of Non-GSE Mortgages, supra note
158, at 73-74.
---------------------------------------------------------------------------
To best conceptualize the NPV model, Treasury describes it
as three discrete calculations, which are combined in the final
stage.\192\ The inputs of the program are used to generate (1)
default rates (for both the modified and unmodified scenarios),
(2) cure rates (for both the modified and unmodified
scenarios), as well as (3) a cash flow equation.\193\ All three
results are then combined in the cash flow equation where the
default and cure rates are used to predict the likelihood for
various cash streams.\194\ The following figure represents a
simplified form of that equation.
---------------------------------------------------------------------------
\192\ Treasury conversations with Panel staff (Oct. 20, 2010).
\193\ The cure rate represents the likelihood that a borrower who
was previously behind in his payments, makes up those missing payments.
\194\ A more complete description of the base NPV model equations
can be found online on the administrative web site for servicers for
HAMP. HAMP Base NPV Model v4.0: Model Documentation, supra note 27, at
31-40 .
---------------------------------------------------------------------------
FIGURE 15: NPV MODEL DIAGRAM \195\
---------------------------------------------------------------------------
\195\ HAMP Base NPV Model v4.0: Model Documentation, supra note 27,
at 6.
[GRAPHIC] [TIFF OMITTED] 62622A.014
In order to calculate the NPV, the model has several inputs
and variables, including the government incentives.\196\
Therefore, both borrowers with a positive NPV in the absence of
the government incentives, as well as those who only have a
positive NPV due to the addition of government incentives, can
participate in HAMP.\197\ Other variables in the NPV model are
as follows:\198\
---------------------------------------------------------------------------
\196\ HAMP Base NPV Model v4.0: Model Documentation, supra note 27.
\197\ HAMP Base NPV Model v4.0: Model Documentation, supra note 27,
at 7-8.
\198\ Some of these variables are fixed and others are updated in
regular intervals depending on the availability of data. For instance,
Treasury indicated that it updates some parameters such as home price
projections and real estate owned (REO) discount rate on a quarterly
basis. However, the redefault rates and cure rates are based on GSE
analytics, and program portfolio data are updated on an as needed
basis. Treasury conversations with Panel staff (Oct. 20, 2010).
---------------------------------------------------------------------------
Modification costs and incentives,
including lower monthly mortgage payments, likelihood
of a redefault, and likelihood that a loan will be paid
off before its term expires;
Property value relative to mortgage value;
Likelihood of foreclosure; and
Foreclosure costs, including legal
expenses, property maintenance costs, and expenses
involved in reselling property, and lost interest.\199\
---------------------------------------------------------------------------
\199\ HAMP Base NPV Model Specifications, supra note 187, at 3;
HAMP Base NPV Model v4.0: Model Documentation, supra note 27, at 36-37;
Treasury conversations with Panel staff (Oct. 20, 2010). See Section
G.2.c, supra, for a discussion of second liens, which are not captured
in Treasury's base NPV model.
---------------------------------------------------------------------------
The NPV model also has a significant amount of inputs.
Post-modification principal and interest payments, loan terms,
and principal amounts are included among the many inputs. Since
many of the PSAs with investors limit the type of modification
that can be offered on a loan, the NPV model needed to be
flexible enough to account for the investor restrictions.\200\
Furthermore, not all the inputs are used in the calculation as
some are strictly for data collection purposes.\201\
---------------------------------------------------------------------------
\200\ See Section G.2.e, supra, for a further discussion on the
limits on modifications placed by these agreements.
\201\ This includes the Borrower's Total Monthly Obligation, an NPV
input closely correlated to back-end debt. Treasury conversations with
Panel staff (Oct. 20, 2010).
---------------------------------------------------------------------------
Servicers are limited in their flexibility to tweak these
inputs and variables; however, they can adjust the discount
rate, the rate at which future cash flows are brought back to
the present time, by adding a risk premium. For non-GSE loans,
all servicers may add a risk premium of up to 250 basis points
to the minimum discount rate set at Freddie Mac's Primary
Mortgage Market Survey rate for 30-year fixed rate conforming
loans.\202\ The number of loans that will qualify for a HAMP
modification will vary depending on the risk premium a servicer
uses in its NPV calculations.\203\ Higher risk premiums
decrease the likelihood of a loan modification.\204\ In the
October 2009 Report, the Panel noted that for a baseline
scenario ``only a one basis point change in the risk premium is
necessary to change the outcome of the test for the baseline
loan from NPV positive to NPV negative.'' \205\ This
demonstrates how sensitive the NPV model is to changes in the
risk premium.
---------------------------------------------------------------------------
\202\ For GSE loans, servicers are not permitted to add a premium.
Servicers only have the flexibility to choose two risk premium rates,
one which they must use for all loans in portfolio (loans owned by the
servicers) and one that they need to apply to all private label loans
(generally, loans that are owned by private label securitization trusts
and not sponsored by any GSE). In addition, the specific risk premiums
are considered to be proprietary and confidential by servicers and are
not publicly disclosed. October 2009 Oversight Report, supra note 15,
at 130; Treasury conversations with Panel staff (Oct. 20, 2010);
Treasury conversations with Panel staff (Dec. 9, 2010).
\203\ U.S. Government Accountability Office, Troubled Asset Relief
Program: Further Actions Needed to Fully and Equitably Implement
Foreclosure Mitigation Programs, at 19 (June 2010) (GAO-10-634) (online
at www.gao.gov/new.items/d10634.pdf) (hereinafter ``GAO Report on
Foreclosure Mitigation Programs'').
\204\ Since the loan modifications push the cash stream further
into the future, discount rates decrease the expected value of the
modified loans more than the unmodified loans, thereby decreasing the
possibility that the NPV will be a positive number. Treasury
conversations with Panel staff (Oct. 20, 2010).
\205\ October 2009 Oversight Report, supra note 15, at 130.
---------------------------------------------------------------------------
Since the NPV model is complex, there are several factors
that could affect the success of the model, including the
design of the model, the implementation of the model, and the
accuracy of the data input by the servicers into the
model.\206\
---------------------------------------------------------------------------
\206\ SIGTARP is in the process of auditing the NPV model,
including the implementation by servicers of the NPV model, Treasury
efforts to ensure quality control, and the procedures that servicers
follow to communicate the reasons for NPV test failure to the
borrowers. SIGTARP Quarterly Report to Congress, supra note 25, at 31.
---------------------------------------------------------------------------
The correct design of the model is critical. If the NPV
model is calibrated correctly, it will get the correct
homeowners into HAMP to prevent avoidable foreclosures.
However, an incorrect calibration could either act as a means
to delay inevitable foreclosures or grant subsidies to those
who would otherwise cure and therefore do not need the extra
help.\207\ In response to servicer feedback as well as changes
to HAMP, Treasury has recently released a new version of the
NPV model. This new model, which is effective as of October 1,
2010, reflects the PRA initiative as well as updates to the
default and prepayment models.\208\
---------------------------------------------------------------------------
\207\ These risks, the risk of delaying inevitable foreclosure (or
redefault risk) and the cure risk, can significantly increase costs
during renegotiation. In evaluating a modification, ``[o]ne must take
into account both the redefault and the self-cure risks.'' Manuel
Adeline, Kristopher Gerardo, and Paul S. Willen, Why Don't Lenders
Renegotiate More Home Mortgages? Redefaults, Self-Cures, and
Securitization, Federal Reserve Bank of Boston Working Paper 09-4, at 7
(July 6, 2009) (online at www.bos.frb.org/economic/ppdp/2009/
ppdp0904.pdf).
\208\ HAMP Base NPV Model v4.0: Model Documentation, supra note 27,
at 4-5.
---------------------------------------------------------------------------
The NPV model, similar to any financial model, can only be
as accurate as its assumptions and inputs.\209\ In creating the
model, Treasury had to make decisions about what to
incorporate, balancing better results with limitations on data.
In the NPV model, among the items that Treasury did not
incorporate were the impact of accounting rules and back-end
debt (including second liens).\210\ Furthermore, the economics
captured in the NPV model are primarily those of the borrower
and the investor, therefore the NPV model does not directly
take into account servicer costs and incentives.\211\
Therefore, it is likely that one reason for these incentives is
to compensate the servicers for some of their costs in
implementing the program.
---------------------------------------------------------------------------
\209\ Many assumptions in the model, including the default,
prepayment and cure rates, are based off of historic data. Treasury
updates these assumptions when it is warranted with the data and when
there is sufficient new information to create reliable models. Treasury
conversations with Panel staff (Oct. 20, 2010); Treasury conversations
with Panel staff (Nov. 15, 2010).
\210\ HAMP Base NPV Model v4.0: Model Documentation, supra note 27,
at 13-20; Treasury conversations with Panel staff (Oct. 20, 2010);
Treasury conversations with Panel staff (Nov. 15, 2010).
\211\ Treasury conversations with Panel staff (Dec. 3, 2010).
---------------------------------------------------------------------------
The Panel previously examined Treasury's base NPV model in
its October 2009 report. Among its recommendations was the need
to increase transparency regarding the methodology of the
model. Treasury has since released documents detailing model
inputs, assumptions, and underlying equations for the expected
cash flows and default probabilities. These documents have been
updated to reflect the new version of the NPV model.\212\
---------------------------------------------------------------------------
\212\ HAMP Base NPV Model v4.0: Model Documentation, supra note 27.
---------------------------------------------------------------------------
A further issue concerning the performance of the model is
the granularity of its inputs. The Panel's October 2009 report
called for ``incorporating more localized information when
determining a mortgage loan's value.'' For example, home price
projections used in the base model are constructed from state
averages rather than values within specific communities or
neighborhoods. Limited granularity within the model is a
possible source of inaccurate NPV calculations, and by
extension, wrongful denials for loan modifications.\213\
---------------------------------------------------------------------------
\213\ Congressional Oversight Panel, Written Testimony of Larry
Litton, president and chief executive officer, Litton Loan Servicing,
Philadelphia Field Hearing on Mortgage Foreclosures, at 3-4 (Sept. 24,
2009) (online at cop.senate.gov/documents/testimony-092409-litton.pdf).
---------------------------------------------------------------------------
The implementation of the model is also a source of
concern. A June 2010 GAO report noted that audits conducted by
MHA Compliance (MHA-C) revealed that 15 of the largest 20 HAMP
servicers were not in compliance with certain guidelines for
implementation of the NPV model. Further, the report discussed
inconsistencies with seven servicers that coded the NPV model
on their internal systems rather than running the model through
the Fannie Mae web portal. Those servicers were required to fix
the coding on their in-house models and until that coding was
fixed, ``MHA-C required the servicers to refrain from denying
permanent modifications because of negative NPV results unless
these results were validated by the Treasury version of the NPV
model housed on the Fannie Mae Web portal.'' \214\ Failing to
code the NPV model properly has likely led to inconsistent
evaluations of similarly situated borrowers.\215\
---------------------------------------------------------------------------
\214\ GAO Report on Foreclosure Mitigation Programs, supra note
203, at 20; Congressional Oversight Panel, Written Testimony of Paul
Heran, program executive, Making Home Affordable--Compliance, Freddie
Mac, COP Hearing on Treasury's Use of Private Contractors, at 5 (Sept.
22, 2010) (online at cop.senate.gov/documents/testimony-092210-
heran.pdf) (hereinafter ``Written Testimony of Paul Heran'') (``[I]f a
recoded NPV model is determined to provide unreliable results, a
servicer may be required to validate results in the Treasury approved
model until the recoded model's reliability can be substantiated.'').
\215\ GAO Report on Foreclosure Mitigation Programs, supra note
203, at 20.
---------------------------------------------------------------------------
Finally, the accuracy of the inputs is vital. Servicers are
responsible for collecting the financial information from the
borrower and inputting the data for the NPV tests. Treasury has
taken steps to improve accuracy, including changing the
guidelines to require that borrowers denied based on a negative
NPV receive letters containing the 33 inputs used in the NPV
model.\216\ Beginning February 1, 2011, these denial letters
will be required and borrowers will then have 30 days from the
notice to correct the NPV inputs.\217\ Moreover, as required by
the Dodd-Frank Wall Street Reform and Consumer Protection Act,
Treasury is in the process of establishing a web portal so that
borrowers can perform their own NPV analysis.\218\ Treasury
indicated that it intends to launch this portal in the spring
of 2011.\219\ Though, Treasury is still ultimately responsible
for ensuring servicer compliance, as the Panel noted in October
2009, ``[m]aking the model publicly available [will] facilitate
negotiations and provide an important check against wrongful
modification denials.'' \220\
---------------------------------------------------------------------------
\216\ The new directive would be effective as of February 1, 2011.
U.S. Department of the Treasury, Making Home Affordable Program--Case
Escalation Process/Dodd-Frank Act NPV Notices, Supplemental Directive
10-15, at 2, 21-24 (Nov. 3, 2010) (online at www.hmpadmin.com/portal/
programs/docs/hamp_servicer/sd1015.pdf) (hereinafter ``Supplemental
Directive 10-15'').
\217\ Id. at 2.
\218\ Though the NPV model will be available to borrowers through a
portal, it will only provide an ``estimated outcome.'' There are many
factors that a borrower may not have access to that will affect the
outcome of the NPV model including information about mortgage insurance
on the loan and restrictions on a loan modification due to PSAs with
investors. Congressional Oversight Panel, Written Testimony of Phyllis
Caldwell, chief of the Homeownership Preservation Office, U.S.
Department of the Treasury, COP Hearing on TARP Foreclosure Mitigation
Programs, at 11 (Oct. 27, 2010) (online at cop.senate.gov/documents/
testimony-102710-caldwell.pdf) (hereinafter ``Written Testimony of
Phyllis Caldwell'') (``As required by the Dodd-Frank Wall Street Reform
and Consumer Protection Act, Treasury is preparing to establish a web
portal that borrowers can access to run a NPV analysis using input data
regarding their own mortgages, and to provide to borrowers who are
turned down for a HAMP modification the input data used in evaluating
the application.''); Supplemental Directive 10-15, supra note 216, at 1
(``However, because a borrower using the Borrower NPV Calculator may
not use exactly the same data used by the servicer, the Borrower NPV
Calculator will only provide an estimated outcome.''); Treasury
conversations with Panel staff (Nov. 15, 2010).
\219\ Supplemental Directive 10-15, supra note 216, at 1; Treasury
conversations with Panel staff (Nov. 15, 2010).
\220\ October 2009 Oversight Report, supra note 15, at 47.
---------------------------------------------------------------------------
g. Sustainability of HAMP Modifications
HAMP modifications make homeownership relatively more
affordable by reducing borrowers' monthly debt burdens. The
Panel has raised concerns, however, as to whether these
modifications make homeownership sufficiently affordable to
avoid foreclosure.\221\
---------------------------------------------------------------------------
\221\ See April 2010 Oversight Report, supra note 1, at 4-5.
---------------------------------------------------------------------------
Treasury has stated that its estimated redefault rate for
HAMP permanent modifications is 40 percent over the five-year
span for a permanent modification.\222\ For the first year of a
modification, the redefault rate has been significantly lower,
although the Panel has previously expressed concerns that the
redefault rate could be higher over the long term. For a full
discussion of redefault rates, see Section I.\223\ Although
HAMP improves affordability, many borrowers' resources are
severely constrained by debt other than monthly mortgage
payments. The program does not consider the homeowner's back-
end DTI ratio--the ratio of total monthly debt payments to
monthly income--and as a result many borrowers with permanent
modifications are still spending a large percentage of their
income on housing and other debt.\224\ After a HAMP
modification, the median borrower is left paying 63 percent of
pre-tax income towards debt.\225\ This in turn implies after-
tax DTI levels that could make modifications
unsustainable.\226\ With high levels of total debt, even a
small disruption of income or increase in expenses could make
repayment impossible for these borrowers. HAMP adopted DTI
targets that may be inappropriately high for long-term
sustainability of loans. The 31 percent DTI target is higher
than the GSEs' own affordability guidelines, and arguably more
important, the program does not have a back-end affordability
target and results in back-end DTIs that are also widely
regarded as too high. By comparison, as discussed above in more
detail, to qualify for an FHA loan, a borrower typically needs
to show a back-end DTI ratio of no more than 41 percent.\227\
The inclusion of a specific and more demanding back-end DTI
target in the NPV model would likely have meant either larger
modification of the first lien loan or concessions from junior
lienholders, raising the question of the treatment of second
lien mortgages. As explained above, the inputs into the NPV
model were designed to exclude back-end DTI as a factor because
of difficulties involved in data collection.\228\
---------------------------------------------------------------------------
\222\ Congressional Oversight Panel, Written Responses of Herbert
M. Allison, Jr., assistant secretary for financial stability, U.S.
Department of the Treasury, COP Hearing with Assistant Treasury
Secretary Herbert M. Allison, Jr., at 3 (Oct. 22, 2009) (online at
cop.senate.gov/documents/testimony-102209-allison-qfr.pdf).
\223\ April 2010 Oversight Report, supra note 1, at 60-62.
\224\ U.S. Department of the Treasury, Making Home Affordable
Program: Handbook for Servicers of Non-GSE Mortgages--Version 3.0, at
19 (Sept. 22, 2010) (online at www.hmpadmin.com/portal/programs/docs/
hamp_servicer/mhahandbook_20.pdf) (hereinafter ``MHA Handbook for
Servicers of Non-GSE Mortgages'').
\225\ Data provided by Treasury.
\226\ See Section D.3.a, supra, for after-tax DTI calculations.
\227\ See Section B.3.a, supra.
\228\ See Section D.3.a, supra, for further discussion.
---------------------------------------------------------------------------
Moreover, interest rates and payments for borrowers who
successfully complete a permanent modification can increase
after the five-year modification period.\229\ This calls into
question whether these mortgages will be sustainable after the
modification period expires. Even among borrowers who complete
a five-year permanent modification, absent a dramatic recovery
in housing prices, some will redefault when their payments
increase at the end of the modification period. The phase-out
of modification terms will be particularly problematic for
those families still underwater on their properties. Unless
housing prices recover to a significant degree or the economy
rebounds notably, redefault rates may be higher than Treasury
currently estimates.
---------------------------------------------------------------------------
\229\ Explaining the rationale underlying HAMP, Ms. Caldwell
testified before the Panel ``[t]he borrower's modified monthly payment
would remain in place for five years, which Treasury expected would
provide sufficient time for the housing market and the financial system
to recover.'' Written Testimony of Phyllis Caldwell, supra note 218, at
4.
---------------------------------------------------------------------------
The state of the broader economy will also have a
significant influence on the redefault rate. Curtailment of
income is the most common reason listed for hardship by
borrowers requesting HAMP modifications, and borrowers who are
unable to find sufficient work will inevitably redefault.
Unemployment and underemployment are particularly problematic
because loan modifications under HAMP require borrowers to stay
in their homes for several years, which may prevent borrowers
from moving to take advantage of better job opportunities.
The deep level of negative equity for many HAMP permanent
modification recipients also makes the sustainability of these
mortgages questionable in the absence of principal reductions.
The median borrower in a HAMP permanent modification has a 125
percent LTV ratio, meaning the family is deeply
underwater.\230\ Even with affordable payments, deeply
underwater borrowers may choose to strategically default or may
be compelled to default in response to core life events, such
as illness, changes in family circumstances, job loss or job
opportunities.
---------------------------------------------------------------------------
\230\ Data provided by Treasury.
---------------------------------------------------------------------------
h. Foreclosure Processing Problems
Charges of servicers' negligence, ranging from lost
paperwork to improper claims of ownership, have plagued the
government's housing rescue programs from nearly the
beginning.\231\ As the Panel discussed in its November 2010
report, some servicers' employees have admitted to approving
foreclosures without properly verifying the accompanying
documents or otherwise following proper procedure.\232\
---------------------------------------------------------------------------
\231\ See April 2010 Oversight Report, supra note 1, at 71.
\232\ See November 2010 Oversight Report, supra note 79, at 10-12.
---------------------------------------------------------------------------
For servicers, foreclosures carry significant costs leading
up to the acquisition of a property's title. Many of the
delinquency costs, including lost principal and interest
payments and lost servicing fee income, are time-dependent
costs that the servicer continues to accrue throughout the
foreclosure.\233\ Servicers are required to pay many of these
funds in advance and often have to borrow money to cover these
costs. In shortening the foreclosure processing period, the
servicer not only pays less in delinquency costs out-of-pocket
compared to a longer, and more thorough, processing period, but
is also able to recoup its costs quicker and repay any amounts
it has borrowed. Foreclosure processing irregularities may tilt
servicers to initiate more foreclosures because pursuing this
option becomes artificially cheaper, although the servicer
should weigh all the relevant costs, including the cost to
replace its guaranteed income from a mortgage under its
management.\234\ Servicers may also be lowering their overhead
costs by forgoing the administrative costs of hiring additional
resources to conduct more robust scrutiny on mortgage
documents.
---------------------------------------------------------------------------
\233\ The national average time between the first missed payment
and the foreclosure sale is approximately one year. See Mortgage
Bankers Association, Lenders' Cost of Foreclosure, at 3 (May 28, 2008)
(online at www.nga.org/Files/pdf/0805FORECLOSUREMORTGAGE.PDF). See also
Congressional Budget Office, Policy Options for the Housing and
Financial Markets, at Chapter 3 (Apr. 2008) (online at www.cbo.gov/
ftpdocs/90xx/doc9078/toc.htm).
\234\ For a more detailed discussion on how servicer incentives
could skew the process toward foreclosures, see Section G.2.d, supra.
---------------------------------------------------------------------------
This difference in cost could also skew HAMP's NPV model.
In determining whether to grant a modification to a borrower, a
servicer uses HAMP's NPV model to compare the net present value
of a modification versus that of a foreclosure.\235\ In valuing
the cost of a foreclosure, the NPV model requires inputs based
on observed costs, i.e. costs calculated based on actual
historical data. By cutting corners in the foreclosure process,
servicers may have been able to lower artificially the cost of
a foreclosure, in particular by reducing the time it would take
to complete a foreclosure. A shorter foreclosure process would
mean decreased costs and that in turn would tilt the model in
favor of foreclosures. In such instances, servicers would have
an incentive to lose paperwork or otherwise deny modifications
that they would be compelled to make under the program
standards. According to Treasury, it is not yet clear whether
the foreclosure irregularities may have altered costs and thus
NPV inputs to the point that it skewed the model toward
foreclosure, however, it is carefully examining the data for
such evidence and will take appropriate steps should it observe
a data break point.\236\
---------------------------------------------------------------------------
\235\ For a discussion of the NPV model, see Section G.2.f, supra.
\236\ Treasury conversations with Panel staff (Oct. 17, 2010).
---------------------------------------------------------------------------
i. Failure To Focus on Root Causes of Foreclosures
The Panel first expressed concerns that HAMP was not
designed to address the root causes of the housing crisis in
March 2009.\237\ In subsequent reports the Panel has raised
serious concerns about Treasury's efforts to address these
problems, noting that HAMP has failed to address foreclosures
caused by factors such as unemployment and negative
equity.\238\ Unemployment can undermine many loan modifications
designed to prevent foreclosure, since these modifications
generally are based on the assumption that the borrower will
stay in place and make payments for several years, and it is
very difficult for unemployed borrowers to pass HAMP's NPV
test. Negative equity similarly leaves borrowers unable to
respond to life events such as changes in family circumstances,
job loss or job opportunities. Homeowners responding to life
events often face the choice of either walking away from their
mortgages or turning down a job opportunity. Those who choose
to leave their homes depress nearby property values, while
those who turn down job opportunities disrupt the labor market.
In either case, the economic impact is negative.
---------------------------------------------------------------------------
\237\ March 2009 Oversight Report, supra note 10.
\238\ April 2010 Oversight Report, supra note 1.
---------------------------------------------------------------------------
As detailed in Section F, in response to the problem of
foreclosures caused by unemployment, Treasury announced changes
to HAMP that will provide temporary assistance to unemployed
homeowners. Treasury also announced an FHA refinance option to
address negative equity, which provides incentive payments and
loss-sharing to encourage voluntary refinancing of underwater
mortgages into FHA mortgages. Additionally, the HHF may offer
assistance to unemployed or underwater borrowers in
participating states.
While the Panel applauds Treasury's efforts to address
unemployment and negative equity, it is unclear that these
initiatives will make significant headway against the scope and
scale of the problem. Julia Gordon, senior policy advisor for
the Center for Responsible Lending, testified before the Panel
that ``over the next several years, the toxic combination of
high unemployment and underwater loans could mean a stunning
total of more than 13 million foreclosures.'' \239\ As
discussed above, these estimates are in line with those of the
Federal Reserve.\240\ The Panel also remains concerned about
the timeliness of Treasury's response. Even if Treasury's
programs succeed, their impact will not be felt until almost
two years after the foreclosure mitigation initiative was first
launched, perhaps proving to be too little, too late.
---------------------------------------------------------------------------
\239\ Congressional Oversight Panel, Testimony of Julia Gordon,
senior policy counsel, Center for Responsible Lending, Transcript: COP
Hearing on TARP Foreclosure Mitigation Programs (Oct. 27, 2010)
(publication forthcoming) (online at cop.senate.gov/hearings/library/
hearing-102710-foreclosure.cfm).
\240\ See Section A, supra.
---------------------------------------------------------------------------
j. Manner in Which HAMP Rollout Happened
Since HAMP was announced in February 2009, Treasury has
initiated half a dozen foreclosure mitigation programs and
announced numerous changes to existing programs. The initial
announcement was referenced but included little specificity
about plans to modify second liens, to modify loans in
geographic areas where home prices have fallen precipitously,
and to encourage alternatives to foreclosure in cases where
modifications are not possible.\241\ In the months that
followed, Treasury announced five additional MHA programs and
released numerous supplemental directives or additional MHA
program guidelines.\242\ Among the changes introduced was an
increase in incentive payments to lenders, servicers, and
borrowers.\243\
---------------------------------------------------------------------------
\241\ U.S. Department of the Treasury, Making Home Affordable
Updated Detailed Program Description (Mar. 4, 2009) (online at
treasury.tpaq.treasury.gov/press/releases/reports/
housing_fact_sheet.pdf).
\242\ MHA Handbook for Servicers of Non-GSE Mortgages, supra note
224.
\243\ MHA Program Enhancements to Offer More Help for Homeowners,
supra note 147, at 2.
---------------------------------------------------------------------------
Although Treasury should be commended for trying new
approaches aimed at preventing foreclosures, loan servicers and
borrower advocates continue to express confusion about the
constant flux of new programs, new standards, new requirements,
and continued changes to HAMP, that make implementation more
complex.\244\ The Panel has also expressed concern with
Treasury's pattern of providing ever-greater incentives to
servicers and lenders. This pattern may create an incentive for
servicers and lenders to delay modifications in hopes that the
government will offer a better deal in the future. The Panel
raised many of these problems in its April 2010 report, and
SIGTARP identified many of these issues in a March 2010 report
examining root causes for HAMP's disappointing results.\245\
---------------------------------------------------------------------------
\244\ Senate Committee on Banking, Housing, and Urban Affairs,
Written Testimony of Barbara Desoer, president, Bank of America Home
Loans, Problems in Mortgage Servicing from Modification to Foreclosure,
at 3 (Nov. 16, 2010) (online at banking.senate.gov/public/
index.cfm?FuseAction=Files.View&FileStore_id=0a4db4b3-b131-4764-8247-
c665b9978e44) (``For example, Treasury, investors, and other
constituencies often change the requirements of their modification
programs. HAMP alone has had nearly 100 major program changes in the
past 20 months. Fannie and Freddie, as investors, have layered on
additional requirements, conditions and restrictions for HAMP
processing. When these changes occur, we and other servicers have to
change our process, train our staff, and update technology.'');
Congressional Oversight Panel, Written Testimony of Julia Gordon,
senior policy counsel, Center for Responsible Lending, COP Hearing on
TARP Foreclosure Mitigation Programs, at 19-20 (Oct. 27, 2010) (online
at cop.senate.gov/documents/testimony-102710-gordon.pdf) (``However,
given the way HAMP was created and implemented, many of these problems
are no surprise. First, the program repeatedly raised public
expectations that were then dashed when programs were not already
operational. This pattern began at the inception of the program, when
HAMP was announced to the public well before its infrastructure was in
place. Servicers were quickly overwhelmed by requests when they were
not yet prepared to qualify people for the program, thereby causing
many homeowners to be very disappointed early on. Despite this initial
bad experience with a lag between public announcement and rollout,
Treasury continued to make every subsequent program change the same
way. Rather than inform the servicers and wait for them to be ready
before informing the public, Treasury's routine was to release the
broad outline of a new initiative or guideline change and then have an
implementation date months away.'').
\245\ SIGTARP linked HAMP's disappointing results to undeveloped
program rules at the time of the initiation, confusion resulting from
continued guideline changes, the decision to permit servicers to start
trial modifications without supporting documentation from borrowers,
and limited marketing efforts by Treasury. SIGTARP Report--Factors
Affecting Implementation of HAMP, supra note 103.
---------------------------------------------------------------------------
Additionally, HAMP could have done a much better job with
borrower outreach. Servicer outreach is inherently problematic,
as servicers generally first contact borrowers in a debt
collector role. A government-run outreach campaign might have
been more effective at ensuring the maximum reach for HAMP. In
its March 2010 report, SIGTARP criticized Treasury for failing
to engage in an effective educational and promotional campaign
on HAMP. They note that a ``lack of basic understanding about
the program has sown confusion'' and servicers have reported
that many borrowers did not even understand that not everyone
is eligible for a HAMP modification.\246\ In addition, SIGTARP
noted that a Treasury-run outreach effort would have provided
the opportunity to educate the public about foreclosure rescue
scams and the dangers of other types of fraud.
---------------------------------------------------------------------------
\246\ SIGTARP Report--Factors Affecting Implementation of HAMP,
supra note 103, at 28.
---------------------------------------------------------------------------
k. Flawed Program Structure
In designing HAMP, Treasury made key choices regarding the
program structure leading to fundamental problems. First, the
basic decision to run HAMP through Treasury was an initial
misstep in the program. Treasury's expertise is in managing the
U.S. government's debt and ensuring financial institution
safety-and-soundness. Treasury had no prior experience running
a program like HAMP, nor did it have experience with housing
issues, with consumer programs, or with mortgage servicers.
Treasury is also the government agency with the closest ties to
financial institutions. It would have been appropriate to
consider the costs and benefits of other agencies, including
HUD, which may have greater experience in the relevant areas
and are less focused on financial institutions. More directly
applicable experience and distance from an interested set of
program participants might have ensured better program design.
A second design flaw arises because HAMP relies on mortgage
servicers as the program's interface with borrowers. As
detailed above, there are a variety of conflicts of interest
and misaligned incentives with mortgage servicers. These make
the decision to permit servicers to be the point of borrower
contact for HAMP modifications questionable. Alternatives would
have been either creating a central borrower interface that
would have intermediated between borrowers and servicers or
putting HAMP modifications up for competitive bidding. The
Panel notes that while it may have been difficult to put in
place an entirely new system in a short amount of time and a
central, government-operated interface would have greatly
increased theprogram's administrative burden, such a system
likely would have given Treasury or another agency much greater control
and information about servicer and borrower behavior.
In addition, Treasury's decision to use Fannie Mae as its
agent for HAMP program administration and Freddie Mac as its
agent for program compliance set up barriers to success.
Because of its own lack of housing policy expertise, Treasury
contracted with the GSEs to serve as its agents for HAMP under
no-bid contracts. Yet, the GSEs are highly conflicted because
they hold the credit risk on most mortgages in the United
States, and have their own operational concerns, raising the
question of why Treasury decided to saddle the GSEs with
oversight of HAMP.
Finally, HAMP tries to be both streamlined and
individualized in its treatment of loans. The result is that it
is neither sufficiently streamlined nor sufficiently
individualized, and gains the virtues of neither.
Part Two: The Future of HAMP
Although HAMP will continue to make trial modifications
until the end of 2012, the October 3, 2010 expiration of the
TARP had important consequences for the program. Throughout
much of 2009 and 2010, Treasury made changes to the program's
structure in an effort to address different aspects of the
foreclosure crisis. Because the TARP has now expired,
programmatic changes to HAMP are no longer possible, and no
additional TARP dollars can be allocated to HAMP, although the
program can continue to spend previously obligated money. This
does not mean, however, that the program will now run on auto-
pilot, or that Treasury is helpless to affect its eventual
impact on the foreclosure crisis.
In late 2010 and beyond, there are two main ways that
Treasury can affect the program's ultimate success. In order to
ensure that HAMP is as successful as possible over the long
term, Treasury should pursue both avenues energetically. First,
it should aggressively monitor Fannie Mae, which Treasury hired
to administer HAMP, and Freddie Mac, which it hired to enforce
compliance with the program's rules, and require these two
contractors to ensure that HAMP servicers are living up to the
obligations of their contracts with Treasury in order to
provide the appropriate outcomes for borrowers. Servicers must
be held accountable, for example, for failing to implement the
HAMP NPV model properly. By ensuring efficient administration
of the program and by requiring HAMP servicers to comply with
the program's rules, Treasury can maximize the number of
borrowers who will eventually benefit from a HAMP modification,
given the existing program structure and constraints.
Second, Treasury should focus more strongly on minimizing
redefaults within HAMP. Ultimately, a HAMP modification is only
a positive outcome if it allows its beneficiary to keep his or
her home, rather than simply delaying a foreclosure.
Preliminary redefault rates for HAMP offer some reason for
hope, but it is still too early to say whether redefault rates
will be lower than expectations over the five-year term of the
program. Data being collected by HAMP servicers can shed light
on what borrower and loan characteristics are correlated with
redefaults. Treasury should make aggressive use of these data
to encourage modifications that are sustainable. For example,
should Treasury find that redefaults are correlated with
unemployment, it can focus on the existing initiative to assist
unemployed borrowers. Similarly, a correlation with negative
equity could be addressed through more focused use of the
existing FHA Short Refinance program.
H. Treasury's Implementation of HAMP and Possible Improvements
To implement HAMP, Treasury has chosen to work primarily
through two financial agents, the GSEs Fannie Mae and Freddie
Mac. To date, there have been a number of problems, both
between the GSEs and the servicers, and between the GSEs and
Treasury. Despite working through agents, however, Treasury
remains ultimately responsible for HAMP's execution. This
section examines the structure that Treasury has constructed to
implement HAMP and what steps Treasury should take to improve
implementation of the program.
1. Role of Fannie Mae and Freddie Mac
a. Selection of Fannie Mae and Freddie Mac as Financial
Agents
Fannie Mae and Freddie Mac are GSEs chartered by Congress
with the mission of providing liquidity, stability, and
affordability to the U.S. housing and mortgage markets. In
2008, Fannie Mae and Freddie Mac combined lost more than $108
billion. In response, the Federal Housing Finance Agency (FHFA)
placed Fannie Mae and Freddie Mac into conservatorship on
September 7, 2008, in order to preserve each company's assets
and to restore them to sound and solvent condition. Since then,
the GSEs' legal status has been unclear. The GSEs remain
responsible for normal business operations and day-to-day
management.\247\ They are, however, effectively owned by the
government; Treasury has guaranteed their debts and FHFA has
all the powers of the management, board, and shareholders of
the GSEs.\248\ Thus, for example, FHFA installed new boards of
directors and CEOs, but could not hire or fire regular
employees. Congress has not yet determined what Fannie Mae and
Freddie Mac's ultimate status will be and is considering a
number of potential options.\249\
---------------------------------------------------------------------------
\247\ House Committee on Financial Services, Written Testimony of
Edward J. DeMarco, acting director, Federal Housing Finance Agency,
Compensation in the Financial Industry--Government Perspectives, at 6
(Feb. 25, 2010) (online at financialservices.house.gov/media/file/
hearings/111/fhfa_acting_director_demarco_testimony_for_2-25-10.pdf).
\248\ House Financial Services, Subcommittee on Capital Markets,
Insurance, and Government-Sponsored Enterprises, Written Testimony of
Edward J. DeMarco, acting director, Federal Housing Finance Agency, The
Future of Housing Finance: A Progress Update on the GSEs, at 2 (Sept.
15, 2010) (online at financialservices.house.gov/Media/file/hearings/
111/DeMarco091510.pdf).
\249\ The Dodd-Frank Wall Street Reform and Consumer Protection Act
requires Treasury to conduct a study and submit recommendations on
ending the conservatorship of Fannie Mae and Freddie Mac no later than
January 31, 2011. Dodd-Frank Wall Street Reform and Consumer Protection
Act, Pub. L. No. 111-203, Sec. 1074 (2010).
---------------------------------------------------------------------------
Treasury has explained its decision to select Fannie Mae
and Freddie Mac as the financial agents responsible for HAMP by
claiming that the GSEs were the only entities with the
experience, resources, and mortgage industry contacts necessary
to implement and administer the program quickly.\250\ The FHFA
also supported the GSEs' selection.\251\ According to Deputy
Assistant Secretary of Treasury Gary Grippo in testimony before
the Panel, ``[s]imply put, we made a determination that there
were no other parties with the capabilities and infrastructure
to operate a national mortgage modification program.'' \252\
The Panel's October 2010 report questions Treasury's selection
of the GSEs.\253\ The GSEs both had to rely heavily on
subcontractors to fulfill their responsibilities, and testimony
from a Panel hearing suggested that Fannie Mae and Freddie
Mac's roles as financial agents were not simply extensions of
their prior work.\254\
---------------------------------------------------------------------------
\250\ Treasury conversations with Panel staff (Sept. 27, 2010). See
also Congressional Oversight Panel, October Oversight Report: Examining
Treasury's Use of Financial Crisis Contracting Authority, at 78-82
(Oct. 14, 2010) (online at cop.senate.gov/documents/cop-101410-
report.pdf) (hereinafter ``October 2010 Oversight Report'').
\251\ Treasury conversations with Panel staff (Oct. 7, 2010)
(during which Mr. Grippo indicated that FHFA ``was involved from the
very beginning,'' provided its explicit authorization for Treasury's
selection, and ``always had firsthand knowledge of everything.''); FHFA
conversations with Panel staff (Oct. 8, 2010).
\252\ Congressional Oversight Panel, Testimony of Gary Grippo,
deputy assistant secretary for fiscal operations and policy, U.S.
Department of the Treasury, Transcript: COP Hearing on Treasury's Use
of Private Contractors (Sept. 22, 2010) (publication forthcoming)
(online at cop.senate.gov/hearings/library/hearing-092210-
contracting.cfm) (hereinafter ``Transcript Testimony of Gary Grippo'').
\253\ October 2010 Oversight Report, supra note 250, at 78-82.
\254\ See Treasury conversations with Panel staff (Sept. 23, 2010);
Written Testimony of Paul Heran, supra note 214, at 2.
---------------------------------------------------------------------------
b. Role of Fannie Mae
Fannie Mae's principal responsibilities under its $127
million agreement include implementing HAMP guidelines and
policies, serving as a point of contact for participating
mortgage servicers and instructing them on how to modify loans,
serving as paying agent to calculate subsidies and compensation
consistent with program guidelines, and coordinating with
Treasury and other parties toward achievement of the program's
goals.\255\
---------------------------------------------------------------------------
\255\ U.S. Department of the Treasury, Financial Agency Agreement
Between U.S. Department of the Treasury and Fannie Mae, at Exhibit A
(Feb. 18, 2009) (Contract No. TOFA-09-FAA-0002) (online at
www.financialstability.gov/docs/ContractsAgreements/
Fannie%20Mae%20FAA%20021809%20.pdf) (hereinafter ``Financial Agency
Agreement Between Treasury and Fannie Mae'').
---------------------------------------------------------------------------
In addition, Fannie Mae serves as the sole data collector
and record keeper for executed loan modifications and program
administration. Fannie Mae collects its data and records in its
``IR2'' database, and Treasury's Homeownership Preservation
Office (HPO) periodically validates the accuracy of IR2's
incentive payment data with the involvement of
PricewaterhouseCoopers. The review tests three elements: (1)
that all incentive payments were paid correctly; (2) that all
the loans for which incentive payments were made meet HAMP
eligibility requirements; and (3) that all of the IR2 data used
is internally consistent. If the validation process reveals
discrepancies, Treasury works with Fannie Mae to resolve them.
In light of recent issues regarding misleading actions by
servicers in potentially tens of thousands of foreclosure
cases,\256\ it is reasonable to raise concerns as to how
accurate Fannie Mae's reporting and recordkeeping is in the
context of HAMP, particularly as Fannie Mae officials have
indicated that primary responsibility for data accuracy lies
with the servicers and that the embedded data checks in IR2
check for completeness but not accuracy.\257\
---------------------------------------------------------------------------
\256\ For example, servicers using ``robo-signers'' filed false
affidavits in foreclosure cases across the country. For a more complete
discussion of foreclosure and mortgage irregularities, see November
2010 Oversight Report, supra note 79.
\257\ Treasury conversations with Panel staff (Oct. 4, 2010).
---------------------------------------------------------------------------
In addition to these recent concerns about servicer data,
servicers have long had problems with losing borrower
documentation.\258\ This persistent failing is of particular
importance to Fannie Mae's administration of HAMP because
incomplete documentation is a leading cause of canceled trial
modifications.\259\ To resolve this issue, the Panel has
repeatedly recommended that Treasury and Fannie Mae develop a
web portal to allow borrowers to submit and track modification
applications, to deliver application documents to servicers,
and to centralize information.\260\ In September 2009, Treasury
testified that it was working to build one,\261\ but in March
2010, Treasury stated that it was still considering whether it
should release a web portal at all.\262\ Treasury has since
decided not to develop its own web portal for three
reasons:\263\ (1) to the extent the portal would facilitate
moving documents between the borrower and the servicers,
Treasury could not guarantee that the documents would be
complete; (2) the security concerns arising from storing so
much personally identifiable information; and
(3) the advent of an industry solution, discussed below, led
Treasury to conclude that developing a Treasury web portal was
not the best use of public money.
---------------------------------------------------------------------------
\258\ See, e.g., SIGTARP Report--Factors Affecting Implementation
of HAMP, supra note 103, at 27; October 2009 Oversight Report, supra
note 15, at 107; Congressional Oversight Panel, Written Testimony of
Irwin Trauss, supervising attorney, Consumer Housing Unit, Philadelphia
Legal Assistance, Philadelphia Field Hearing on Mortgage Foreclosures
(Sept. 24, 2009) (online at cop.senate.gov/documents/testimony-092409-
trauss.pdf) (hereinafter ``Philadelphia Field Hearing on Mortgage
Foreclosures''); Congressional Oversight Panel, Written Testimony of
Deborah Goldberg, director, Hurricane Relief Project, National Fair
Housing Alliance, Philadelphia Field Hearing on Mortgage Foreclosures
(Sept. 24, 2009) (online at cop.senate.gov/documents/testimony-092409-
goldberg.pdf) (hereinafter ``Written Testimony of Deborah Goldberg'');
Floyd Norris, Are Banks Losing Lots of Documents?, New York Times (Dec.
4, 2009) (online at norris.blogs.nytimes.com/2009/12/04/are-banks-
losing-lots-of-documents/); Mary Kane, White House, Loan Servicers
Point Fingers as Foreclosure Plan Fails, The Minnesota Independent
(Jan. 4, 2010) (online at minnesotaindependent.com/52967/white-house-
loan-servicers-point-fingers-as-foreclosure-plan-fails); Bendix
Anderson, Second Chance for Loan Modifications, Housing Watch (Mar. 23,
2010) (online at www.housingwatch.com/2010/03/23/second-chance-for-
loan-modifications/); Bendix Anderson, HAMP Offers New Home for
Borrowers, Housing Watch (May 28, 2010) (online at
www.housingwatch.com/2010/05/28/hamp-offers-new-hope-for-borrowers/);
Stephanie Armour, More Homeowners Get Help Outside of Federal Program,
USAToday (July 23, 2010) (online at www.usatoday.com/money/economy/
housing/2010-07-23-mortgages23_CV_N.htm).
\259\ See, e.g., MHA Servicer Performance Report, supra note 38, at
5. There were 196,835 trial modifications that did not convert to
permanent modifications because of lack of documentation. Data provided
by Treasury. Given the extensive anecdotal evidence noted above, in
footnote 258, it is reasonable to believe that at least some of these
conversions would not have been denied absent servicers losing borrower
documentation.
\260\ March 2009 Oversight Report, supra note 10, at 52; October
2009 Oversight Report, supra note 15, at 111; April 2010 Oversight
Report, supra note 1, at 83.
\261\ Congressional Oversight Panel, Written Testimony of Seth
Wheeler, senior advisor, U.S. Department of the Treasury, Philadelphia
Field Hearing on Mortgage Foreclosures, at 6 (Sept. 24, 2009) (online
at cop.senate.gov/documents/testimony-092409-wheeler.pdf) (hereinafter
``Written Testimony of Seth Wheeler'').
\262\ Treasury conversations with Panel staff (Mar. 24, 2010).
\263\ The web portal referred to here should not be confused with
the NPV web portal Treasury is required to develop by the Dodd-Frank
Wall Street Reform and Consumer Protection Act. The web portal being
discussed is not for NPV calculations, but to centralize borrower
documentation and submit it to servicers.
---------------------------------------------------------------------------
The Panel is pleased to note, however, that in November
2009, HOPE NOW, a coalition of mortgage companies, investors,
counselors, and other mortgage market participants,\264\
piloted the HOPE LoanPort web portal.\265\ At present, the
LoanPort is primarily a ``counselor's portal.'' In order to use
the LoanPort, borrowers first approach a participating housing
counselor to modify a loan with a participating servicer. The
counselor works with the borrower to build a loan modification
application package, and then the counselor goes on the
LoanPort. The LoanPort has certain required fields and
mandatory documents that must be provided in order to submit
the application. Once the counselor has entered all the
required information,\266\ the application package is delivered
to the servicer. The servicer is then required to provide the
counselor with updates on the status of the application every
10 days. The process from borrower approaching a housing
counselor to the servicer reaching a decision on the
application takes an average of 44 days (10 days for the
housing counselor to complete the application package with the
borrower and 34 days for the servicer to reach a
decision).\267\ There are currently 12 servicers,\268\ and over
1,800 housing counselors from more than 420 organizations
participating in the LoanPort.\269\ As of November 19, 2010,
counselors had started a total of 8,585 applications on the
LoanPort. Of those, 3,320 were still being completed, and 4,784
had been finished and submitted to the servicers.\270\ Though
an important start, this annual volume of applications is a
tiny fraction of the 672,439 HAMP trial modifications and the
929,148 proprietary modifications started last year.\271\ Nor,
unfortunately, is it significant in comparison to the 1,344,337
foreclosures initiated last year.\272\
---------------------------------------------------------------------------
\264\ HOPE NOW Alliance, About Us (online at www.hopenow.com/
hopenow-aboutus.php) (accessed Dec. 10, 2010).
\265\ The HOPE LoanPort is now operated by its own non-profit
group. The LoanPort completed its pilot phase in June 2010. Testimony
of Faith Schwartz, supra note 78.
\266\ Documents are date stamped as they are submitted, providing a
record to minimize later documentation problems.
\267\ HOPE LoanPort conversations with Panel staff (Nov. 15, 2010).
\268\ HOPE LoanPort conversations with Panel staff (Nov. 15, 2010).
The servicers are: American Home Mortgage Servicing, Bank of America,
Bayview Loan Servicing, Chase, Citi, GMAC, Ocwen Loan Servicing,
OneWest Bank, PNC, Saxon, Sun Trust Mortgage, and Wells Fargo. HOPE NOW
Alliance, Partners (online at www.hopenow.com/partners.php) (accessed
Dec. 10, 2010). An additional three servicers, Nation Star, Met Life,
and SPS Servicing, will be live on the LoanPort by the end of the year.
HOPE LoanPort conversations with Panel staff (Nov. 19, 2010).
\269\ HOPE LoanPort conversations with Panel staff (Nov. 15, 2010).
An additional three servicers will be live on the LoanPort by the end
of the year. The 1,800 housing counselors are located in 47 states.
Before allowing a housing counselor access, HOPE LoanPort provides
webinar trainings. HOPE LoanPort conversations with Panel staff (Nov.
15, 2010).
\270\ The remaining 481 had a variety of dispositions. E-mail from
Larry Gilmore, president and CEO, HOPE LoanPort to Panel staff (Nov.
19, 2010).
\271\ Data provided by Treasury; MHA Servicer Performance Report,
supra note 38, at 2 (year measured from November 2009 through October
2010); OCC/OTS Mortgage Metrics Report, Second Quarter 2010, supra note
31, at 2.
\272\ OCC/OTS Mortgage Metrics Report, Second Quarter 2010, supra
note 31, at 2.
---------------------------------------------------------------------------
HOPE LoanPort has also developed but not deployed a
``borrower portal.'' \273\ The borrower portal allows borrowers
to interact directly with the LoanPort, uploading documents and
information without working through housing counselors. The
LoanPort would then ensure the applications were complete and
verify the borrower's information. After this was finished,
LoanPort would deliver the application package to the servicer.
This borrower portal has not been deployed nationwide because
of lack of funding. HOPE LoanPort estimates it would cost under
$30 million.\274\
---------------------------------------------------------------------------
\273\ A variant of the borrower portal has been deployed in
Arizona. In Arizona, a borrower submits information through the
LoanPort. That information is then transferred to a housing counselor
in the borrower's area, who will contact the borrower and work with
them to complete and verify the application package. From then on, the
counselor will use the standard counselor's portal. HOPE LoanPort
conversations with Panel staff (Nov. 15, 2010).
\274\ HOPE LoanPort conversations with Panel staff (Nov. 15, 2010).
---------------------------------------------------------------------------
Going forward, HOPE LoanPort is trying to increase the
number of borrowers using the system and to expand the number
of participating housing counselor organizations to over 1,000
by June 2011. It is also working with state housing agencies to
develop partnerships. In addition, it is seeking to attract
funding to deploy the borrower portal nationwide. Further, the
LoanPort is in conversations with the GSEs to engage the
investor community and move them to push servicers to use the
LoanPort. Though 12 servicers currently participate, they do
not use the LoanPort as their primary point of entry for
modification applications. Freddie Mac has expressed interest
but Fannie Mae has been reluctant. Fannie Mae has also refused
to allowHOPE LoanPort to adapt Fannie Mae's counselor software,
Home Counselor Online, to the LoanPort system in order to avoid the
need to enter application information on both systems.\275\
---------------------------------------------------------------------------
\275\ HOPE LoanPort conversations with Panel staff (Nov. 15, 2010).
---------------------------------------------------------------------------
Treasury is supportive of the new portal,\276\ and was
consulted during its development, but was not responsible for
its creation.\277\ It stays abreast of new developments with
the LoanPort but has no input in how it is run.\278\
---------------------------------------------------------------------------
\276\ Transcript Testimony of Phyllis Caldwell, supra note 108.
\277\ HOPE LoanPort, About Hope LoanPort (online at
www.hopeloanportal.org/aboutloanport.php#) (accessed Dec. 10, 2010).
\278\ Treasury conversations with Panel staff (Nov. 9, 2010).
---------------------------------------------------------------------------
Though the LoanPort as it currently stands is an important
step forward, only a few borrowers are currently using the
system. Deploying the borrower portal is essential and should
be completed as soon as possible. Allowing direct borrower
access to the system will both increase the LoanPort's
utilization and most effectively solve documentation problems.
The Panel recommends that Treasury examine ways to facilitate
the LoanPort's development in this direction, including
exploring funding the LoanPort's borrower portal. Treasury
should also work with HAMP servicers to encourage them to push
the LoanPort as their primary point of entry for applications.
In addition, Treasury should work with FHFA to determine why
Fannie Mae has been reluctant to use the LoanPort and to
overcome their resistance.
Fannie Mae has additional functions relating to servicer
support and operates the HOPE Hotline. The hotline provides
free housing assistance to borrowers and is the primary means
for borrowers to escalate concerns about how a servicer handled
their claim.\279\ Treasury, however, has not explicitly
informed borrowers that the hotline can be used to escalate
complaints.\280\ If borrowers assert that their modifications
were wrongly denied, complaints are escalated to HUD-approved
housing counselors at MHA Help. That counselor will set up a
three-way call with the servicer and borrower to attempt to
resolve the issue. If the counselor is unable to broker a
solution, and further intervention is needed, the complaint can
then be sent to the counseling agencies' management, which will
consult with higher-level officials at the servicer level.\281\
If the issue still cannot be resolved, the complaint will be
escalated to the HAMP Solution Center at Fannie Mae.\282\
Throughout this process, the housing counselors and Fannie Mae
do not have independent authority to resolve a borrower's
complaint. They rely on the servicer to make the ultimate
decision as to whether the initial modification decision was
wrong.\283\ If a complaint has been escalated through to the
HAMP Solution Center and alleges that the servicer did not
follow HAMP guidelines, the complaint can be escalated one more
time, to the MHA Compliance Committee at Treasury, which does
have authority to require the servicer to change a modification
decision.\284\ Since Fannie Mae began to report the information
in June 2010, the number of complaints to the hotline, though
small, has increased every month.\285\
---------------------------------------------------------------------------
\279\ GAO Report on Foreclosure Mitigation Programs, supra note
203, at 23-26.
\280\ GAO Report on Foreclosure Mitigation Programs, supra note
203, at 25-26. Treasury does inform borrowers through a number of
channels that the hotline can be used for problems related to HAMP, but
not that it can be used to escalate complaints. Treasury has explained
this decision by stating that it wanted a single number for everything
and that Treasury did not wish to confuse the market by identifying
that number as a place to escalate complaints. Treasury conversations
with Panel staff (Nov. 16, 2010).
\281\ Treasury conversations with Panel staff (Nov. 16, 2010).
\282\ GAO Report on Foreclosure Mitigation Programs, supra note
203.
\283\ Treasury conversations with Panel staff (Nov. 16, 2010). On
November 3, 2010, Treasury issued HAMP Supplemental Directive 10-15.
The Directive requires servicers to have written procedures and
sufficient personnel in place to provide timely responses to escalated
complaints. The Directive further requires that, for large servicers,
the personnel who review escalated cases must be independent from the
servicer personnel who made the initial modification decision. It also
compels improved communications between the borrower and servicer
during the escalation process. The Directive will take effect on
February 1, 2011. Supplemental Directive 10-15, supra note 216.
\284\ Treasury conversations with Panel staff (Nov. 16, 2010).
\285\ U.S. Department of the Treasury, Making Home Affordable
Program: Servicer Performance Report Through June 2010, at 8 (Aug. 6,
2010) (online at www.financialstability.gov/docs/
June%20MHA%20Public%20Revised%20080610.pdf); U.S. Department of the
Treasury, Making Home Affordable Program: Servicer Performance Report
Through July 2010, at 8 (Aug. 20, 2010) (online at
www.financialstability.gov/docs/JulyMHAPublic2010.pdf); U.S. Department
of the Treasury, Making Home Affordable Program: Servicer Performance
Report Through August 2010, at 8 (Sept. 22, 2010) (online at
www.financialstability.gov/docs/AugustMHAPublic2010.pdf) (hereinafter
``Making Home Affordable Program: Servicer Performance Report Through
August 2010''); U.S. Department of the Treasury, Making Home Affordable
Program: Servicer Performance Report Through September 2010, at 9 (Oct.
25, 2010) (online at www.financialstability.gov/docs/
Sept%20MHA%20Public%202010.pdf) (hereinafter ``MHA Servicer Performance
Report Through September 2010''); MHA Servicer Performance Report,
supra note 38, at 8. In the most recent report, for October 2010, the
complaint rate was at 5.7 percent.
---------------------------------------------------------------------------
c. Role of Freddie Mac
Treasury obligated $92 million to retain Freddie Mac to act
as the HAMP compliance agent, responsible for ensuring that
participating servicers satisfy their obligations under the
HAMP SPAs.\286\ Freddie Mac is required, among other tasks, to
``conduct examinations and review servicer compliance with the
published rules for the program and report results to the
Treasury.'' \287\
---------------------------------------------------------------------------
\286\ U.S. Department of the Treasury, Financial Agency Agreement
Between the U.S. Department of the Treasury and Freddie Mac, at Exhibit
A (Feb. 18, 2009) (Contract No. TOFA-09-FAA-0003) (online at
www.financialstability.gov/docs/ContractsAgreements/
Freddie%20Mac%20Financial%20Agency%20Agreement.pdf) (hereinafter
``Financial Agency Agreement Between Treasury and Freddie Mac'').
\287\ Id. at Exhibit A.
---------------------------------------------------------------------------
Freddie Mac conducts periodic on-site examinations of
servicers to evaluate readiness, governance, and implementation
of HAMP requirements. In addition, Freddie Mac reviews the
disbursements of incentive payments to servicers. Freddie Mac
also performs periodic assessments of the use of the NPV model
and an ongoing review of all servicers with recoded NPV models.
Where issues are identified in these recoded models, servicers
are asked to validate their results in the Treasury model.\288\
If servicers' models do not meet Treasury's NPV specifications,
Freddie Mac will require the servicers to discontinue use of
their models and revert back to the NPV application available
from Treasury through the MHA Servicer Portal.\289\
---------------------------------------------------------------------------
\288\ Id. at Exhibit A.
\289\ TARP Two Year Retrospective, supra note 98, at 75.
---------------------------------------------------------------------------
Freddie Mac is also authorized to conduct both unannounced
and announced audits of servicers. It has chosen not to do any
unannounced audits, stating that they are not practical because
of the considerable coordination required to arrange a
productive audit.\290\ Freddie Mac, however, does conduct
regularly scheduled announced audits of servicers. The
frequency of these audits is determined by the size, risk, and
volume of the servicer.\291\ The ten largest servicers are
subject to nearly continuous review and there are 42 servicers
who have not yet been examined on any aspect of their
performance. These 42 servicers are the smallest servicers and
represent a tiny fraction of the total volume of loans.\292\
Freddie Mac stated that they will review these servicers at
some point, but not necessarily within the first year. The
schedule of reviews has been approved by Treasury.\293\
---------------------------------------------------------------------------
\290\ Treasury and Freddie Mac conversations with Panel staff
(Sept. 27, 2010).
\291\ The larger, riskier, and higher volume servicers are audited
more frequently. Treasury and Freddie Mac conversations with Panel
staff (Sept. 27, 2010).
\292\ Congressional Oversight Panel, Written Responses of Phyllis
Caldwell, chief of the Homeownership Preservation Office, U.S.
Department of the Treasury, COP Hearing on TARP Foreclosure Mitigation
Programs, at 1 (Oct. 27, 2010) (online at cop.senate.gov/documents/
testimony-102710-caldwell-qfr.pdf).
\293\ Treasury and Freddie Mac conversations with Panel staff
(Sept. 27, 2010).
---------------------------------------------------------------------------
In cases of noncompliance, Freddie Mac may direct servicers
to perform remediation activities in consultation with
Treasury. For example, Freddie Mac may direct a servicer who
failed to comply with its solicitation obligations to restrict
foreclosure activities. It is Treasury's MHA Compliance
Committee that can make decisions to impose financial remedies,
which may include withholding or reducing incentive payments to
servicers, requiring repayments of prior incentive payments
made to servicers with respect to affected loans, or requiring
additional servicer oversight.\294\ Treasury, though, has
eschewed levying penalties in favor of directing servicers to
change their processes.\295\ In some instances, though, this
committee has ordered additional oversight, for example,
requiring servicers to assess their solicitations' impact on
borrowers.\296\ Treasury also temporarily withheld incentive
payments on approximately 1,400 loans until servicers resolved
several data issues. After servicers complied, the payments
that had been withheld were paid out, though there are still
132 loans with outstanding problems. For these 132 loans,
Treasury is considering permanently withholding the incentive
payments.\297\
---------------------------------------------------------------------------
\294\ TARP Two Year Retrospective, supra note 98, at 76.
\295\ House Committee on Financial Services, Transcript: Robo-
Signing, Chain of Title, Loss Mitigation, and Other Issues in Mortgage
Servicing (Nov. 18, 2010) (publication forthcoming) (online at
financialservices.house.gov/Hearings/hearingDetails.aspx?NewsID=1376).
Treasury has stated that it may begin to use financial penalties more
frequently as the programs mature. House Committee on Financial
Services, Written Testimony of Phyllis Caldwell, chief of the
Homeownership Preservation Office, U.S. Department of the Treasury,
Robo-signing, Chain of Title, Loss Mitigation, and Other Issues in
Mortgage Servicing, at 8-9 (Oct. 27, 2010)
(financialservices.house.gov/Media/file/hearings/111/
Caldwell111810.pdf).
\296\ Treasury conversations with Panel staff (Oct. 21, 2010).
\297\ Treasury conversations with Panel staff (Oct. 21, 2010).
---------------------------------------------------------------------------
In addition, Treasury asked Freddie Mac to develop a
``second look'' process to audit a sample of HAMP modification
files in order to double-check the servicer's determination on
the request.\298\ Freddie Mac retrieves servicer source
documents and loan files on a sample of their non-performing
loans, as well as data from Fannie Mae's IR2 database of HAMP
records. According to Treasury, individual borrowers whose
files are being reviewed are not contacted in order to avoid
added stress to borrowers. Freddie Mac then independently
reviews the documents to ensure that it agrees with the
decision the servicer reached on the modification.\299\ If
Freddie Mac disagrees with a decision, it can require further
action, including requiring servicers to reevaluate loans that
were not offered HAMP modifications,\300\ to submit further
documentation, to clarify a loan status, or to engage in
process remediation, training, or policy clarification.\301\
---------------------------------------------------------------------------
\298\ U.S. Department of the Treasury, Making Home Affordable
Program On Pace To Offer Help to Millions of Homeowners (Aug. 4, 2009)
(online at www.financialstability.gov/latest/tg252.html).
\299\ Treasury conversations with Panel staff (Sept. 27, 2010).
Second look reviews are frequently done on-site.
\300\ If the home is in foreclosure when Freddie Mac requires the
servicer to reevaluate the modification decision, the servicer must
forestall the foreclosure sale. MHA Servicer Performance Report Through
September 2010, supra note 285, at 1.
\301\ MHA Servicer Performance Report Through September 2010, supra
note 285, at 1.
---------------------------------------------------------------------------
Freddie Mac publicly releases some results from these
second look reviews, including specific results for several
servicers.\302\ This disclosure, however, is limited. Freddie
Mac only notes whether it disagreed with, could not yet
determine, or agreed with a servicer's determination. This is
not enough information for the public to evaluate the servicers
effectively. For example, the public is unable to tell if
Freddie Mac's determination that it disagrees with the servicer
is the result of a technical difference or a gross violation.
Furthermore, Freddie Mac has not focused follow-up reviews on
the servicers with the worst records. Some servicers were
reviewed twice, others once, but the servicers with the most
questionable decisions in their first review were not those who
were reviewed again.
---------------------------------------------------------------------------
\302\ See U.S. Department of the Treasury, Making Home Affordable
Program: Servicer Performance Report Through May 2010, at 6 (June 21,
2010) (online at www.financialstability.gov/docs/
May%20MHA%20Public%20062110.pdf); MHA Servicer Performance Report
Through September 2010, supra note 285, at 11.
---------------------------------------------------------------------------
More troublingly, recent statements raise questions about
Freddie Mac's willingness to pursue servicers who violate
program guidelines. In response to revelations that servicers
have been using ``robo-signers'' to submit false affidavits in
thousands of foreclosure cases, Freddie Mac noted that ``we
believe that our seller/servicers would be in violation of
their servicing contracts with us to the extent that they
improperly executed documents in foreclosure or bankruptcy
proceedings.'' \303\ Trying to enforce Freddie Mac contractual
rights, however, ``may negatively impact our relationships with
these seller/servicers, some of which are among our largest
sources of mortgage loans.'' \304\ The Panel condemns this
sentiment. If Freddie Mac is hesitant to jeopardize their
relationships with servicers to enforce their rights in their
own book of business, it is reasonable to worry that they may
be similarly unwilling to risk these relationships on
Treasury's behalf by aggressively overseeing HAMP servicers. It
is important to note, though, that both Freddie Mac and Fannie
Mae have threatened to penalize financially thousands of
servicers if they did not fix their foreclosure practices,\305\
and that FHFA is having the GSEs pursue repurchase requests
from many of these same institutions.\306\ Nevertheless, the
potential conflict of interest raises concerns.
---------------------------------------------------------------------------
\303\ Federal Home Loan Mortgage Corporation, Form 10-Q for the
Quarterly Period Ended September 30, 2010, at 190 (Nov. 3, 2010)
(online at www.sec.gov/Archives/edgar/data/1026214/000102621410000053/
f71398e10vq.htm) (hereinafter ``Federal Home Loan Mortgage Corporation
Form 10-Q'').
\304\ Federal Home Loan Mortgage Corporation Form 10-Q, supra note
303, at 190 (``While we believe that our seller/servicers would be in
violation of their servicing contracts with us to the extent that they
improperly executed documents in foreclosure or bankruptcy proceedings,
as such contracts require that foreclosure proceedings be conducted in
accordance with applicable law, it may be difficult, expensive, and
time consuming for us to enforce our contractual rights. Our efforts to
enforce our contractual rights may negatively impact our relationships
with these seller/servicers, some of which are among our largest
sources of mortgage loans.'').
\305\ American Bankers Association, Fannie, Freddie Place Lenders
on Notice in Foreclosure Debacle, 2 Mortgage Lending Bulletin 15, at 5-
6 (Oct. 21, 2010) (online at www.aba.com/NR/rdonlyres/A73585B8-6541-
4A77-8184-1BEFB4AB2E60/69412/10222010MortgageLendingBulletin.pdf).
\306\ Senate Committee on Banking, Housing, and Urban Affairs,
Testimony of Edward DeMarco, acting director, Federal Housing Finance
Agency, Transcript: Problems in Mortgage Servicing From Modification to
Foreclosure, Part II (Dec. 1, 2010) (publication forthcoming)
(online at banking.senate.gov/public/
index.cfm?FuseAction=Hearings.LiveStream&Hearing_id=ea6d7672-f492-4b1f-
be71-b0b658b48bef).
---------------------------------------------------------------------------
2. Current Oversight Mechanisms
Treasury has a comprehensive oversight structure in place
to monitor Fannie Mae and Freddie Mac. Primary responsibility
inside Treasury for overseeing the GSEs rests with the Office
of Financial Agents (OFA), which monitors the GSEs in their
role as financial agents. However, OFA has a collaborative
relationship with three other offices inside Treasury's Office
of Financial Stability (OFS): HPO, the Office of the Chief
Financial Officer, and OFS-Compliance. HPO is particularly
important in this collaboration as it is the program office and
therefore makes the policy decisions on how HAMP is structured
and implemented. In addition, Treasury has a Making Home
Affordable program committee that meets weekly as well as
several working committees (centered on compliance, budgeting,
and governance issues) to oversee the GSEs. These committees
meet on a regular basis and include interlocking membership
from each of the different Treasury offices (referenced above)
that are tasked with monitoring and oversight responsibilities
for Fannie Mae and Freddie Mac.\307\ Treasury's monitoring and
supervision of the GSEs are also closely coordinated with
general oversight and risk assessment by FHFA as part of the
conservatorship process. Members of the FHFA conservatorship
team continuously oversee Fannie Mae's and Freddie Mac's
financial agency agreements, monitor the tasks that Treasury
asks the GSEs to perform as a risk assessment measure, and help
ensure that they are compensated appropriately for their
work.\308\
---------------------------------------------------------------------------
\307\ Treasury and Fannie Mae conversations with Panel staff (Oct.
4, 2010).
\308\ FHFA conversations with Panel staff (Oct. 4, 2010).
---------------------------------------------------------------------------
Treasury uses several methods to evaluate and manage Fannie
Mae's and Freddie Mac's performance and compliance with their
financial agency agreements and the performance of their
fiduciary obligation to Treasury. On the performance side,
these include qualitative measures (such as assessments of cost
containment, responsiveness, and nature of their business
relationship with Treasury), and quantitative measures (such as
how they process transactions, the timeliness and accuracy of
their reports, and the number of servicer reviews
conducted).\309\ Treasury staff are also in frequent informal
contact with GSE staff at many levels within each
organization.\310\
---------------------------------------------------------------------------
\309\ Transcript Testimony of Gary Grippo, supra note 252.
\310\ Treasury and Freddie Mac conversations with Panel staff
(Sept. 27, 2010); Treasury and Fannie Mae conversations with Panel
staff (Oct. 4, 2010).
---------------------------------------------------------------------------
On the compliance side, the GSEs are required to report to
Treasury on internal controls, risk assessments, information
technology security, employee training, and how they have
revisited their conflicts of interest mitigation plans.\311\
The financial agency agreements also require that Fannie Mae
and Freddie Mac self-certify annually that they are complying
with 11 selected terms of the agreements and review the
effectiveness of their internal controls on an annual
basis.\312\ Also on an annual basis, Treasury staff conduct on-
site visits to review the processes and controls of each agent
at their offices.\313\ Treasury also requires agents to submit
information regarding conflicts of interest, which it reviews
on an ongoing basis.\314\
---------------------------------------------------------------------------
\311\ Financial Agency Agreement Between Treasury and Fannie Mae,
supra note 255; Financial Agency Agreement Between Treasury and Freddie
Mac, supra note 286.
\312\ Financial Agency Agreement Between Treasury and Fannie Mae,
supra note 255, at Sec. 16, Exhibit D; Financial Agency Agreement
Between Treasury and Freddie Mac, supra note 286, at Sec. 16, Exhibit
D.
\313\ Congressional Oversight Panel, Joint Written Testimony of
Gary Grippo, deputy assistant secretary for fiscal operations and
policy, and Ronald W. Backes, director of procurement services, U.S.
Department of the Treasury, COP Hearing on Treasury's Use of Private
Contractors, at 6 (Sept. 22, 2010) (online at cop.senate.gov/documents/
testimony-092210-treasury.pdf).
\314\ Treasury conversations with Panel staff (Sept. 16, 2010). For
a more complete discussion of Treasury's monitoring of contractor and
agent conflicts of interest, see October 2010 Oversight Report, supra
note 250, at Sections B.3 & H.
---------------------------------------------------------------------------
3. Implementation Failures and Ways To Improve
Despite the intricate oversight mechanisms Treasury has
developed to ensure the GSEs' performance, HAMP has suffered a
number of implementation problems. First, there have been
problems with the GSEs' oversight of the servicers. Secretary
Geithner himself noted that ``servicers have done a terrible
job of making sure that they are doing everything they can to
meet the needs of their customers . . . And they still have
some distance to go to try to make up for that series of basic,
how should I say it, mistakes, inadequacies, in performance.''
\315\ Since this statement on June 22, 2010, servicers have
made little progress. Of particular note are long-standing
public complaints that servicers lose borrower
documentation.\316\ For example, the California Reinvestment
Coalition surveyed more than 50 housing counselors from 40
different housing counseling agencies and found that 100
percent of the counselors said it was very common for servicers
to request documents the counselors had already submitted.\317\
A similar study by the National Counsel of La Raza found that
60 percent of counselors reported that servicers usually or
always lost borrower documents.\318\ Additionally, GAO reported
in March 2010 that different servicers were applying different
criteria in determining whether particular borrowers are at
risk of imminent default, and therefore if they were eligible
for HAMP, introducing inconsistencies into a standardized
national program.\319\ In June 2010, GAO found further servicer
errors that created inconsistencies, including that 15 of the
largest 20 participating servicers did not comply with all
program guidelines when implementing the NPV model.\320\ And
there is mounting evidence of widespread irregularities by
servicers in foreclosure proceedings, including the use of
``robo-signers.'' The failure of Fannie Mae and Freddie Mac to
detect foreclosure irregularities by the servicers they hire as
part of their own business raises questions as to their
credibility in overseeing the servicers' adherence to HAMP
standards.
---------------------------------------------------------------------------
\315\ Congressional Oversight Panel, Testimony of Timothy Geithner,
secretary, U.S. Department of the Treasury, Transcript: COP Hearing
With Treasury Secretary Timothy Geithner (June 22, 2010) (publication
forthcoming) (online at cop.senate.gov/hearings/library/hearing-062210-
geithner.cfm).
\316\ See, e.g., October 2009 Oversight Report, supra note 15, at
107; Written Testimony of Deborah Goldberg, supra note 258; Written
Testimony of Deborah Goldberg, supra note 258.
\317\ Chasm Between Words and Deeds IV: HAMP Is Not Working, supra
note 139, at 4.
\318\ National Council of La Raza, Saving Homes and Homeownership:
Perspectives From Housing Counselors, at 3 (Apr. 14, 2010) (online at
www.nclr.org/index.php/publications/
saving_homes_and_homeownership_perspectives_From_housing_counselors).
Another survey, a random sample of 42 active cases conducted by a
housing counselor organization, found that servicers had lost
documentation in 28 percent of cases. E-mail from Cheyenne Martinez-
Boyette, homeownership program lead, Mission Economic Development
Agency, to Panel staff (Dec. 8, 2010).
\319\ U.S. Government Accountability Office, Troubled Asset Relief
Program: Home Affordable Modification Program Continues To Face
Implementation Challenges, at 13-14 (Mar. 25, 2010) (GAO-10-556T)
(online at www.gao.gov/new.items/d10556t.pdf).
\320\ GAO Report on Foreclosure Mitigation Programs, supra note
203, at 20-21.
---------------------------------------------------------------------------
There have also been several implementation failures on the
part of the GSEs themselves. OFS initially had a number of
concerns about Freddie Mac, including the use of unqualified
staff to perform audits.\321\ Treasury and Freddie Mac
developed a detailed remediation plan, and Treasury has not
reported any problems since.\322\ In addition, in its October
2010 report, the Panel noted an instance when Fannie Mae
publicly released a table with incorrect information. The error
led Treasury to acknowledge that they lacked adequate controls
with respect to the communication of program requirements and
the validation of data.\323\ Since then, Treasury has taken
steps to improve its oversight, hiring MITRE Corporation to
correct the table in question as well as to validate and
improve Treasury's data production process for all HAMP
reports. Though both GSEs have resolved these issues as they
are identified, the fact that such significant errors were able
to occur raises questions about how closely Treasury is
overseeing the GSEs.
---------------------------------------------------------------------------
\321\ Office of the Special Inspector General for the Troubled
Asset Relief Program, Quarterly Report to Congress, at 102 (Oct. 21,
2009) (online at www.sigtarp.gov/reports/congress/2009/
October2009_Quarterly_Report_to_Congress.pdf).
\322\ Id. at 102.
\323\ Treasury conversations with Panel staff (Sept. 23, 2010).
---------------------------------------------------------------------------
Notably, these problems are faults in implementation, not
program design. Treasury must work to ensure that these sorts
of errors do not continue to occur. These situations underscore
the need to improve oversight of the GSEs, which will help
prevent future problems and maximize HAMP's impact.
For example, OFA evaluates the GSEs on both qualitative
measures (such as assessments of cost containment,
responsiveness, and nature of their business relationship with
Treasury), and quantitative measures (such as how they process
transactions, the timeliness and accuracy of their reports, and
the number of servicer reviews conducted).\324\ These metrics
are focused completely on the interaction between Treasury and
the GSEs, not on how the servicers deal with borrowers. This
incentivizes the GSEs to focus on developing intricate
processes and not necessarily on ensuring that borrowers' needs
are met. Treasury should instead adopt measures that track the
end results Treasury wishes to see, such as the number of
modifications a servicer is making as compared to its peers or
the number of complaints a servicer receives. Treasury itself
initially proposed several metrics, including ``average
borrower wait time in response to inquiries and response time
for completed applications,'' but never adopted them.\325\
Evaluating the GSEs on the end results rather than on the
processes they follow will encourage them to push servicers to
do the same.
---------------------------------------------------------------------------
\324\ Transcript Testimony of Gary Grippo, supra note 252.
\325\ Written Testimony of Seth Wheeler, supra note 261, at 6.
---------------------------------------------------------------------------
Another simple change Treasury could make is requiring
Freddie Mac to call borrowers or their housing counselors
during its Second Look reviews. At present, Freddie Mac looks
at a servicer's source data and loan file, but does not call
the borrowers to avoid adding to their stress. Calling
borrowers or their counselors, however, is critical to
assessing the accuracy of a servicer's determination. For
example, without direct contact there is no way to validate
that the servicer has all of the documentation the borrower
sent. Establishing this fact is particularly important in light
of the fact that servicers' losing documentation has been a
consistent problem and that one of the most common causes of
modifications being denied is missing documentation.
Similarly, Treasury should reform its process for
escalating complaints. Despite having three different layers of
review for complaints, it is the servicers themselves who
ultimately choose whether to alter the borrower's modification
decision.\326\ The escalation process thus preserves servicers'
discretion to decide on a modification, relying on them to
determine if they made a mistake initially. Though the
requirement in Supplemental Directive 10-15 that personnel who
review escalated complaints in large servicers be independent
from the staff that made the initial modification decision is a
step in the right direction, it is not sufficient.\327\
Treasury should remove the decision from the servicers' hands
in favor of an independent and enforceable review. To maximize
the HOPE Hotline's impact, moreover, Treasury should clearly
inform borrowers that the hotline can be used to escalate
complaints. For example, Treasury could easily post a
notification to that effect on its Making Home Affordable Web
site. In addition, Treasury should include the information in
other communications with borrowers. The combination of these
two changes would improve servicer accountability to borrowers,
as it would provide borrowers a clear and independent pathway
to challenge incorrect determinations. If Fannie Mae or
Treasury were to track successful complaints, these changes
would also allow Fannie Mae and Treasury to identify those
servicers with the best, and the worst, decision-making
records. By then rewarding or punishing them accordingly,
servicers could be appropriately incentivized to focus on
borrower outcomes.
---------------------------------------------------------------------------
\326\ GAO Report on Foreclosure Mitigation Programs, supra note
203, at 25-26; Treasury conversations with Panel staff (Nov. 16, 2010).
\327\ Supplemental Directive 10-15, supra note 216, at 3.
---------------------------------------------------------------------------
Treasury should also make more prominent use of its
available enforcement mechanisms against servicers. Treasury
has emphasized the effectiveness of temporarily withholding
incentive payments in resolving data issues with approximately
1,400 loans, but has only ever withheld incentive payments in
that one instance.\328\ It has not yet permanently stopped any
incentive payments. As the Panel noted as early as October
2009, ``[m]onitoring alone is ineffective unless accompanied by
meaningful penalties for failure to comply.'' \329\ Treasury
should be more willing to use penalties to resolve long-
standing problems, such as servicers losing borrower
documentation.\330\
---------------------------------------------------------------------------
\328\ Treasury conversations with Panel staff (Oct. 21, 2010).
\329\ October 2009 Oversight Report, supra note 15, at 109.
\330\ As noted in Section G.2.a, supra, the voluntary nature of the
program generates the understandable concern that if servicers are
penalized too much, they will simply leave the program. Treasury,
however, raised the possibility that it may not be so easy for
servicers to exit from HAMP. The SPAs do not contain any provision that
allows servicer withdrawal. Treasury has not fully explored the
ramifications of this absence, but did state that Treasury had
discussed suing a servicer for specific performance if they attempted
to withdraw unilaterally from the program. While Treasury may decide
not to sue, or such a suit may not be successful, this possibility
would complicate a servicer's exit. It may, therefore, create more
space for Treasury to penalize servicers. Treasury conversations with
Panel staff (Oct. 21, 2010).
---------------------------------------------------------------------------
The Panel is pleased to note that Treasury has indicated
that it will increase its public disclosure of particular
servicers with problems.\331\ Such ``naming and shaming'' can
be an effective enforcement mechanism. Moreover, as the Panel
noted in its October 2010 report, ``[i]n order for compliance
and enforcement to function as a deterrence mechanism and be
exercised effectively, they must be sufficiently robust and
transparent.'' Publicly noting problem servicers and their
sanctions will help build a credible deterrent. Public exposure
may also be the most effective enforcement mechanism for
policing the GSEs. Monetary inducements are ineffective because
Fannie Mae and Freddie Mac are performing their work at
cost,\332\ and even if Treasury were to fine one of the GSEs
for failing to perform, this sanction would ultimately be paid
by the taxpayers, not by the now almost non-existent GSE
equity-holders.\333\ Congress, however, is in the process of
deciding on the GSEs' ultimate fate, so Fannie Mae and Freddie
Mac may be sensitive to disclosures that would cast a negative
light on their abilities. Treasury should regularly publish a
detailed scorecard of the GSEs' performance, as well as
disclose any particular problems or failures.
---------------------------------------------------------------------------
\331\ Treasury conversations with Panel staff (Oct. 21, 2010).
\332\ See October 2010 Oversight Report, supra note 250, at 66.
\333\ The original agreements with the GSEs included the
possibility of incentive payments, but no incentive payments have been
made, and Treasury has indicated that it has taken the incentive
payment clause off the table indefinitely. Treasury conversations with
Panel staff (Sept. 23, 2010). See also Congressional Oversight Panel,
Testimony of Joy Cianci, senior vice president, Making Home Affordable
Program, Fannie Mae, Transcript: COP Hearing on Treasury's Use of
Private Contractors (Sept. 22, 2010) (publication forthcoming) (online
at cop.senate.gov/hearings/library/hearing-092210-contracting.cfm)
(stating that ``[t]here was a provision in the original contract that
provided for the potential for incentives. We have not received
incentives to date. And we're in the process of working through a
revision to that contract. My understanding is that there will not be
an incentive framework forward.'').
---------------------------------------------------------------------------
Alternatively, Treasury could work with FHFA to intervene
more directly with Fannie Mae and Freddie Mac. FHFA can
exercise the powers of the management and board of both the
GSEs. These powers could be used to generate internal pressure
to improve performance, even to the extent of holding senior
officials individually responsible for poor outcomes.
I. Redefaults of Modified Mortgages
To ensure HAMP's success, Treasury should focus attention
on trying to minimize redefaults of HAMP-modified loans.
Redefaults, which occur when borrowers who have entered
permanent modifications become delinquent on their loans,
present a large potential pitfall for HAMP. This is in large
part because every borrower who stops making monthly mortgage
payments likely will not be able to keep his or her home in the
long run. (There may be some benefit from the delay in
foreclosures that results from failed modifications, since
delays ease the downward pressure that foreclosures put on
housing prices. But delays that do not address the underlying
issues associated with mass foreclosures will not provide
solutions.) On top of that, every redefault under HAMP
represents a loss to taxpayers. This is a result of HAMP's
structure: the program begins making incentive payments to
borrowers, servicers, and investors after a loan is permanently
modified. These payments continue for up to five years, so if a
borrower redefaults after three years, the government will have
paid thousands of dollars for an ultimately unsuccessful
modification. In this sense, redefaults are more costly than
HAMP trial modifications that fail to convert to permanent
modifications, since no tax dollars are spent on trial
modifications.
Treasury's initial estimated redefault rate over the five-
year span of HAMP permanent modifications was 40 percent, but
some private analysts estimate that the program's redefault
rate will be higher. Barclays Capital has projected a 60
percent redefault rate for HAMP.\334\ In addition, Standard &
Poor's estimates that only 20 percent of HAMP trial
modifications will ultimately succeed, an estimate that
includes both the program's rate of conversion from trial
modifications to permanent modifications and its redefault
rate.\335\
---------------------------------------------------------------------------
\334\ Barclays Capital e-mail to Panel staff (Nov. 1, 2010).
\335\ Standard & Poor's, U.S. Government Cost To Resolve and
Relaunch Fannie Mae and Freddie Mac Could Approach $700 Billion (Nov.
4, 2010).
---------------------------------------------------------------------------
Because HAMP permanent modifications have only been in
effect for a maximum of about a year, it is difficult to
evaluate the program's redefault rate in the context of
Treasury's five-year estimate of a 40 percent redefault rate.
So far, the 60+ day delinquency rate after three months in the
program is 4.6 percent. After six months, the redefault rate is
9.8 percent. After nine months, the rate is 15.6 percent. And
after 12 months, the rate is 25.4 percent.\336\ (For a fuller
discussion of HAMP redefaults, see Section I.4, below.)
---------------------------------------------------------------------------
\336\ Data provided by Treasury.
---------------------------------------------------------------------------
In the context of the falling number of new HAMP permanent
modifications, redefaults are an even greater concern. If
present trends continue, the monthly number of new permanent
modifications could actually fall below the monthly number of
redefaults, as Figure 16 shows. If this happens, the overall
number of homeowners being helped by HAMP would begin to fall.
FIGURE 16: MONTHLY REDEFAULTS AND NEW PERMANENT MODIFICATIONS IF
CURRENT TRENDS CONTINUE \337\
---------------------------------------------------------------------------
\337\ ``Monthly New Permanent Modifications'' and ``Monthly New
Redefaults'' are derived from cumulative ``All Permanent Modifications
Started'' and ``Permanent Modifications Canceled'' (excluding loans
paid off) levels from March 2010 to October 2010 recorded in the Making
Home Affordable Program's monthly Servicer Performance Reports. For
these monthly reports, see Treasury Reports and Documents, supra note
48.
[GRAPHIC] [TIFF OMITTED] 62622A.015
To understand better how Treasury might seek to minimize
redefaults, the Panel examined (1) the incidence of redefaults
during the Great Depression, the only previous era in which the
U.S. government instituted a nationwide foreclosure-prevention
program, in order to draw lessons to apply in today's
environment; (2) redefault rates for non-HAMP modifications
during the current crisis; and (3) HAMP redefaults, analyzing
the data by amount of equity held by the borrower and
affordability, among other factors.
1. The Great Depression-Era Home Owners' Loan Corporation
The 1920s and 1930s represent one the largest boom-and-bust
real-estate cycles in U.S. history. From 1920 until 1930,
prices of owner-occupied homes rose by an average of 45
percent. Then between 1930 and 1940, nominal prices in those
same cities fell by an average of 48.6 percent.\338\ The market
shocks were particularly severe in the early 1930s. Housing
prices fell by 30 to 40 percent nationwide between 1929 and
1932.\339\ The non-farm foreclosure rate reached 13 percent in
1933.\340\
---------------------------------------------------------------------------
\338\ Price V. Fishback et al., The Influence of the Home Owner's
Loan Corporation on Housing Markets During the 1930s, National Bureau
of Economic Research Working Paper No. 15824, at 5-6 (Mar. 2010)
(online at www.nber.org/papers/w15824) (hereinafter ``NBER Working
Paper No. 15824''). These figures were derived from a sample of 278 of
the largest cities across the country. In real terms, the price
declines were lower because the U.S. economy was experiencing
deflation. The Consumer Price Index fell by 16 percent between 1930 and
1940.
\339\ Id. at 5-6. The shocks were even more severe in certain
regions; in Manhattan, home prices fell by a staggering 66 percent
during the same four-year period. See also Tom Nicholas and Anna
Scherbina, Real Estate Prices During the Roaring Twenties and the Great
Depression, U.C. Davis Graduate School of Management Research Paper No.
18-09, at 16 (Jan. 29, 2010) (online at papers.ssrn.com/sol3/
papers.cfm?abstract_id=1470448).
\340\ David Wheelock, The Federal Response to Home Mortgage
Distress: Lessons from the Great Depression, Federal Reserve Bank of
St. Louis Review, at 139 (May/June 2008) (online at
research.stlouisfed.org/publications/review/08/05/Wheelock.pdf)
(hereinafter ``Federal Reserve Bank of St. Louis Review Paper'').
---------------------------------------------------------------------------
In June 1933, in response to the housing crisis and an
unemployment rate of 25 percent, Congress established the Home
Owners' Loan Corporation (HOLC).\341\ The HOLC was intended to
assist homeowners who were in trouble largely through no fault
of their own. It did so by purchasing mortgages from private
lenders and offering homeowners refinanced mortgages that were
intended to be more sustainable.\342\ Initially, HOLC loans had
5 percent interest rates and amortizing, 15-year terms, which
was a substantial improvement for borrowers.\343\ HOLC loans
also represented better terms than the private mortgage market
was offering at the time. (Annex I provides an extended
discussion of the HOLC.)
---------------------------------------------------------------------------
\341\ Robert Van Giezen and Albert E. Schwenk, Compensation From
Before World War I Through the Great Depression, Bureau of Labor
Statistics Paper (Jan. 30, 2003) (online at www.bls.gov/opub/cwc/
cm20030124ar03p1.htm).
\342\ NBER Working Paper No. 15824, supra note 338, at 6-7.
\343\ Residential mortgages in the 1920s typically required very
large down payments and had relatively high interest rates and short
loan terms, so borrowers were required to make balloon payments at the
end of the term. By lowering the required down payment, extending the
loan term, and lowering the interest rate, the HOLC was able to
refinance many loans even in an environment of much lower property
values. As Annex I discusses, the interest-only and negatively
amortizing loans of the 2000s arguably offer a similar opportunity
today.
---------------------------------------------------------------------------
How did the HOLC's refinanced mortgages perform? In the
HOLC's early years, the performance was relatively poor. In
June 1936, 62.6 percent of HOLC borrowers were at least one
month delinquent, and 39.5 percent of them were at least three
months delinquent. But over time the delinquency statistics
showed improvement. By June 1939, the one-month delinquency
rate was 46.8 percent, and the three-month delinquency rate was
24.3 percent. By June 1942, the one-month delinquency rate was
down to 28.2 percent, and the three-month delinquency rate was
just 5.1 percent.\344\
---------------------------------------------------------------------------
\344\ C. Lowell Harriss, History and Policies of the Home Owners'
Loan Corporation, at 201-202, National Bureau of Economic Research
(1951) (online at www.nber.org/books/harr51-1) (hereinafter ``NBER
Research Paper''). See Annex I for more detailed data.
---------------------------------------------------------------------------
This improvement was likely in part the result of improving
economic conditions in the 1940s. It was also partly the result
of the HOLC foreclosing on delinquent borrowers; once the loans
were foreclosed, the HOLC no longer counted them as part of its
loan inventory. Altogether, throughout the 18-year life of the
HOLC, about 200,000 loans, or roughly 20 percent of the HOLC's
portfolio, went into foreclosure or were voluntarily
transferred by the borrower to the HOLC.
Another potential reason why the percentage of defaulted
loans fell between 1936 and 1950 is that the HOLC went to great
lengths to keep homeowners in their houses. At its peak, the
HOLC had about 20,000 employees and offices in 48 states. In
servicing loans, the HOLC relied heavily on personal contact
aimed at helping distressed homeowners. Servicing practices
varied by state and over time, but there were some common
themes. In the second and third months of delinquency, the HOLC
would insert special notices with the homeowner's monthly bill.
If the homeowner was unresponsive, a form letter followed. Next
came a personal letter. If there was no response, a HOLC
staffer would make an in-person visit to the home. In some
cases, HOLC employees helped homeowners or their relatives find
jobs, and to collect insurance claims, unpaid debts, and
pensions. HOLC employees even suggested ways of finding tenants
or foster children to defray the homeowner's monthly mortgage
payment.\345\ C. Lowell Harriss, the author of a 1951 book
about the HOLC, wrote: ``The assistance given by the HOLC's
service representatives is difficult to summarize adequately.
The closest parallel, perhaps, is found in the social worker's
helping individuals and families adjust to their own problems
and to the community around them.'' \346\ All of this stands in
marked contrast, of course, to the highly automated, impersonal
mortgage servicing practices of today. These differences are an
important consideration in a comparison of redefault rates
under the HOLC and HAMP.
---------------------------------------------------------------------------
\345\ Id. at 66-67.
\346\ Id. at 67.
---------------------------------------------------------------------------
It is difficult to analyze systematically the factors that
drove HOLC delinquencies and foreclosures, since there is a
relative lack of useful data. For example, the HOLC collected
borrower income data only at the time it made the loan, not on
an ongoing basis,\347\ which makes it impossible to study the
connection between mortgage affordability and delinquency.
There has been research, though, on the connection between
foreclosures and HOLC borrowers' equity stake in their homes. A
study of a sample of HOLC loans in the New York region found
that when the borrower's equity was equal to or greater than
the amount of the HOLC loan, the foreclosure rate was 12
percent. That number rose as the borrower's equity got smaller,
though. For borrowers whose equity was less than 25 percent of
the loan value, the foreclosure rate topped 40 percent.\348\
---------------------------------------------------------------------------
\347\ Id. at 87-88.
\348\ Id. at 98.
---------------------------------------------------------------------------
The finding that HOLC loans were more likely to end in
foreclosure if the borrower had little or no equity has
important implications for HAMP. It suggests that borrowers
with less equity or negative equity will be more likely to
redefault on their modified loans, and thereby underscores the
importance of principal reductions to the program's long-term
success.
2. Redefaults in Other Loan Modification Efforts
Shortly after the FDIC took IndyMac Bank into receivership
in July 2008, it instituted a mortgage modification program for
delinquent borrowers. The program, also discussed in Section B,
was applied to more than 60,000 residential mortgages that were
60 days or more past due.\349\ The FDIC's program is similar to
HAMP in several ways. Like HAMP, it uses an NPV test to
determine whether a mortgage should be modified. And it uses
interest rate reductions, term extensions and principal
forbearance to make mortgages more affordable. The FDIC program
is somewhat less aggressive than HAMP in seeking affordability;
it reduces interest rates as low as 3 percent, rather than 2
percent under HAMP, and it requires that first-lien mortgage
payments exceed no more than 38 percent of income, rather than
31 percent under HAMP.\350\
---------------------------------------------------------------------------
\349\ Federal Deposit Insurance Corporation, The FDIC Loan
Modification Program at IndyMac Federal Savings Bank, presented by
Richard A. Brown, Chief Economist, at the Mortgages and the Future of
Housing Finance conference (Oct. 25, 2010).
\350\ Federal Deposit Insurance Corporation, FDIC Loan Modification
Program, at 3 (online at www.fdic.gov/consumers/loans/loanmod/
FDICLoanMod.pdf) (accessed Dec. 10, 2010).
---------------------------------------------------------------------------
The FDIC estimated that 33 percent of the IndyMac loans
that it modified would eventually redefault.\351\ So far, in
the context of the IndyMac modifications, that projection has
proven to be too optimistic. The FDIC recently published
certain redefault data, and they show that the program's
redefault rate rises steadily as modifications age. After
modifications have been in the program for six months, the
redefault rate is 18.5 percent. At one year, the redefault rate
has risen to 42.1 percent. And at 18 months, the redefault rate
is 59.0 percent.
---------------------------------------------------------------------------
\351\ FDIC conversations with Panel staff (Oct. 28, 2010).
---------------------------------------------------------------------------
The FDIC's data also show that depending on various
characteristics of the loan and the borrower, the 18-month
redefault rate may be as low as 40 percent or as high as 70
percent. For example, 18-month redefault rates on IndyMac loans
modified when they were more than 180 days delinquent were 66.5
percent, while the redefault rate on loans modified when they
were 60 days delinquent was 43.9 percent. Similarly, there was
a divergence of redefault rates based on the borrower's
original credit score. Homeowners with credit scores in the
highest range had a redefault rate of 45.7 percent, while
borrowers with credit scores in the lowest range had a
redefault rate of 63.2 percent. There was also a large gap in
redefault rates based on monthly payment reductions. For
homeowners whose monthly payments dropped by 40 to 50 percent,
the 18-month redefault rate was 42.8 percent. But for those
whose payments fell by less than 20 percent, the redefault rate
was around 70 percent. Perhaps the best news from the FDIC data
is that redefault rates have been lower for more recent loan
modifications than they were for earlier ones. For loans
modified between September 2008 and April 2009, the 10-month
redefault rate was 38.6 percent. But for loans modified during
the following six months, the 10-month redefault rate fell to
27.1 percent.\352\
---------------------------------------------------------------------------
\352\ FDIC data provided to Panel staff (Dec. 8, 2010).
---------------------------------------------------------------------------
3. Current Trends in Loan Modifications and Redefaults
Two federal financial regulatory agencies, the Office of
the Comptroller of the Currency (OCC) and the Office of Thrift
Supervision (OTS), publish a quarterly statistical report on
the U.S. mortgage market. This report, the OCC/OTS Mortgage
Metrics Report, provides data on the performance of first-lien
residential mortgages serviced by federally regulated banks and
thrifts, which comprise 65 percent of all U.S. mortgages. So
while the Mortgage Metrics Report is the most reliable source
of nationwide data on mortgage delinquencies, loan
modifications, and foreclosures, it does not include data from
35 percent of the mortgage market. The most recent Mortgage
Metrics Report includes data through June 30, 2010. It shows
that the number of loan modifications that are happening
outside of HAMP, and therefore without incentive payments from
the government, has exceeded the number of HAMP modifications
each quarter since HAMP began in 2009. In the second quarter of
2010, there were 164,473 permanent non-HAMP modifications and
108,946 permanent HAMP modifications.\353\ It is still too
early to draw any conclusions about the sustainability of
modification programs that began in the last two years, since
the ultimate success of those programs will hinge on whether
the borrowers are able to stay in their homes over the long
term. Still, it is instructive to review redefault data on
modifications from non-government programs, since they shed
light on how to make HAMP as effective as possible.
---------------------------------------------------------------------------
\353\ OCC/OTS Mortgage Metrics Report, Second Quarter 2010, supra
note 31, at 21.
---------------------------------------------------------------------------
One must be cautious when comparing non-HAMP modifications
with HAMP modifications because they do not necessarily have
the same characteristics. For example, HAMP modifications
generally result in larger decreases in monthly payments, and
consequently result in better affordability, than non-HAMP
modifications do. The Mortgage Metrics Report shows that in the
second quarter of 2010, HAMP modifications resulted in an
average monthly payment reduction of $608, compared to $307 for
non-HAMP modifications.\354\ Furthermore, homeowners in non-
HAMP modifications, as a group, may be more likely to redefault
than those in HAMP modifications, since certain factors that
kept them from being approved for HAMP may also indicate that
their personal finances are more precarious.
---------------------------------------------------------------------------
\354\ OCC/OTS Mortgage Metrics Report, Second Quarter 2010, supra
note 31, at 30-32.
---------------------------------------------------------------------------
The data from the Mortgage Metrics Report show that
redefault rates for all permanent loan modifications--a
category that includes both HAMP and non-HAMP modifications--
have been dropping each quarter since the start of 2009. For
example, in the first quarter of 2009, 30.8 percent of modified
loans were at least 60 days delinquent within just three months
of the modification; by the fourth quarter of 2009, this figure
had fallen to 11.4 percent. Similarly, in the first quarter of
2009, 42.8 percent of modified loans were at least 60 days
delinquent within six months of the modification; by the fourth
quarter, that figure had dropped to 20.7 percent.\355\ Figure
17 shows the improvements in redefault rates since the start of
2009.
---------------------------------------------------------------------------
\355\ OCC/OTS Mortgage Metrics Report, Second Quarter 2010, supra
note 31, at 33.
FIGURE 17: PERMANENT MODIFICATIONS AT LEAST 60 DAYS DELINQUENT, BY QUARTER \356\
----------------------------------------------------------------------------------------------------------------
3-month Redefault 6-month Redefault 29-month 12-month
Rate Rate Redefault Rate Redefault Rate
----------------------------------------------------------------------------------------------------------------
First Quarter 2009.................. 30.8 42.8 51.5 55.0
Second Quarter 2009................. 18.7 33.5 40.9 43.2
Third Quarter 2009.................. 14.7 27.7 32.7 --
Fourth Quarter 2009................. 11.4 20.7 -- --
First Quarter 2010.................. 11.1 -- -- --
----------------------------------------------------------------------------------------------------------------
\356\ OCC/OTS Mortgage Metrics Report, Second Quarter 2010, supra note 31, at 33.
Data from nine mortgage servicers collected by the State
Foreclosure Prevention Working Group confirm the notion that
redefault rates on loan modifications have been improving.
Rates of serious delinquency after six months fell from 30.8
percent for loans modified in August and September of 2008 to
15.3 percent for loans modified in August and September of
2009.\357\ The State Foreclosure Prevention Working Group also
found that in the same time period redefault rates have fallen
by more than 60 percent for both modifications that result in
significant payment reductions and for those that result in
significant principal reductions.\358\
---------------------------------------------------------------------------
\357\ State Foreclosure Prevention Working Group, Redefault Rates
Improve for Recent Loan Modifications, at 5 (Aug. 2010) (online at
www.csbs.org/regulatory/Documents/SFPWG/DataReportAug2010.pdf).
\358\ Id. at 6 (``A comparison of five reporting servicers
demonstrates how the improvement in redefault rate is evident even when
controlling for the type of loan modification. For instance, the
redefault rate at six months for loans with significant payment
reductions fell from almost 31.4% for loans modified in August to
September of 2008 to just 11.8% for loans modified in August to
September of 2009, a more than 62% reduction. Similarly, the redefault
rate for loans with significant principal reductions fell from 35.4% to
12.9%, over a 63% reduction.'').
---------------------------------------------------------------------------
The Mortgage Metrics Report also shows that modifications
resulting in large reductions in monthly payments have lower
redefault rates than other modifications. This correlation is
not surprising, given that HAMP, which produces greater
reductions in monthly payments than other loan modification
programs, has lower redefault rates than those programs, as
shown below in Section I.4. The 2009 data show that 19.6
percent of modifications that resulted in payment reductions of
at least 20 percent were at least 60 days delinquent within six
months. The 60-day delinquency rates are higher for
modifications that result in less relief--or in some cases, a
greater payment burden--for borrowers.\359\ Figure 18 shows 60+
day delinquency rates by change in monthly payment.
---------------------------------------------------------------------------
\359\ OCC/OTS Mortgage Metrics Report, Second Quarter 2010, supra
note 31, at 39.
---------------------------------------------------------------------------
FIGURE 18: 60+ DAY DELINQUENCY RATES OF LOANS MODIFIED IN 2009 BY
CHANGE IN PAYMENT \360\
---------------------------------------------------------------------------
\360\ OCC/OTS Mortgage Metrics Report, Second Quarter 2010, supra
note 31, at 40.
[GRAPHIC] [TIFF OMITTED] 62622A.016
4. HAMP Redefaults
a. Redefault Rates
Overall, through October 2010, 35,815 of 519,648 HAMP
permanent modifications have redefaulted. This yields an
overall redefault rate of 6.9 percent.\361\ This statistic is
misleading, however, in terms of predicting the ultimate
redefault rate on HAMP permanent modifications. More than half
of all HAMP permanent modifications were made between April and
October 2010. Thus, the denominator in the overall redefault
rate is tilted toward more recent modifications. As most
redefaults do not happen immediately, the numerator is tilted
toward older modifications, which are fewer.
---------------------------------------------------------------------------
\361\ Data provided by Treasury.
---------------------------------------------------------------------------
A more complete picture of HAMP redefault rates emerges
from the redefault rate for HAMP modifications at set numbers
of months post-modification, as shown in Figure 19. Figure 19
shows that although only around 1 percent of permanent
modifications are 90+ days delinquent within their first three
months, the number jumps to 5.5 percent by month six and 11
percent by month nine; within a year, 21 percent of HAMP
permanent modifications are 90 or more days delinquent, at
which point they are disqualified from the program. This
compares to a 40 percent redefault projection over five years.
HAMP does not yet have redefault rates past one year, and it is
not likely that the redefault rate will plateau at one year.
Certain characteristics common to HAMP permanent modifications,
such as balloon payments, deep negative equity, borrowers with
severely damaged credit scores (which increase the borrowers'
cost of credit for other obligations), and interest rates and
payments that can rise after five years, may leave HAMP
borrowers vulnerable to redefaults that peak over a somewhat
longer time period. Doubtless, the state of the economy and
unemployment will have a bearing on the performance of the
modifications over the coming years. The 12-month redefault
rate is well below Treasury's 40-percent assumption, but it is
important to remember that the 40-percent number reflects the
redefault rate over the five-year span of the modification.
Only time will tell whether the assumption is too high,
accurate, or too low.
FIGURE 19: HAMP REDEFAULT RATES BY MONTHS POST-MODIFICATION \362\
---------------------------------------------------------------------------
\362\ Data provided by Treasury.
[GRAPHIC] [TIFF OMITTED] 62622A.017
HAMP's trial period serves as an effective tool in reducing
the investment of HAMP dollars towards borrowers who are more
likely to redefault, as HAMP redefault rates do not include
borrowers who default during the trial period. Without such a
screening mechanism, many additional loans would likely have
redefaulted. In addition to the modifications experiencing
traditional payment default during the permanent modification
period (redefault), 146,031 HAMP trials were disqualified
because of payment default. Had Treasury not utilized a trial
period, these payment defaults would likely have combined with
the traditional redefaults to yield a 44 percent redefault rate
at 12 months and a 50 percent redefault rate at 15 months, as
shown in Figure 20. The policy decision by Treasury to include
a trial period helped screen out these borrowers who could not
support a modified payment and contributed to more sustainable
outcomes within HAMP. Recent changes made in June within HAMP
requiring up-front verification to receive a trial modification
further prevent less prepared borrowers from entering the trial
period.
FIGURE 20: 90+ DAY DELINQUENCY RATE FOR ALL QUALIFIED HAMP
MODIFICATIONS, TRIAL AND PERMANENT \363\
---------------------------------------------------------------------------
\363\ Data provided by Treasury. This figure assumes that there was
only a three month trial period, which is what was called for by the
original terms of HAMP. In fact some trial periods were extended
considerably longer.
[GRAPHIC] [TIFF OMITTED] 62622A.018
Data available in the OCC/OTS Mortgage Metrics report allow
a comparison of HAMP modifications and non-HAMP modifications
over the same time period. It is important to note, however,
that the OCC/OTS data only cover approximately two-thirds of
the mortgage market. According to that data, HAMP modifications
are redefaulting at lower rates than other loan modifications,
which likely stems from the more substantial relief generally
received under the program as compared to non-HAMP
modifications, as well as differences in borrower
characteristics between borrowers in HAMP and borrowers
receiving proprietary modifications. Among HAMP modifications
initiated in the fourth quarter of 2009, 10.8 percent were at
least 60 days delinquent within six months. This compares to
22.4 percent for non-HAMP modifications started during the same
period.\364\ Figure 21 compares the redefault rates for
permanent HAMP modifications with those for permanent non-HAMP
modifications.
---------------------------------------------------------------------------
\364\ OCC/OTS Mortgage Metrics Report, Second Quarter 2010, supra
note 31, at 37.
FIGURE 21: HAMP VS. NON-HAMP 60+ DAY DELINQUENCY RATES \365\
----------------------------------------------------------------------------------------------------------------
3 Months after 6 Months after
Number of Modification Modification
Modifications (Percent) (Percent)
----------------------------------------------------------------------------------------------------------------
HAMP Fourth Quarter 2009............................... 20,679 7.9 10.8
Other Fourth Quarter 2009.............................. 103,617 12.1 22.4
HAMP First Quarter 2010................................ 100,269 10.5 --
Other First Quarter 2010............................... 131,207 11.6 --
----------------------------------------------------------------------------------------------------------------
\365\ OCC/OTS Mortgage Metrics Report, Second Quarter 2010, supra note 31, at 37.
b. Relationship Between Modification Characteristics and
Redefault Rates
A full exploration of the relationship between modification
characteristics and redefault rates would necessitate careful
statistical analysis of loan-level data and is beyond the scope
of this report. The Panel is surprised that Treasury has not
undertaken such a statistical analysis itself, as it is
fundamental to understanding what is and what is not working
with HAMP. At this point there is already over a year's worth
of performance history on many modifications. A better
understanding of the relationship between modification
characteristics and redefaults is critical for optimizing
modifications and making the best use of TARP funds. The Panel
strongly urges Treasury to undertake such analysis as soon as
possible, with due attention to LTV ratios, back-end DTI
ratios, the presence of second liens, hardship reasons, and
state and ZIP code of the property's location, as well as these
variables' interactions, and to make its findings public.
The Panel also urges Treasury to make loan-level data on
HAMP modifications publicly available in a form that is readily
accessible for data analysis. Enabling public analysis of the
data (including analysis by researchers at other government
agencies) will help provide feedback to Treasury that can be
used to improve HAMP as well as to optimize the design of non-
HAMP foreclosure mitigation programs. The Panel recognizes that
there are legitimate concerns about protecting borrower
privacy, but simply removing borrower names, Social Security
numbers, and street addresses from the data would provide
borrowers with the same level of privacy as is provided for
data released under the Home Mortgage Disclosure Act.
Although this report does not undertake a loan-level
statistical analysis, in this section it does explore the
relationship between HAMP redefault rates and certain
modification characteristics.
Vintage
HAMP's original vintage--permanent modifications from the
third quarter of 2009--have performed consistently worse than
more recent vintages. (See Figure 22.) The reason for this is
not clear.
FIGURE 22: REDEFAULT RATES (90+ DAYS DELINQUENT) BY QUARTERLY VINTAGE
\366\
---------------------------------------------------------------------------
\366\ Data provided by Treasury.
[GRAPHIC] [TIFF OMITTED] 62622A.019
Mark-to-Market Loan-to-Value Ratio
Mark-to-market loan-to-value ratio at the time of the
modification has a strong impact on redefault rates. More
deeply underwater loans redefault at higher rates. This
suggests that if the goal is to lower redefaults, HAMP
modifications should place greater emphasis on principal
reduction instead of principal forbearance; however, that goal
must be weighed against the possibility that fewer borrowers
would receive modifications, unless the NPV model were changed.
PRA attempts to consider these points, as it requires servicers
to run two NPV analyses--one general analysis and one featuring
principal reduction. Thus, the success of PRA could positively
influence the success of HAMP.
Figure 25 shows that the distribution of loan-to-value
ratios for permanent modifications is heavily tilted toward
deeply underwater loans. For all of the categories shown in
Figure 23, redefaults trend upward with time. However, at each
point in time, the loans with higher LTVs had higher redefault
rates as compared to the loans with lower LTVs. The trend
becomes even more pronounced for 90+ day delinquencies, as
depicted in Figure 24. This conclusion is worrisome, given that
most HAMP modifications retain a high LTV post-modification.
Also notable in Figures 23 and 24 is that there is little
difference in performance once LTVs exceed 120 percent.
Permanent modifications with LTVs of 120-150 percent performed
similarly to those with LTVs of above 150 percent. These
figures do not necessarily represent a break point for
performance based on LTV, but they do indicate that if LTV is
low enough, there is a noticeable improvement in long-term
performance. Figure 25 reveals that the number of HAMP
modifications with LTVs of over 100 percent, meaning that the
homeowner is underwater, far exceed the number of modifications
in which the homeowner is above-water. Unless housing prices
increase, or unless negative equity is addressed in some other
manner, LTV represents a risk for HAMP going forward.\367\
---------------------------------------------------------------------------
\367\ Data provided by Treasury.
---------------------------------------------------------------------------
FIGURE 23: 60+ DAY REDEFAULT RATES BY MARK-TO-MARKET LOAN-TO-VALUE
RATIO \368\
---------------------------------------------------------------------------
\368\ Data provided by Treasury.
[GRAPHIC] [TIFF OMITTED] 62622A.020
FIGURE 24: 90+ DAY REDEFAULT RATES BY MARK-TO-MARKET LOAN-TO-VALUE
RATIO \369\
---------------------------------------------------------------------------
\369\ Data provided by Treasury.
[GRAPHIC] [TIFF OMITTED] 62622A.021
FIGURE 25: NUMBER OF PERMANENT MODIFICATIONS BY MARK-TO-MARKET LOAN-TO-
VALUE RATIO \370\
---------------------------------------------------------------------------
\370\ Data provided by Treasury.
[GRAPHIC] [TIFF OMITTED] 62622A.022
Back-End DTI Ratio
Surprisingly, back-end DTI ratios, which provide a measure
of homeowners' total indebtedness, seem to have no correlation
with redefault rates. (See Figures 26 and 27.) \371\ Redefaults
are actually more frequent on modifications for borrowers who
have lower back-end DTI ratios than they are for those with
higher ratios.\372\ It is not clear what implications to draw
from this. It is possible that borrowers given a second chance
through HAMP but still saddled with high total debt have
managed to stay current for the term of the modification to
date, generally less than a year. After all, many borrowers
will prioritize a mortgage payment, especially a reduced
payment, lest the family lose its home. It is unclear whether
homeowners will be able to stay afloat under such a debt load
over a longer term, particularly if the economy and
unemployment do not improve. While DTI has not proven to be
correlated with redefault to date, this is an important point
to monitor over the coming years, as the situation could change
dramatically. A counter intuitive result is often the result of
a hidden correlation, which is why it is so important for
Treasury to engage in sophisticated analysis of redefault data.
It is striking, as Figure 28 shows, that nearly one-third of
HAMP permanent modifications have back-end DTI ratios of more
than 80 percent.\373\
---------------------------------------------------------------------------
\371\ Data provided by Treasury.
\372\ See also Federal Reserve Bank of Atlanta Working Paper, supra
note 88 (finding that borrowers' DTI ratios at their loans'
originations are not strong predictors of the likelihood that they will
default on their mortgages).
\373\ Data provided by Treasury.
---------------------------------------------------------------------------
FIGURE 26: 60+ DAY REDEFAULT RATES BY BACK-END DEBT-TO-INCOME RATIO
\374\
---------------------------------------------------------------------------
\374\ Data provided by Treasury.
[GRAPHIC] [TIFF OMITTED] 62622A.023
FIGURE 27: 90+ DAY REDEFAULT RATES BY BACK-END DEBT-TO-INCOME RATIO
\375\
---------------------------------------------------------------------------
\375\ Data provided by Treasury.
[GRAPHIC] [TIFF OMITTED] 62622A.024
FIGURE 28: NUMBER OF PERMANENT MODIFICATIONS BY BACK-END DEBT-TO-INCOME
RATIO \376\
---------------------------------------------------------------------------
\376\ Data provided by Treasury.
[GRAPHIC] [TIFF OMITTED] 62622A.025
Payment Reduction
The magnitude of the borrower's payment reduction under
HAMP also clearly affects redefault rates. As Figures 29 and 30
show, redefault rates are much lower on loans with a larger
percentage decrease in payments. It is not clear why the
percentage of payment reduction would matter, whereas the
absolute debt burden level would not, as discussed above. As
Figure 31 shows, the majority of permanent modifications at
least three months old in September 2010 had payment reductions
of more than 30 percent.\377\
---------------------------------------------------------------------------
\377\ Data provided by Treasury.
---------------------------------------------------------------------------
FIGURE 29: 60+ DAY REDEFAULT RATES BY REDUCTION IN MONTHLY PAYMENT
\378\
---------------------------------------------------------------------------
\378\ Data provided by Treasury.
[GRAPHIC] [TIFF OMITTED] 62622A.026
FIGURE 30: 90+ DAY REDEFAULT RATES BY REDUCTION IN MONTHLY PAYMENT
\379\
---------------------------------------------------------------------------
\379\ Data provided by Treasury.
[GRAPHIC] [TIFF OMITTED] 62622A.027
FIGURE 31: NUMBER OF PERMANENT MODIFICATIONS BY PAYMENT REDUCTION
PERCENTAGE \380\
---------------------------------------------------------------------------
\380\ Data provided by Treasury.
[GRAPHIC] [TIFF OMITTED] 62622A.028
Conclusions and Recommendations
In completing its fourth report on foreclosure mitigation
efforts, the Panel remains concerned that the choices made by
Treasury concerning issues such as program structure,
transparency, and data collection have not left borrowers well
served. Nearly two years after the announcement of MHA and
HAMP, foreclosures remain largely unabated, and while the
foreclosure level may be leveling off in some regions, it
remains extremely high in others.
Under EESA, Treasury still has an obligation to use TARP
funds in a manner that ``protects home values'' and ``preserves
homeownership.'' \381\ The expiration of its ability to
allocate additional money or create new programs does not mean
that it cannot make strides toward meeting that mandate.
---------------------------------------------------------------------------
\381\ 12 U.S.C. Sec. 5201(2)(A)-(B). For a discussion of the
authority of the Secretary of the Treasury to use TARP funds to create
a program such as HAMP, please see Appendix III of the Panel's April
2010 Oversight Report, supra note 1, at 147.
---------------------------------------------------------------------------
1. Treasury Should Announce Clear, Measurable Goals for HAMP
Nearly two years after HAMP was announced, it remains
virtually impossible for oversight bodies and the public to
determine whether the program is a success because Treasury has
failed to offer a definition of success. This has been
especially frustrating, given the clear shortcomings of the
program. Yet, because tradeoffs are inevitable, any foreclosure
mitigation program will be imperfect. However the fact that
tradeoffs are inevitable does not mean that all of HAMP's flaws
are acceptable. Unfortunately, the Panel is hamstrung in its
attempts to distinguish between these types of problems--the
unavoidable and the avoidable--because Treasury has provided so
few goals and metrics for foreclosure prevention.
Most fundamentally, how many foreclosures was HAMP intended
to prevent? What percentage of temporary modifications did
Treasury intend to convert to permanent status? What redefault
rate did Treasury consider acceptable for permanent
modifications? How frequently were servicers expected to
misplace paperwork? In short, how many of HAMP's shortcomings
were expected and inevitable, and how many were unexpected and
potentially resolvable? Despite repeated urgings from the Panel
and others, Treasury still has not answered these questions.
Treasury continues to have the ability to resolve this
problem. It should announce clear, measurable goals for HAMP.
Specifically, Treasury should announce a clear metric regarding
how many foreclosures will be prevented, the only real measure
of success. Up to 13 million foreclosures are expected over the
coming years, and the American people should know how many will
be averted with the $30 billion Treasury says it intends to
spend on HAMP.
Announcing clear, measurable goals will help create much
needed transparency for the program. Because the program lacks
any metrics, Treasury has continued to focus on unrealistic
expectations. For example, Treasury continues to state that
HAMP will expend approximately $30 billion in TARP funds, yet
CBO recently estimated that Treasury will spend only $12
billion out of the $45.6 billion allocated for all TARP
foreclosure mitigation programs, including both HAMP and the
Hardest Hit Fund. While CBO did not publish any detail on how
it expects the projected $12 billion to be spent, if one
assumes that states that are recipients of Hardest Hit Fund
grants will spend the $7.6 billion allocated to them, that
would leave only $4.4 billion to be spent on HAMP and the other
TARP housing programs. Had Treasury faced up to HAMP's problems
before the TARP expired, it could have taken more concrete
steps, such as making material program changes or reallocating
money. Absent a dramatic and unexpected increase in HAMP
enrollment, many billions of dollars set aside for foreclosure
mitigation will be left unused because Treasury failed to
recognize HAMP's shortcomings.
2. Treasury Should Collect More Data on HAMP's Progress and on Loan
Modifications
Treasury is to be commended for its improvement in HAMP
data collection, yet additional data would provide valuable new
information. In outlining the need for federal data collection,
the Panel's March 2009 report noted, ``While there is a clear
picture of rising foreclosures and loss mitigation efforts that
fail to keep pace, they do not provide sufficient information
to determine why so many loans are defaulting and why
foreclosure, rather than workouts, have been the dominant
response and why modifications have often been unsuccessful. .
. . Absent more complete and accurate information, legislators,
regulators, and market participants are flying blind.'' \382\
In particular, Treasury still does not collect sufficient
information about why loans are moving to foreclosure rather
than workouts, nor does it monitor closely enough any loan
modifications performed outside of HAMP. Treasury should also
explore further public reporting of compliance matters,
findings, and remediation activities by servicers.
---------------------------------------------------------------------------
\382\ March 2009 Oversight Report, supra note 10, at 15.
---------------------------------------------------------------------------
While Treasury should collect and report more data
generally, the Panel is particularly frustrated that data
sufficient for analysis is lacking for any of Treasury's
foreclosure mitigation programs, save HAMP. This is the Panel's
fourth foreclosure mitigation report, and the paucity of
available data has been a theme in each. Nearly two years after
the general foreclosure mitigation initiative was announced,
Treasury indicated that it does not have data available on its
second lien programs.\383\ More than six months after some of
its add-on programs were announced, Treasury still does not
even have a system of record to accept data.\384\ For other
areas, such as redefaults, Treasury is still considering what
data it wishes to collect.\385\ From the perspective of being
able to demonstrate results and fix shortcomings, most Treasury
programs are still at the very beginning stages. Because the
add-on programs represent efforts to supplement HAMP's narrow
focus, it is critical that they demonstrate results. Additional
delays in data collection and reporting could mean that many
programs do not even begin reporting data until the foreclosure
mitigation programs are scheduled to stop making additional
modifications at the end of 2012.
---------------------------------------------------------------------------
\383\ Transcript Testimony of Phyllis Caldwell, supra note 108.
\384\ SIGTARP Quarterly Report to Congress, supra note 25, at 202.
\385\ Treasury conversations with Panel staff (Oct. 28, 2010).
---------------------------------------------------------------------------
Further, because redefaults of permanent modifications pose
a particular risk to HAMP's ultimate success, Treasury should
focus its data analysis on identifying borrower characteristics
that correlate to a higher risk of redefault. Treasury must
also ensure that servicers are complying with data reporting
requirements related to redefaults. At this point, some
servicers are taking six months or longer to report new
modifications to the system of record. Should redefaults be
reported in a similarly tardy fashion, Treasury could end up
improperly paying incentives to servicers for loans that have
already redefaulted. Finally, Treasury should expand the range
and frequency of its data reporting on redefaults.
3. Treasury Should Use HAMP's Existing Authorities as Effectively as
Possible To Prevent Foreclosures
The TARP's expiration has ended Treasury's ability to
change HAMP's structure or to dedicate additional money to the
program. Even so, Treasury can focus on preventing as many
foreclosures as possible under the existing program structure.
Treasury Should Enable Borrowers To Apply for HAMP as
Easily as Possible. Treasury should ensure that the web portal
is expanded to provide direct access for borrowers, allowing
them to apply for modifications and to track the status of
their applications online. A one-stop website for HAMP would
end one of the biggest obstacles that borrowers have identified
to their participation in the program. In addition, Treasury
should work with HAMP servicers to encourage them to push the
LoanPort as their primary point of entry for applications.
Treasury Must Hold Servicers Accountable for Failing To
Complete Loan Modifications Appropriately. For example, to
determine how frequently loan servicers are losing paperwork,
Freddie Mac should expand its Second Look loan reviews to
include contacting borrowers or their representatives to verify
whether documents were submitted properly. Further, although
Fannie Mae and Freddie Mac serve as Treasury's agents in
administering HAMP, Treasury bears ultimate responsibility for
the program's success or failure. As such, it should take
greater steps to hold Fannie Mae and Freddie Mac accountable.
This is especially critical in light of statements from the
GSEs indicating a potential conflict of interest between their
own business interests and their role as financial agents.
Performance reviews of Fannie Mae and Freddie Mac should be
expanded to include borrower oriented qualitative and
quantitative measures. Compliance activities must focus on
borrowers and outcomes, not merely process. Treasury should
also be more willing to use its power to withhold or clawback
incentive payments. It should then publicly detail its
sanctions for non-compliance by both its financial agents and
HAMP servicers.
Treasury Should Provide a Meaningful, Independent Appeals
Process from Servicer Decisions. Currently, the primary
mechanism for escalating complaints, the HOPE Hotline,
preserves servicers' discretion to decide on a modification,
relying on them to determine if they make a mistake initially.
Treasury should remove the decision from the servicers' hands
in favor of an independent and enforceable review. As
previously suggested by the Panel, this would be an appropriate
role for the Office of Homeowner Advocate or an ombudsman. In
addition, to maximize the impact of the appeals process,
Treasury should clearly inform borrowers that the hotline can
be used to escalate complaints.
Treasury Must Address the Obstacles Presented by Second
Liens. Since its initial report on foreclosure mitigation in
March 2009, the Panel has consistently highlighted the
modification obstacles created by second liens. Treasury is to
be commended for creating programs to address second liens,
such as 2MP and 2LP; however, it is extremely disappointing
that nearly two years later the programs have no track record
of success. It is critical that Treasury get the programs fully
operational and produce data on which they can be evaluated.
Further, given the important role second liens play in
affecting a loan modification, Treasury should explore the
implications of adding borrower-specific junior lien
information directly into the NPV model. In particular,
Treasury should consider the effect on the number of borrowers
served and the impact on modification sustainability.
Treasury Should Consider Ways To Increase Participation.
Although Treasury no longer has the ability to make material
changes to the foreclosure mitigation programs, it can make
more modest changes designed to incentivize participation
further. In rolling out PRA, Treasury held discussions with the
industry to find ways to increase participation in the new
initiative. Based on that feedback, Treasury included
authorization, but not a requirement, for equity sharing
arrangements subject to borrower protection provisions.
Treasury should monitor PRA to determine whether authorization
for equity sharing does indeed appear to increase
participation. If so, Treasury should consider authorizing
equity sharing arrangements in other programs.
Treasury Should Encourage Loan Servicers To Offer More
Effective, Sustainable Modifications. The Panel remains
concerned regarding the long-term sustainability of HAMP
modifications. High, persistent unemployment continues to
present problems for many borrowers. HAMP modifications leave
borrowers with continuing high levels of negative equity, and
even after receiving a modification, half of HAMP borrowers are
still paying 63 percent of pre-tax income towards debt.
Treasury must continue to adapt its programs to address more
effectively these root causes of foreclosure. Treasury can
determine, based on data collected to date, which types of
modifications have a lower correlation of redefault. Treasury
should encourage servicers to make more of these types of
modifications and fewer of the types of modifications that tend
to end in redefault. Redefaults, after all, represent the worst
failure of HAMP, as each redefault represents thousands of
taxpayer dollars that have been spent merely to delay rather
than prevent a foreclosure.
Treasury Should Identify and Consider Intervening in Cases
of Potential Redefault. Once borrowers make it into the
modification program, Treasury must focus on keeping them
there, as unchecked redefaults have the potential to undermine
even the modest progress made by HAMP. One lesson to be taken
from the HOLC program during the Great Depression was the
importance of ongoing intervention to hold down redefault
rates. The HOLC experienced success with early intervention
upon delinquency, such as targeted borrower outreach, and
Treasury should encourage the same approach. Delinquencies that
are flagged in their early stages can potentially be brought
current through a repayment plan. Such early intervention could
be done at minimal cost. In more serious situations or when
early intervention is ineffective, Treasury should assess the
cost of any additional intervention and the likelihood that
such intervention would keep borrowers in their homes before
spending additional resources. As part of this evaluation
Treasury must consider whether resources are best directed to
providing additional assistance to those already in the program
or helping borrowers who have not yet entered the program.
Although HAMP has managed to prevent some number of
foreclosures, the program as currently structured will never
have the reach necessary to put an appreciable dent into the
foreclosure crisis. Nonetheless, HAMP continues to have the
authority to spend $30 billion in taxpayer funds. Treasury must
ensure that every dollar spent is used as effectively as
possible to prevent foreclosures.
ANNEX I: LESSONS FROM THE HOME OWNERS' LOAN CORPORATION OF THE 1930s
AND 1940s
Prior to the current foreclosure crisis, the 1930s is the
most recent era when U.S. housing prices have fallen on a
sustained nationwide basis. The Depression-era decline in home
prices led to a flood of foreclosures, which caused tremendous
harm to families and communities. The federal government
responded by enacting numerous policies aimed at helping
affected homeowners and supporting the faltering mortgage
market. One of the most significant steps was the establishment
in June 1933 of the Home Owners' Loan Corporation (HOLC). The
HOLC's goal was to provide relief to borrowers who were in
trouble largely through no fault of their own. It did so by
purchasing mortgages from private lenders and offering
refinancing to homeowners on more favorable terms. This section
examines the HOLC in the context of the U.S. mortgage market
when it was established, compares the HOLC with HAMP, and draws
lessons from the HOLC that provide perspective on challenges
facing HAMP today.
A. Background
The mortgage market of the 1920s was substantially
different than the modern system to which Americans have become
accustomed. There was no national mortgage market, as there is
today, meaning that interest rates could vary substantially in
different parts of the country. In some geographic areas, even
as home prices soared during the 1920s, banks did not offer
real estate loans, so local residents turned to insurance
companies for mortgages.\386\ In addition, borrowers typically
had to make down payments equal to 40 to 60 percent of the
property's value. Loans typically lasted 10 years or less and
required payment of interest only; after the loan term, a
balloon payment equal to the loan's principal was due. Because
most borrowers did not have the cash to pay off the principal,
they typically refinanced into a new loan.\387\ The booming
real-estate market of the 1920s also had some features that
would be familiar to modern-day Americans. Lending standards
became looser during the 1920s, with many homeowners taking out
amortizing second liens that allowed them to borrow an
additional 30 percent of their home's value.\388\
---------------------------------------------------------------------------
\386\ Senate Committee on Banking and Currency, Subcommittee on
Home Mortgages, Etc., Testimony of Horace Russell, general counsel,
Federal Home Loan Bank Board of Atlanta, Home Owners Loan Act, 73rd
Congress, at 7-8 (Apr. 20 and 22, 1933) (hereinafter ``Testimony of
Horace Russell'') (estimating that in 1933, one-third of U.S. counties
did not have banks that made real estate loans, and noting that in many
of these communities there had never been banks that made such loans).
\387\ While some lenders did offer amortizing loans, even those had
relatively short terms. NBER Working Paper No. 15824, supra note 338,
at 6-7.
\388\ Kenneth A. Snowden, The Anatomy of a Residential Mortgage
Crisis: A Look Back to the 1930s, National Bureau of Economic Research
Working Paper No. 16244, at 9-10 (July 2010) (online at www.nber.org/
papers/w16244.pdf).
---------------------------------------------------------------------------
When property values contracted sharply in the early 1930s,
obtaining a refinanced mortgage became much more difficult. As
in the current crisis, relatively few homeowners received the
kinds of private mortgage modifications that made it easier for
them to keep their homes.\389\ One study of mortgage
modifications in the New York metropolitan area found that 5
percent of loans originated from 1920-1939 received a
modification that might be considered a concession by the
lender, while almost 14 percent of those loans ended in a
foreclosure or an agreement by the borrower to provide a deed
in lieu of foreclosure.\390\ Lenders may have been cautious
about offering concessions to homeowners because they had
trouble distinguishing between the mortgages that needed a
modification in order to avert a foreclosure and those that did
not. Modern credit scores were not available in the 1930s,\391\
and DTI ratios, which today's lenders use to determine whether
a borrower can afford a mortgage, were apparently not widely
used.\392\
---------------------------------------------------------------------------
\389\ See, e.g., Jonathan D. Rose, The Incredible HOLC? Mortgage
Relief During the Great Depression, at 25 (Nov. 9, 2009) (online at
www.uncg.edu/bae/econ/seminars/2010/Rose.pdf) (hereinafter ``The
Incredible HOLC? Mortgage Relief During the Great Depression'')
(``Mortgage lenders were reluctant during the Depression to engage in
much serious refinancing, especially debt reductions, and they appear
similarly reluctant today.'')
\390\ Andra C. Ghent, Residential Mortgage Renegotiation During the
Great Depression, at 20 (June 15, 2010) (online at papers.ssrn.com/
sol3/papers.cfm?abstract_id=1604664) (hereinafter ``Residential
Mortgage Renegotiation During the Great Depression'').
\391\ There were credit reports at the time, and the HOLC used them
to obtain information on applicants. NBER Research Paper, supra note
344, at 2, National Bureau of Economic Research.
\392\ Residential Mortgage Renegotiation During the Great
Depression, supra note 390.
---------------------------------------------------------------------------
By 1933, the government estimated that 20 to 25 percent of
the nation's $20 billion in home mortgage debt ($336 billion in
today's dollars) was in default.\393\
---------------------------------------------------------------------------
\393\ Testimony of Horace Russell, supra note 386, at 6, 9.
---------------------------------------------------------------------------
B. The HOLC's Operations
The HOLC was established by the Home Owners Loan Act of
1933 as a government corporation to be administered by the
recently established Federal Home Loan Bank Board. The
legislation, which had broad support in Congress,\394\ was
proposed by President Roosevelt. Roosevelt argued at the time
that the ``broad interests of the Nation require that specific
safeguards should be thrown around home ownership as a
guarantee of social and economic stability, and that to protect
home owners from inequitable enforced liquidation in a time of
general distress is a proper concern of the Government.'' \395\
The HOLC was capitalized with $200 million ($3.4 billion in
today's dollars). It was also given the authority to issue up
to $2 billion in government bonds ($33 billion today).
---------------------------------------------------------------------------
\394\ The vote in the House of Representatives was 383-4. No record
vote was taken in the Senate.
\395\ Franklin D. Roosevelt, Message to Congress on Small Home
Mortgage Foreclosures (Apr. 13, 1933) (online at
www.presidency.ucsb.edu/ws/?pid=14618).
---------------------------------------------------------------------------
HOLC refinance loans were restricted to borrowers in
default.\396\ Homeowners whose homes were worth $20,000 or
less--this figure is equivalent to $336,000 today, and the vast
majority of U.S. homeowners qualified under the standard--were
eligible to apply for a refinancing. The government provided
refinancings of up to 80 percent of the property's value, or
$14,000, whichever was higher. Initially, under the terms of
the refinanced loans, the homeowner paid 5 percent interest,
which was 1-2 percent below prevailing market rates,\397\ and
the loans fully amortized over 15 years. Second liens could
also be refinanced, although total obligations on the property
could not exceed 100 percent of the appraisal value.\398\
Delinquent property tax payments could be financed, including
for people who owned their homes outright, which helped boost
revenue for local governments.\399\ The HOLC addressed the
problem of unemployment by allowing three-year term extensions
for borrowers who were unable to pay. The program did not offer
loans for new home purchases,\400\ but it did lend money for
the reconditioning of homes.\401\
---------------------------------------------------------------------------
\396\ NBER Research Paper, supra note 344, at 1, National Bureau of
Economic Research.
\397\ Testimony of Horace Russell, supra note 386, at 15, 17.
\398\ NBER Research Paper, supra note 344, at 35-39, National
Bureau of Economic Research.
\399\ Rosalind Tough, The Life Cycle of the Home Owners' Loan
Corporation, Land Economics (1951) (hereinafter ``The Life Cycle of the
Home Owners' Loan Corporation'').
\400\ Home Owner's Loan Corporation, Statement by Chairman William
F. Stevenson Relative to the Method and Procedure of Procuring Loans
from the Federal Home Owners' Loan Corporation (June 6, 1933)
(hereinafter ``Statement by Chairman William F. Stevenson'').
\401\ NBER Research Paper, supra note 344, at 4, National Bureau of
Economic Research.
---------------------------------------------------------------------------
Lenders who agreed to sell their mortgages to the
government received the appraised value of the property.
Appraisals were also used to determine whether a homeowner had
the 20 percent equity needed to qualify for a HOLC mortgage.
Because housing markets were not functioning properly at the
time, determining appraised values was difficult. The HOLC's
chairman, William Stevenson, acknowledged this problem when the
program was established, saying, ``The matter of appraisal is
the most difficult problem to be dealt with by the Corporation
on account of the chaotic condition of the country with
reference to values.'' \402\ HOLC appraisals were supposed to
weigh three factors equally: the property's estimated current
market price; the cost of a similar lot, plus the cost of
reproducing the building, minus depreciation; and the
capitalization of the reasonable monthly rental value for the
last 10 years.\403\ Despite the establishment of this standard,
research has shown that ensuring accurate appraisals was a
problem for the HOLC, and appraisal standards varied
significantly across the country.\404\
---------------------------------------------------------------------------
\402\ Statement by Chairman William F. Stevenson, supra note 400.
\403\ NBER Research Paper, supra note 344, at 2, National Bureau of
Economic Research.
\404\ The Incredible HOLC? Mortgage Relief During the Great
Depression, supra note 389, at 23.
---------------------------------------------------------------------------
Jonathan Rose, an economist with the Federal Reserve Board,
concludes in a recent paper that the HOLC set appraisals at
high levels,\405\ which benefitted private lenders, since they
received more than they expected to earn from their delinquent
mortgages, and resulted in fewer principal reductions for
homeowners than would have occurred at lower appraisal values.
Mr. Rose concludes that the inflated appraisals seem to have
been deliberate, with the likely goals having been to encourage
lenders to participate and to support prices in the housing
market. ``The conclusion is that in many ways the HOLC was a
lenders' program,'' Mr. Rose writes. ``Fundamentally, with a
median decline in housing prices of 33 percent, large
adjustments were needed to debts undertaken during the 1920s.
Under the HOLC, the bulk of this adjustment was left to the
borrowers, while many lenders were absolved completely.
Borrowers certainly benefitted from the HOLC's lenient mortgage
structure, but lenders also benefitted greatly from the removal
of poorly performing assets off of their balance sheets.''
\406\
---------------------------------------------------------------------------
\405\ Mr. Rose adds: ``The fact that generous appraisals were made
was not a secret, although it is not widely known today. A 1933
pamphlet published by the HOLC to give information to potential
borrowers described the appraisal as being an estimate of `fair worth'
rather than `technical market value.' The Federal Home Loan Bank Review
. . . stated that HOLC loans `were permitted to be equal to 80 per cent
of liberal appraisals. They were intended to be generous and may have
frequently approached or sometimes exceeded market values at that
time.''' The Incredible HOLC? Mortgage Relief During the Great
Depression, supra note 389, at 1-2.
\406\ The Incredible HOLC? Mortgage Relief During the Great
Depression, supra note 389, at 1-2. See also The Life Cycle of the Home
Owners' Loan Corporation, supra note 399 (``What actually happened
between 1933 and 1936--the emergency period during which the H.O.L.C.
extended loans--was that the new organization bailed out not only the
home owners of the United States but also the banking institutions.'').
---------------------------------------------------------------------------
Private lenders who sold mortgages to the HOLC usually were
not paid in cash. Instead, they generally received government
bonds. Initially, these were 18-year bonds that paid 4 percent
interest. Although these bonds were often worth less than the
face value of the original mortgages, the private lenders
benefited from the government's guarantee of the bonds, as well
as their favorable tax treatment.\407\ If a lender refused to
sell the loan except for cash, and the loan was worth 40
percent or less of the property's value, the government would
pay cash, and then refinance the borrower into a 6-percent
interest loan.\408\
---------------------------------------------------------------------------
\407\ The bonds were exempt from all federal, state, and local
taxes except for estate taxes, surtaxes, inheritance taxes, and gift
taxes. NBER Research Paper, supra note 344, at 11, National Bureau of
Economic Research.
\408\ Statement by Chairman William F. Stevenson, supra note 400.
---------------------------------------------------------------------------
When Congress established the HOLC, it was aware that the
program's thin spread--the HOLC was collecting 5 percent
interest, and its bonds paid 4 percent--combined with the
likelihood of defaults, meant that the program might prove to
be unprofitable. Given the harsh economic circumstances,
Congress was willing to accept losses from the program.\409\ As
it turned out, the HOLC benefitted greatly from the low
interest rates of the era. Although the HOLC initially paid 4
percent interest on its bonds, its average borrowing cost
between 1933 and 1949 was 2.24 percent.\410\ Over the life of
the program, even though the interest rate charged to HOLC
borrowers fell to 4.5 percent in 1939, the HOLC's spread was
about 2.5 percentage points.\411\
---------------------------------------------------------------------------
\409\ Senate Committee on Banking and Currency, Subcommittee on
Home Mortgages, Etc., Statement of Senator James Couzens, Home Owners
Loan Act, 73rd Cong., at 15 (Apr. 20 and 22, 1933).
\410\ NBER Research Paper, supra note 344, at 5.
\411\ NBER Research Paper, supra note 344, at 162-163.
---------------------------------------------------------------------------
The HOLC established offices in every state, and its
employees worked with homeowners to keep them in their homes.
The window of time to obtain a HOLC loan was relatively short--
from August 1933 until June 1936. By the end, the HOLC received
1.886 million applications. The applicants sought a total of
$6.2 billion in loans, or an average of $3,272 per application.
According to one estimate, these applications accounted for
about 20 percent of the non-farm, owner-occupied homes in the
United States, and about 40 percent of mortgaged properties
that qualified for the HOLC. Close to half of the applications
were withdrawn or rejected, either because the homeowner was
deemed not to be in sufficient trouble or because the loan was
deemed too risky.\412\ The HOLC ultimately issued 1.02 million
refinance loans. The loans averaged $3,039, for a total of $3.1
billion.\413\ According to one historian who studied the HOLC,
applications from people who were unemployed would probably
have been rejected, but most of the people who qualified had
relatively modest incomes, and most had suffered financially
during the Depression.\414\ The HOLC reported in 1938 that,
``Almost without exception, the cost to the home owner under
the HOLC amortized mortgage is less than rent for a home of
corresponding value. In addition, it permits the borrower
actually to acquire final ownership free of debt.'' \415\
---------------------------------------------------------------------------
\412\ NBER Working Paper No. 15824, supra note 338, at 6-8.
\413\ NBER Research Paper, supra note 344, at 1.
\414\ NBER Research Paper, supra note 344, at 50.
\415\ Federal Home Loan Bank Board, Sixth Annual Report for the
Period Covering July 1, 1937-June 30, 1938, at 95 (Oct. 1, 1938)
(online at fraser.stlouisfed.org/publications/holc/issue/3013/download/
40596/1937_38annualrpt.pdf).
---------------------------------------------------------------------------
In 1933, the HOLC accounted for 12 percent of all new
mortgages on one-to-four-family homes. That figure rose to 71
percent in 1934 before falling to 26 percent in 1935, and to 6
percent in 1936. The HOLC's share of the nation's outstanding
mortgage debt peaked at 19 percent in 1935.\416\ The HOLC
purchased $770 million in mortgages from building & loans, many
of which became savings & loans around this time; $525 million
from commercial banks; $410 million from mutual savings banks;
$165 million from insurance companies; and $880 million from
other mortgage holders, including individuals.\417\
---------------------------------------------------------------------------
\416\ Federal Reserve Bank of St. Louis Review Paper, supra note
340, at 142.
\417\ Charles Courtemanche and Kenneth Snowden, Repairing a
Mortgage Crisis: HOLC Lending and Its Impact on Local Housing Markets,
National Bureau of Economic Research Working Paper No. 16245, at 30
(July 2010) (online at www.nber.org/papers/w16245) (hereinafter ``NBER
Working Paper No. 16245'').
---------------------------------------------------------------------------
Over time, some of the HOLC's initial parameters changed.
In 1934, Congress extended the government guarantee of the
HOLC's bonds, which originally only covered interest, to
include principal.\418\ In 1935, Congress increased the HOLC's
bonding authority to $4.75 billion. In 1939, the interest rate
on HOLC loans was reduced from 5 percent to 4.5 percent, and
Congress authorized an extension of HOLC loan terms from 15
years to 25 years.\419\
---------------------------------------------------------------------------
\418\ Federal Home Loan Bank Board, Second Annual Report of the
Home Owners' Loan Corporation Covering the Year 1934, at 81 (Feb. 11,
1935) (online at fraser.stlouisfed.org/publications/holc/issue/3008/
download/40940/1934_annualrpt_pt2.pdf). This decision was seen as an
important way to encourage participation by lenders; after the
government's guarantee was extended, the HOLC's bonds were considered
as credit-worthy as U.S. Treasury bonds.
\419\ Federal Home Loan Bank Board, Eighth Annual Report for the
Period July 1, 1939, through June 30, 1940, at 126 (Oct. 1, 1940)
(online at fraser.stlouisfed.org/publications/holc/issue/3016/download/
40989/1939_40annualrpt_pt7.pdf). By the end of 1942, 30 percent of
outstanding loans had received these extensions. NBER Research Paper,
supra note 344, at 136.
---------------------------------------------------------------------------
As discussed in Section I.1, above, the delinquency rate on
HOLC mortgages was initially high, but it fell sharply over the
life of the program. Figure 32 shows the fall in three-month
delinquency rates until 1950, shortly before the HOLC finished
liquidating its portfolio. Likely contributors to the falling
delinquency rate include the improving economy of the 1940s and
the fact that HOLC foreclosures pushed down the program's
default rate by eliminating those loans from the overall pool
of HOLC mortgages. In addition, the HOLC had a reputation as a
lenient loan servicer, and often went to great lengths in an
effort to keep families in their homes.\420\ Starting in 1937,
for example, the HOLC extended many defaulted loans by adding
arrearages to the loan and, in many cases, giving the borrower
several years to become current, with the goal of giving the
borrower a fresh start psychologically.\421\
---------------------------------------------------------------------------
\420\ For more detail about the HOLC's loan servicing practices,
see Section I.1, supra.
\421\ NBER Research Paper, supra note 344, at 69.
FIGURE 32: PERCENTAGE OF HOLC MORTGAGES AT LEAST THREE MONTHS DELINQUENT, BY YEAR \422\
----------------------------------------------------------------------------------------------------------------
Percentage of
Year Loans 3-plus Total Loans Loans 3-plus
Months Late Months Late
----------------------------------------------------------------------------------------------------------------
1936................................................... 397,533 1,005,988 39.5
1937................................................... 331,664 930,049 35.7
1938................................................... 270,144 878,017 30.8
1939................................................... 205,582 845,630 24.3
1940................................................... 100,027 854,233 11.7
1941................................................... 57,348 843,175 6.8
1942................................................... 41,607 808,219 5.1
1943................................................... 25,942 741,390 3.5
1944................................................... 16,009 641,446 2.5
1945................................................... 11,405 532,495 2.1
1946................................................... 9,349 430,307 2.2
1947................................................... 8,672 351,127 2.5
1948................................................... 9,407 278,189 3.4
1949................................................... 7,017 200,782 3.5
1950................................................... 2,420 73,965 3.3
----------------------------------------------------------------------------------------------------------------
\422\ All data were reported in June of the specified year.
Despite its efforts to avoid foreclosure, the HOLC
eventually acquired nearly 200,000 homes from borrowers who
failed to make payments. These property acquisitions were
mostly through foreclosure proceedings, though some involved a
voluntary transfer of the property by the borrower. More than
half of HOLC foreclosures involved borrowers who were a year
and a half or more delinquent.\423\ The HOLC often rented the
foreclosed homes it owned, spending an average of $51 on
reconditioning and $135 on maintenance before selling the
homes.\424\ The HOLC sold the foreclosed homes for an average
of 93 percent of the original HOLC loan amount. Its average
loss per foreclosed property was $1,568, for a total net loss
of $310 million on properties acquired.\425\
---------------------------------------------------------------------------
\423\ NBER Research Paper, supra note 344, at 3, 101 (``There was
no precedent for a real estate management situation of this size and
complexity. Most of the foreclosed properties presented difficult
problems of repair, reconditioning, rental, insurance, tax payment, and
eventual sale. They were widely distributed geographically, many were
twenty years old when acquired, and virtually all had been neglected by
their defaulting owners.'').
\424\ NBER Research Paper, supra note 344, at 3. The HOLC also
offered loans in connection with the sale of foreclosed properties.
\425\ NBER Research Paper, supra note 344, at 4.
---------------------------------------------------------------------------
As discussed in Section I.1, research indicates that when
HOLC borrowers had less equity, they were more likely to lose
their homes in foreclosures. Figure 33 shows this pattern in a
sample of HOLC loans from the New York region.
FIGURE 33: FORECLOSURE RATES BY BORROWER'S EQUITY FOR A SAMPLE OF HOLC
LOANS IN THE NEW YORK REGION
------------------------------------------------------------------------
Borrower's Equity as Percentage of Loan Foreclosure
Amount Loans Made Rate (Percent)
------------------------------------------------------------------------
Less than 0%............................ 561 46
0-24%................................... 968 40
25-49%.................................. 920 37
50-74%.................................. 498 22
75-99%.................................. 258 22
100% or more............................ 405 12
------------------------------------------------------------------------
Mr. Rose recently reached a related conclusion. Mr. Rose's
2009 study looked at loan-level HOLC data from New York, New
Jersey, and Connecticut. There was wide variation between HOLC
foreclosure rates in different states, ranging from just 4.4
percent in Nevada to 42.9 percent in New York and 38.4 percent
in New Jersey. One reason why foreclosures on HOLC loans may
have been so common in New York and New Jersey, the paper
concludes, is that those states used appraisal practices that
overvalued the properties being refinanced, even in comparison
to other states.\426\ The paper notes that HOLC offices in New
York and New Jersey paid more generous prices than elsewhere
and concludes that these practices ``likely contributed to the
weak performance'' of HOLC loans in those states.\427\ HOLC
borrowers in New York and New Jersey had less equity on average
than HOLC borrowers elsewhere, and their loans foreclosed at
higher rates.
---------------------------------------------------------------------------
\426\ Other factors likely contributed to the wide variations in
foreclosure rates. The HOLC attributed the differences, at least in
part, to regional variations in real-estate price levels, in the
duration and severity of the economic downturn, in real-estate tax
rates, and in levels of mortgage indebtedness prior to 1933. The Life
Cycle of the Home Owners' Loan Corporation, supra note 399, at 327.
\427\ The Incredible HOLC? Mortgage Relief During the Great
Depression, supra note 389, at 1-3, 8, 13. For the 3,032 loans that Mr.
Rose reviewed from New York, New Jersey, and Connecticut, the average
mark-up from the estimated market price to the final appraisal was 4.2
percent. Using the estimated market price rather than the appraised
value, Mr. Rose calculated that 50.1 percent of the homeowners in this
sample had less than 20 percent equity in their properties, and 19
percent had negative equity.
---------------------------------------------------------------------------
There had been fears that the HOLC would lose up to $500
million, and these concerns occasionally led to the
introduction of bills in Congress to force the early
liquidation of the HOLC's loans.\428\ Such calls went unheeded,
though, and when the HOLC was liquidated in 1951, it reported a
net profit of $14.2 million, based on net income of $352.2
million and losses of $338 million.\429\ This accounting,
however, did not include certain costs. The HOLC had borrowed
from Treasury, and the costs of that borrowing, which were
later pegged at $91.9 million, were not included in the HOLC's
calculations. One recent estimate pegged the HOLC's losses in
the general vicinity of $100 million.\430\ In addition to the
higher than expected spread between the HOLC's borrowing costs
and its earnings, the HOLC also benefited from the nation's
rising prosperity during World War II. Incomes rose, which led
to falling delinquencies and lower servicing costs.\431\
---------------------------------------------------------------------------
\428\ The Life Cycle of the Home Owners' Loan Corporation, supra
note 399, at 329-300.
\429\ NBER Research Paper, supra note 344, at 160.
\430\ NBER Working Paper No. 15824, supra note 338, at 9. Also
omitted were the HOLC's free use of the U.S. Postal Service, which
saved it $6 million and cost the government $3 million. Furthermore,
the HOLC benefitted by virtue of its exemption from state and local
business taxes, as well as from Social Security taxes. NBER Research
Paper, supra note 344, at 160-162, National Bureau of Economic
Research.
\431\ NBER Research Paper, supra note 344, at 163-165.
---------------------------------------------------------------------------
Recent research by Kenneth Snowden and Charles Courtemanche
of the University of North Carolina Greensboro found that the
HOLC increased home values and home ownership rates--by
repairing credit channels and by shutting off the destructive
cycle by which foreclosures hurt nearby home values, which
leads to more foreclosures. The researchers also found that the
HOLC did not lead to an increase in home building.\432\ The
HOLC played a key role in the gradual reshaping of the U.S.
mortgage market. Over time, the five-year interest-only loans
of the 1920s were replaced by 15-year, and later, 25-year,
government-backed amortizing loans. And 20 percent down
payments became a new standard in the mortgage industry.
---------------------------------------------------------------------------
\432\ NBER Working Paper No. 16245, supra note 417, at 6, 26.
---------------------------------------------------------------------------
C. How the HOLC Compares to HAMP
The differences between the HOLC and HAMP are considerable.
This section distills some of the most important differences.
1. Scale of Programs
The HOLC operated on a significantly larger scale in
relation to the U.S. housing market than HAMP has operated to
date. The HOLC provided refinancing for about 10 percent of all
non-farm, owner-occupied homes in the country, and for about 20
percent of all mortgaged homes.\433\ For HAMP to achieve a
comparable scale, it would have to yield between 7.6 million-
10.1 million permanent modifications.\434\ Through September
2010, less than 500,000 homeowners have received permanent HAMP
modifications.\435\ Of course, the scale of the 1930s mortgage
crisis and the larger economic crisis of the Great Depression
were comparatively larger than those of today. The number of
non-farm foreclosures between 1929 and 1938 was 1.89
million,\436\ at a time when the United States had
approximately 10 million non-farm, owner-occupied homes.\437\
Since July 2007, there have been approximately 8 million
foreclosure starts in the United States,\438\ while the number
of U.S. owner-occupied homes is around 76 million.\439\
---------------------------------------------------------------------------
\433\ NBER Research Paper, supra note 344, at 1-2.
\434\ In 2009, there were 76.4 million owner-occupied homes in the
United States. Ten percent of those homes equals 7.6 million. There
were 50.5 million mortgaged homes in the United States. Twenty percent
of those homes equals 10.1 million. U.S. Census Bureau, American
Housing Survey National Tables: 2009, Table 2-1 and Table 3-15 (online
at www.census.gov/hhes/www/housing/ahs/ahs09/ahs09.html) (accessed Dec.
10, 2010) (hereinafter ``American Housing Survey National Tables:
2009'').
\435\ MHA Servicer Performance Report Through September 2010, supra
note 285, at 2.
\436\ U.S. Department of Commerce, Bureau of the Census, Historical
Statistics of the United States: Colonial Times to 1970, Part 2, at 651
(online at www2.census.gov/prod2/statcomp/documents/CT1970p2-02.pdf).
\437\ See NBER Research Paper, supra note 344, at 1-2, National
Bureau of Economic Research (stating that the HOLC made roughly one
million loans, which provided aid to the owners of approximately one
out of 10 non-farm, owner-occupied homes).
\438\ See HOPE NOW Alliance, National Data July 2007 to November
2009, Appendix--Mortgage Loss Mitigation Statistics, at 7 (online at
www.hopenow.com/industry-data/
HOPE%20NOW%20National%20Data%20July07%20to%20Nov09%20v2%20%282%29.pdf)
(showing 5.91 million foreclosure starts between July 2007 and November
2009); HOPE NOW Alliance, Data Report May 2010, at 8 (online at
www.hopenow.com/industry-data/HOPE NOW Data Report (May) 06-21-2010
DRAF62.pdf) (showing approximately 890,000 foreclosure starts between
December 2009 and March 2010); HOPE NOW Alliance Industry
Extrapolations and Metrics, supra note 4 (showing approximately 1.04
million foreclosure starts between April 2010 and September 2010).
\439\ American Housing Survey National Tables: 2009, supra note
434.
---------------------------------------------------------------------------
2. Nature of Government's Role
The HOLC, by purchasing mortgages from private lenders in
an effort to prevent foreclosures, directly intervened in the
mortgage market. HAMP, by contrast, intervenes in the market in
an indirect way: by providing incentive payments to private
lenders that agree to modify their loans within certain
parameters. As a result of the fact that HAMP's market
intervention is indirect, the government is not in a good
position to use loan servicing policies as a tool for
preventing foreclosures, as the HOLC did during the Depression.
(It is important to note that two other government programs
during the current foreclosure crisis, HOPE for Homeowners and
the FHA Short Refinance Program, aim to purchase mortgages from
private lenders and thus represent direct interventions in the
mortgage market. Neither program, though, has made a
significant impact thus far.)
3. Enrollment Rate
The HOLC received 1.89 million applications, of which 1.02
million resulted in refinanced loans, for an enrollment rate of
54 percent. Because HAMP is structured differently than the
HOLC, it does not have a directly comparable enrollment rate.
The most comparable measures involve HAMP's permanent
modifications started as a percentage of its trial modification
offers, and HAMP's permanent modifications started as a
percentage of its trial modifications started. The former rate
is 32 percent; the latter rate is 37 percent.\440\
---------------------------------------------------------------------------
\440\ Through October 2010, 1,647,474 trial modifications had been
offered; 1,395,543 trial modifications had started; and 519,648
permanent modifications had started. MHA Servicer Performance Report,
supra note 38, at 2; Data provided by Treasury.
---------------------------------------------------------------------------
4. Homeowner Equity
Homeowners who participated in the HOLC had an average LTV
ratio of 68.6 percent,\441\ which suggests that their equity at
the time of the refinancing was equal to nearly one-third of
their homes' value. This figure is dependent, though, on the
accuracy of HOLC appraisals, and those appraisals were often
higher than what the housing market actually would bear at the
time.\442\ But even after accounting for inflated appraisals,
it seems reasonable to assume that in most cases the homeowner
had at least some equity at the time of the HOLC refinancing.
This is in sharp contrast to HAMP, where most homeowners have
substantial negative equity after their mortgage is modified.
As of October 2010, half of all homeowners entering HAMP had an
LTV ratio of greater than 120 percent. Their homes would have
to appreciate by 20 percent before they would have any equity
at all. This disparity between the HOLC and HAMP is critical
because, as the data in Section I.4.b. illustrate, homeowners
with little or no equity have less incentive to try to keep
their homes, and are more likely to redefault on modified loans
than homeowners who have more equity.
---------------------------------------------------------------------------
\441\ NBER Research Paper, supra note 344, at 25.
\442\ NBER Research Paper, supra note 344, at 25 (``In most areas
appraisals were sufficiently generous to permit loans nearly as large--
possibly larger--than current market price.'').
---------------------------------------------------------------------------
5. Duration of Relief
Because the HOLC refinanced loans, participating homeowners
got new loans that generally offered permanently lower interest
rates than in their previous mortgages. By contrast, despite
the fact that Treasury refers to HAMP modifications as
permanent, HAMP-modified loans can reset after five years to a
higher interest rate.
6. Risk/Reward for Government
HOLC loans represented a substantial risk to the
government, but one with the possibility of a profit. If a HOLC
borrower defaulted, the government stood to lose a considerable
amount of money in a foreclosure. If the borrower paid off the
loan, the government not only got its money back, it made a
small profit. HAMP's structure is quite different, with both
less potential downside and less potential upside. The
government's potential liability for each loan is limited to a
series of incentive payments over a five-year period, but it
has no opportunity to earn any of that money back.
7. Data Availability
The HOLC closed its doors in 1951, prior to the
popularization of computers. The HOLC therefore relied on
employees using paper to collect data about borrowers and
loans, a process that was inefficient, time-consuming, and
inaccurate by today's standards. One important advantage that
HAMP has over the HOLC is the availability of tools to collect
and analyze vast quantities of data. These tools should allow
Treasury and outside observers to gain a better understanding
of what is working and what is not working in HAMP, in a way
that was not possible during the 1930s and 1940s.
D. Lessons From the HOLC
In the context of today's foreclosure crisis, the size and
breadth of the HOLC's work is striking. The HOLC bought up 10
percent of the residential mortgages in the United States; it
hired 20,000 employees at a time when the U.S. population was
only about 40 percent of today's population; and by amortizing
and extending mortgages, it reshaped the mortgage contract in a
fundamental way. This is not to say that a comparably large
government intervention would be appropriate in the present-day
situation. Indeed, the need for relief was substantially
greater during the 1930s than it is today. And the scale of the
present-day U.S. foreclosure problem will depend in large part
on the future trajectory of home prices.
It is at least arguable that the residential mortgage
market of the 1920s--which in some ways resembles the market
for construction loans today--offered a better opportunity for
government intervention than today's residential mortgage
market. Because many homeowners with mortgages from the 1920s
were not building equity, they had little incentive to continue
making payments when the price of their homes plummeted. By
adjusting home values to more realistic levels, and by giving
homeowners a chance to build equity, the HOLC gave homeowners
much more reason to try to stay in their homes. On the other
hand, certain reckless mortgage products in the 2000s, such as
interest-only loans and negatively amortizing loans with rate
resets, do offer low-hanging fruit for the government. Indeed,
these are the kinds of loans that HAMP was designed to address.
Lastly, the HOLC's servicing practices, which involved
extraordinary efforts by the government to avoid foreclosures,
shed light on some of the problems that HAMP has encountered.
Whereas the HOLC was a purely governmental effort, HAMP, by its
design, relies on private servicers to carry out public-policy
aims. This creates a principal-agent problem--there may be a
divergence of interests between Treasury and the servicers it
is paying--which could be problematic even if HAMP were a
mandatory program. Unfortunately, the voluntary nature of HAMP
exacerbates the problem, because Treasury has few tools
available to ensure that servicers fulfill their obligations
under the program. As was described earlier in the report, if
Treasury enforces the servicers' contractual obligations more
stringently, the servicers may have the option of simply
dropping out of the program.
SECTION TWO: ADDITIONAL VIEWS
A. J. Mark McWatters and Professor Kenneth Troske
We concur with the issuance of the December report and
offer the additional observations below. We appreciate the
efforts the Panel staff made incorporating our suggestions
offered during the drafting of the report.
The issue discussed in this month's report--foreclosures
and the government's efforts to help keep families in their
homes--remain quite contentious and fraught with strong
feelings among people debating this issue. However, when
considering the effectiveness of programs designed to mitigate
foreclosures, it is important to keep in mind that one of our
primary goals should be returning the economy to a place where
it can begin to grow at a pace that helps everyone currently in
distress.
Certainly all of us would like to return to a world where
we have steadily rising housing prices, low unemployment rates,
and an economy that is growing at 4 percent to 5 percent per
year. However, this is not the world in which we currently
live. Instead, we are in an economy where housing prices
nationwide have fallen by 14 percent from their peak,\443\
where prices in the largest metropolitan areas have fallen by
almost one-third,\444\ and annual existing home sales have
plunged by over 40 percent.\445\ Without a doubt, the housing
market has been in disequilibrium for several years, even
before the recent discoveries of problems with foreclosures.
The important question is what are the best policies for
helping the housing market return to stability? Because until
we achieve stability in the housing market, the economy will
continue to limp along at 1 percent to 2 percent growth per
year and unemployment will remain unacceptably high.
---------------------------------------------------------------------------
\443\ Federal Housing Finance Agency, U.S. and Census Division
Monthly Purchase Only Index (Instrument: USA, Seasonally Adjusted)
(online at www.fhfa.gov/Default.aspx?Page=87) (accessed Dec. 10, 2010)
(hereinafter ``U.S. and Census Division Monthly Purchase Only Index'').
\444\ Standard and Poor's, S&P/Case-Shiller Home Price Indices
(Instrument: Case-Shiller 20-City Composite Seasonally Adjusted,
Frequency: Monthly) (online at www.standardandpoors.com/ indices/sp-
case-shiller-home-price-indices/en/us/?indexId=spusa-cashpidff_p-us_)
(accessed Dec. 10, 2010) (hereinafter ``S&P/Case-Shiller Home Price
Indices'').
\445\ Data accessed through Bloomberg Data Service (Dec. 10, 2010).
---------------------------------------------------------------------------
One of the main problems with the housing market is that in
2005 and 2006 many people borrowed money to purchase houses, or
took out home-equity loans, predicated on the belief that
housing prices would continue rising. It is important to note
that few of these borrowers were first-time home buyers.
Instead these were people who had a mortgage and decided to
refinance in order to extract some of the equity they had built
up in their house to purchase other goods. As long as home
values kept rising, homeowners and other investors could
refinance these loans at lower rates based on the accumulation
of equity. When housing prices started to decline in 2006, many
of these people were left with mortgages where the amount they
owed was less than the value of the home. The question is, what
if anything should the government do to fix this problem?
As we point out in the report, the Administration's
foreclosure mitigation programs--primarily HAMP--have failed to
provide meaningful relief to distressed homeowners and,
disappointingly, the Administration has inadvertently created a
sense of false expectations among millions of homeowners who
reasonably anticipated that they would have the opportunity to
modify or refinance their troubled mortgage loans under HAMP.
In our view, the primary reason for HAMP's lack of success lies
in the confusing and illogical basis for the program. Under
HAMP, the government pays lenders and borrowers to modify a
mortgage only when the estimated value of the modified mortgage
(estimated using a procedure specified by the government)
exceeds the estimated value of the foreclosed loan (again
estimated using government rules). In short, under HAMP,
Treasury is planning on paying $30 billion to lenders and
borrowers to do something that they should be willing to do
without receiving any money from the government. The fact that
a program which should be an unmitigated success--paying people
money for nothing--has had such limited success should be a
clue that the situation is far more complicated then it
appears.
To begin with, the structure of HAMP indicates that it is
likely to have only limited success. HAMP works by reducing the
monthly mortgage payments of borrowers through a capitalization
of arrearages, a term extension, forbearance, and/or a
reduction of interest rates or principal for up to five years.
Then the program ends and the interest rate will gradually rise
to the prevailing rate in place at the time the modification
was made. Given the structure of the program, it seems unlikely
that borrowers, especially those with negative equity, will be
able to keep their homes unless we see dramatic improvements in
the housing market, which also seems unlikely. The median
borrower in the program had monthly debt payments equal to 80
percent of their pre-tax income.\446\ On an after-tax basis,
even after all the modifications have been done, after making
their new monthly mortgage payment and all the other payments
to lenders, the typical HAMP participant has $444 per month
left over for expenses such as food, clothing, and health care,
so it is hard to imagine how any modification is going to be
successful.\447\ Additionally, instead of being directed at
borrowers who are in trouble because of some sudden, unexpected
occurrence, such as losing a job or having the value of their
home fall below the balance of their mortgage, this program is
primarily focused on borrowers who can't make their monthly
payments even though they are currently employed and not under
water. This despite evidence from researchers at the Federal
Reserve Banks of Atlanta and Boston showing that helping
workers who have experienced temporary shocks is much more
likely to result in those owners keeping their homes.\448\
---------------------------------------------------------------------------
\446\ Data provided by Treasury.
\447\ See footnote 52 supra, for further discussion of after-tax
debt-to-income ratios.
\448\ Federal Reserve Bank of Atlanta Working Paper, supra note 88.
---------------------------------------------------------------------------
There are also a myriad of details and rules that limit the
ability and/or willingness of lenders to modify loans. For
example, for loans in which there are multiple liens, if the
first lien holder modifies the loan without reaching an
agreement with the other lien holders, then the first lien
holder might have to take a subordinated position to the other
lien holders. Given that over 40 percent of current mortgages
have two or more liens; this significantly increases the cost
of modifying a mortgage.\449\ In addition, since a lender must
recognize losses once a loan is modified, for banks holding a
large number of underwater mortgages, this has the potential to
impose a significant financial strain on the institution, a
strain they will try to avoid.
---------------------------------------------------------------------------
\449\ Amherst Securities Group LP, Amherst Mortgage Insight, 2nd
Liens--How Important, at 3 (Jan. 29, 2010).
---------------------------------------------------------------------------
We are also troubled that HAMP itself may have exacerbated
the mortgage loan delinquency and foreclosure problem by
encouraging homeowners to refrain from remitting their monthly
mortgage installments based upon the expectation that they
would ultimately receive a favorable restructure or principal
reduction subsidized by the taxpayers. The curious incentives
offered by HAMP arguably converts the concept of home ownership
into the economic equivalent of a ``put option''--as long as a
homeowner's residence continues to appreciate in value, the
homeowner will not exercise the put option, but as soon as the
residence falls in value, the homeowner will elect to exercise
the put option and walk away--or threaten to walk away--if a
favorable bailout is not offered.
We remain unconvinced that government-sponsored foreclosure
mitigation programs are necessarily capable of lifting millions
of American families out of their underwater home mortgage
loans. From our perspective, the best foreclosure mitigation
tool is a steady job at a fair wage and not a hodgepodge of
government-subsidized programs that create and perpetuate moral
hazard risks and all but establish the government as the
implicit guarantor of distressed homeowners. In the end it
appears that, for most participants, HAMP will only postpone
the inevitable.
So, what would be the downside if all HAMP does is postpone
foreclosures for a few years? Well, as one of us has pointed
out in an earlier Panel report,\450\ despite all the attention
they have received, homeowners with unaffordable mortgages were
not the only group hurt by the financial crisis. Millions of
homeowners who didn't have mortgages or who had affordable
mortgages saw the value of their home plummet, and this was
devastating for those who were going to use the equity in their
home to finance their retirement. Millions of others saw the
value of their retirement savings decline significantly, and
families lost substantial amounts in their children's college
savings accounts. For all of these people, relief will only
come once the economy starts growing again. That growth will
only occur once the housing market has stabilized, and that
stability will not develop until people move out of homes with
mortgages they cannot afford and into housing they can afford.
So to the extent that HAMP simply kicks the foreclosure can
down the road, it ends up hurting all of the people who are
desperate for the economy to start growing again so that their
lives can return to normal.
---------------------------------------------------------------------------
\450\ April 2010 Oversight Report, supra note 1, at 179-180 (from
the additional views of J. Mark McWatters).
---------------------------------------------------------------------------
HAMP carries a 100 percent subsidy rate according to the
Congressional Budget Office (CBO).\451\ This means that the
U.S. government expects to recover none of the $30 billion of
taxpayer-sourced TARP funds invested in HAMP. Since Treasury is
charged with protecting the interests of the taxpayers who fund
HAMP and the other TARP programs, we recommend that Treasury's
foreclosure mitigation efforts be structured so as to
incorporate an effective exit strategy by allowing Treasury to
participate in any subsequent appreciation in the home equity
of any mortgagor whose loan is modified under HAMP or any other
taxpayer-subsidized program. An equity appreciation right--the
functional equivalent of a warrant in a non-commercial
transaction--will also mitigate the moral hazard risk of
homeowners who may undertake risky loans in the future based on
the assumption that the government will act as a backstop with
no strings attached.
---------------------------------------------------------------------------
\451\ CBO Report on the TARP--November 2010, supra note 113, at 7.
---------------------------------------------------------------------------
This analysis is in no way intended to diminish the
financial hardship that many Americans are suffering as they
attempt to modify or refinance their underwater home mortgage
loans, and we fully acknowledge and empathize with the stress
and economic uncertainty created from the bursting of the
housing bubble. It is particularly frustrating--although not
surprising--that many of the hardest hit housing markets are
also suffering from seemingly intractable rates of unemployment
and underemployment. We also recognize that there have been
serious mistakes, and perhaps fraud, committed by servicers and
lenders in the lending and foreclosure process, and any illegal
activity on the part of banks needs to be fully prosecuted. In
addition, we know that many homeowners are rightfully
frustrated and angry over the treatment they have received by
lenders and servicers once they begin to experience financial
distress. As such, we encourage each mortgage loan and
securitized debt investor and servicer to work with each of
their borrowers in a good faith, transparent, and accountable
manner to reach an economically reasonable resolution prior to
pursuing foreclosure. In our view, foreclosure should serve as
the exception to the rule that only follows from the
transparent and objective failure of the parties to modify or
refinance a troubled mortgage loan pursuant to market-based
terms. It is regrettable that HAMP creates disincentives for
investors and servicers as well as homeowners by rewarding
their dilatory and inefficient behavior with the expectation of
enhanced taxpayer-funded subsidies. Since any intermediate to
long-term resolution of the housing crisis must reside
substantially with the private sector lenders and investors who
hold the mortgage notes and liens, instead of spending an
additional $30 billion on a government-sponsored foreclosure
mitigation effort, we believe Treasury would be best served by
strongly encouraging these participants to engage in good
faith, market-based negotiations with their distressed
borrowers. In our opinion, this is the best way to bring
stability to the housing market so that the economy can start
growing again.
SECTION THREE: CORRESPONDENCE WITH TREASURY
Patricia Geoghegan, the Special Master for TARP Executive
Compensation, sent a letter to Senator Ted Kaufman, the Panel's
Chairman, on November 18, 2010.\452\ The letter responds to a
series of questions presented by the Panel seeking additional
information about TARP executive compensation restrictions
following the Panel's October 21, 2010 hearing on the
topic.\453\
---------------------------------------------------------------------------
\452\ See Appendix I, infra.
\453\ See Appendix I of the Panel's November 2010 Oversight Report,
supra note 79, at 125.
SECTION FOUR: TARP UPDATES SINCE LAST REPORT
A. GM IPO
General Motors held an initial public offering on November
18, 2010, after three quarters of posting profit. Treasury
received $11.7 billion in net proceeds on November 23 for the
sale of 358,546,795 shares of common stock. As widely expected,
underwriters exercised their over-allotment option, and on
December 2, Treasury received $1.8 billion in net proceeds from
the sale of 53,782,019 additional shares of common stock. In
total, these sales reduced Treasury's share of GM's outstanding
common stock from 60.8 percent to 33.3 percent. Treasury's
remaining stock in General Motors consists of 500,065,254
shares of common stock. Treasury's total receipt from the IPO
was $13.5 billion.
B. Stress-Tested Banks
The Federal Reserve has requested that bank holding
companies that participated in the Supervisory Capital
Assessment Program (SCAP) consult with Federal Reserve staff
before taking actions that could result in a diminished capital
base, such as increasing dividends, repurchasing common stock,
and other planned capital actions. Stress-tested BHCs have been
requested to file a Comprehensive Capital Plan by January 7,
2011. This plan should incorporate a stress-testing framework
that will estimate potential capital needs under a range of
circumstances from normal to very severe. The Federal Reserve
will use this and supervisory actions, in addition to firms'
risk profiles, in order to assess the BHC's capital adequacy.
Stress-tested BHCs are expected to complete reimbursement of
U.S. government investment before undertaking any other capital
actions. Five SCAP institutions--Fifth Third Bancorp, SunTrust,
Regions Financial, KeyCorp, and GMAC/Ally Financial--have yet
to repay their TARP assistance.
C. Citigroup Stock Sale
Treasury announced on December 6, 2010 that it was
commencing an underwritten public offering of approximately 2.4
billion shares of Citigroup Inc. common stock. Treasury
converted its common shares at $3.25 per share in July 2009 and
the stock was sold to investors at $4.35 on December 6, 2010,
which will result in an estimated profit of $2.64 billion. This
offering will dispose of Treasury's remaining shares of
Citigroup common stock, although Treasury will still hold
warrants and be entitled to receive up to $800 million of
CitigroupTrust Preferred securities from the FDIC. As of the
time of the writing of this report, the sale has not officially
closed, and thus the Panel has not incorporated the results of
this sale in this month's financial update section.
D. Metrics
Each month, the Panel's report highlights a number of
metrics that the Panel and others, including Treasury, the
Government Accountability Office (GAO), Special Inspector
General for the Troubled Asset Relief Program (SIGTARP), and
the Financial Stability Oversight Board, consider useful in
assessing the effectiveness of the Administration's efforts to
restore financial stability and accomplish the goals of EESA.
This section discusses changes that have occurred in several
indicators since the release of the Panel's November 2010
report.
1. Financial Indices
a. Overview
The St. Louis Financial Stress Index, a proxy for financial
stress in the U.S. economy, has remained relatively stable
since the Panel's November report.\454\ The index has decreased
more than 60 percent since its post-crisis peak in June 2010.
The recent trend in the index suggests that financial stress
continues moving toward its long-run norm. The index has
decreased by more than three standard deviations since EESA was
enacted in October 2008.
---------------------------------------------------------------------------
\454\ Federal Reserve Bank of St. Louis, Series STLFSI: Business/
Fiscal: Other Economic Indicators (Instrument: St. Louis Financial
Stress Index, Frequency: Weekly) (online at research.stlouisfed.org/
fred2/series/STLFSI) (accessed Dec. 1, 2010). The index includes 18
weekly data series, beginning in December 1993 to the present. The
series are: effective federal funds rate, 2-year Treasury, 10-year
Treasury, 30-year-Treasury, Baa-rated corporate, Merrill Lynch High
Yield Corporate Master II Index, Merrill Lynch Asset-Backed Master BBB-
rated, 10-year Treasury minus 3-month Treasury, Corporate Baa-rated
bond minus 10-year Treasury, Merrill Lynch High Yield Corporate Master
II Index minus 10-year Treasury, 3-month LIBOR-OIS spread, 3-month TED
spread, 3-month commercial paper minus 3-month Treasury, the J.P.
Morgan Emerging Markets Bond Index Plus, Chicago Board Options Exchange
Market Volatility Index, Merrill Lynch Bond Market Volatility Index (1-
month), 10-year nominal Treasury yield minus 10-year Treasury Inflation
Protected Security yield, and Vanguard Financials Exchange-Traded Fund
(equities). The index is constructed using principal components
analysis after the data series are de-meaned and divided by their
respective standard deviations to make them comparable units. The
standard deviation of the index is set to 1. For more details on the
construction of this index, see Federal Reserve Bank of St. Louis,
National Economic Trends Appendix: The St. Louis Fed's Financial Stress
Index (Jan. 2010) (online at research.stlouisfed.org/publications/net/
NETJan2010Appendix.pdf).
---------------------------------------------------------------------------
FIGURE 34: ST. LOUIS FEDERAL RESERVE FINANCIAL STRESS INDEX
[GRAPHIC] [TIFF OMITTED] 62622A.029
Stock market volatility, as measured by the Chicago Board
Options Exchange Volatility Index (VIX), continues to decrease.
The VIX has fallen by more than half since the post-crisis peak
in May 2010 and has decreased 4 percent since the Panel's
November report. However, as of December 1, 2010, volatility
was 35 percent higher than its post-crisis low on April 12,
2010.
FIGURE 35: CHICAGO BOARD OPTIONS EXCHANGE VOLATILITY INDEX \455\
---------------------------------------------------------------------------
\455\ Data accessed through Bloomberg data service (Dec. 1, 2010).
The CBOE VIX is a key measure of market expectations of near-term
volatility. Chicago Board Options Exchange, The CBOE Volatility Index--
VIX, 2009 (online at www.cboe.com/micro/vix/vixwhite.pdf) (accessed
Dec. 1, 2010).
[GRAPHIC] [TIFF OMITTED] 62622A.030
b. Interest Rates, Spreads, and Issuance
As of December 1, 2010, the 3-month and 1-month London
Interbank Offer Rates (LIBOR), the prices at which banks lend
and borrow from each other, were 0.30 and 0.27,
respectively.\456\ Rates have increased slightly since the
Panel's November report. However, the 3-month and 1-month LIBOR
remain below their post-crisis highs in June 2010. Over the
longer term, however, interest rates remain extremely low
relative to pre-crisis levels, reflecting the impact of the
actions of central banks and institutions' perceptions of
reduced risk in lending to other banks.
---------------------------------------------------------------------------
\456\ Data accessed through Bloomberg data service (Dec. 1, 2010).
FIGURE 36: 3-MONTH AND 1-MONTH LIBOR RATES (AS OF DECEMBER 1, 2010)
----------------------------------------------------------------------------------------------------------------
Percent Change From Data
Indicator Current Rates Available at Time of Last
Report (11/3/2010)
----------------------------------------------------------------------------------------------------------------
3-Month LIBOR \457\............................................ 0.30 3.4
1-Month LIBOR \458\............................................ 0.27 8.0
----------------------------------------------------------------------------------------------------------------
\457\ Data accessed through Bloomberg data service (Dec. 1, 2010).
\458\ Data accessed through Bloomberg data service (Dec. 1, 2010).
As of December 1, 2010, the conventional mortgage rate
spread, which measures the difference between 30-year mortgage
rates and 10-year Treasury bond yields, remained unchanged
since the Panel's November report.\459\ The TED spread, which
captures the difference between the 3-month LIBOR and the 3-
month Treasury bill rates, serves as an indicator for perceived
risk in the financial markets.\460\ As of December 1, 2010, the
spread was 14.3 basis points, declining approximately two basis
points in November. As shown in Figure 37 below, the spread
remains below pre-crisis levels.
---------------------------------------------------------------------------
\459\ Board of Governors of the Federal Reserve System, Federal
Reserve Statistical Release H.15: Selected Interest Rates: Historical
Data (Instrument: Conventional Mortgages, Frequency: Weekly) (online at
www.federalreserve.gov/releases/h15/data/Weekly_Thursday_/
H15_MORTG_NA.txt) (accessed Dec. 1, 2010) (hereinafter ``Federal
Reserve Statistical Release H.15'').
\460\ Federal Reserve Bank of Minneapolis, Measuring Perceived
Risk--The TED Spread (Dec. 2008) (online at www.minneapolisfed.org/
publications_papers/pub_display.cfm?id=4120).
---------------------------------------------------------------------------
The LIBOR-OIS (Overnight Index Swap) spread serves as an
indicator of the health of the banking system, as it reflects
what banks believe to be the risk of default associated with
interbank lending.\461\ The spread increased over threefold
from early April to July, before falling in mid-July.\462\
Decreases in the LIBOR-OIS spread and the TED spread suggest
that hesitation among banks to lend to counterparties has
receded. The LIBOR-OIS spread remained fairly constant since
the Panel's November report, averaging approximately 11 basis
points during the month.
---------------------------------------------------------------------------
\461\ Federal Reserve Bank of St. Louis, What the LIBOR-OIS Spread
Says (May 11, 2009) (online at research.stlouisfed.org/publications/es/
09/ES0924.pdf).
\462\ Data accessed through Bloomberg data service (Dec. 1, 2010).
---------------------------------------------------------------------------
FIGURE 37: TED SPREAD \463\
---------------------------------------------------------------------------
\463\ Data accessed through Bloomberg data service (Dec. 1, 2010).
[GRAPHIC] [TIFF OMITTED] 62622A.031
FIGURE 38: LIBOR-OIS SPREAD \464\
---------------------------------------------------------------------------
\464\ Data accessed through Bloomberg data service on (Dec. 1,
2010).
[GRAPHIC] [TIFF OMITTED] 62622A.032
The interest rate spread for AA asset-backed commercial
paper, which is considered mid-investment grade, decreased by
more than five percent since the Panel's November report. The
interest rate spread on A2/P2 commercial paper, a lower grade
investment than AA asset-backed commercial paper, fell by
approximately 10 percent. These declining spreads indicate
healthier fundraising conditions for corporations.
FIGURE 39: INTEREST RATE SPREADS (AS OF DECEMBER 1, 2010)
------------------------------------------------------------------------
Percent Change Since
Indicator Current Last Report (11/1/
Spread 2010)
------------------------------------------------------------------------
Conventional mortgage rate spread 1.56 0.0
\465\............................
TED Spread (basis points)......... 14.34 (8.0)
Overnight AA asset-backed 0.07 (5.4)
commercial paper interest rate
spread \466\.....................
Overnight A2/P2 nonfinancial 0.12 (10.1)
commercial paper interest rate
spread \467\.....................
------------------------------------------------------------------------
\465\ Federal Reserve Statistical Release H.15, supra note 459; Board of
Governors of the Federal Reserve System, Federal Reserve Statistical
Release H.15: Selected Interest Rates: Historical Data (Instrument:
U.S. Government Securities/Treasury Constant Maturities/Nominal 10-
Year, Frequency: Weekly) (online at www.federal reserve.gov/releases/
h15/data/Weekly_Friday_/H15_ TCMNOM_Y10.txt) (accessed Dec. 1, 2010).
\466\ The overnight AA asset-backed commercial paper interest rate
spread reflects the difference between AA asset-backed commercial
paper discount rate and the AA nonfinancial commercial paper discount
rate. Board of Governors of the Federal Reserve System, Federal
Reserve Statistical Release: Commercial Paper Rates and Outstandings:
Data Download Program (Instrument: AA Asset-Backed Discount Rate,
Frequency: Daily) (online at www.federalreserve.gov/DataDownload/
Choose.aspx?rel=CP) (accessed Dec. 1, 2010); Board of Governors of the
Federal Reserve System, Federal Reserve Statistical Release:
Commercial Paper Rates and Outstandings: Data Download Program
(Instrument: AA Nonfinancial Discount Rate, Frequency: Daily) (online
at www.federalreserve.gov/DataDownload/Choose.aspx?rel=CP) (accessed
Dec. 1, 2010). In order to provide a more complete comparison, this
metric utilizes the average of the interest rate spread for the last
five days of the month.
\467\ The overnight A2/P2 nonfinancial commercial paper interest rate
spread reflects the difference between A2/P2 nonfinancial commercial
paper discount rate and the AA nonfinancial commercial paper discount
rate. Board of Governors of the Federal Reserve System, Federal
Reserve Statistical Release: Commercial Paper Rates and Outstandings:
Data Download Program (Instrument: A2/P2 Nonfinancial Discount Rate,
Frequency: Daily) (online at www.federalreserve.gov/DataDownload/
Choose.aspx?rel=CP) (accessed Dec. 1, 2010); Board of Governors of the
Federal Reserve System, Federal Reserve Statistical Release:
Commercial Paper Rates and Outstandings: Data Download Program
(Instrument: AA Nonfinancial Discount Rate, Frequency: Daily) (online
at www.federalreserve.gov/DataDownload/Choose.aspx?rel=CP) (accessed
Dec. 1, 2010). In order to provide a more complete comparison, this
metric utilizes the average of the interest rate spread for the last
five days of the month.
The spread between Moody's Baa Corporate Bond Yield Index
and 30-year constant maturity U.S. Treasury Bond, which
indicates the difference in perceived risk between corporate
and government bonds, doubled from late April to mid-June 2010.
During November, the spread declined over 5 percent, and has
decreased 22 percent since its post-crisis peak in mid-June.
The declining spread could indicate waning concerns about the
riskiness of corporate bonds.
FIGURE 40: MOODY'S BAA CORPORATE BOND INDEX AND 30-YEAR U.S. TREASURY
YIELD \468\
---------------------------------------------------------------------------
\468\ Federal Reserve Bank of St. Louis, Series DGS30: Selected
Interest Rates (Instrument: 30-Year Treasury Constant Maturity Rate,
Frequency: Daily) (online at research.stlouisfed.org/fred2/) (accessed
Dec. 1, 2010) (hereinafter ``Series DGS30: Selected Interest Rates'').
Corporate Baa rate data accessed through Bloomberg data service (Dec.
1, 2010).
[GRAPHIC] [TIFF OMITTED] 62622A.033
c. Condition of the Banks
During November, year-to-date bank failures surpassed the
2009 level of 140 failures. As of November 25, 2010, 149 banks
have been placed into receivership. Despite exceeding the total
number of bank failures in 2009, banks that have failed in 2010
thus far had $90.5 billion in total assets, which represents
only half of the total assets of failed institutions in
2009.\469\ Most failures in 2010 involved institutions with
less than $10 billion in assets. Of the 10 banks that failed in
November, two were CPP recipients.
---------------------------------------------------------------------------
\469\ Federal Deposit Insurance Corporation, Failures & Assistance
Transactions (online at www2.fdic.gov/hsob/SelectRpt.asp?EntryTyp=30)
(accessed Dec. 1, 2010) (hereinafter ``Failures & Assistance
Transactions'').
---------------------------------------------------------------------------
FIGURE 41: BANK FAILURES AS A PERCENTAGE OF TOTAL BANKS AND BANK
FAILURES BY TOTAL ASSETS (1990-2010) \470\
---------------------------------------------------------------------------
\470\ The disparity between the number of and total assets of
failed banks in 2008 is driven primarily by the failure of Washington
Mutual Bank, which held $307 billion in assets. The 2010 year-to-date
percentage of bank failures includes failures through November. The
total number of FDIC-insured institutions as of September 30, 2010 is
7,760 commercial banks and savings institutions, which represents a
decline of 70 institutions since June 30, 2010. Failures & Assistance
Transactions, supra note 469; Federal Deposit Insurance Corporation,
Quarterly Banking Profile, Third Quarter 2010: Statistics At A Glance
(online at www.fdic.gov/bank/statistical/stats/2010sep/industry.pdf)
(accessed Dec. 10, 2010). Asset totals have been adjusted for deflation
into 2005 dollars using the GDP implicit price deflator. The quarterly
values were averaged into a yearly value. Series DGS30: Selected
Interest Rates, supra note 468.
[GRAPHIC] [TIFF OMITTED] 62622A.034
2. Unemployment and Underemployment
The unemployment rate increased in November to 9.8 percent
after three consecutive months at 9.6 percent, while the
underemployment rate remained unchanged at 17.0 percent. The
median duration of unemployment increased by approximately half
a week, to 21.6 weeks, in November.
FIGURE 42: UNEMPLOYMENT, UNDEREMPLOYMENT, AND MEDIAN DURATION OF
UNEMPLOYMENT \471\
---------------------------------------------------------------------------
\471\ It is important to note that the measures of unemployment and
underemployment do not include people who have stopped actively looking
for work altogether. While the Bureau of Labor Statistics (BLS) does
not have a distinct metric for ``underemployment,'' the U-6 category of
Table A-15 ``Alternative Measures of Labor Underutilization'' is used
here as a proxy. BLS defines this measure as: ``Total unemployed, plus
all persons marginally attached to the labor force, plus total employed
part time for economic reasons, as a percent of the civilian labor
force plus all persons marginally attached to the labor force.'' U.S.
Department of Labor, International Comparisons of Annual Labor Force
Statistics (online at www.bls.gov/webapps/legacy/cpsatab15.htm)
(accessed Dec. 3, 2010); Series DGS30: Selected Interest Rates, supra
note 468.
[GRAPHIC] [TIFF OMITTED] 62622A.035
3. Housing Indices
New home sales saw a month-over-month decrease in October,
declining 8 percent during the month. New home sales as
measured by the U.S. Census Bureau totaled 283,000 units. With
respect to existing home sales, National Association of
Realtors estimates a 2 percent month-over-month decline in
October, to 4.4 million homes sold. Although existing home
sales in October remained below the 10-year historical average,
current levels are above the July 2010 level, when existing
home sales reached their lowest point in more than a decade.
FIGURE 43: NEW AND EXISTING HOME SALES (2000-2010) \472\
---------------------------------------------------------------------------
\472\ Data accessed through Bloomberg Data Service (Dec. 1, 2010).
Spikes in both new and existing home sales in January 2009 and November
2009 correlate with the tax credits extended to first-time and repeat
home buyers during these periods. After both tax credits were
extinguished on April 30, 2010, existing home sales dropped to 3.8
million homes in July, their lowest level in a decade. National
Association of Realtors, July Existing-Home Sales Fall as Expected but
Prices Rise (Aug. 24, 2010) (online at www.realtor.org/press_room/
news_releases/2010/08/ehs_fall).
[GRAPHIC] [TIFF OMITTED] 62622A.036
Foreclosure actions, which consist of default notices,
scheduled auctions, and bank repossessions, increased 4.4
percent in October to 332,172.\473\ Since the enactment of
EESA, there have been approximately 8.1 million foreclosure
filings.\474\ Both the Case-Shiller Composite 20-City Composite
Home Price Index and the FHFA Housing Price Index decreased
approximately 1 percent in September 2010. The Case-Shiller and
FHFA indices are 7 percent and 6 percent, respectively, below
their October 2008 levels.\475\
---------------------------------------------------------------------------
\473\ RealtyTrac, Foreclosure Activity Decreases 4 Percent in
October (Nov. 11, 2010) (online at www.realtytrac.com/content/press-
releases/foreclosure-activity-decreases-4-percent-in-october-6182)
(hereinafter ``Foreclosure Activity Decreases 4 Percent in October'').
\474\ Data accessed through Bloomberg data service (Dec. 1, 2010).
\475\ The most recent data available are for September 2010. See
S&P/Case-Shiller Home Price Indices, supra note 444; U.S. and Census
Division Monthly Purchase Only Index, supra note 443. S&P has cautioned
that the seasonal adjustment is probably being distorted by irregular
factors. These factors could include distressed sales and the various
government programs. See Standard and Poor's, S&P/Case-Shiller Home
Price Indices and Seasonal Adjustment (Apr. 2010) (online at
www.standardandpoors. com/ servlet/ BlobServer?blobheadername3= MDT-
Type&blobcol=urldata&blobtable= MungoBlobs&blobheadervalue2=
inline;+filename%3DCase
Shiller_SeasonalAdjustment2,0.pdf&blobheadername2=Content-
Disposition&blobheadervalue1= application/
pdf&blobkey=id&blobheadername1=content-
type&blobwhere=1243679046081&blobheadervalue3=UTF-8). For a discussion
of the differences between the Case-Shiller Index and the FHFA Index,
see April 2010 Oversight Report, supra note 1, at 98.
---------------------------------------------------------------------------
Case-Shiller futures prices indicate a market expectation
that home-price values for the major Metropolitan Statistical
Areas (MSAs) will decrease through 2011.\476\ These futures are
cash-settled to a weighted composite index of U.S. housing
prices in the top 10 MSAs, as well as to those specific
markets. They are used to hedge by businesses whose profits and
losses are related to any area of the housing industry, and to
balance portfolios by businesses seeking exposure to an
uncorrelated asset class. As such, futures prices are a
composite indicator of market information known to date and can
be used to indicate market expectations for home prices.
---------------------------------------------------------------------------
\476\ Data accessed through Bloomberg data service on December 1,
2010. The Case-Shiller Futures contract is traded on the Chicago
Mercantile Exchange (CME) and is settled to the Case-Shiller Index two
months after the previous calendar quarter. For example, the February
contract will be settled against the spot value of the S&P Case-Shiller
Home Price Index values representing the fourth calendar quarter of the
previous year, which is released in February one day after the
settlement of the contract. Note that most close observers believe that
the accuracy of these futures contracts as forecasts diminishes the
farther out one looks.
A Metropolitan Statistical Area is defined as a core area
containing a substantial population nucleus, together with adjacent
communities having a high degree of economic and social integration
with the core. U.S. Census Bureau, About Metropolitan and Micropolitan
Statistical Areas (online at www.census.gov/population/www/metroareas/
aboutmetro.html) (accessed Dec. 10, 2010).
FIGURE 44: HOUSING INDICATORS
----------------------------------------------------------------------------------------------------------------
Percent Change from Percent Change
Indicator Most Recent Data Available at Time Since October
Monthly Data of Last Report 2008
----------------------------------------------------------------------------------------------------------------
Monthly foreclosure actions \477\................... 332,172 (4.4) 18.8
S&P/Case-Shiller Composite 20 Index \478\........... 145.47 (1.0) (6.9)
FHFA Housing Price Index \479\...................... 190.47 (1.2) (5.7)
----------------------------------------------------------------------------------------------------------------
\477\ Foreclosure Activity Decreases 4 Percent in October, supra note 473. The most recent data available are
for October 2010.
\478\ S&P/Case-Shiller Home Price Indices, supra note 444. The most recent data available are for September
2010.
\479\ U.S. and Census Division Monthly Purchase Only Index, supra note 443. The most recent data available are
for September 2010.
FIGURE 45: CASE-SHILLER HOME PRICE INDEX AND FUTURES VALUES \480\
---------------------------------------------------------------------------
\480\ All data normalized to 100 at January 2000. Futures data
accessed through Bloomberg data service on December 1, 2010. S&P/Case-
Shiller Home Price Indices, supra note 444.
[GRAPHIC] [TIFF OMITTED] 62622A.037
E. Financial Update
Each month, the Panel summarizes the resources that the
federal government has committed to the rescue and recovery of
the financial system. The following financial update provides:
(1) an updated accounting of the TARP, including a tally of
dividend income, repayments, and warrant dispositions that the
program has received as of October 31, 2010; and (2) an updated
accounting of the full federal resource commitment as of
November 26, 2010.
1. The TARP
a. Program Updates \481\
---------------------------------------------------------------------------
\481\ U.S. Department of the Treasury, Cumulative Dividends,
Interest and Distributions Report as of September 30, 2010 (Oct. 11,
2010) (online at financialstability.gov/docs/dividends-interest-
reports/September%202010%20Dividends%20&%20Interest%20 Report.pdf)
(hereinafter ``Cumulative Dividends, Interest and Distributions
Report''); U.S. Department of the Treasury, Troubled Asset Relief
Program Transactions Report for the Period Ending November 26, 2010
(Nov. 30, 2010) (online at financialstability.gov/docs/transaction-
reports/11-30-10%20Transactions%20Report%20as%20of%2011-26-10.pdf)
(hereinafter ``Treasury Transactions Report'').
---------------------------------------------------------------------------
Treasury's spending authority under the TARP officially
expired on October 3, 2010. Though it can no longer make new
funding commitments, Treasury can continue to provide funding
for programs for which it has existing contracts and previous
commitments. To date, $395.1 billion has been spent under the
TARP's $475 billion ceiling.\482\ Of the total amount
disbursed, $223.0 billion has been repaid. Treasury has also
incurred $6.1 billion in losses associated with its CPP and
Automotive Industry Financing Program (AIFP) investments. A
significant portion of the $166.7 billion in TARP funds
currently outstanding relates to Treasury's investments in AIG
and assistance provided to the automotive industry.
---------------------------------------------------------------------------
\482\ The original $700 billion TARP ceiling was reduced by $1.26
billion as part of the Helping Families Save Their Homes Act of 2009.
12 U.S.C. Sec. 5225(a)-(b); Helping Families Save Their Homes Act of
2009, Pub. L. No. 111-22 Sec. 202(b) (2009) (online at
financialservices.house.gov/FinancialSvcsDemMedia/file/public%20laws/
111-22.pdf). On June 30, 2010, the House-Senate Conference Committee
agreed to reduce the amount authorized under the TARP from $700 billion
to $475 billion as part of the Dodd-Frank Wall Street Reform and
Consumer Protection Act that was signed into law on July 21, 2010. See
Dodd-Frank Wall Street Reform and Consumer Protection Act, Pub. L. No.
111-203 (2010); The White House, Remarks by the President at Signing of
Dodd-Frank Wall Street Reform and Consumer Protection Act (July 21,
2010) (online at www.whitehouse.gov/the-press-office/remarks-president-
signing-dodd-frank-wall-street-reform-and-consumer-protection-act).
---------------------------------------------------------------------------
CPP Repayments
As of November 26, 2010, 114 of the 707 banks that
participated in the CPP have fully redeemed their preferred
shares either through capital repayment or exchanges for
investments under the Community Development Capital Initiative
(CDCI). During the month of November, Treasury received an
$11.3 million full repayment from Central Jersey Bancorp, a
$5.83 million full repayment from Leader Bancorp, Inc. and a
$6.25 million partial repayment from Horizon Bancorp. A total
of $152.9 billion has been repaid under the program, leaving
$49.4 billion in funds currently outstanding.
b. Income: Dividends, Interest, and Warrant Sales
In conjunction with its preferred stock investments under
the CPP and the TIP, Treasury generally received warrants to
purchase common equity.\483\ As of November 26, 2010, 45
institutions have repurchased their warrants from Treasury at
an agreed upon price. Treasury has also sold warrants for 15
other institutions at auction. To date, income from warrant
dispositions totals $8.1 billion.
---------------------------------------------------------------------------
\483\ For its CPP investments in privately held financial
institutions, Treasury also received warrants to purchase additional
shares of preferred stock, which it exercised immediately. Similarly,
Treasury also received warrants to purchase additional subordinated
debt that were also immediately exercised along with its CPP
investments in subchapter S corporations. Treasury Transactions Report,
supra note 481, at 14.
---------------------------------------------------------------------------
In addition to warrant proceeds, Treasury also receives
dividend payments on the preferred shares that it holds under
the CPP, 5 percent per annum for the first five years and 9
percent per annum thereafter.\484\ For preferred shares issued
under the TIP, Treasury received a dividend of 8 percent per
annum.\485\ In total, Treasury has received approximately $25.8
billion in net income from warrant repurchases, dividends,
interest payments, and other proceeds deriving from TARP
investments (after deducting losses).\486\ For further
information on TARP profit and loss, see Figure 46.
---------------------------------------------------------------------------
\484\ U.S. Department of the Treasury, Capital Purchase Program
(Oct. 3, 2010) (online at www. financial stability.gov/
roadtostability/capitalpurchaseprogram.html).
\485\ U.S. Department of the Treasury, Targeted Investment Program
(Oct. 3, 2010) (online at www. financial stability.gov/roadtostability/
targetedinvestmentprogram.html).
\486\ Cumulative Dividends, Interest and Distributions Report,
supra note 481; Treasury Transactions Report, supra note 481. Treasury
also received an additional $1.2 billion in participation fees from its
Guarantee Program for Money Market Funds. U.S. Department of the
Treasury, Treasury Announces Expiration of Guarantee Program for Money
Market Funds (Sept. 18, 2009) (online at www.ustreas.gov/press/
releases/tg293.htm).
---------------------------------------------------------------------------
c. TARP Accounting
FIGURE 46: TARP ACCOUNTING (AS OF NOVEMBER 26, 2010)
[Dollars in billions] i
----------------------------------------------------------------------------------------------------------------
Maximum Total Repayments/ Funding
Program Amount Actual Reduced Total Currently Funding
Allotted Funding Exposure Losses Outstanding Available
----------------------------------------------------------------------------------------------------------------
Capital Purchase Program $204.9 $204.9 ii $(152.9) iii $(2.6) $49.5 $0
(CPP)......................
Targeted Investment Program 40.0 40.0 (40.0) 0 0 0
(TIP)......................
Asset Guarantee Program 5.0 iv 5.0 v (5.0) 0 0 0
(AGP)......................
AIG Investment Program 69.8 vi 47.5 0 0 47.5 22.3
(AIGIP)....................
Auto Industry Financing 81.3 81.3 (24.3) vii (3.5) viii 53.6 0
Program (AIFP).............
Auto Supplier Support 0.4 0.4 (0.4) 0 0 0
Program (ASSP) ix..........
Term Asset-Backed Securities x 4.3 xi 0.1 0 0 0.1 4.2
Loan Facility (TALF).......
Public-Private Investment 22.4 xiii 14.9 xiv (0.4) 0 14.4 7.5
Program (PPIP) xii.........
SBA 7(a) Securities Purchase 0.4 xv 0.4 0 0 0.4 xvi 0
Home Affordable Modification 29.9 0.7 0 0 0.7 29.2
Program (HAMP).............
Hardest Hit Fund (HHF)...... xvii 7.6 xviii 0.1 0 0 0.1 7.5
FHA Refinance Program....... 8.1 xix 0.1 0 0 0.1 8.0
Community Development 0.8 xx 0.6 0 0 0.6 0
Capital Initiative (CDCI)..
Total................... $475.0 $395.9 $(223.0) $(6.0) $167.0 $78.8
----------------------------------------------------------------------------------------------------------------
i Figures affected by rounding. Unless otherwise noted, data in this table are from the following source: U.S.
Department of the Treasury, Troubled Asset Relief Program Transactions Report for the Period Ending November
26, 2010 (Nov. 30, 2010) (online at financialstability.gov/docs/transaction-reports/11-30-
10%20Transactions%20Report%20as%20of%2011-26-10.pdf).
ii As of October 29, 2010, Treasury had sold 4.1 billion Citigroup common shares for $16.4 billion in gross
proceeds. Amount repaid under CPP includes $13.4 billion Treasury received as part of its sales of Citigroup
common stock. The difference between these two numbers represents the $3.0 billion in net profit Treasury has
received from the sale of Citigroup common stock. In June 2009, Treasury exchanged $25 billion in Citigroup
preferred stock for 7.7 billion shares of the company's common stock at $3.25 per share.
Total CPP repayments also include amounts repaid by institutions that exchanged their CPP investments for
investments under the CDCI, as well as proceeds earned from the sale of preferred stock issued by South
Financial Group, Inc. and TIB Financial Corp and warrants. See U.S. Department of the Treasury, Troubled Asset
Relief Program Transactions Report for the Period Ending November 26, 2010, at 2, 13-15 (Nov. 30, 2010)
(online at financialstability.gov/docs/transaction-reports/11-30-10%20Transactions%20Report%20as%20of%2011-26-
10.pdf); U.S. Department of the Treasury, Troubled Asset Relief Program: Two-Year Retrospective, at 25 (Oct.
2010) (online at www.financialstability.gov/docs/
TARP%20Two%20Year%20Retrospective_10%2005%2010_transmittal%20letter.pdf); U.S. Department of the Treasury,
Treasury Commences Plan to Sell Citigroup Common Stock (Apr. 26, 2010) (online at ustreas.tpaq.treasury.gov/
press/releases/tg660.htm).
iii On the TARP Transactions Report, Treasury has classified the investments it made in two institutions, CIT
Group ($2.3 billion) and Pacific Coast National Bancorp ($4.1 million), as losses. In addition, Treasury sold
its preferred ownership interests, along with warrants, in South Financial Group, Inc. and TIB Financial Corp.
to non-TARP participating institutions. These shares were sold at prices below the value of the original CPP
investment, at respective losses of $217 million and $25 million. Therefore, Treasury's net current CPP
investment is $49.5 billion due to the $2.6 billion in losses thus far. See U.S. Department of the Treasury,
Troubled Asset Relief Program Transactions Report for the Period Ending November 26, 2010, at 13-14 (Nov. 30,
2010) (online at financialstability.gov/docs/transaction-reports/11-30-
10%20Transactions%20Report%20as%20of%2011-26-10.pdf).
iv The $5.0 billion AGP guarantee for Citigroup was unused since Treasury was not required to make any guarantee
payments during the life of the program. U.S. Department of the Treasury, Troubled Asset Relief Program: Two-
Year Retrospective, at 31 (Oct. 2010) (online at www.financialstability.gov/docs/
TARP%20Two%20Year%20Retrospective_10%2005%2010_transmittal%20letter.pdf); U.S. Department of the Treasury,
Troubled Asset Relief Program Transactions Report for the Period Ending November 26, 2010, at 20 (Nov. 30,
2010) (online at financialstability.gov/docs/transaction-reports/11-30-
10%20Transactions%20Report%20as%20of%2011-26-10.pdf).
v Although this $5.0 billion is no longer exposed as part of the AGP, Treasury did not receive a repayment in
the same sense as with other investments. Treasury did receive other income as consideration for the
guarantee, which is not a repayment and is accounted for in Figure 46. See U.S. Department of the Treasury,
Troubled Asset Relief Program Transactions Report for the Period Ending November 26, 2010, at 20 (Nov. 30,
2010) (online at financialstability.gov/docs/transaction-reports/11-30-
10%20Transactions%20Report%20as%20of%2011-26-10.pdf).
vi AIG has completely utilized the $40 billion that was made available on November 25, 2008, in exchange for the
company's preferred stock. See U.S. Department of the Treasury, Troubled Asset Relief Program Transactions
Report for the Period Ending November 26, 2010, at 21 (Nov. 30, 2010) (online at financialstability.gov/docs/
transaction-reports/11-30-10%20Transactions%20Report%20as%20of%2011-26-10.pdf). It has also drawn down $7.5
billion of the $29.8 billion made available on April 17, 2009. American International Group, Inc., Form 10-Q
for the Fiscal Year Ended September 30, 2010, at 119 (Nov. 5, 2010) (online at sec.gov/Archives/edgar/data/
5272/000104746910009269/a2200724z10-q.htm). This figure does not include $1.6 billion in accumulated but
unpaid dividends owed by AIG to Treasury due to the restructuring of Treasury's investment from cumulative
preferred shares to non-cumulative shares. See U.S. Department of the Treasury, Troubled Asset Relief Program
Transactions Report for the Period Ending November 26, 2010, at 21 (Nov. 30, 2010) (online at
financialstability.gov/docs/transaction-reports/11-30-10%20Transactions%20Report%20as%20of%2011-26-10.pdf).
AIG expects to draw down up to $22.3 billion in unutilized funds from the TARP as part of its plan to repay
the revolving credit facility provided by the Federal Reserve Bank of New York. American International Group,
Inc., AIG Announces Plan to Repay U.S. Government (Sept. 30, 2010) (online at www.aigcorporate.com/newsroom/
2010_September/AIGAnnouncesPlantoRepay30Sept2010.pdf);
vii On May 14, 2010, Treasury accepted a $1.9 billion settlement payment for its $3.5 billion loan to Chrysler
Holding. The payment represented a $1.6 billion loss from the termination of the debt obligation. See U.S.
Department of the Treasury, Chrysler Financial Parent Company Repays $1.9 Billion in Settlement of Original
Chrysler Loan (May 17, 2010) (online at www.financialstability.gov/latest/pr_05172010c.html); U.S. Department
of the Treasury, Troubled Asset Relief Program Transactions Report for the Period Ending November 26, 2010, at
18-19 (Nov. 30, 2010) (online at www.financialstability.gov/docs/transaction-reports/11-30-
10%20Transactions%20Report%20as%20of%2011-26-10.pdf). Also, following the bankruptcy proceedings for Old
Chrysler, which extinguished the $1.9 billion debtor-in-possession (DIP) loan provided to Old Chrysler,
Treasury retained the right to recover the proceeds from the liquidation of specified collateral. Although
Treasury does not expect a significant recovery from the liquidation proceeds, Treasury is not yet reporting
this loan as a loss in the Transaction Report. To date, Treasury has collected $40.2 million in proceeds from
the sale of collateral. Treasury includes these proceeds as part of the $10.8 billion repaid under the AIFP.
U.S. Department of the Treasury, Troubled Assets Relief Program Monthly 105(a) Report--September 2010 (Oct.
12, 2010) (online at financialstability.gov/docs/105CongressionalReports/September 105(a) report_FINAL.pdf);
Treasury conversations with Panel staff (Aug. 19, 2010 and Nov. 29, 2010); U.S. Department of the Treasury,
Troubled Asset Relief Program Transactions Report for the Period Ending November 26, 2010, at 18 (Nov. 30,
2010) (online at financialstability.gov/docs/transaction-reports/11-30-
10%20Transactions%20Report%20as%20of%2011-26-10.pdf).
viii On the TARP Transactions Report, the $1.9 billion Chrysler debtor-in-possession loan, which was
extinguished April 30, 2010, was deducted from Treasury's AIFP investment amount. U.S. Department of the
Treasury, Troubled Asset Relief Program Transactions Report for the Period Ending November 26, 2010, at 18
(Nov. 30, 2010) (online at financialstability.gov/docs/transaction-reports/11-30-
10%20Transactions%20Report%20as%20of%2011-26-10.pdf). See endnote vii, supra, for details on losses from
Treasury's investment in Chrysler.
ix On April 5, 2010, Treasury terminated its commitment to lend to the GM SPV under the ASSP. On April 7, 2010,
it terminated its commitment to lend to the Chrysler SPV. In total, Treasury received $413 million in
repayments from loans provided by this program ($290 million from the GM SPV and $123 million from the
Chrysler SPV). Further, Treasury received $101 million in proceeds from additional notes associated with this
program. U.S. Department of the Treasury, Troubled Asset Relief Program Transactions Report for the Period
Ending November 26, 2010, at 19 (Nov. 30, 2010) (online at financialstability.gov/docs/transaction-reports/11-
30-10%20Transactions%20Report%20as%20of%2011-26-10.pdf).
x For the TALF program, $1 of TARP funds was committed for every $10 of funds obligated by the Federal Reserve.
The program was intended to be a $200 billion initiative, and the TARP was responsible for the first $20
billion in loan-losses, if any were incurred. The loan was incrementally funded. When the program closed in
June 2010, a total of $43 billion in loans was outstanding under the TALF program, and the TARP's commitments
constituted $4.3 billion. The Federal Reserve Board of Governors agreed that it was appropriate for Treasury
to reduce TALF credit protection from TARP to $4.3 billion. Board of Governors of the Federal Reserve System,
Federal Reserve Announces Agreement with the Treasury Department Regarding a Reduction of Credit Protection
Provided for the Term Asset-Backed Securities Loan Facility (TALF) (July 20, 2010) (online at
www.federalreserve.gov/newsevents/press/monetary/20100720a.htm).
xi As of December 1, 2010, Treasury had provided $106 million to TALF LLC. This total is net of accrued interest
payable to Treasury. Board of Governors of the Federal Reserve System, Factors Affecting Reserve Balances
(H.4.1) (Dec. 2, 2010) (online at www.federalreserve.gov/releases/h41/20101202/).
xii As of September 30, 2010, the total value of securities held by the PPIP managers was $19.3 billion. Non-
agency Residential Mortgage-Backed Securities represented 82 percent of the total; Commercial Mortgage-Backed
Securities represented the balance. U.S. Department of the Treasury, Legacy Securities Public-Private
Investment Program, Program Update--Quarter Ended September 30, 2010, at 4 (Oct. 20, 2010) (online at
financialstability.gov/docs/External%20Report%20-%2009-10%20vFinal.pdf).
xiii U.S. Department of the Treasury, Troubled Assets Relief Program Monthly 105(a) Report--October 2010, at 4
(Nov. 10, 2010) (online at www.financialstability.gov/docs/October 105(a) Report.pdf).
xiv As of November 26, 2010, Treasury has received $428 million in capital repayments from two PPIP fund
managers. U.S. Department of the Treasury, Troubled Asset Relief Program Transactions Report for the Period
Ending November 26, 2010, at 23 (Nov. 30, 2010) (online at financialstability.gov/docs/transaction-reports/11-
30-10%20Transactions%20Report%20as%20of%2011-26-10.pdf).
xv As of November 26, 2010, Treasury's purchases under the SBA 7(a) Securities Purchase Program totaled $364.2
million. U.S. Department of the Treasury, Troubled Asset Relief Program Transactions Report for the Period
Ending November 26, 2010, at 22 (Nov. 30, 2010) (online at financialstability.gov/docs/transaction-reports/11-
30-10%20Transactions%20Report%20as%20of%2011-26-10.pdf).
xvi Treasury will not make additional purchases pursuant to the expiration of its purchasing authority under
EESA. U.S. Department of the Treasury, Troubled Asset Relief Program: Two-Year Retrospective, at 43 (Oct.
2010) (online at www.financialstability.gov/docs/
TARP%20Two%20Year%20Retrospective_10%2005%2010_transmittal%20letter.pdf).
xvii On June 23, 2010, $1.5 billion was allocated to mortgage assistance through the Hardest Hit Fund (HHF).
Another $600 million was approved on August 3, 2010. U.S. Department of the Treasury, Obama Administration
Approves State Plans for $600 million of `Hardest Hit Fund' Foreclosure Prevention Assistance (Aug. 3, 2010)
(online at www.financialstability.gov/latest/pr_08042010.html). As part of its revisions to TARP allocations
upon enactment of the Dodd-Frank Wall Street Reform and Consumer Protection Act, Treasury allocated an
additional $2 billion in TARP funds to mortgage assistance for unemployed borrowers through the HHF. U.S.
Department of the Treasury, Obama Administration Announces Additional Support for Targeted Foreclosure-
Prevention Programs to Help Homeowners Struggling with Unemployment (Aug. 11, 2010) (online at
www.financialstability.gov/latest/pr_08112010.html). Another $3.5 billion was allocated among the 18 states
and the District of Columbia currently participating in HHF. The amount each state received during this round
of funding is proportional to its population. U.S. Department of the Treasury, Troubled Asset Relief Program:
Two Year Retrospective, at 72 (Oct. 2010) (online at www.financialstability.gov/docs/
TARP%20Two%20Year%20Retrospective_10%2005%2010_transmittal%20letter.pdf).
xviii As of December 1, 2010, a total of $103.6 million has been disbursed to 12 state Housing Finance Agencies
(HFAs). Data provided by Treasury (Dec. 2, 2010).
xix This figure represents the amount Treasury disbursed to fund the advance purchase account of the Letter of
Credit issued under the FHA Short Refinance Program. The $53.3 million in the FHA Short Refinance program is
broken down as follows: $50 million for a deposit into an advance purchase account as collateral to the
initial $50 million Letter of Credit, $2.9 million for the closing and funding of the Letter of Credit,
$115,000 in trustee fees, $175,000 in claims processor fees, and $156,000 for an unused commitment fee for the
Letter of Credit. Data provided by Treasury (Dec. 2, 2010).
xx U.S. Department of the Treasury, Troubled Asset Relief Program Transactions Report for the Period Ending
November 26, 2010, at 1-13, 16-17 (Nov. 30, 2010) (online at financialstability.gov/docs/transaction-reports/
11-30-10%20Transactions%20Report%20as%20of%2011-26-10.pdf). Treasury closed the program on September 30, 2010,
after investing $570 million in 84 CDFIs. U.S. Department of the Treasury, Treasury Announces Special
Financial Stabilization Initiative Investments of $570 Million in 84 Community Development Financial
Institutions in Underserved Areas (Sept. 30, 2010) (online at financialstability.gov/latest/
pr_09302010b.html).
FIGURE 47: TARP PROFIT AND LOSS
[Dollars in millions]
--------------------------------------------------------------------------------------------------------------------------------------------------------
Warrant
Dividends xxii Interest xxiii Disposition Other Proceeds
TARP Initiative xxi (as of 10/31/ (as of 10/31/ Proceeds xxiv (as of 10/31/ Losses xxv (as Total
2010) 2010) (as of 11/26/ 2010) of 11/26/2010)
2010)
--------------------------------------------------------------------------------------------------------------------------------------------------------
Total...................................................... $16,725 $1,061 $8,160 $5,852 ($6,034) $25,764
CPP........................................................ 9,860 49 6,905 xxvi 3,015 (2,576) 17,252
TIP........................................................ 3,004 -- 1,256 -- -- 4,260
AIFP....................................................... xxvii 3,418 931 -- xxviii 15 (3,458) 906
ASSP....................................................... -- 15 -- xxix 101 -- 116
AGP........................................................ 443 -- -- xxx 2,246 -- 2,689
PPIP....................................................... -- 66 -- xxxi 199 -- 264
SBA 7(a)................................................... -- 1 -- -- -- 1
Bank of America Guarantee.................................. -- -- -- xxxii 276 -- 276
--------------------------------------------------------------------------------------------------------------------------------------------------------
xxi AIG is not listed in this table because no profit or loss has been recorded to date for AIG. Its missed dividends were capitalized as part of the
issuance of Series E preferred shares and are not considered to be outstanding. Treasury currently holds non-cumulative preferred shares, meaning AIG
is not penalized for non-payment. Therefore, no profit or loss has been realized on Treasury's AIG investment to date.
HAMP is not listed in this table because HAMP is a 100% subsidy program and there is no profit expected.
xxii U.S. Department of the Treasury, Cumulative Dividends, Interest and Distributions Report as of October 31, 2010 (Nov. 11, 2010) (online at
financialstability.gov/docs/dividends-interest-reports/October%202010%20Dividends%20&%20Interest%20Report.pdf).
xxiii U.S. Department of the Treasury, Cumulative Dividends, Interest and Distributions Report as of October 31, 2010 (Nov. 11, 2010) (online at
financialstability.gov/docs/dividends-interest-reports/October%202010%20Dividends%20&%20Interest%20Report.pdf).
xxiv U.S. Department of the Treasury, Troubled Asset Relief Program Transactions Report for the Period Ending November 26, 2010 (Nov. 30, 2010) (online
at financialstability.gov/docs/transaction-reports/11-30-10%20Transactions%20Report%20as%20of%2011-26-10.pdf).
xxv In the TARP Transactions Report, Treasury classified the investments it made in two institutions, CIT Group ($2.3 billion) and Pacific Coast
National Bancorp ($4.1 million), as losses. Treasury has also sold its preferred ownership interests and warrants from South Financial Group, Inc. and
TIB Financial Corp. This represents a $241.7 million loss on its CPP investments in these two banks. Two TARP recipients, UCBH Holdings, Inc. ($298.7
million) and a banking subsidiary of Midwest Banc Holdings, Inc. ($89.4 million), are currently in bankruptcy proceedings. As of November 26, three
TARP recipients, Pierce County Bancorp, Sonoma Valley Bancorp, and Tifton Banking Company, had entered receivership. Cumulatively, they had received
$19.3 million in TARP funding. U.S. Department of the Treasury, Troubled Asset Relief Program Transactions Report for the Period Ending November 26,
2010 (Nov. 30, 2010) (online at financialstability.gov/docs/transaction-reports/11-30-10%20Transactions%20Report%20as%20of%2011-26-10.pdf).
xxvi This figure represents net proceeds to Treasury from the sale of Citigroup common stock to date. For details on Treasury's sales of Citigroup
common stock, see endnote ii, supra. U.S. Department of the Treasury, Troubled Asset Relief Program Transactions Report for the Period Ending November
26, 2010, at 15 (Nov. 30, 2010) (online at financialstability.gov/docs/transaction-reports/11-30-10%20Transactions%20Report%20as%20of%2011-26-10.pdf);
U.S. Department of the Treasury, Troubled Asset Relief Program: Two-Year Retrospective, at 25 (Oct. 2010) (online at www.financialstability.gov/docs/
TARP%20Two%20Year%20Retrospective_10%2005%2010_transmittal%20letter.pdf).
xxvii This figure includes $815 million in dividends from Ally preferred stock, trust preferred securities, and mandatory convertible preferred shares.
The dividend total also includes a $748.6 million senior unsecured note from Treasury's investment in General Motors. U.S. Department of the Treasury,
Cumulative Dividends, Interest and Distributions Report as of October 31, 2010 (Nov. 11, 2010) (online at financialstability.gov/docs/dividends-
interest-reports/October%202010%20Dividends%20&%20Interest%20Report.pdf); Data provided by Treasury (May 7, 2010).
xxviii Treasury received proceeds from an additional note connected with the loan made to Chrysler Financial on January 16, 2009. U.S. Department of the
Treasury, Troubled Asset Relief Program Transactions Report for the Period Ending November 26, 2010, at 18 (Nov. 30, 2010) (online at
financialstability.gov/docs/transaction-reports/11-30-10%20Transactions%20Report%20as%20of%2011-26-10.pdf).
xxix This represents the total proceeds from additional notes connected with Treasury's investments in GM Supplier Receivables LLC and Chrysler
Receivables SPV LLC. U.S. Department of the Treasury, Troubled Asset Relief Program Transactions Report for the Period Ending November 26, 2010, at 19
(Nov. 30, 2010) (online at financialstability.gov/docs/transaction-reports/11-30-10%20Transactions%20Report%20as%20of%2011-26-10.pdf).
xxx As a fee for taking a second-loss position of up to $5 billion on a $301 billion pool of ring-fenced Citigroup assets as part of the AGP, Treasury
received $4.03 billion in Citigroup preferred stock and warrants. Treasury exchanged these preferred stocks for trust preferred securities in June
2009. Following the early termination of the guarantee in December 2009, Treasury cancelled $1.8 billion of the trust preferred securities, leaving
Treasury with $2.23 billion in Citigroup trust preferred securities. On September 30, 2010, Treasury sold these securities for $2.25 billion in total
proceeds. At the end of Citigroup's participation in the FDIC's TLGP, the FDIC may transfer $800 million of $3.02 billion in Citigroup Trust Preferred
Securities it received in consideration for its role in the AGP to Treasury. U.S. Department of the Treasury, Troubled Asset Relief Program
Transactions Report for the Period Ending November 26, 2010, at 20 (Nov. 30, 2010) (online at financialstability.gov/docs/transaction-reports/11-30-
10%20Transactions%20Report%20as%20of%2011-26-10.pdf); U.S. Department of the Treasury, Board of Governors of the Federal Reserve System, Federal
Deposit Insurance Corporation, and Citigroup Inc., Termination Agreement, at 1 (Dec. 23, 2009) (online at www.financialstability.gov/docs/
Citi%20AGP%20Termination%20Agreement%20-%20Fully%20Executed%20Version.pdf); U.S. Department of the Treasury, Treasury Announces Further Sales of
Citigroup Securities and Cumulative Return to Taxpayers of $41.6 Billion (Sept. 30, 2010) (online at financialstability.gov/latest/pr_09302010c.html);
Federal Deposit Insurance Corporation, 2009 Annual Report, at 87 (June 30, 2010) (online at www.fdic.gov/about/strategic/report/2009annualreport/
AR09final.pdf).
xxxi As of October 31, 2010, Treasury has earned $264.2 million in membership interest distributions from the PPIP. Additionally, Treasury has earned
$20.6 million in total proceeds following the termination of the TCW fund. See U.S. Department of the Treasury, Cumulative Dividends, Interest and
Distributions Report as of October 31, 2010, at 14 (Nov. 11, 2010) (online at financialstability.gov/docs/dividends-interest-reports/
October%202010%20Dividends%20&%20Interest%20Report.pdf); U.S. Department of the Treasury, Troubled Asset Relief Program Transactions Report for the
Period Ending November 26, 2010, at 23 (Nov. 30, 2010) (online at financialstability.gov/docs/transaction-reports/11-30-
10%20Transactions%20Report%20as%20of%2011-26-10.pdf).
xxxii lthough Treasury, the Federal Reserve, and the FDIC negotiated with Bank of America regarding a similar guarantee, the parties never reached an
agreement. In September 2009, Bank of America agreed to pay each of the prospective guarantors a fee as though the guarantee had been in place during
the negotiations period. This agreement resulted in payments of $276 million to Treasury, $57 million to the Federal Reserve, and $92 million to the
FDIC. U.S. Department of the Treasury, Board of Governors of the Federal Reserve System, Federal Deposit Insurance Corporation, and Bank of America
Corporation, Termination Agreement, at 1-2 (Sept. 21, 2009) (online at www.financialstability.gov/docs/AGP/BofA%20-%20Termination%20Agreement%20-
%20executed.pdf).
d. CPP Unpaid Dividend and Interest Payments \487\
---------------------------------------------------------------------------
\487\ U.S. Department of the Treasury, Cumulative Dividends,
Interest, and Distributions Report as of October 31, 2010, at 20 (Nov.
10, 2010) (online at www.financialstability.gov/docs/dividends-
interest-reports/
October%202010%20Dividends%20&%20Interest%20Report.pdf) (hereinafter
``Cumulative Dividends, Interest, and Distributions Report as of
October 31, 2010'').
---------------------------------------------------------------------------
As of October 31, 2010, 123 institutions have missed at
least one dividend payment on preferred stock issued under CPP
outstanding.\488\ Among these institutions, 97 are not current
on cumulative dividends, amounting to $110.8 million in missed
payments. Another 26 banks have not paid $8 million in non-
cumulative dividends. Of the $49.5 billion currently
outstanding in CPP funding, Treasury's investments in banks
with non-current dividend payments total $3.8 billion. A
majority of the banks that remain delinquent on dividend
payments have under $1 billion in total assets on their balance
sheets. Also, there are 22 institutions that no longer have
outstanding unpaid dividends, after previously deferring their
quarterly payments.\489\
---------------------------------------------------------------------------
\488\ Does not include banks with missed dividend payments that
have either repaid all delinquent dividends, exited TARP, gone into
receivership, or filed for bankruptcy.
\489\ Includes institutions that have either (a) fully repaid their
CPP investment and exited the program or (b) entered bankruptcy or its
subsidiary was placed into receivership. Cumulative Dividends,
Interest, and Distributions Report as of October 31, 2010, supra note
487, at 20.
---------------------------------------------------------------------------
Six banks have failed to make six dividend payments, while
one bank has missed all seven quarterly payments. These
institutions have received a total of $207.1 million in CPP
funding. Under the terms of the CPP, after a bank fails to pay
dividends for six periods, Treasury has the right to elect two
individuals to the company's board of directors.\490\ Figure 48
below provides further details on the distribution and the
number of institutions that have missed dividend payments.
---------------------------------------------------------------------------
\490\ U.S. Department of the Treasury, Frequently Asked Questions
Capital Purchase Program (CPP): Related to Missed Dividend (or
Interest) Payments and Director Nomination (online at
www.financialstability.gov/docs/CPP/CPP%20Directors%20FAQs.pdf)
(accessed Dec. 10, 2010).
---------------------------------------------------------------------------
In addition, eight CPP participants have missed at least
one interest payment, representing $3.6 million in cumulative
unpaid interest payments. Treasury's total investments in these
non-public institutions represent less than $1 billion in CPP
funding.
FIGURE 48: CPP MISSED DIVIDEND PAYMENTS (AS OF OCTOBER 31, 2010) \491\
----------------------------------------------------------------------------------------------------------------
Number of Missed Payments 1 2 3 4 5 6 7 Total
----------------------------------------------------------------------------------------------------------------
Cumulative Dividends
Number of Banks, by asset size.. 30 20 18 16 10 3 0 97
Under $1B................... 22 17 14 11 7 1 0 72
$1B-$10B.................... 6 3 3 5 3 2 0 22
Over $10B................... 2 0 1 0 0 0 0 3
Non-Cumulative Dividends
Number of Banks, by asset size.. 2 5 7 3 5 3 1 26
Under $1B................... 1 5 6 3 5 3 1 24
$1B-$10B.................... 1 0 1 0 0 0 0 2
Over $10B................... 0 0 0 0 0 0 0 0
-------------------------------------------------------------------------------
Total Missed Payments........... ........ ........ ........ ........ ........ ........ ........ 123
----------------------------------------------------------------------------------------------------------------
\491\ Cumulative Dividends, Interest, and Distributions Report as of October 31, 2010, supra note 487, at 17-20.
Data on total bank assets compiled using SNL Financial data service (accessed Nov. 3, 2010).
e. CPP Losses
As of November 26, 2010, Treasury has realized a total of
$2.6 billion in losses from investments in four CPP
participants. CIT Group Inc. and Pacific Coast National Bancorp
have both completed bankruptcy proceedings, and the preferred
stock and warrants issued by the South Financial Group and TIB
Financial Corp. were sold to third-party institutions at a
discount. Excluded from Treasury's total losses are investments
in institutions that have pending receivership or bankruptcy
proceedings, as well as an institution that is currently the
target of an acquisition.\492\ Settlement of these transactions
and proceedings would increase total losses in the CPP to $3.0
billion. Figure 49 below details settled and unsettled
investment losses from CPP participants that have declared
bankruptcy, been placed into receivership, or renegotiated the
terms of their CPP contracts.
---------------------------------------------------------------------------
\492\ Treasury Transactions Report, supra note 481, at 13.
FIGURE 49: CPP SETTLED AND UNSETTLED LOSSES \493\
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Investment Disposition Warrant Disposition Possible Losses/
Institution Investment Amount Amount Amount Dividends & Interest Reduced Exposure Action
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Cadence Financial Corporation........ $44,000,000 $38,000,000 - $2,970,000 $(6,000,000) 10/29/2010: Treasury agreed
to sell preferred stock and
warrants issued by Cadence
Financial to Community
Bancorp LLC for $38 million
plus accrued and unpaid
dividends. Completion of
the sale subject to
fulfillment of certain
closing conditions.
Capital Bank Corporation \494\ 41,279,000 - - 3,457,117 (20,639,500) 11/9/2010: Capital Bank
Corp. is seeking to enter
an agreement with Treasury
pursuant to which the
company will repurchase
outstanding TARP preferred
shares at 50 percent of
liquidation value, plus
accrued unpaid dividends.
The company will use cash
proceeds from its
acquisition by North
American Financial Holdings
Inc. As of Nov. 30, 2010,
no agreement has been
reached between Capital
Bank Corp. and Treasury.
CIT Group Inc.*...................... 2,330,000,000 - - 43,687,500 (2,330,000,000) 12/10/2009: Bankruptcy
reorganization plan for CIT
Group Inc. became
effective. CPP preferred
shares and warrants were
extinguished and replaced
with contingent value
rights (CVR). On Feb. 8,
2010, the CVRs expired
without value.
Midwest Banc Holdings, Inc........... 89,388,000 - - 824,289 (89,388,000) 5/14/2010: Midwest Banc
Holdings, Inc. subsidiary,
Midwest Bank and Trust,
Co., placed into
receivership. Midwest Banc
Holdings is currently in
bankruptcy proceedings.
Pacific Coast National Bancorp.* 4,120,000 - - 18,088 (4,120,000) 2/11/2010: Pacific Coast
National Bancorp dismissed
its bankruptcy proceedings
without recovery to
creditors or investors.
Investments, including
Treasury's CPP investments,
were extinguished.
Pierce County Bancorp................ 6,800,000 - - 207,948 (6,800,000) 11/5/2010: Pierce County
Bancorp subsidiary, Pierce
Commercial Bank, placed
into receivership.
Sonoma Valley Bancorp................ 8,653,000 - - 347,164 (8,653,000) 8/20/2010: Sonoma Valley
Bancorp subsidiary, Sonoma
Valley Bank, placed into
receivership.
South Financial Group*............... 347,000,000 130,179,219 $400,000 16,386,111 (216,820,781) 9/30/2010: Preferred stock
and warrants sold to
Toronto-Dominion Bank.
The Bank of Currituck................ 4,021,000 1,752,850 - 169,834 (2,268,150) 11/5/2010: Treasury agreed
to sell all preferred stock
(including preferred stock
received upon exercise of
warrants) to the Bank of
Currituck.
TIB Financial Corp.*................. 37,000,000 12,119,637 40,000 1,284,722 (24,880,363) 9/30/2010: Preferred stock
and warrants sold to North
American Financial
Holdings.
Tifton Banking Company............... 3,800,000 - - 223,208 (3,800,000) 11/12/2010: Tifton Banking
Company placed into
receivership.
UCBH Holdings, Inc................... 298,737,000 - - 7,509,920 (298,737,000) 11/6/2009: United Commercial
Bank, a wholly-owned
subsidiary of UCBH
Holdings, Inc., was placed
into receivership. UCBH
Holdings is currently in
bankruptcy proceedings.
Total............................ $3,214,798,000 $182,051,706 440,000 77,085,901 $(3,012,106,794) ............................
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
\493\ Treasury Transactions Report, supra note 481, at 14. The asterisk (``*'') denotes recognized losses on Treasury's Transactions Report.
\494\ Capital Bank Corporation, Schedule 14A, at 5 (Nov. 19, 2010) (online at www.sec.gov/Archives/edgar/data/1071992/000095012310107474/g25191ddef14a.htm).
f. Rate of Return
As of December 2, 2010, the average internal rate of return
for all public financial institutions that participated in the
CPP and fully repaid the U.S. government (including preferred
shares, dividends, and warrants) remained at 8.4 percent, as no
institutions exited the program in November.\495\ The internal
rate of return is the annualized effective compounded return
rate that can be earned on invested capital.
---------------------------------------------------------------------------
\495\ Calculation of the internal rate of return (IRR) also
includes CPP investments in public institutions not repaid in full (for
reasons such as acquisition by another institution) in the Transaction
Report, e.g., The South Financial Group and TIB Financial Corporation.
The Panel's total IRR calculation now includes CPP investments in
public institutions recorded as a loss on the TARP Transaction Report
due to bankruptcy, e.g., CIT Group Inc. Going forward, the Panel will
continue to include losses due to bankruptcy when Treasury determines
that any associated contingent value rights have expired without value.
When excluding CIT Group from the calculation, the resulting IRR is
10.4 percent. Treasury Transactions Report, supra note 481.
---------------------------------------------------------------------------
g. Warrant Disposition
FIGURE 50: WARRANT REPURCHASES/AUCTIONS FOR FINANCIAL INSTITUTIONS WHO HAVE FULLY REPAID CPP FUNDS (AS OF DECEMBER 2, 2010)
--------------------------------------------------------------------------------------------------------------------------------------------------------
Panel's Best
Warrant Warrant Valuation Price/
Institution Investment Repurchase Repurchase/ Sale Estimate at Estimate IRR
Date Date Amount Disposition Ratio (Percent)
Date
--------------------------------------------------------------------------------------------------------------------------------------------------------
Old National Bancorp.............................................. 12/12/2008 5/8/2009 $1,200,000 $2,150,000 0.558 9.3
Iberiabank Corporation............................................ 12/5/2008 5/20/2009 1,200,000 2,010,000 0.597 9.4
Firstmerit Corporation............................................ 1/9/2009 5/27/2009 5,025,000 4,260,000 1.180 20.3
Sun Bancorp, Inc.................................................. 1/9/2009 5/27/2009 2,100,000 5,580,000 0.376 15.3
Independent Bank Corp............................................. 1/9/2009 5/27/2009 2,200,000 3,870,000 0.568 15.6
Alliance Financial Corporation.................................... 12/19/2008 6/17/2009 900,000 1,580,000 0.570 13.8
First Niagara Financial Group..................................... 11/21/2008 6/24/2009 2,700,000 3,050,000 0.885 8.0
Berkshire Hills Bancorp, Inc...................................... 12/19/2008 6/24/2009 1,040,000 1,620,000 0.642 11.3
Somerset Hills Bancorp............................................ 1/16/2009 6/24/2009 275,000 580,000 0.474 16.6
SCBT Financial Corporation........................................ 1/16/2009 6/24/2009 1,400,000 2,290,000 0.611 11.7
HF Financial Corp................................................. 11/21/2008 6/30/2009 650,000 1,240,000 0.524 10.1
State Street...................................................... 10/28/2008 7/8/2009 60,000,000 54,200,000 1.107 9.9
U.S. Bancorp...................................................... 11/14/2008 7/15/2009 139,000,000 135,100,000 1.029 8.7
The Goldman Sachs Group, Inc...................................... 10/28/2008 7/22/2009 1,100,000,000 1,128,400,000 0.975 22.8
BB&T Corp......................................................... 11/14/2008 7/22/2009 67,010,402 68,200,000 0.983 8.7
American Express Company.......................................... 1/9/2009 7/29/2009 340,000,000 391,200,000 0.869 29.5
Bank of New York Mellon Corp...................................... 10/28/2008 8/5/2009 136,000,000 155,700,000 0.873 12.3
Morgan Stanley.................................................... 10/28/2008 8/12/2009 950,000,000 1,039,800,000 0.914 20.2
Northern Trust Corporation........................................ 11/14/2008 8/26/2009 87,000,000 89,800,000 0.969 14.5
Old Line Bancshares Inc........................................... 12/5/2008 9/2/2009 225,000 500,000 0.450 10.4
Bancorp Rhode Island, Inc......................................... 12/19/2008 9/30/2009 1,400,000 1,400,000 1.000 12.6
Centerstate Banks of Florida Inc.................................. 11/21/2008 10/28/2009 212,000 220,000 0.964 5.9
Manhattan Bancorp................................................. 12/5/2008 10/14/2009 63,364 140,000 0.453 9.8
CVB Financial Corp................................................ 12/5/2008 10/28/2009 1,307,000 3,522,198 0.371 6.4
Bank of the Ozarks................................................ 12/12/2008 11/24/2009 2,650,000 3,500,000 0.757 9.0
Capital One Financial............................................. 11/14/2008 12/3/2009 148,731,030 232,000,000 0.641 12.0
JPMorgan Chase & Co............................................... 10/28/2008 12/10/2009 950,318,243 1,006,587,697 0.944 10.9
CIT Group Inc..................................................... 12/31/2008 -- -- -- -- (97.2)
TCF Financial Corp................................................ 1/16/2009 12/16/2009 9,599,964 11,825,830 0.812 11.0
LSB Corporation................................................... 12/12/2008 12/16/2009 560,000 535,202 1.046 9.0
Wainwright Bank & Trust Company................................... 12/19/2008 12/16/2009 568,700 1,071,494 0.531 7.8
Wesbanco Bank, Inc................................................ 12/5/2008 12/23/2009 950,000 2,387,617 0.398 6.7
Union First Market Bankshares Corporation (Union Bankshares 12/19/2008 12/23/2009 450,000 1,130,418 0.398 5.8
Corporation).....................................................
Trustmark Corporation............................................. 11/21/2008 12/30/2009 10,000,000 11,573,699 0.864 9.4
Flushing Financial Corporation.................................... 12/19/2008 12/30/2009 900,000 2,861,919 0.314 6.5
OceanFirst Financial Corporation.................................. 1/16/2009 2/3/2010 430,797 279,359 1.542 6.2
Monarch Financial Holdings, Inc................................... 12/19/2008 2/10/2010 260,000 623,434 0.417 6.7
\496\ 10/28/
2008
\497\1/9/2009
Bank of America................................................... \498\ 1/14/ 3/3/2010 1,566,210,714 1,006,416,684 1.533 6.5
2009
Washington Federal Inc./Washington Federal Savings & Loan 11/14/2008 3/9/2010 15,623,222 10,166,404 1.537 18.6
Association......................................................
Signature Bank.................................................... 12/12/2008 3/10/2010 11,320,751 11,458,577 0.988 32.4
Texas Capital Bancshares, Inc..................................... 1/16/2009 3/11/2010 6,709,061 8,316,604 0.807 30.1
Umpqua Holdings Corp.............................................. 11/14/2008 3/31/2010 4,500,000 5,162,400 0.872 6.6
City National Corporation......................................... 11/21/2008 4/7/2010 18,500,000 24,376,448 0.759 8.5
First Litchfield Financial Corporation............................ 12/12/2008 4/7/2010 1,488,046 1,863,158 0.799 15.9
PNC Financial Services Group Inc.................................. 12/31/2008 4/29/2010 324,195,686 346,800,388 0.935 8.7
Comerica Inc...................................................... 11/14/2008 5/4/2010 183,673,472 276,426,071 0.664 10.8
Valley National Bancorp........................................... 11/14/2008 5/18/2010 5,571,592 5,955,884 0.935 8.3
Wells Fargo Bank.................................................. 10/28/2008 5/20/2010 849,014,998 1,064,247,725 0.798 7.8
First Financial Bancorp........................................... 12/23/2008 6/2/2010 3,116,284 3,051,431 1.021 8.2
Sterling Bancshares, Inc./Sterling Bank........................... 12/12/2008 6/9/2010 3,007,891 5,287,665 0.569 10.8
SVB Financial Group............................................... 12/12/2008 6/16/2010 6,820,000 7,884,633 0.865 7.7
Discover Financial Services....................................... 3/13/2009 7/7/2010 172,000,000 166,182,652 1.035 17.1
Bar Harbor Bancshares............................................. 1/16/2009 7/28/2010 250,000 518,511 0.482 6.2
Citizens & Northern Corporation................................... 1/16/2009 8/4/2010 400,000 468,164 0.854 5.9
Columbia Banking System, Inc...................................... 11/21/2008 8/11/2010 3,301,647 3,291,329 1.003 7.3
Hartford Financial Services Group, Inc............................ 6/26/2009 9/21/2010 713,687,430 472,221,996 1.511 30.3
Lincoln National Corporation...................................... 7/10/2009 9/16/2010 216,620,887 181,431,183 1.194 27.1
Fulton Financial Corporation...................................... 12/23/2008 9/8/2010 10,800,000 15,616,013 0.692 6.7
The Bancorp, Inc./The Bancorp Bank................................ 12/12/2008 9/8/2010 4,753,985 9,947,683 0.478 12.8
South Financial Group, Inc./Carolina First Bank................... 12/5/2008 9/30/2010 400,000 1,164,486 0.343 (34.2)
TIB Financial Corp/TIB Bank....................................... 12/5/2008 9/30/2010 40,000 235,757 0.170 (38.0)
Total............................................................. .............. ........... $8,148,332,166 $7,999,280,713 1.019 8.4
--------------------------------------------------------------------------------------------------------------------------------------------------------
\496\ Investment date for Bank of America in CPP.
\497\ Investment date for Merrill Lynch in CPP.
\498\ Investment date for Bank of America in TIP.
FIGURE 51: VALUATION OF CURRENT HOLDINGS OF WARRANTS (AS OF DECEMBER 2, 2010)
[Dollars in millions]
----------------------------------------------------------------------------------------------------------------
Warrant Valuation
Financial Institutions with Warrants Outstanding -----------------------------------------------
Low Estimate High Estimate Best Estimate
----------------------------------------------------------------------------------------------------------------
Citigroup, Inc.\499\............................................ $74.68 $1,383.79 $229.25
SunTrust Banks, Inc............................................. 13.52 322.84 103.76
Regions Financial Corporation................................... 4.61 157.75 87.58
Fifth Third Bancorp............................................. 92.59 373.07 177.87
KeyCorp......................................................... 20.60 152.96 68.87
AIG............................................................. 401.00 1,977.51 744.11
All Other Banks................................................. 555.16 1,868.68 1,102.97
Total........................................................... $1,162.16 $6,236.60 $2,514.41
----------------------------------------------------------------------------------------------------------------
\499\ Includes warrants issued under CPP, AGP, and TIP.
2. Federal Financial Stability Efforts
a. Federal Reserve and FDIC Programs
In addition to the direct expenditures Treasury has
undertaken through the TARP, the federal government has engaged
in a much broader program directed at stabilizing the U.S.
financial system. Many of these initiatives explicitly augment
funds allocated by Treasury under specific TARP initiatives,
such as FDIC and Federal Reserve asset guarantees for
Citigroup, or operate in tandem with Treasury programs. Other
programs, like the Federal Reserve's extension of credit
through its Section 13(3) facilities and special purpose
vehicles (SPVs) and the FDIC's Temporary Liquidity Guarantee
Program (TLGP), operate independently of the TARP.
b. Total Financial Stability Resources
Beginning in its April 2009 report, the Panel broadly
classified the resources that the federal government has
devoted to stabilizing the economy through myriad new programs
and initiatives such as outlays, loans, or guarantees. With the
reductions in funding for certain TARP programs, the Panel
calculates the total value of these resources to be over $2.5
trillion. However, this would translate into the ultimate
``cost'' of the stabilization effort only if: (1) assets do not
appreciate; (2) no dividends are received, no warrants are
exercised, and no TARP funds are repaid; (3) all loans default
and are written off; and (4) all guarantees are exercised and
subsequently written off.
With respect to the FDIC and Federal Reserve programs, the
risk of loss varies significantly across the programs
considered here, as do the mechanisms providing protection for
the taxpayer against such risk. As discussed in the Panel's
November 2009 report, the FDIC assesses a premium of up to 100
basis points on TLGP debt guarantees.\500\ In contrast, the
Federal Reserve's liquidity programs are generally available
only to borrowers with good credit, and the loans are over-
collateralized and with recourse to other assets of the
borrower. If the assets securing a Federal Reserve loan realize
a decline in value greater than the ``haircut,'' the Federal
Reserve is able to demand more collateral from the borrower.
Similarly, should a borrower default on a recourse loan, the
Federal Reserve can turn to the borrower's other assets to make
the Federal Reserve whole. In this way, the risk to the
taxpayer on recourse loans only materializes if the borrower
enters bankruptcy.
---------------------------------------------------------------------------
\500\ Congressional Oversight Panel, November Oversight Report:
Guarantees and Contingent Payments in TARP and Related Programs, at 36
(Nov. 6, 2009) (online at cop.senate.gov/
documents/cop-110609-report.pdf).
---------------------------------------------------------------------------
c. Mortgage Purchase Programs
On September 7, 2008, Treasury announced the GSE Mortgage
Backed Securities Purchase Program. The Housing and Economic
Recovery Act of 2008 provided Treasury with the authority to
purchase MBS guaranteed by GSEs through December 31, 2009.
Treasury purchased approximately $225 billion in GSE MBS by the
time its authority expired.\501\ As of November 2010, there was
approximately $149.7 billion in MBS still outstanding under
this program.\502\
---------------------------------------------------------------------------
\501\ U.S. Department of the Treasury, FY2011 Budget in Brief, at
138 (Feb. 2010) (online at www.treasury.gov/about/budget-performance/
budget-in-brief/Documents/FY%202011%20BIB% 20(2).pdf).
\502\ U.S. Department of the Treasury, MBS Purchase Program:
Portfolio by Month (online at www.financialstability.gov/docs/
November%202010%20Portfolio%20by%20month.pdf) (accessed Dec. 3, 2010).
Treasury has received $65.7 billion in principal repayments and $14.3
billion in interest payments from these securities. See U.S. Department
of the Treasury, MBS Purchase Program Principal and Interest Received
(online at www.financialstability.gov/docs/
November%202010%20MBS%20Principal%20and%20Interest%20Monthly%20Breakout.
pdf) (accessed Dec. 3, 2010).
---------------------------------------------------------------------------
In March 2009, the Federal Reserve authorized purchases of
$1.25 trillion MBS guaranteed by Fannie Mae, Freddie Mac, and
Ginnie Mae, and $200 billion of agency debt securities from
Fannie Mae, Freddie Mac, and the Federal Home Loan Banks.\503\
The intended purchase amount for agency debt securities was
subsequently decreased to $175 billion.\504\ All purchasing
activity was completed on March 31, 2010. As of December 1,
2010, the Federal Reserve held $1.02 trillion of agency MBS and
$148 billion of agency debt.\505\
---------------------------------------------------------------------------
\503\ Board of Governors of the Federal Reserve System, Federal
Reserve System Monthly Report on Credit and Liquidity Programs and the
Balance Sheet, at 5 (Nov. 2010) (online at federalreserve.gov/
monetarypolicy/files/monthlyclbsreport201011.pdf).
\504\ Id. at 5.
\505\ Board of Governors of the Federal Reserve System, Factors
Affecting Reserve Balances (H.4.1) (Dec. 2, 2010) (online at
www.federalreserve.gov/releases/h41/20101202/) (hereinafter ``Factors
Affecting Reserve Balances (H.4.1)'').
---------------------------------------------------------------------------
d. Federal Reserve Treasury Securities Purchases \506\
---------------------------------------------------------------------------
\506\ Board of Governors of the Federal Reserve System, Press
Release--FOMC Statement (Nov. 3, 2010) (online at
www.federalreserve.gov/newsevents/press/monetary/20101103a.htm);
Federal Reserve Bank of New York, Statement Regarding Purchases of
Treasury Securities (Nov. 3, 2010) (online at www.federalreserve.gov/
newsevents/press/monetary/monetary20101103a1.pdf).
---------------------------------------------------------------------------
On November 3, 2010, the Federal Open Market Committee
(FOMC) announced that it has directed FRBNY to begin purchasing
an additional $600 billion in longer-term Treasury securities.
In addition, FRBNY will reinvest $250 billion to $300 billion
in principal payments from agency debt and agency MBS in
Treasury securities.\507\ The additional purchases and
reinvestments will be conducted through the end of the second
quarter of 2011, meaning the pace of purchases will be
approximately $110 billion per month. In order to facilitate
these purchases, FRBNY will temporarily lift its System Open
Market Account per-issue limit, which prohibits the Federal
Reserve's holdings of an individual security from surpassing 35
percent of the outstanding amount.\508\ As of December 1, 2010,
the Federal Reserve held $917 billion in Treasury
securities.\509\
---------------------------------------------------------------------------
\507\ On August 10, 2010, the Federal Reserve began reinvesting
principal payments on agency debt and agency MBS holdings in longer-
term Treasury securities in order to keep the amount of their
securities holdings in their System Open Market Account portfolio at
their then-current level. Board of Governors of the Federal Reserve
System, FOMC Statement (Aug. 10, 2010) (online at
www.federalreserve.gov/newsevents/press/monetary/20100810a.htm).
\508\ Federal Reserve Bank of New York, FAQs: Purchases of Longer-
term Treasury Securities (Nov. 3, 2010) (online at www.newyorkfed.org/
markets/lttreas_faq.html).
\509\ Factors Affecting Reserve Balances (H.4.1), supra note 505.
FIGURE 52: FEDERAL GOVERNMENT FINANCIAL STABILITY EFFORT (AS OF DECEMBER 1, 2010) xxxiii
[Dollars in billions]
----------------------------------------------------------------------------------------------------------------
Program Treasury (TARP) Federal Reserve FDIC Total
----------------------------------------------------------------------------------------------------------------
Total............................... $475 $1,345.3 $690.9 $2,511.2
Outlays xxxiv................... 218.7 1,196.9 188.9 1,604.6
Loans........................... 23.4 148.4 0 171.8
Guarantees xxxv................. 4.3 0 502 506.3
Repaid and Unavailable TARP 228.6 0 0 228.6
Funds..........................
AIG xxxvi........................... 69.8 82.6 0 152.4
Outlays......................... xxxvii 69.8 xxxviii 26.1 0 95.9
Loans........................... 0 xxxix 56.5 0 56.5
Guarantees...................... 0 0 0 0
Citigroup........................... 11.6 0 0 11.6
Outlays......................... xl 11.6 0 0 11.6
Loans........................... 0 0 0 0
Guarantees...................... 0 0 0 0
Capital Purchase Program (Other).... 37.8 0 0 37.8
Outlays......................... xli 37.8 0 0 37.8
Loans........................... 0 0 0 0
Guarantees...................... 0 0 0 0
Capital Assistance Program.......... N/A 0 0 xlii N/A
TALF................................ 4.3 38.7 0 43.0
Outlays......................... 0 0 0 0
Loans........................... 0 xliv 38.7 0 38.7
Guarantees...................... xliii 4.3 0 0 4.3
PPIP (Loans) xlv.................... 0 0 0 0
Outlays......................... 0 0 0 0
Loans........................... 0 0 0 0
Guarantees...................... 0 0 0 0
PPIP (Securities)................... xlvi 22.4 0 0 22.4
Outlays......................... 7.5 0 0 7.5
Loans........................... 14.9 0 0 14.9
Guarantees...................... 0 0 0 0
Making Home Affordable Program/ 45.6 0 0 45.6
Foreclosure Mitigation.............
Outlays......................... xlvii 45.6 0 0 45.6
Loans........................... 0 0 0 0
Guarantees...................... 0 0 0 0
Automotive Industry Financing xlvii 53.6 0 0 53.6
Program............................
Outlays......................... 45.5 0 0 45.5
Loans........................... 8.1 0 0 8.1
Guarantees...................... 0 0 0 0
Automotive Supplier Support Program. 0.4 0 0 0.4
Outlays......................... 0 0 0 0
Loans........................... xlix 0.4 0 0 0.4
Guarantees...................... 0 0 0 0
SBA 7(a) Securities Purchase........ l 0.36 0 0 0.36
Outlays......................... 0.36 0 0 0.36
Loans........................... 0 0 0 0
Guarantees...................... 0 0 0 0
Community Development Capital li 0.57 0 0 0.57
Initiative.........................
Outlays......................... 0 0 0 0
Loans........................... 0.57 0 0 0.57
Guarantees...................... 0 0 0 0
Temporary Liquidity Guarantee 0 0 502.0 502.0
Program............................
Outlays......................... 0 0 0 0
Loans........................... 0 0 0 0
Guarantees...................... 0 0 lii 502.0 502.0
Deposit Insurance Fund.............. 0 0 188.9 188.9
Outlays......................... 0 0 liii 188.9 188.9
Loans........................... 0 0 0 0
Guarantees...................... 0 0 0 0
Other Federal Reserve Credit 0 1,224.0 0 1,224.0
Expansion..........................
Outlays......................... 0 liv 1,170.8 0 1,170.8
Loans........................... 0 lv 53.2 0 53.2
Guarantees...................... 0 0 0 0
----------------------------------------------------------------------------------------------------------------
xxxiii Unless otherwise noted, all data in this figure are as of November 26, 2010.
xxxiv The term ``outlays'' is used here to describe the use of Treasury funds under the TARP, which are broadly
classifiable as purchases of debt or equity securities (e.g., debentures, preferred stock, exercised warrants,
etc.). These values were calculated using (1) Treasury's actual reported expenditures, and (2) Treasury's
anticipated funding levels as estimated by a variety of sources, including Treasury statements and GAO
estimates. Anticipated funding levels are set at Treasury's discretion, have changed from initial
announcements, and are subject to further change. Outlays used here represent investment and asset purchases--
as well as commitments to make investments and asset purchases--and are not the same as budget outlays, which
under section 123 of EESA are recorded on a ``credit reform'' basis.
xxxv Although many of the guarantees may never be exercised or will be exercised only partially, the guarantee
figures included here represent the federal government's greatest possible financial exposure.
xxxvi U.S. Department of the Treasury, Treasury Update on AIG Investment Valuation (Nov. 1, 2010) (online at
financialstability.gov/latest/pr_11012010.html). AIG values exclude accrued dividends on preferred interests
in the AIA and ALICO SPVs and accrued interest payable to FRBNY on the Maiden Lane LLCs.
xxxvii This number includes investments under the AIGIP/SSFI Program: a $40 billion investment made on November
25, 2008, and a $30 billion investment made on April 17, 2009 (less a reduction of $165 million representing
bonuses paid to AIG Financial Products employees). As of November 1, 2010, AIG had utilized $47.5 billion of
the available $69.8 billion under the AIGIP/SSFI. U.S. Department of the Treasury, Treasury Update on AIG
Investment Valuation (Nov. 1, 2010) (online at www.financialstability.gov/latest/pr_11012010.html); U.S.
Department of the Treasury, Troubled Asset Relief Program Transactions Report for the Period Ending November
26, 2010, at 13 (Nov. 30, 2010) (online at financialstability.gov/docs/ transactionreports/11-30-
10%20Transactions %20Report%20as%20of%2011-26-10.pdf).
xxxviii As part of the restructuring of the U.S. government's investment in AIG announced on March 2, 2009, the
amount available to AIG through the Revolving Credit Facility was reduced by $25 billion in exchange for
preferred equity interests in two special purpose vehicles, AIA Aurora LLC and ALICO Holdings LLC. Board of
Governors of the Federal Reserve System, Federal Reserve System Monthly Report on Credit and Liquidity
Programs and the Balance Sheet, at 18 (Nov. 2010) (online at www.federalreserve.gov/monetarypolicy/ files/
monthlyclbsreport201011.pdf). These SPVs were established to hold the common stock of two AIG subsidiaries:
American International Assurance Company Ltd. (AIA) and American Life Insurance Company (ALICO). As of
December 1, 2010, the book value of the Federal Reserve Bank of New York's holdings in AIA Aurora LLC and
ALICO Holdings LLC was $26.1 billion in preferred equity ($16.7 billion in AIA and $9.4 billion in ALICO).
Federal Reserve Bank of New York, Factors Affecting Reserve Balances (H.4.1) (Dec. 2, 2010) (online at
www.federalreserve.gov/releases/h41/20101202/).
xxxix This number represents the full $29.2 billion made available to AIG through its Revolving Credit Facility
(RCF) with FRBNY ($21.3 billion had been drawn down as of December 1, 2010) and the outstanding principal of
the loans extended to the Maiden Lane II and III SPVs to buy AIG assets (as of December 1, 2010, $13.3 billion
and $13.9 billion, respectively). Federal Reserve Bank of New York, Factors Affecting Reserve Balances (H.4.1)
(Dec. 2, 2010) (online at www.federalreserve.gov/releases/ h41/20101202/); Board of Governors of the Federal
Reserve System, Federal Reserve System Monthly Report on Credit and Liquidity Programs and the Balance Sheet
(Nov. 2010) (online at www.federalreserve.gov/monetarypolicy/ files/monthlyclbsreport201011.pdf). The amounts
outstanding under the Maiden Lane II and III facilities do not reflect the accrued interest payable to FRBNY.
Income from the purchased assets is used to pay down the loans to the SPVs, reducing the taxpayers' exposure
to losses over time. Board of Governors of the Federal Reserve System, Federal Reserve System Monthly Report
on Credit and Liquidity Programs and the Balance Sheet, at 15 (Nov. 2010) (online at www.federalreserve.gov/
monetarypolicy/ files/monthlyclbsreport201011.pdf).
The maximum amount available through the RCF decreased from $34.4 billion to $29.3 billion between March and
September 2010, as a result of the sale of several subsidiaries. The reduced ceiling also reflects a $3.95
billion repayment to the RCF from proceeds earned from a debt offering by the International Lease Finance
Corporation (ILFC), an AIG subsidiary. The balance on the AIG Revolving Credit Facility increased $0.3 billion
between September 29 and October 27, 2010, primarily due to recapitalized interest and fees as principal
repayments. Board of Governors of the Federal Reserve System, Federal Reserve System Monthly Report on Credit
and Liquidity Programs and the Balance Sheet, at 17, 19-20 (Nov. 2010) (online at www.federalreserve.gov/
monetarypolicy/files/monthlyclbsreport201011.pdf).
xl This figure represents Treasury's $25 billion investment in Citigroup, minus $13.4 billion applied as a
repayment for CPP funding. The amount repaid comes from the $16.4 billion in gross proceeds Treasury received
from the sale of 4.1 billion Citigroup common shares. See endnote ii, supra, for further details of the sales
of Citigroup common stock to date. U.S. Department of the Treasury, Troubled Asset Relief Program Transactions
Report for the Period Ending November 26, 2010, at 1, 13 (Nov. 30, 2010) (online at financialstability.gov/
docs/ transaction-reports/11-30-10%20 Transactions%20Report%20as%20of%2011-26-10.pdf).
xli This figure represents the $204.9 billion Treasury disbursed under the CPP, minus the $25 billion investment
in Citigroup identified above, $139.5 billion in repayments (excluding the amount repaid for the Citigroup
investment) that are in ``repaid and unavailable'' TARP funds, and losses under the program. This figure does
not account for future repayments of CPP investments and dividend payments from CPP investments. U.S.
Department of the Treasury, Troubled Asset Relief Program Transactions Report for the Period Ending November
26, 2010, at 13 (Nov. 30, 2010) (online at financialstability.gov/docs/transaction-reports/11-30-
10%20Transactions%20Report%20as%20of%2011-26-10.pdf).
xlii On November 9, 2009, Treasury announced the closing of the CAP and that only one institution, GMAC (now
Ally Financial), was in need of further capital from Treasury. GMAC, however, received further funding through
the AIFP. Therefore, the Panel considers CAP unused. U.S. Department of the Treasury, Treasury Announcement
Regarding the Capital Assistance Program (Nov. 9, 2009) (online at www.financialstability.gov/latest/
tg_11092009.html).
xliii This figure represents the $4.3 billion adjusted allocation to the TALF SPV. However, as of October 27,
2010, TALF LLC had drawn only $105 million of the available $4.3 billion. Board of Governors of the Federal
Reserve System, Factors Affecting Reserve Balances (H.4.1) (Oct. 28, 2010) (online at www.federalreserve.gov/
releases/h41/20101028/); U.S. Department of the Treasury, Troubled Asset Relief Program Transactions Report
for the Period Ending November 26, 2010, at 21 (Nov. 30, 2010) (online at financialstability.gov/docs/
transaction- reports/11-30-10%20Transactions%20Report%20as%20of%2011-26-10.pdf). On June 30, 2010, the Federal
Reserve ceased issuing loans collateralized by newly issued CMBS. As of this date, investors had requested a
total of $73.3 billion in TALF loans ($13.2 billion in CMBS and $60.1 billion in non-CMBS) and $71 billion in
TALF loans had been settled ($12 billion in CMBS and $59 billion in non-CMBS). Earlier, it ended its issues of
loans collateralized by other TALF-eligible newly issued and legacy ABS (non-CMBS) on March 31, 2010. Federal
Reserve Bank of New York, Term Asset-Backed Securities Loan Facility: Terms and Conditions (online at
www.newyorkfed.org/markets/talf_terms.html) (accessed Dec. 10, 2010); Federal Reserve Bank of New York, Term
Asset-Backed Securities Loan Facility: CMBS (online at www.newyorkfed.org/markets/cmbs_operations.html)
(accessed Dec. 10, 2010); Federal Reserve Bank of New York, Term Asset-Backed Securities Loan Facility: CMBS
(online at www.newyorkfed.org/markets/CMBS_recent_operations.html) (accessed Dec. 10, 2010); Federal Reserve
Bank of New York, Term Asset-Backed Securities Loan Facility: non-CMBS (online at www.newyorkfed.org/markets/
talf_operations.html) (accessed Dec. 10, 2010); Federal Reserve Bank of New York, Term Asset-Backed Securities
Loan Facility: non-CMBS (online at www.newyorkfed.org/markets/TALF_recent_operations.html) (accessed Dec. 10,
2010).
xliv This number is derived from the unofficial 1:10 ratio of the value of Treasury loan guarantees to the value
of Federal Reserve loans under the TALF. U.S. Department of the Treasury, Fact Sheet: Financial Stability
Plan, at 4 (Feb.10, 2009) (online at financialstability.gov/docs/fact-sheet.pdf) (describing the initial $20
billion Treasury contribution tied to $200 billion in Federal Reserve loans and announcing potential expansion
to a $100 billion Treasury contribution tied to $1 trillion in Federal Reserve loans). Since only $43 billion
in TALF loans remained outstanding when the program closed, Treasury is currently responsible for reimbursing
the Federal Reserve Board only up to $4.3 billion in losses from these loans. Thus, the Federal Reserve's
maximum potential exposure under the TALF is $38.7 billion. See Board of Governors of the Federal Reserve
System, Federal Reserve Announces Agreement with Treasury Regarding Reduction of Credit Protection Provided
for the Term Asset-Backed Securities Loan Facility (TALF) (July 20, 2010) (online at www.federalreserve.gov/
newsevents/ press/monetary/20100720a.htm); Board of Governors of the Federal Reserve System, Factors Affecting
Reserve Balances (H.4.1) (Oct. 28, 2010) (online at www.federalreserve.gov/releases/h41/20101028/).
xlv No TARP resources were expended under the PPIP Legacy Loans Program, a TARP program that was announced in
March 2009 but never launched. Since no TARP funds were allocated for the program by the time the TARP expired
in October 2010, this or a similar program cannot be implemented unless another source of funding is
available.
xlvi This figure represents Treasury's final adjusted investment amount in the Legacy Securities Public-Private
Investment Program (PPIP). As of November 26, 2010, Treasury reported commitments of $14.9 billion in loans
and $7.5 billion in membership interest associated with PPIP. See U.S. Department of the Treasury, Troubled
Asset Relief Program Transactions Report for the Period Ending November 26, 2010, at 23 (Nov. 30, 2010)
(online at financialstability.gov/docs/transaction-reports/11-30-10%20Transactions%20Report%20as%20of%2011-26-
10.pdf). On January 4, 2010, Treasury and one of the nine fund managers, UST/TCW Senior Mortgage Securities
Fund, L.P. (TCW), entered into a ``Winding-Up and Liquidation Agreement.'' U.S. Department of the Treasury,
Winding Up and Liquidation Agreement Between the United States Department of the Treasury and UST/TCW Senior
Mortgage Securities Fund, L.P. (Jan. 4, 2010) (online at financialstability.gov/docs/
TCW%20Winding%20Up%20Agmt%20(Execution%20Copy)%20Redacted.pdf). Treasury's final investment amount in TCW
totaled $356 million. Following the liquidation of the fund, Treasury's initial $3.3 billion obligation to TCW
was reallocated among the eight remaining funds on March 22, 2010. See U.S. Department of the Treasury,
Troubled Asset Relief Program Transactions Report for the Period Ending November 26, 2010, at 23 (Nov. 30,
2010) (online at financialstability.gov/docs/transaction-reports/11-30-
10%20Transactions%20Report%20as%20of%2011-26-10.pdf).
On October 20, 2010, Treasury released its fourth quarterly report on PPIP. The report indicates that as of
September 30, 2010, all eight investment funds have realized an internal rate of return since inception (net
of any management fees or expenses owed to Treasury) above 19 percent. The highest performing fund, thus far,
is AG GECC PPIF Master Fund, L.P., which has a net internal rate of return of 52 percent. U.S. Department of
the Treasury, Legacy Securities Public-Private Investment Program, at 7 (Oct. 20, 2010) (online at
financialstability.gov/docs/External%20Report%20-%2009-10%20vFinal.pdf).
xlvii As of November 26, 2010, the total cap for HAMP was $29.9 billion. The total amount of TARP funds
committed to HAMP is $29.9 billion. However, as of December 2, 2010, only $728.9 million in non-GSE payments
have been disbursed under HAMP. U.S. Department of the Treasury, Troubled Asset Relief Program Transactions
Report for the Period Ending November 26, 2010, at 45 (Nov. 30, 2010) (online at financialstability.gov/docs/
transaction-reports/11-30-10%20Transactions%20Report%20as%20of%2011-26-10.pdf); U.S. Department of the
Treasury, Troubled Assets Relief Program Monthly 105(a) Report--October 2010, at 4 (Nov. 10, 2010) (online at
financialstability.gov/docs/October%20105(a)%20Report.pdf); Data provided by Treasury (Dec. 3, 2010).
xlviii A substantial portion of the total $81.3 billion in debt instruments extended under the AIFP has since
been converted to common equity and preferred shares in restructured companies. $8.1 billion has been retained
as first-lien debt (with $1 billion committed to Old GM and $7.1 billion to Chrysler). This figure ($53.6
billion) represents Treasury's current obligation under the AIFP after repayments and losses. U.S. Department
of the Treasury, Troubled Asset Relief Program Transactions Report for the Period Ending November 26, 2010, at
18 (Nov. 30, 2010) (online at financialstability.gov/docs/transaction-reports/11-30-
10%20Transactions%20Report%20as%20of%2011-26-10.pdf).
xlix This figure represents Treasury's total adjusted investment amount in the ASSP. U.S. Department of the
Treasury, Troubled Asset Relief Program Transactions Report for the Period Ending November 26, 2010, at 19
(Nov. 30, 2010) (online at financialstability.gov/docs/transaction-reports/11-30-
10%20Transactions%20Report%20as%20of%2011-26-10.pdf).
l U.S. Department of the Treasury, Troubled Asset Relief Program: Two Year Retrospective, at 43 (Oct. 2010)
(online at www.financialstability.gov/docs/
TARP%20Two%20Year%20Retrospective_10%2005%2010_transmittal%20letter.pdf).
li U.S. Department of the Treasury, Troubled Asset Relief Program Transactions Report for the Period Ending
November 26, 2010, at 17 (Nov. 30, 2010) (online at financialstability.gov/docs/transaction-reports/11-30-
10%20Transactions%20Report%20as%20of%2011-26-10.pdf).
lii This figure represents the current maximum aggregate debt guarantees that could be made under the program,
which is a function of the number and size of individual financial institutions participating. $286.8 billion
of debt subject to the guarantee is currently outstanding, which represents approximately 57.1 percent of the
current cap. Federal Deposit Insurance Corporation, Monthly Reports Related to the Temporary Liquidity
Guarantee Program: Debt Issuance Under Guarantee Program (Oct. 31, 2010) (online at www.fdic.gov/regulations/
resources/tlgp/total_issuance10-10.html). The FDIC has collected $10.4 billion in fees and surcharges from
this program since its inception in the fourth quarter of 2008. Federal Deposit Insurance Corporation, Monthly
Reports Related to the Temporary Liquidity Guarantee Program: Fees Under Temporary Liquidity Guarantee Debt
Program (Oct. 31, 2010) (online at www.fdic.gov/regulations/resources/tlgp/fees.html).
liii This figure represents the FDIC's provision for losses to its deposit insurance fund attributable to bank
failures in the third and fourth quarters of 2008; the first, second, third, and fourth quarters of 2009; and
the first and second quarters of 2010. Federal Deposit Insurance Corporation, Chief Financial Officer's (CFO)
Report to the Board: DIF Income Statement--Second Quarter 2010 (Sept. 23, 2010) (online at www.fdic.gov/about/
strategic/corporate/cfo_report_2ndqtr_10/income.html). For earlier reports, see Federal Deposit Insurance
Corporation, Chief Financial Officer's (CFO) Report to the Board (Sept. 23, 2010) (online at www.fdic.gov/
about/strategic/corporate/index.html). This figure includes the FDIC's estimates of its future losses under
loss-sharing agreements that it has entered into with banks acquiring assets of insolvent banks during these
eight quarters. Under a loss-sharing agreement, as a condition of an acquiring bank's agreement to purchase
the assets of an insolvent bank, the FDIC typically agrees to cover 80 percent of an acquiring bank's future
losses on an initial portion of these assets and 95 percent of losses on another portion of assets. See, e.g.,
Federal Deposit Insurance Corporation, Purchase and Assumption Agreement--Whole Bank, All Deposits--Among
FDIC, Receiver of Guaranty Bank, Austin, Texas, Federal Deposit Insurance Corporation and Compass Bank, at 65-
66 (Aug. 21, 2009) (online at www.fdic.gov/bank/individual/failed/guaranty-tx_p_and_a_w_addendum.pdf).
liv Outlays are comprised of the Federal Reserve Mortgage Related Facilities. The Federal Reserve balance sheet
accounts for these facilities under federal agency debt securities and mortgage-backed securities held by the
Federal Reserve. Board of Governors of the Federal Reserve System, Factors Affecting Reserve Balances (H.4.1)
(Dec. 2, 2010) (online at www.federalreserve.gov/releases/h41/20101202/) (accessed Dec. 3, 2010). Although the
Federal Reserve does not employ the outlays, loans, and guarantees classification, its accounting clearly
separates its mortgage-related purchasing programs from its liquidity programs. See, e.g., Board of Governors
of the Federal Reserve System, Factors Affecting Reserve Balances (H.4.1), at 2 (Dec. 2, 2010) (online at
www.federalreserve.gov/releases/h41/20101202) (accessed Dec. 3, 2010).
lv Federal Reserve Liquidity Facilities classified in this table as loans include primary credit, secondary
credit, central bank liquidity swaps, Asset-Backed Commercial Paper Money Market Mutual Fund Liquidity
Facility, loans outstanding to Commercial Paper Funding Facility LLC, seasonal credit, term auction credit,
the Term Asset-Backed Securities Loan Facility, and loans outstanding to Bear Stearns (Maiden Lane LLC). Board
of Governors of the Federal Reserve System, Factors Affecting Reserve Balances (H.4.1) (Dec. 2, 2010) (online
at www.federalreserve.gov/releases/h41/20101202/) (accessed Dec. 3, 2010). For further information, please see
the data that the Federal Reserve recently disclosed on these programs pursuant to its obligations under the
Dodd-Frank Wall Street Reform and Consumer Protection Act. Board of Governors of the Federal Reserve System,
Credit and Liquidity Programs and the Balance Sheet: Overview (May 11, 2010) (online at www.federalreserve.gov/
monetarypolicy/bst.htm); Board of Governors of the Federal Reserve System, Credit and Liquidity Programs and
the Balance Sheet: Reports and Disclosures (Aug. 24, 2010) (online at www.federalreserve.gov/monetarypolicy/
bst_reports.htm); Board of Governors of the Federal Reserve System, Usage of Federal Reserve Credit and
Liquidity Facilities (Dec. 3, 2010) (online at www.federalreserve.gov/newsevents/reform_transaction.htm).
SECTION FIVE: OVERSIGHT ACTIVITIES
The Congressional Oversight Panel was established as part
of the Emergency Economic Stabilization Act (EESA) and formed
on November 26, 2008. Since then, the Panel has produced 25
oversight reports, as well as a special report on regulatory
reform, issued on January 29, 2009, and a special report on
farm credit, issued on July 21, 2009.
Upcoming Reports and Hearings
The Panel will release its next oversight report in
January. The report will provide an update on government
support for the domestic automotive industry via the TARP's
Automotive Industry Financing Program. This will be the Panel's
third report focusing on the AIFP, following its September 2009
and March 2010 oversight reports.\510\
---------------------------------------------------------------------------
\510\ See Congressional Oversight Panel, September Oversight
Report: The Use of TARP Funds in Support and Reorganization of the
Domestic Automotive Industry (Sept. 9, 2009) (online at cop.senate.gov/
documents/cop-090909-report.pdf); Congressional Oversight Panel, March
Oversight Report: The Unique Treatment of GMAC Under TARP (Mar. 11,
2010) (online at cop.senate.gov/documents/cop-031110-report.pdf).
---------------------------------------------------------------------------
The Panel will hold a hearing with Secretary Geithner in
Washington on December 16, 2010. The Panel will ask the
Secretary for a general update on the TARP, for information
regarding the future plans for TARP investments following
expiration of the program's authority on October 3, 2010, and
for specific information pertaining to the topics of the
Panel's recently published and forthcoming oversight reports.
This will be the Secretary's fifth appearance before the Panel;
his most recent appearance was on June 22, 2010.
SECTION SIX: ABOUT THE CONGRESSIONAL OVERSIGHT PANEL
In response to the escalating financial crisis, on October
3, 2008, Congress provided Treasury with the authority to spend
$700 billion to stabilize the U.S. economy, preserve home
ownership, and promote economic growth. Congress created the
Office of Financial Stability (OFS) within Treasury to
implement the TARP. At the same time, Congress created the
Congressional Oversight Panel to ``review the current state of
financial markets and the regulatory system.'' The Panel is
empowered to hold hearings, review official data, and write
reports on actions taken by Treasury and financial institutions
and their effect on the economy. Through regular reports, the
Panel must oversee Treasury's actions, assess the impact of
spending to stabilize the economy, evaluate market
transparency, ensure effective foreclosure mitigation efforts,
and guarantee that Treasury's actions are in the best interests
of the American people. In addition, Congress instructed the
Panel to produce a special report on regulatory reform that
analyzes ``the current state of the regulatory system and its
effectiveness at overseeing the participants in the financial
system and protecting consumers.'' The Panel issued this report
in January 2009. Congress subsequently expanded the Panel's
mandate by directing it to produce a special report on the
availability of credit in the agricultural sector. The report
was issued on July 21, 2009.
On November 14, 2008, Senate Majority Leader Harry Reid and
the Speaker of the House Nancy Pelosi appointed Richard H.
Neiman, Superintendent of Banks for the State of New York,
Damon Silvers, Director of Policy and Special Counsel of the
American Federation of Labor and Congress of Industrial
Organizations (AFL-CIO), and Elizabeth Warren, Leo Gottlieb
Professor of Law at Harvard Law School, to the Panel. With the
appointment on November 19, 2008, of Congressman Jeb Hensarling
to the Panel by House Minority Leader John Boehner, the Panel
had a quorum and met for the first time on November 26, 2008,
electing Professor Warren as its chair. On December 16, 2008,
Senate Minority Leader Mitch McConnell named Senator John E.
Sununu to the Panel. Effective August 10, 2009, Senator Sununu
resigned from the Panel, and on August 20, 2009, Senator
McConnell announced the appointment of Paul Atkins, former
Commissioner of the U.S. Securities and Exchange Commission, to
fill the vacant seat. Effective December 9, 2009, Congressman
Jeb Hensarling resigned from the Panel and House Minority
Leader John Boehner announced the appointment of J. Mark
McWatters to fill the vacant seat. Senate Minority Leader Mitch
McConnell appointed Kenneth Troske, Sturgill Professor of
Economics at the University of Kentucky, to fill the vacancy
created by the resignation of Paul Atkins on May 21, 2010.
Effective September 17, 2010, Elizabeth Warren resigned from
the Panel, and on September 30, 2010, Senate Majority Leader
Harry Reid announced the appointment of Senator Ted Kaufman to
fill the vacant seat. On October 4, 2010, the Panel elected
Senator Kaufman as its chair.
APPENDIX I: LETTER FROM SPECIAL MASTER PATRICIA GEOGHEGAN TO CHAIRMAN
TED KAUFMAN RE: FOLLOW UP TO EXECUTIVE COMPENSATION HEARING, DATED
NOVEMBER 18, 2010
[GRAPHIC] [TIFF OMITTED] 62622A.038
[GRAPHIC] [TIFF OMITTED] 62622A.039
[GRAPHIC] [TIFF OMITTED] 62622A.040
[GRAPHIC] [TIFF OMITTED] 62622A.041
[GRAPHIC] [TIFF OMITTED] 62622A.042
[GRAPHIC] [TIFF OMITTED] 62622A.043
[GRAPHIC] [TIFF OMITTED] 62622A.044
[GRAPHIC] [TIFF OMITTED] 62622A.045
[GRAPHIC] [TIFF OMITTED] 62622A.046
[GRAPHIC] [TIFF OMITTED] 62622A.047
[GRAPHIC] [TIFF OMITTED] 62622A.048
[GRAPHIC] [TIFF OMITTED] 62622A.049
[GRAPHIC] [TIFF OMITTED] 62622A.050
[GRAPHIC] [TIFF OMITTED] 62622A.051
[GRAPHIC] [TIFF OMITTED] 62622A.052
[GRAPHIC] [TIFF OMITTED] 62622A.053
[GRAPHIC] [TIFF OMITTED] 62622A.054
[GRAPHIC] [TIFF OMITTED] 62622A.055
[GRAPHIC] [TIFF OMITTED] 62622A.056
[GRAPHIC] [TIFF OMITTED] 62622A.057
[GRAPHIC] [TIFF OMITTED] 62622A.058
[GRAPHIC] [TIFF OMITTED] 62622A.059
[GRAPHIC] [TIFF OMITTED] 62622A.060