Troubled Asset Relief Program
Treasury Needs to Strengthen Its Decision-Making Process on the Term Asset-Backed Securities Loan Facility Gao ID: GAO-10-25 February 5, 2010The Term Asset-Backed Securities Loan Facility (TALF) was created by the Board of Governors of the Federal Reserve System (Federal Reserve) to help meet consumer and small business credit needs by supporting issuance of asset-backed securities (ABS) and commercial mortgage-backed securities (CMBS). This report assesses (1) the risks TALF-eligible assets pose to the Troubled Asset Relief Program (TARP), (2) Department of the Treasury's (Treasury) role in decision making for TALF, and (3) the condition of securitization markets before and after TALF. GAO reviewed program documents, analyzed data from prospectuses and other sources, and interviewed relevant agency officials and TALF participants.
TALF contains a number of risk management features that in turn likely reduce the risk of loss to TARP funds, but risks remain. TALF was designed to reopen the securitization markets in an effort to improve access to credit for consumers and businesses. The Federal Reserve Bank of New York (FRBNY), which manages TALF, is authorized to lend up to $200 billion to certain eligible borrowers in return for collateral in the form of securities that are forfeited if the loans are not repaid. To assist in this effort, Treasury has pledged $20 billion of TARP funds in the form of credit protection to the program in the event the loans are not repaid. As of December 2009, FRBNY has made about $61.6 billion in TALF loans, of which $47.5 billion remained outstanding. For most TALF-eligible collateral, FRBNY will stop providing new TALF loans in March 2010, while new-issue CMBSs will be accepted as collateral on new TALF loans through June 2010. Treasury and FRBNY analyses project minimal, if any, use of TARP funds for TALF-related losses, and Treasury currently anticipates a profit. While GAO found that the overall risks TALF poses to TARP funds are likely minimal, GAO analyses showed that CMBSs potentially pose higher risk of loss than ABSs. As shown in figure 1, ongoing uncertainty in the commercial real estate market and TALF exposure to legacy CMBSs warrant ongoing monitoring. Finally, TALF may present risks beyond the potential risks to TARP, such as the risk that FRBNY might fail to identify material noncompliance with program requirements by TALF participants. Because the Federal Reserve views TALF as a monetary policy tool, however, statutory limitations on GAO's authority prohibited GAO from auditing FRBNY's role in administering TALF. Treasury has not fully documented its rationale, as part of its decision-making processes, for reaching final decisions related to the risks of TALF--including decisions involving other agencies. For example, the outcomes of Treasury's internal analysis of the amount of equity that TALF borrowers should hold in TALF ABS collateral, along with other TALF program terms, sometimes differed from FRBNY's. However, there was no clear documentation or explanation of how the discrepancies were resolved or how final decisions were made with FRBNY. Documenting the rationale and basis for these decisions would increase transparency and strengthen internal controls for TALF decision-making processes. Moreover, a sound decision-making process would help ensure that TALF objectives are being met and that it is functioning as intended. Unless Treasury documents the basis for major program decisions that it made with the Federal Reserve, it cannot demonstrate accountability for meeting the goals of TALF and could unnecessarily place TARP funds at risk.
RecommendationsOur recommendations from this work are listed below with a Contact for more information. Status will change from "In process" to "Open," "Closed - implemented," or "Closed - not implemented" based on our follow up work.
Director: Team: Phone:GAO-10-25, Troubled Asset Relief Program: Treasury Needs to Strengthen Its Decision-Making Process on the Term Asset-Backed Securities Loan Facility
This is the accessible text file for GAO report number GAO-10-25
entitled 'Troubled Asset Relief Program: Treasury Needs to Strengthen
Its Decision-Making Process on the Term Asset-Backed Securities Loan
Facility' which was released on February 5, 2010.
This text file was formatted by the U.S. Government Accountability
Office (GAO) to be accessible to users with visual impairments, as
part of a longer term project to improve GAO products' accessibility.
Every attempt has been made to maintain the structural and data
integrity of the original printed product. Accessibility features,
such as text descriptions of tables, consecutively numbered footnotes
placed at the end of the file, and the text of agency comment letters,
are provided but may not exactly duplicate the presentation or format
of the printed version. The portable document format (PDF) file is an
exact electronic replica of the printed version. We welcome your
feedback. Please E-mail your comments regarding the contents or
accessibility features of this document to Webmaster@gao.gov.
This is a work of the U.S. government and is not subject to copyright
protection in the United States. It may be reproduced and distributed
in its entirety without further permission from GAO. Because this work
may contain copyrighted images or other material, permission from the
copyright holder may be necessary if you wish to reproduce this
material separately.
Report to Congressional Committees:
United States Government Accountability Office:
GAO:
February 2010:
Troubled Asset Relief Program:
Treasury Needs to Strengthen Its Decision-Making Process on the Term
Asset-Backed Securities Loan Facility:
GAO-10-25:
GAO Highlights:
Highlights of GAO-10-25, a report to congressional committees.
Why GAO Did This Study:
The Term Asset-Backed Securities Loan Facility (TALF) was created by
the Board of Governors of the Federal Reserve System (Federal Reserve)
to help meet consumer and small business credit needs by supporting
issuance of asset-backed securities (ABS) and commercial mortgage-
backed securities (CMBS). This report assesses (1) the risks TALF-
eligible assets pose to the Troubled Asset Relief Program (TARP), (2)
Department of the Treasury‘s (Treasury) role in decision making for
TALF, and (3) the condition of securitization markets before and after
TALF. GAO reviewed program documents, analyzed data from prospectuses
and other sources, and interviewed relevant agency officials and TALF
participants.
What GAO Found:
TALF contains a number of risk management features that in turn likely
reduce the risk of loss to TARP funds, but risks remain. TALF was
designed to reopen the securitization markets in an effort to improve
access to credit for consumers and businesses. The Federal Reserve
Bank of New York (FRBNY), which manages TALF, is authorized to lend up
to $200 billion to certain eligible borrowers in return for collateral
in the form of securities that are forfeited if the loans are not
repaid. To assist in this effort, Treasury has pledged $20 billion of
TARP funds in the form of credit protection to the program in the
event the loans are not repaid. As of December 2009, FRBNY has made
about $61.6 billion in TALF loans, of which $47.5 billion remained
outstanding. For most TALF-eligible collateral, FRBNY will stop
providing new TALF loans in March 2010, while new-issue CMBSs will be
accepted as collateral on new TALF loans through June 2010. Treasury
and FRBNY analyses project minimal, if any, use of TARP funds for TALF-
related losses, and Treasury currently anticipates a profit. While GAO
found that the overall risks TALF poses to TARP funds are likely
minimal, GAO analyses showed that CMBSs potentially pose higher risk
of loss than ABSs. As shown in figure 1, ongoing uncertainty in the
commercial real estate market and TALF exposure to legacy CMBSs
warrant ongoing monitoring. Finally, TALF may present risks beyond the
potential risks to TARP, such as the risk that FRBNY might fail to
identify material noncompliance with program requirements by TALF
participants. Because the Federal Reserve views TALF as a monetary
policy tool, however, statutory limitations on GAO‘s authority
prohibited GAO from auditing FRBNY‘s role in administering TALF.
Figure 1: Commercial Real Estate Prices and CMBS Delinquencies from
January 2004 through October 2009:
[Refer to PDF for image: multiple line graph]
Year: 2004;
January:
Moody's/REAL Price Index: 121.08;
Percentage delinquency: 1.29.
February:
Moody's/REAL Price Index: 123.84;
Percentage delinquency: 1.32.
March:
Moody's/REAL Price Index: 126.04;
Percentage delinquency: 1.33.
April:
Moody's/REAL Price Index: 126.91;
Percentage delinquency: 1.19.
May:
Moody's/REAL Price Index: 127.52;
Percentage delinquency: 1.24.
June:
Moody's/REAL Price Index: 127.78;
Percentage delinquency: 1.22.
July:
Moody's/REAL Price Index: 132.18;
Percentage delinquency: 1.15.
August:
Moody's/REAL Price Index: 133.38;
Percentage delinquency: 1.04.
September:
Moody's/REAL Price Index: 134.66;
Percentage delinquency: 1.01.
October:
Moody's/REAL Price Index: 137.28;
Percentage delinquency: 1.01.
November:
Moody's/REAL Price Index: 138.54;
Percentage delinquency: 1.
December:
Moody's/REAL Price Index: 140.05;
Percentage delinquency: 0.95.
Year: 2005;
January:
Moody's/REAL Price Index: 145.45;
Percentage delinquency: 0.96.
February:
Moody's/REAL Price Index: 146.7;
Percentage delinquency: 0.93.
March:
Moody's/REAL Price Index: 150.36;
Percentage delinquency: 0.9.
April:
Moody's/REAL Price Index: 151.29;
Percentage delinquency: 0.85.
May:
Moody's/REAL Price Index: 154.28;
Percentage delinquency: 0.78.
June:
Moody's/REAL Price Index: 156.26;
Percentage delinquency: 0.77.
July:
Moody's/REAL Price Index: 158.31;
Percentage delinquency: 0.75.
August:
Moody's/REAL Price Index: 161.62;
Percentage delinquency: 0.7.
September:
Moody's/REAL Price Index: 162.76;
Percentage delinquency: 0.68.
October:
Moody's/REAL Price Index: 161.63;
Percentage delinquency: 0.65.
November:
Moody's/REAL Price Index: 162.89;
Percentage delinquency: 0.64.
December:
Moody's/REAL Price Index: 160.64;
Percentage delinquency: 0.56.
Year: 2006;
January:
Moody's/REAL Price Index: 166.57;
Percentage delinquency: 0.55.
February:
Moody's/REAL Price Index: 169.36;
Percentage delinquency: 0.54.
March:
Moody's/REAL Price Index: 169.19;
Percentage delinquency: 0.48.
April:
Moody's/REAL Price Index: 166.1;
Percentage delinquency: 0.49.
May:
Moody's/REAL Price Index: 168.32;
Percentage delinquency: 0.49.
June:
Moody's/REAL Price Index: 170.24;
Percentage delinquency: 0.44.
July:
Moody's/REAL Price Index: 168.1;
Percentage delinquency: 0.41.
August:
Moody's/REAL Price Index: 168.24;
Percentage delinquency: 0.39.
September:
Moody's/REAL Price Index: 168.54;
Percentage delinquency: 0.38.
October:
Moody's/REAL Price Index: 170.52;
Percentage delinquency: 0.37.
November:
Moody's/REAL Price Index: 170.78;
Percentage delinquency: 0.34.
December:
Moody's/REAL Price Index: 174.08;
Percentage delinquency: 0.29.
Year: 2007;
January:
Moody's/REAL Price Index: 179.25;
Percentage delinquency: 0.29;
February:
Moody's/REAL Price Index: 183.47;
Percentage delinquency: 0.28;
March:
Moody's/REAL Price Index: 185.2;
Percentage delinquency: 0.25.
April:
Moody's/REAL Price Index: 186.37;
Percentage delinquency: 0.24;
May:
Moody's/REAL Price Index: 185.59;
Percentage delinquency: 0.23.
June:
Moody's/REAL Price Index: 187.25;
Percentage delinquency: 0.24.
July:
Moody's/REAL Price Index: 188.17;
Percentage delinquency: 0.22.
August:
Moody's/REAL Price Index: 191.11;
Percentage delinquency: 0.24.
September:
Moody's/REAL Price Index: 188.89;
Percentage delinquency: 0.24.
October: Prices peaked in October 2007:
Moody's/REAL Price Index: 191.87;
Percentage delinquency: 0.23.
November:
Moody's/REAL Price Index: 191.4;
Percentage delinquency: 0.27.
December:
Moody's/REAL Price Index: 188.51;
Percentage delinquency: 0.29.
Year: 2008;
January:
Moody's/REAL Price Index: 187.31;
Percentage delinquency: 0.31.
February:
Moody's/REAL Price Index: 191.24;
Percentage delinquency: 0.32.
March:
Moody's/REAL Price Index: 186.92;
Percentage delinquency: 0.35.
April:
Moody's/REAL Price Index: 181.23;
Percentage delinquency: 0.36.
May:
Moody's/REAL Price Index: 174.97;
Percentage delinquency: 0.43.
June:
Moody's/REAL Price Index: 169.28;
Percentage delinquency: 0.46.
July:
Moody's/REAL Price Index: 169.96;
Percentage delinquency: 0.48.
August:
Moody's/REAL Price Index: 169.74;
Percentage delinquency: 0.5.
September:
Moody's/REAL Price Index: 173.92;
Percentage delinquency: 0.54.
October:
Moody's/REAL Price Index: 169.76;
Percentage delinquency: 0.6.
November:
Moody's/REAL Price Index: 164;
Percentage delinquency: 0.75.
December:
Moody's/REAL Price Index: 160.46;
Percentage delinquency: 0.95.
Year: 2009;
January:
Moody's/REAL Price Index: 151.58;
Percentage delinquency: 1.16.
February:
Moody's/REAL Price Index: 150.63;
Percentage delinquency: 1.32.
March:
Moody's/REAL Price Index: 148.07;
Percentage delinquency: 1.59.
April:
Moody's/REAL Price Index: 135.31;
Percentage delinquency: 2.27.
May:
Moody's/REAL Price Index: 125.04;
Percentage delinquency: 2.67.
June:
Moody's/REAL Price Index: 123.82;
Percentage delinquency: 3.02.
July:
Moody's/REAL Price Index: 117.56;
Percentage delinquency: 3.23.
August:
Moody's/REAL Price Index: 114.06;
Percentage delinquency: 3.64.
September:
Moody's/REAL Price Index: 109.6;
Percentage delinquency: 4.01.
October:
Moody's/REAL Price Index: 107.98;
Percentage delinquency: 4.47.
November:
Moody's/REAL Price Index: 109.1;
Percentage delinquency: 4.9.
As of November 2009, prices show a drop of 43% since peak. November
was the first month in which prices increased since September 2008.
Note: Moody‘s/REAL Commercial Property Price Index is a report showing
the change in commercial real estate asset sales. Moody‘s CMBS
Delinquency Tracker is a monthly report showing delinquency rates for
commercial mortgage-backed securities.
Source: GAO presentation of Moody‘s/REAL Commercial Property Price
Index and Moody‘s CMBS Delinquency Tracker data.
[End of figure]
Treasury has not fully documented its rationale, as part of its
decision-making processes, for reaching final decisions related to the
risks of TALF”including decisions involving other agencies. For
example, the outcomes of Treasury‘s internal analysis of the amount of
equity that TALF borrowers should hold in TALF ABS collateral, along
with other TALF program terms, sometimes differed from FRBNY‘s.
However, there was no clear documentation or explanation of how the
discrepancies were resolved or how final decisions were made with
FRBNY. Documenting the rationale and basis for these decisions would
increase transparency and strengthen internal controls for TALF
decision-making processes. Moreover, a sound decision-making process
would help ensure that TALF objectives are being met and that it is
functioning as intended. Unless Treasury documents the basis for major
program decisions that it made with the Federal Reserve, it cannot
demonstrate accountability for meeting the goals of TALF and could
unnecessarily place TARP funds at risk.
FRBNY, in consultation with Treasury, monitors various economic
indicators to measure TALF's impact. Our analysis of such indicators”-
including securitization volumes, interest rates, and TALF loan volume
trends”-suggests that the securitization markets improved for the more
frequently traded TALF-eligible sectors after the program‘s first
activity in March 2009, which is illustrated in figure 2. Figure 2
also shows that ABS issuances in all of the most liquid TALF-eligible
sectors dropped sharply in 2008 from their peak levels in 2006 and
2007. Consumer credit rates have not changed significantly since TALF
started. FRBNY officials said that it is possible that without TALF,
interest rates on loans to consumers and small businesses would be
much higher than they are now. While Treasury bears the first-loss
risk from any assets that TALF borrowers might surrender in
conjunction with unpaid loans, it has not developed measures to
analyze and publicly report on the potential purchase, management, and
sale of such assets. Without such a plan Treasury cannot measure TALF‘
s success in meeting its goals under TARP with respect to any
collateral that is placed in TALF LLC. Finally, without a plan for
communicating the findings that result from tracking and analyzing
such metrics, Treasury may not fully inform the public of how the
assets are managed and financed, undermining Treasury‘s efforts to be
fully transparent about TARP activities.
Figure 2: Fluctuations in ABS Issuances, First Quarter of 2005 through
the Fourth Quarter of 2009:
[Refer to PDF for image: multiple line graph]
2005, Q1;
Credit card: $12.0 billion;
Auto: $9.59 billion;
Student loan: $5.9 billion;
CMBS: $31.4 billion.
2005, Q2;
Credit card: $25.9 billion;
Auto: $11.4 billion;
Student loan: $8.1 billion;
CMBS: $30.4 billion.
2005, Q3;
Credit card: $17.9 billion;
Auto: $14.2 billion;
Student loan: $9.4 billion;
CMBS: $38.1 billion.
2005, Q4;
Credit card: $21.1 billion;
Auto: $17.8 billion;
Student loan: $22.8 billion;
CMBS: $52.3 billion.
2006, Q1;
Credit card: $18.6 billion;
Auto: $17.3 billion;
Student loan: $16.5 billion;
CMBS: $45.6 billion.
2006, Q2;
Credit card: $18.0 billion;
Auto: $15.5 billion;
Student loan: $21.7 billion;
CMBS: $42.5 billion.
2006, Q3;
Credit card: $26.6 billion;
Auto: $16.0 billion;
Student loan: $16.4 billion;
CMBS: $44.0 billion.
2006, Q4;
Credit card: $25.0 billion;
Auto: $11.6 billion;
Student loan: $15.5 billion;
CMBS: $69.3 billion.
2007, Q1;
Credit card: $13.4 billion;
Auto: $25.0 billion;
Student loan: $23.4 billion;
CMBS: $46.0 billion.
2007, Q2;
Credit card: $27.1 billion;
Auto: $23.0 billion;
Student loan: $9.6 billion;
CMBS: $74.4 billion.
2007, Q3;
Credit card: $14.5 billion;
Auto: $22.1 billion;
Student loan: $10.0 billion;
CMBS: $54.2 billion.
2007, Q4;
Credit card: $16.0 billion;
Auto: $24.3 billion;
Student loan: $7.6 billion;
CMBS: $25.4 billion.
2008, Q1;
Credit card: $10.3 billion;
Auto: $28.8 billion;
Student loan: $8.4 billion;
CMBS: $4.4 billion.
2008, Q2;
Credit card: $14.0 billion;
Auto: $28.2 billion;
Student loan: $14.6 billion;
CMBS: $2.0 billion.
2008, Q3;
Credit card: $3.2 billion;
Auto: $10.4 billion;
Student loan: $6.4 billion;
CMBS: $3.1 billion.
2008, Q4;
Credit card: $1.6 billion;
Auto: 0;
Student loan: 0;
CMBS: 0.
2009, Q1;
Credit card: $7.6 billion;
Auto: $6.5 billion;
Student loan: $2.0 billion;
CMBS: 0.
2009, Q2;
Credit card: $15.0 billion;
Auto: $20.4 billion;
Student loan: $8.3 billion;
CMBS: $3.43 billion.
2009, Q3;
Credit card: $19.3 billion;
Auto: $15.5 billion;
Student loan: $6.1 billion;
CMBS: 0.
2009, Q4;
Credit card: $9.4 billion;
Auto: $3.6 billion;
Student loan: $4.8 billion;
CMBS: $2.2 billion.
Source: GAO analysis of Thomson Reuters IFR Markets data.
[End of section]
What GAO Recommends:
Treasury should (1) give increased attention to reviewing risks posed
by CMBS, (2) strengthen its TALF decision-making process, and (3)
determine which data are needed to track the management and sale of
assets TALF borrowers might surrender. To enable more effective review
of TARP, Congress also should grant GAO authority to audit Federal
Reserve TALF operational and administrative activities. Treasury
appreciated GAO‘s recommendations but said GAO understated TALF‘s
success and overstated the risk of CMBS. The Federal Reserve disagreed
that there are limitations on GAO‘s authority to audit TALF
activities. GAO continues to believe that its depiction of TALF is
accurate and its recommended actions are necessary to strengthen the
oversight and operations of TALF.
View [hyperlink, http://www.gao.gov/products/GAO-10-25] or key
components. For more information, contact Orice Williams Brown at
(202)512-8678 or williamso@gao.gov.
[End of section]
Contents:
Letter:
Background:
TALF Includes Features That Mitigate Potential Losses, but CMBSs Could
Pose Greater Risks Than Other Asset Classes:
Treasury Worked with the Federal Reserve and FRBNY to Analyze Risks
Related to TALF but Did Not Fully Document Analysis Supporting Final
Decisions:
Treasury's and FRBNY's Indicators Suggest That Credit Market
Conditions Have Shown Some Improvement, but Treasury Lacks Performance
Indicators in the Event that It Must Purchase TALF Assets:
Conclusions:
Matter for Congressional Consideration:
Recommendations for Executive Action:
Agency Comments and Our Evaluation:
Appendix I: Objectives, Scope, and Methodology:
Appendix II: Methodology for Market Value Analysis of ABSs and CMBSs:
Appendix III: The Securitization Process Explained:
Appendix IV: Descriptions of Asset-Backed Securities:
Appendix V: Credit Enhancement:
Appendix VI: Additional Information on TALF Compliance:
Appendix VII: New Securitization Volumes Have Increased since the
Inception of the TALF Program:
Appendix VIII: Comments from the Department of the Treasury:
Appendix IX: Comments from the Board of Governors of the Federal
Reserve System:
Appendix X: Analysis of Legal Comments Submitted by the Federal
Reserve:
Appendix XI: GAO Contact and Staff Acknowledgments:
Tables:
Table 1: Timetable of Asset Classes Accepted As Eligible Collateral:
Table 2: TALF Haircuts by Asset Class:
Table 3: TALF Loans Disbursed by Asset Class, March 2009 through
December 2009:
Table 4: GAO Sample Selection for ABS Extreme Market Value Loss
Analysis:
Table 5: Auto Loan Securitization Volume, 2005 through Fourth Quarter
2009:
Table 6: Credit Card Receivables Securitization Volume, 2005 through
Fourth Quarter 2009:
Table 7: Student Loan Securitization Volume, 2005 through Fourth
Quarter 2009:
Table 8: CMBS Securitization Volume, 2005 Through Fourth Quarter 2009:
Figures:
Figure 1: Status of TARP Apportioned Funds as of December 31, 2009:
Figure 2: TALF Loan Origination and Repayment Process:
Figure 3: Funding Required if a Borrower Walks Away from a Loan:
Figure 4: Overall TALF Borrower Returns at Issuance for Select Asset
Classes Accepted as TALF Collateral Generally Decreased from March
through September 2009:
Figure 5: Increases in Levels of Credit Enhancement on ABSs Since the
Credit Crisis Began, March 2009 through September 2009:
Figure 6: Commercial Real Estate Prices and CMBS Delinquencies from
January 2004 through October 2009:
Figure 7: Fluctuations in ABS Issuances, First Quarter 2005 through
Fourth Quarter 2009:
Figure 8: Trends in Average Finance Rates for Credit Cards, Auto Loans
at Banks, and Auto Finance Companies, June 2008 through December 2009:
Figure 9: Composition of TALF Loans Disbursed by Asset Class as of
December 2009:
Figure 10: Trends for AAA ABS and CMBS Spreads from January 2005
through September 2009:
Figure 11: The Securitization Process:
Abbreviations:
ABS: asset-backed security:
AIG: American International Group, Inc.
AUP: agreed upon procedure:
CBLI: Consumer and Business Lending Initiative:
CMBS: commercial mortgage-backed security:
CUSIP: Committee on Uniform Security Identification Procedures:
EESA: Emergency Economic Stabilization Act of 2008:
Federal Reserve: Board of Governors of the Federal Reserve System:
FFELP: Federal Family Education Loan Program:
FICO: Fair Isaac Corporation:
FRBNY: Federal Reserve Bank of New York:
LIBOR: London Interbank Offered Rate:
NRSRO: nationally recognized statistical rating organization:
OIS: Overnight Indexed Swap:
PIMCO: Pacific Investment Management Company, LLC:
RMBS: residential mortgage-backed security:
SBA: Small Business Administration:
SIGTARP: Special Inspector General for TARP:
SUBI: special units of beneficial interest:
TALF: Term Asset-Backed Securities Loan Facility:
TARP: Troubled Asset Relief Program:
Treasury: Department of the Treasury:
[End of section]
United States Government Accountability Office:
Washington, DC 20548:
February 5, 2010:
Congressional Committees:
The recent financial crisis disrupted the securitization markets--
which have funded an increasing share of consumer credit and small
business loans in recent years--making credit less available to
households and small businesses.[Footnote 1] This "credit crunch" has
been a major factor behind the weak economy and slow recovery. The
Term Asset-Backed Securities Loan Facility (TALF), which the Board of
Governors of the Federal Reserve System (Federal Reserve) first
announced in November 2008, was created to help restore the
securitization markets. The Department of the Treasury (Treasury)
provides credit protection for this program under the Troubled Asset
Relief Program (TARP), as authorized under the Emergency Economic
Stabilization Act of 2008 (EESA).[Footnote 2] The Federal Reserve Bank
of New York (FRBNY), which manages TALF, is authorized to lend up to
$200 billion in 3-or 5-year nonrecourse loans to investors to purchase
AAA-rated asset-backed securities (ABS) and commercial mortgage-backed
securities (CMBS), which are, in turn, pledged as collateral for the
loans.[Footnote 3] Nonrecourse loans are provided against collateral,
and if the loans are not repaid, FRBNY's only recourse is to assume
control of the pledged assets.[Footnote 4] If borrowers do not repay
their TALF loans, Treasury will fund the purchase of up to $20 billion
of the collateral backing the loans from FRBNY, creating exposure that
places TARP funds, and therefore taxpayers, at risk.
EESA, which authorized TARP, also provided GAO with broad oversight
authorities for actions taken under TARP and requires GAO to report at
least every 60 days on TARP activities and performance.[Footnote 5] To
fulfill our statutorily mandated responsibilities, we have been
monitoring and providing updates on TARP programs, including TALF, in
prior 60-day reports.[Footnote 6] This report expands on previously
reported TALF activities.
The Federal Reserve created TALF under Section 13(3) of the Federal
Reserve Act, which permits the Federal Reserve to authorize a Federal
Reserve Bank to lend to nondepository institutions in "unusual and
exigent circumstances." The Federal Reserve viewed creation of TALF as
part of its monetary policy activities because, having already reduced
the federal funds target rate to close to zero, it needed additional
mechanisms, like TALF, to provide liquidity directly to borrowers and
investors in key credit markets. GAO is prohibited by statute from
auditing deliberations, decisions, or actions by the Federal Reserve
or any Federal Reserve bank that are related to monetary policy
matters.[Footnote 7] Accordingly, although the Federal Reserve and
FRBNY voluntarily gave us access to TALF-related documents and staff,
our review of TALF was limited and we did not evaluate the actions of
the Federal Reserve with respect to TALF. For example, we did not
assess FRBNY's system of internal controls or the role of TALF
participants--such as agents, borrowers, or auditors--in certifying
and validating compliance with the program's terms.[Footnote 8] We
also conducted only limited verification of factual information
provided by Federal Reserve officials, which we used for background
description purposes and to assess Treasury's involvement in TALF.
The objectives of this report are to (1) analyze the risks that TALF
presents to TARP funds and therefore to taxpayers, (2) evaluate how
Treasury analyzed the risk of TALF assets and used this information in
making decisions on TALF with the Federal Reserve and FRBNY, and (3)
assess changes in securitization and credit market conditions before
and after TALF's implementation, based on indicators tracked by
Treasury and FRBNY.
To meet the report objectives, we analyzed data from TALF prospectuses
and data providers to understand securitization volumes, prices, and
spreads. We also reviewed program operation and design documents,
including documentation regarding the selection of eligible asset
classes, the amount of equity a TALF borrower is required to hold in a
security, and requirements for borrowers and collateral.[Footnote 9]
In addition, we analyzed agreements between the relevant agencies on
TALF. We interviewed officials from the federal agencies responsible
for the program, along with policy analysts, economists, and
attorneys. We also interviewed current and potential TALF program
participants, including investors, underwriters, issuers, and
sponsors. For additional information on the scope and methodology for
this engagement, see appendix I.
We conducted this performance audit from May 2009 to January 2010 in
accordance with generally accepted government auditing standards.
Those standards require that we plan and perform the audit to obtain
sufficient, appropriate evidence to provide a reasonable basis for our
findings and conclusions based on our audit objectives. We believe
that the evidence obtained provides a reasonable basis for our
findings and conclusions based on our audit objectives.
Background:
Securitization is a process by which similar debt instruments--such as
loans, leases, or receivables--are aggregated into pools, and interest-
bearing securities backed by such pools are then sold to investors.
These ABSs provide a source of liquidity for consumers and small
businesses because financial institutions can take assets that they
would otherwise hold on their balance sheets, sell them as securities,
and use the proceeds to originate new loans, among other
purposes.[Footnote 10] During the recent financial crisis, the value
of many ABSs dropped precipitously, bringing the securitization
markets to a virtual halt. As a result, households and small
businesses found themselves unable to access the credit that they
needed to, among other things, buy homes and expand inventories.
TALF was designed to reopen the securitization markets in an effort to
improve access to credit for consumers and businesses. The program
provides loans to certain institutions and business entities in return
for collateral in the form of securities that are forfeited if the
loans are not repaid. To assist in this effort, Treasury provides
credit protection for the program as part of TARP's Financial
Stability Plan under the Consumer and Business Lending Initiative
(CBLI). Treasury has pledged $20 billion for TALF LLC--a special
purpose vehicle created by FRBNY--to purchase the underlying
collateral associated with TALF loans in the event the loans are not
repaid. Because of Treasury's role in protecting these taxpayer funds
committed through TARP, Treasury has consulted with the Federal
Reserve on TALF's design. Under TALF, FRBNY is currently authorized to
extend up to $200 billion in nonrecourse loans to eligible borrowers
pledging eligible collateral. TALF is authorized to extend new loans
against nonmortgage-backed ABSs and legacy CMBS collateral until March
31, 2010, and against newly issued CMBS collateral until June 30,
2010. As of December 2009, FRBNY has made about $61.6 billion in loans
under TALF. Of that amount, $47.5 billion in TALF loans remained
outstanding as of the end of December 2009. The amount of loans
outstanding may be less than the amount of loans extended due to loan
prepayments by the TALF borrower or paydown of principal.
TALF accounts for a small proportion of TARP funds (see figure 1). As
of December 31, 2009--of the $20 billion committed--Treasury had
loaned TALF LLC $100 million: $16 million for administrative expenses
and $84 million for potential asset purchases. This amount is less
than 1 percent of the $25.3 billion apportioned to the CBLI program,
which itself represents 5 percent of apportioned TARP funds. TALF will
receive TARP funds beyond the $100 million already loaned if
additional funding is required by TALF LLC to purchase surrendered or
seized collateral resulting from unpaid TALF loans.
Figure 1: Status of TARP Apportioned Funds as of December 31, 2009:
[Refer to PDF for image: pie-chart]
Capital Purchase Program: $208 billion (38%);
Other[A]: $115 billion (17%);
Automotive Industry Financing Program: $91.8 billion (17%);
Public Private Investment Program: $68.0 (12%);
Home Affordable Modification Program: $45.5 (8%);
Consumer and Business Lending Initiative: $25.3 billion (5%): Of the
$25.3 in Consumer and Business Lending Initiative funds apportioned,
less than 1%, or $100 million, has gone to TALF.
Source: GAO presentation of Treasury Office of Financial Stability
data.
[A] The "other" category consists of the Targeted Investment Program,
American International Group Investment, and the Asset Guarantee
Program.
Note: Percentages may not total due to rounding.
[End of figure]
FRBNY created TALF LLC, a special purpose vehicle, to purchase and
manage any assets that TALF borrowers surrender or the FRBNY seizes.
[Footnote 11] TALF LLC also holds any excess accumulated interest from
TALF loans and the $100 million funded portion of the Treasury loan
for administrative expenses and collateral purchases, plus interest
earned from permitted investments.
A portion of the interest earned by FRBNY on all TALF loans--called
the "excess interest"--is paid to TALF LLC as a fee for TALF LLC's
commitment to purchase the assets received by FRBNY in conjunction
with a TALF loan. As of December 31, 2009, TALF LLC had accumulated
approximately $198 million in excess interest, with roughly $30
million added each month (according to FRBNY officials), based on the
current loan portfolio.
If accumulated fees and interest earned on TALF LLC's investments are
insufficient to cover any asset purchases, Treasury will provide
additional TARP funds to TALF LLC to finance up to $20 billion of
asset purchases. Subsequently, TALF LLC will finance any additional
purchases by borrowing funds from FRBNY. The TARP loan is subordinate
to the FRBNY loan, thus the TARP funds provide credit protection to
FRBNY.
Asset Classes Eligible for Use as TALF Collateral:
When TALF was first announced, the Federal Reserve made a number of
asset classes eligible for use as collateral in consultation with
Treasury, adding more as the program evolved (see table 1). Initially,
securities backed by automobile, credit card, and student loans, as
well as loans guaranteed by the Small Business Administration (SBA),
were deemed eligible because of the need to make credit in these
sectors more widely available.[Footnote 12] For most TALF-eligible
collateral, FRBNY will stop providing new TALF loans in March 31,
2010, while new-issue CMBSs will be accepted as collateral on new TALF
loans through June 30, 2010.
Table 1: Timetable of Asset Classes Accepted As Eligible Collateral:
Asset class: Auto, credit card receivables, student loans, and small
business[A];
Description:
* Auto includes retail loans and leases relating to cars, light
trucks, motorcycles, and other recreational vehicles originated on or
after October 1, 2007;
* Credit card receivables include both consumer and corporate credit
card receivables maturing in 2009 or the first quarter of 2010;
* Student loans include federally guaranteed (including consolidation
loans) and private student loans with a disbursement date on or after
May 1, 2007;
* Small business includes fully guaranteed SBA 7(a) and 504 loans,
debentures or pools originated on or after January 1, 2008;
Date announced: November 2008.
Asset class: Mortgage servicing advances, business equipment, vehicle
fleet, and floor plan[B];
Description:
* Mortgage servicing advances are receivables created by principal and
interest, tax and insurance, and corporate advances made by Fannie Mae-
or Freddie Mac-approved residential mortgage servicers for mortgages
originated on or after January 1, 2007;
* Business equipment includes retail loans and leases originated on or
after October 1, 2007;
* Vehicle fleet include commercial and government fleets and
commercial loans secured by vehicles and the related fleet leases and
subleases of such vehicles to rental car companies, all originated on
or after October 1, 2007;
* TALF-eligible floor plan ABS include revolving lines of credit to
finance dealer inventories originated on or after January 1, 2009;
Date announced: March 2009.
Asset class: Insurance premium finance, new-issue CMBSs, and legacy
CMBSs[C];
Description:
* Insurance premium finance includes loans originated for the purposes
of paying premiums on property and casualty insurance originated on or
after January 1, 2009;
* New-issue CMBSs issued on or after January 1, 2009;
* Legacy CMBSs issued before January 1, 2009;
Date announced: May 2009.
Source: GAO presentation of information gathered from the FRBNY Web
site.
[A] Under the 7(a) program, SBA provides lenders with guarantees on up
to 85 percent of the value of loans to qualifying small businesses in
exchange for fees to help offset the costs of the program. Under the
504 program, which generally applies to small business real estate and
other fixed assets, SBA also provides certified development companies
with a guarantee on up to 40 percent of the financing of the projects'
costs in exchange for fees, while the small business borrowers and
other lenders provide the remaining 60 percent of the financing with
no guarantee. For additional information on these programs, see GAO,
Small Business Administration's Implementation of Administrative
Provisions in the American Recovery and Reinvestment Act, GAO-10-507R
(Washington, DC: Apr. 16, 2009).
[B] Floor plan loans are made to auto and nonauto dealers to finance
their inventories.
[C] CMBSs are securities backed by mortgages for commercial real
estate, such as office buildings or shopping centers.
[End of table]
All TALF-eligible ABSs must be denominated in U.S. dollars, must be
rated AAA by at least two TALF-eligible nationally recognized
statistical rating organizations (NRSRO), must not have a credit
rating below the highest investment-grade rating category from a TALF-
eligible NRSRO, and must not be on review for a potential rating
downgrade.[Footnote 13] In general, borrowers must be U.S.-based
businesses, investment funds, or U.S.-insured depository institutions,
although foreign banks with U.S. branches that maintain reserves with
a Federal Reserve bank are also eligible. However, all or
substantially all of the eligible collateral's underlying credit
exposures must be:
* for newly issued ABSs--originated by U.S.-organized entities or
institutions, or U.S. branches or agencies of foreign banks--and:
* for all ABSs--made to U.S.-domiciled obligors or located in the U.S.
or one of its territories, in the case of real property.
Interest rates for TALF loans are either fixed or floating and vary
according to the collateral securing the loan, as has been determined
by FRBNY.
In order to constitute eligible collateral for a TALF loan, both the
issuers and sponsors of the proposed collateral must provide
certification documents stating that, among other things,
* they have reviewed TALF's terms and conditions;
* the collateral is TALF-eligible;
* an independent accounting firm was provided consent, in certain
cases, to contact the TALF compliance fraud hotline if it suspects
fraud or illegal acts; and:
* purchasers of the securities that are affiliated with the
originators, issuers, or sponsors cannot use these securities as TALF
collateral.[Footnote 14]
In addition, external auditors review certain representations made by
issuers and sponsors about the TALF eligibility of the collateral.
[Footnote 15] If any of these representations change, the issuer and
sponsor must provide public notice. If any of the certifications are
found to be incorrect, TALF LLC and FRBNY can recover damages and the
issuer and sponsor will be subject to review by Treasury, the Special
Inspector General for TARP (SIGTARP), and GAO.[Footnote 16]
How TALF Works:
A number of entities help administer the TALF program.
* TALF agents, which are primary dealers or designated broker-dealers
that operate under FRBNY's Master Loan and Security Agreement. The
agents' responsibilities include conducting due diligence on TALF
borrowers and making representations to FRBNY regarding eligibility of
TALF borrowers and their collateral, submitting TALF loan requests and
supporting documentation to FRBNY and the TALF custodian on behalf of
borrowers, delivering administrative fees and collateral from TALF
borrowers to FRBNY, and distributing the TALF borrower's share of
principal and interest payments paid on the collateral backing the
TALF loan.[Footnote 17]
* The Bank of New York Mellon, which serves as custodian of the
program and is responsible for administering TALF loans, holding and
reviewing collateral, collecting payments and administrative fees,
disbursing cash flows, maintaining the program's books and records,
and assisting other TALF entities with the pricing of collateral.
* Collateral monitors--Trepp LLC and Pacific Investment Management
Company LLC (PIMCO)--which check the pricing and ratings of
securities; provide valuation, modeling, reporting, and analytical
support; and advise on related matters.
* CW Capital, which provides underwriting advisory services related to
certain commercial mortgage loans backing newly issued CMBSs.
FRBNY announces monthly subscription periods, during which potential
borrowers apply for loans and funds are disbursed. FRBNY has a
precertification process to streamline the process for certain
eligible borrowers. TALF precertification documents indicate that
borrowers must be deemed to be top-tier financial entities--that is,
they must be seen as industry leaders and be ranked among the largest
entities in the industry or have some of the largest
operations.[Footnote 18] During our audit, FRBNY officials told us
that they review loan requests from all borrowers that meet general
eligibility criteria, and that not all borrowers need to be
precertified; indeed, most are not precertified. Applicants must work
through a TALF agent throughout the application process. Because the
agents must demonstrate that they know a potential borrower and must
vouch for its reputation, they put applicants through a "Know Your
Customer" program based on provisions in the Patriot Act.[Footnote 19]
Once this process is completed, TALF agents submit a loan request
package to FRBNY that includes:
* borrower identification information, such as name, address, and tax
identification number;
* loan information, such as the requested loan amount ($10 million
minimum), the term of the loan, the loan rate, and the type of
interest rate (fixed or floating) that corresponds to the type of
collateral offered;
* collateral information, such as the CUSIPs of the securities,
[Footnote 20] asset class and subclass, price and face value of the
collateral, the weighted average life of the collateral,[Footnote 21]
and the haircut amount--a percentage assigned to the loan based on the
asset class, or subsector where appropriate, of the collateral and its
weighted average life;
* any appropriate filing documents, including a prospectus and
offering documents of the securities expected to be pledged; and:
* proof of purchase for the ABSs and CMBSs that are being offered as
collateral.
Next, the Bank of New York Mellon and FRBNY verify the data that the
TALF agents provide and, among other things, ensure that the ratings
submitted for the securities are the most recent. For legacy CMBS
collateral, FRBNY evaluates whether the price the TALF borrower paid
was reasonable based on pricing information FRBNY receives from the
collateral monitors and Bank of New York Mellon.[Footnote 22] The
collateral monitors, Trepp LLC and PIMCO, also conduct stress tests on
pledged legacy CMBS collateral to help ensure that the loan amounts
will not exceed the stressed value of the pledged securities. Three
weeks prior to each ABS subscription, a newly issued ABS undergoes a
risk assessment by FRBNY (with PIMCO's support) to determine if it is
likely to be accepted as TALF collateral. The issuer must provide
FRBNY any information it provided the NRSROs so FRBNY can conduct its
own credit risk assessment. The issuer must also consent to permitting
NRSROs to discuss all aspects of the rating with FRBNY, including
credit quality of the ABS, modeling, and methodology, among other
things.
On each TALF loan's settlement date, the borrower must deliver the
loan collateral and administrative fees to FRBNY's custodian, the Bank
of New York Mellon, which holds the collateral for the life of the
loan. The administrative fees vary by asset class and cover FRBNY's
administrative costs for the facility.[Footnote 23]
Determining the Loan Amount:
If FRBNY deems the collateral eligible, it will lend an amount
calculated by subtracting a designated haircut percentage from the
lesser of either par or market value of the pledged collateral (or, in
the case of legacy CMBSs, a value based on an internal risk
assessment). This percentage, or "haircut," in effect sets the amount
of equity the TALF borrower holds in the collateral. Haircuts vary by
FRBNY's assessment of market risks for each sector and subsector.
[Footnote 24] The haircut represents the difference between the value
of the proposed collateral and the value of the loan (table 2).
[Footnote 25]
Table 2: TALF Haircuts by Asset Class:
Sector: Auto;
Subsector: Prime retail lease;
Weighted average life in years for ABSs:
0 to less than 1: 10%;
1 to less than 2: 11%;
2 to less than 3: 12%;
3 to less than 4: 13%;
4 to less than 5: 14%;
5 to less than 6: [Empty];
6 to less than 7: [Empty].
Subsector: Prime retail loan;
Weighted average life in years for ABSs:
0 to less than 1: 6%;
1 to less than 2: 7%;
2 to less than 3: 8%;
3 to less than 4: 9v;
4 to less than 5: 10%;
5 to less than 6: [Empty];
6 to less than 7: [Empty].
Subsector: Subprime retail loan;
Weighted average life in years for ABSs:
0 to less than 1: 9%;
1 to less than 2: 10%;
2 to less than 3: 11%;
3 to less than 4: 12%;
4 to less than 5: 13%;
5 to less than 6: [Empty];
6 to less than 7: [Empty].
Subsector: Motorcycle or other recreational vehicles;
Weighted average life in years for ABSs:
0 to less than 1: 7%;
1 to less than 2: 8%;
2 to less than 3: 9%;
3 to less than 4: 10%;
4 to less than 5: 11%;
5 to less than 6: [Empty];
6 to less than 7: [Empty].
Subsector: Commercial and government fleets;
Weighted average life in years for ABSs:
0 to less than 1: 9%;
1 to less than 2: 10%;
2 to less than 3: 11%;
3 to less than 4: 12%;
4 to less than 5: 13v;
5 to less than 6: [Empty];
6 to less than 7: [Empty].
Subsector: Rental fleets;
Weighted average life in years for ABSs:
0 to less than 1: 12%;
1 to less than 2: 13%;
2 to less than 3: 14%;
3 to less than 4: 15%;
4 to less than 5: 16%;
5 to less than 6: [Empty];
6 to less than 7: [Empty].
Sector: Credit Card;
Subsector: Prime;
Weighted average life in years for ABSs:
0 to less than 1: 5%;
1 to less than 2: 5%;
2 to less than 3: 6%;
3 to less than 4: 7%;
4 to less than 5: 8%;
5 to less than 6: [Empty];
6 to less than 7: [Empty].
Subsector: Subprime;
Weighted average life in years for ABSs:
0 to less than 1: 6%;
1 to less than 2: 7%;
2 to less than 3: 8%;
3 to less than 4: 9%;
4 to less than 5: 10%;
5 to less than 6: [Empty];
6 to less than 7: [Empty].
Sector: Equipment;
Subsector: Loans and leases;
Weighted average life in years for ABSs:
0 to less than 1: 5%;
1 to less than 2: 6%;
2 to less than 3: 7%;
3 to less than 4: 8%;
4 to less than 5: 9%;
5 to less than 6: [Empty];
6 to less than 7: [Empty].
Sector: Floor plan;
Subsector: Auto;
Weighted average life in years for ABSs:
0 to less than 1: 12%;
1 to less than 2: 13%;
2 to less than 3: 14%;
3 to less than 4: 15%;
4 to less than 5: 16%;
5 to less than 6: [Empty];
6 to less than 7: [Empty].
Subsector: Nonauto;
Weighted average life in years for ABSs:
0 to less than 1: 11%;
1 to less than 2: 12%;
2 to less than 3: 13%;
3 to less than 4: 14%;
4 to less than 5: 15%;
5 to less than 6: [Empty];
6 to less than 7: [Empty].
Sector: Premium finance;
Subsector: Property and casualty;
Weighted average life in years for ABSs:
0 to less than 1: 5%;
1 to less than 2: 6%;
2 to less than 3: 7%;
3 to less than 4: 8%;
4 to less than 5: 9v;
5 to less than 6: [Empty];
6 to less than 7: [Empty].
Sector: Servicing advances;
Subsector: Residential mortgages;
Weighted average life in years for ABSs:
0 to less than 1: 12%;
1 to less than 2: 13%;
2 to less than 3: 14%;
3 to less than 4: 15%;
4 to less than 5: 16%;
5 to less than 6: [Empty];
6 to less than 7: [Empty].
Sector: Small business;
Subsector: SBA loans;
Weighted average life in years for ABSs:
0 to less than 1: 5%;
1 to less than 2: 5%;
2 to less than 3: 5%;
3 to less than 4: 5%;
4 to less than 5: 5%;
5 to less than 6: 6%;
6 to less than 7: 6v.
Sector: Student loan;
Subsector: Private;
Weighted average life in years for ABSs:
0 to less than 1: 8%;
1 to less than 2: 9%;
2 to less than 3: 10%;
3 to less than 4: 11%;
4 to less than 5: 12%;
5 to less than 6: 13%;
6 to less than 7: 14%.
Subsector: Government guaranteed;
Weighted average life in years for ABSs:
0 to less than 1: 5%;
1 to less than 2: 5%;
2 to less than 3: 5%;
3 to less than 4: 5%;
4 to less than 5: 5%;
5 to less than 6: 6%;
6 to less than 7: 6%.
Sector: New-issue CMBSs;
Weighted average life in years for ABSs:
0 to less than 1: 15%;
1 to less than 2: 15%;
2 to less than 3: 15%;
3 to less than 4: 15%;
4 to less than 5: 15%;
5 to less than 6: [A];
6 to less than 7: [A].
Sector: Legacy CMBSs;
Weighted average life in years for ABSs:
0 to less than 1: 15%;
1 to less than 2: 15%;
2 to less than 3: 15%;
3 to less than 4: 15%;
4 to less than 5: 15v;
5 to less than 6: [B];
6 to less than 7: [B].
Source: GAO presentation of information gathered from the FRBNY Web
site.
Note: For ABSs benefiting from a government guarantee with average
lives of 5 years and beyond, haircuts will increase by 1 percentage
point for every 2 additional years (or portion thereof) of average
life at or beyond 5 years. For all other ABSs with average lives of 5
years and beyond, haircuts will increase by 1 percentage point for
each additional year (or portion thereof) of average life at or beyond
5 years.
[A] For newly issued CMBSs with average lives beyond 5 years,
collateral haircuts increase by 1 percentage point of par for each
additional year of average life. No newly issued CMBS may have an
average life of more than 10 years.
[B] For legacy CMBSs with average lives beyond 5 years, haircuts
increase by 1 percentage point of par for each additional year of
average life. No legacy CMBS may have an average life of more than 10
years.
[End of table]
Loan Repayment Process:
As discussed earlier, the Bank of New York Mellon holds the collateral
throughout the life of the loan. As the collateral securities generate
cash flows, the Bank of New York Mellon makes all principal and
interest payments to FRBNY on behalf of the TALF borrower. The
borrower may earn returns from the collateral after all loan
obligations have been satisfied. As shown in figure 2, any returns
that the collateral assets earn beyond the required principal and
interest payments is delivered to the borrower via the TALF agent
after all monthly loan payment obligations are met.[Footnote 26] FRBNY
retains a portion of the interest that is calculated using Overnight
Indexed Swap (OIS) rates plus 25 basis points (approximately the
FRBNY's cost of funds).[Footnote 27] The remaining part of the TALF
loan interest payment is transferred to TALF LLC, which would use the
funds to purchase surrendered collateral before accessing funds from
Treasury. This accumulated interest from TALF loans is held in an
account called the cash collateral account, and the Bank of New York
Mellon is authorized to invest these funds on behalf of TALF LLC to
earn interest income.
Figure 2: TALF Loan Origination and Repayment Process:
[Refer to PDF for image: illustration]
TALF Borrower:
* Assets pledged as collateral to Custodian and administrator, Bank of
New York through TALF agent;
(flow of TALF loan from FRBNY TALF account through TALF agent);
* Principal and interest payment to FRBNY TALF account through TALF
agent;
* Returns from collateral after meeting loan payment obligations[A]
through TALF agent.
Custodian and administrator, Bank of New York Mellon:
* FRBNY TALF account: principal and interest received;
* TALF loan to FRBNY;
* Portion of interest payment retained by FRBNY;
* Excess interest paid to TALF LLC.
Source: GAO.
[A] If the collateral is not generating enough interest to satisfy the
loan obligation, the borrower must eventually make up the difference
or surrender collateral.
[End of figure]
Process If Borrower Walks Away from the Loan:
TALF offers nonrecourse loans, which allow borrowers to walk away--or
stop paying a loan--with no personal exposure for the unpaid portion
of the debt.[Footnote 28] In this case, borrowers would surrender
their collateral through a TALF agent, which would submit a collateral
surrender form to FRBNY. Within 10 days of receiving the surrender
notice, the Bank of New York Mellon cancels the outstanding balance of
the loan and transfers the related collateral to FRBNY. FRBNY has the
option to sell the collateral to TALF LLC (see figure 3) at a price
equal to the then outstanding principal amount of the TALF loan plus
accrued but unpaid interest. This process has several steps:
* FRBNY sends a purchase notice to TALF LLC. To purchase surrendered
assets, TALF LLC first uses funds that have been accumulating in the
cash collateral account (as of December 31, 2009, the account
contained approximately $198 million).[Footnote 29]
* When these funds are exhausted, TALF LLC borrows money to purchase
any surrendered assets. Its first source of borrowed funds is
Treasury, which has committed up to $20 billion in TARP funds to the
TALF program for this purpose. To obtain these funds, TALF LLC must
submit a request to Treasury (as the subordinate lender) one business
day prior to the desired funding date. Treasury's loan rate is equal
to the 1-month London Interbank Offered Rate (LIBOR) rate plus 300
basis points.
* If the $20 billion TARP loan commitment is fully exhausted, TALF LLC
must ask FRBNY for a loan to purchase any additional surrendered
assets. FRBNY has committed to provide up to $180 billion to TALF LLC
for this purpose. TALF LLC must submit a borrowing request to FRBNY
(as senior lender) one business day prior to the desired funding date.
The FRBNY's:
* loan rate is equal to the 1-month LIBOR rate plus 100 basis points.
The principal on FRBNY's loan would be repaid before the principal of
the $20 billion TARP loan.
Figure 3: Funding Required if a Borrower Walks Away from a Loan:
[Refer to PDF for image: illustration]
TALF borrower:
form surrendering collateral[A];
goes to:
Custodian, BNYM:
Collateral, goes to:
FRBNY:
Collateral and senior loan;
goes to:
TALF LLC:
Funds for purchasing collateral, in order of use:
- excess accumulated interest;
- $20 billion sub loan (TARP);
- $180 billion senior loan (FRBNY);
Repayment goes to FRBNY;
Loan repayment[B] goes to TARP;
Sunloan from TARP to TALF LLC.
Source: GAO.
[A] The form goes through the TALF agent and is provided to the
custodian and FRBNY simultaneously.
[B] To be supplemented by 90 percent of any remaining TALF LLC assets
after meeting all obligations, as noted in the order of payment shown
below.
[End of figure]
Alternatively, if a TALF borrower stops payment on a loan and does not
submit a collateral surrender form, under certain circumstances FRBNY
has rights to seize the collateral.
Management of TALF LLC:
With the Federal Reserve's and Treasury's agreement, TALF LLC may
dispose of assets it has acquired. Agency officials told us that there
were no formal guidelines on when to sell any acquired assets and that
the decision to sell would be made on a case-by-case basis as is
necessary given the differences in types of assets that could be
acquired by TALF LLC. However, they added that factors such as market
rates and the nature of the underlying collateral would likely play a
role in any decision to hold or sell assets purchased by TALF LLC.
According to Treasury officials, because the Federal Reserve's
monetary policy considerations may not align with Treasury's
investment interests in the TALF program, Treasury plans to obtain
independent advice on the disposition of investments.
TALF LLC has certain sources of cash inflows, such as the excess
interest from TALF loans, interest paid from permitted investments,
principal and interest paid on ABS holdings, and proceeds from
possible asset sales of ABS holdings. These inflows are distributed
according to the order defined in an agreement among Treasury, FRBNY,
the Bank of New York Mellon, and TALF LLC. The payment order is as
follows:
* Pay TALF LLC expenses.
* Fund expense reimbursement account up to $15 million.
* Repay outstanding principal on any FRBNY senior loans to TALF LLC.
* Fund the cash collateral account up to the senior loan commitment
(currently $180 billion).
* Repay outstanding principal on any Treasury loans.
* Repay FRBNY loan interest.
* Repay Treasury loan interest.
* Repay any other secured obligations that may arise that have not
been specified yet by the agencies.
* Pay Treasury and FRBNY (90 percent and 10 percent, respectively) any
residual amounts but only after the above requirements are satisfied.
As of December 2009, cash inflows have been used to pay TALF LLC
expenses and fund the expense reimbursement and cash collateral
accounts.
TALF Includes Features That Mitigate Potential Losses, but CMBSs Could
Pose Greater Risks Than Other Asset Classes:
Under some scenarios, TALF borrowers may have economic incentives to
stop payment on their loans and surrender collateral. Treasury and
FRBNY assessments, along with our analyses, however, suggest that a
number of factors reduce the likelihood of this occurring. First,
certain TALF features are designed to protect TARP funds and also
limit taxpayer exposure. Specifically, FRBNY officials told us that
risks in TALF are managed based on four pillars: credit protection,
credit ratings, FRBNY due diligence, and market discipline. FRBNY also
sought market perspectives on the level of returns from TALF-eligible
securities that would be required to encourage market participation
while also ensuring proper due diligence by TALF participants, which
in turn reduces the risks to taxpayers. Second, most ABSs issued since
the credit crisis began contain features such as increased levels of
subordination and overcollateralization that reduce the likelihood
that TALF borrowers will stop payment on their loans. Third, Treasury
and FRBNY analyses project minimal, if any, likelihood that TARP funds
will be used for TALF-related purchases, and Treasury currently
projects a profit from TALF. While TALF poses minimal risks to TARP
even in adverse market conditions, our analyses showed that CMBSs held
as collateral as of September 2009 potentially pose higher risks than
ABSs and under adverse conditions losses could exceed $500 million.
TALF presents a range of other taxpayer risks beyond those presented
to TARP funds. Such risks include the risk that FRBNY might not
identify instances of material noncompliance with program requirements
by TALF participants. However, because of statutory audit limitations
on GAO, this review has been limited to presenting only descriptive
information about FRBNY's role in TALF and we cannot evaluate FRBNY's
TALF operations.
Some Scenarios Could Provide TALF Borrowers with Economic Incentives
to Stop Payment on Their Loans:
As of January 8, 2010, there have been no collateral surrenders by any
TALF borrowers, but in some instances, a TALF borrower could have an
economic incentive to stop payment on a loan and surrender the
underlying collateral.[Footnote 30] A number of scenarios could result
in a borrower walking away from a loan. For example, the collateral
could lose value so that the loan amount exceeded the value of the
collateral. Or, the expected returns from the collateral could be less
than the cost of financing the loan. Also, interest rates could rise
across the board, decreasing the market value of the collateral or
making refinancing more difficult.
A borrower can lose equity in the collateral if the collateral's value
falls below the outstanding loan balance. As discussed earlier, TALF's
established haircuts determine the amount of equity borrowers have in
their collateral. This equity represents the amount of money that a
TALF borrower would lose by surrendering the collateral and not
repaying the loan.[Footnote 31] For example, if the ABSs a borrower
seeks to use as collateral on a TALF loan were initially valued at
$100,000, a haircut of 10 percent would provide the TALF borrower with
a $90,000 loan and require $10,000 of its own funds to acquire the
securities. This $10,000 represents the borrower's equity in the
securities. If the value of the collateral were to decline--such that
the borrower could sell the securities only for some amount less than
$90,000--and especially if it declined dramatically, the borrower
could decide to cut further losses by stopping payments on the loan
and surrendering the collateral. The borrower is particularly likely
to make this choice if such a situation occurs at the point when the
loan matures.
As TALF currently works, TALF borrowers earn the difference between
the ABS's return and the cost of borrowing from FRBNY on the TALF
loan, multiplied by the inverse of the borrower's percent equity
stake. However, if the returns on TALF collateral are less than
initially anticipated, the TALF loan costs could exceed them,
providing another incentive to stop making payments. This situation
would be most likely to develop if the underlying loans in the
securities defaulted or otherwise failed to meet the terms of the
original loan agreement. In this situation, if the returns were less
than the total cost of the TALF loan--the interest and principal due--
a borrower might stop payment on the loan or surrender the collateral
to FRBNY.
Investors might also choose to surrender collateral at the maturity of
their TALF loan if overall interest rates on credit increased
significantly above levels available at the time the underlying
securities were issued. A large increase in interest rates would lower
the market value of the securities, especially for those with a fixed
interest rate, because future cash flows would be worth less. In this
scenario, a borrower could wind up owing more to FRBNY at maturity
than the securities were worth. The borrower would need to raise funds
at higher interest rates to pay back the TALF loan, but the collateral
would be worth less than the new loan, potentially making it difficult
to find a lender. This situation would provide an economic incentive
for the borrower to surrender the collateral and walk away from the
loan. Nevertheless, many of the longer-term securities pledged as
collateral for TALF loans have floating interest rates. Generally,
floating rate securities do not decline in value when interest rates
increase because the interest rates on the securities also increase.
TALF borrowers will not necessarily stop payment on a loan even with
one or more of these economic incentives. For example, a market
participant and a Treasury official told us that TALF borrowers wanted
to avoid "reputation risk" by not walking away from their loans--even
if it might be in their financial interest to do so. By continuing to
pay on their loans, even at some loss, borrowers could protect their
reputations. Other market participants we spoke with, however, did not
mention concerns about reputation, though they thought it was unlikely
that TALF borrowers would stop payment on their loans.
Moreover, FRBNY and Treasury officials believe that the most important
disincentive for borrowers to stop payment on a TALF loan and
surrender collateral is maintaining a positive return based on the
difference between the cost of the TALF loan and the return of the
underlying ABS collateral. Even if the value of the collateral
declines and removes the borrower's equity, these officials stated
that borrowers were not likely to walk away from the loan if they were
still receiving positive cash flow from the asset. While this
assumption may be reasonable, it is conceivable that investors subject
to mark-to-market accounting might choose to walk away from an ABS
that was still paying the required scheduled interest and principal
payments but had lost sufficient market value. Investors would be most
likely to walk away if they owed significantly more on the loan than
the current value of the asset so that walking away from the asset
would have positive implications for the borrower's reported
profitability. As of January 8, 2010, no TALF borrowers had
surrendered collateral to FRBNY. Nevertheless, because the behavior of
individual borrowers is difficult to predict, it remains unclear
whether and why borrowers might stop payment on TALF loans based on
their own investment strategies and other objectives.
TALF Has Several Features That Likely Protect Taxpayers:
Certain TALF features help protect TALF funds and also taxpayer
exposures through TARP. We discussed these protections with FRBNY
officials but did not evaluate FRBNY compliance with them because of
the limitation of our statutory audit authority. FRBNY officials told
us that risks in TALF are managed based on four pillars: credit
protection, credit ratings, FRBNY due diligence, and market discipline.
* Credit protection is provided primarily by the borrower's equity in
the security (set by the haircuts) and a portion of the interest rate
on TALF loans that provides accumulated excess interest in TALF LLC.
FRBNY officials said that haircuts were designed to approximate
multiples of stressed historical impairment rates for ABSs. The size
of the haircut is important, because if it is set too low, the
borrower will have less equity in the collateral and, in the
circumstances that we have discussed, could have an incentive to walk
away. The excess interest from TALF loans accumulated in TALF LLC also
protects taxpayers, because this money would be used before TARP funds
to purchase collateral surrendered from unpaid loans. FRBNY officials
said that such features were designed to ensure that the haircuts and
excess interest would result in no credit losses for Treasury or the
Federal Reserve.
* The Federal Reserve requires that TALF collateral be rated AAA or
its equivalent by two of the nationally recognized statistical rating
organizations that it deems eligible to provide credit ratings for
TALF, among other requirements for credit ratings.[Footnote 32]
Collateral that is offered that has a lower rating is not eligible for
TALF. The rating requirement helps ensure that the securities TALF
accepts as collateral present minimal credit risks.
* FRBNY due diligence serves as another pillar of taxpayer protection.
As discussed earlier, FRBNY, with the support of its collateral
monitors, reviews the credit risks related to individual assets FRBNY
might consider accepting as TALF collateral. In addition, for legacy
CMBSs, FRBNY reserves the right to reject any collateral and has not
disclosed its selection criteria to reduce the likelihood that only
the poorest-performing collateral is put forward for TALF loans.
* The final pillar, market discipline, includes the TALF borrowers'
due diligence conducted on the risks related to the underlying
collateral, given their equity in the collateral, as set by the
haircut. Such discipline includes reviewing the prospectuses and
understanding how the structure of the underlying securities impacts
its overall risks. Investors purchasing ABSs would generally review
the terms of the security, the anticipated risks, and the likely
return. In addition, TALF borrowers help to facilitate price discovery.
These design features are intended to help ensure that TALF borrowers
hold equity in the underlying TALF collateral, that such collateral is
highly rated, and that TALF borrowers carry out the same due diligence
on TALF collateral that they would conduct on any other security.
Rates of Return for TALF Borrowers on TALF Collateral Have Declined
for Most TALF Collateral from Their Highs Earlier in 2009:
As noted earlier, TALF borrowers earn the difference between the ABS's
coupon rate of return and the cost of borrowing from FRBNY on the TALF
loan. According to FRBNY officials, they, along with officials from
the Federal Reserve and Treasury, were aware of the importance of
striking a balance between achieving returns on equity that would
encourage program participation in the stressed market conditions when
TALF was announced, while also ensuring that investors had incentives
for due diligence and that the program would be less attractive during
times of less market stress.[Footnote 33] If borrowers saw an
opportunity to earn excessive returns due to a poorly designed
program, borrowers might not conduct appropriate due diligence on the
underlying securities, which could put taxpayers at risk. Moreover,
because the loans are nonrecourse much of the borrowers' risk is
shifted to FRBNY and Treasury, while most of the earnings potential
remains with borrowers. FRBNY gathered information on the rates of
return that would entice potential TALF participants by surveying
market participants about expected returns for TALF-eligible asset
classes.[Footnote 34]
To understand the changes in returns on equity over time and the
reasonableness of those returns from the perspective of helping to
ensure that taxpayers were not subsidizing high returns, we analyzed
fluctuations in returns on TALF eligible collateral from March 2009
through September 2009. As seen in figure 4, returns generally
decreased for select classes of TALF-eligible collateral between the
first TALF operation in March 2009 and September 2009, with limited
exceptions.[Footnote 35]
Figure 4: Overall TALF Borrower Returns at Issuance for Select Asset
Classes Accepted as TALF Collateral Generally Decreased from March
through September 2009:
[Refer to PDF for image: series of line graphs]
Prime Auto Loans: All;
March: 21%;
April: 18%;
May: 14%;
June: 7%;
July: 6%;
September: 2%;
Change in return: -19 percentage points.
Prime Auto Loans: About 1 year;
March: 13%;
April: 7%;
May: 6%;
June: 6%;
July: 1%;
September: -5%;
Change in return: -18 percentage points.
Prime Auto Loans: About 2 years;
March: 19%;
April: 15%;
May: 9%;
June: 8%;
July: 7%;
September: 1%;
Change in return: -18 percentage points.
Prime Auto Loans: Greater than 2 years;
March: 37%;
April: 32%;
May: 23%;
June: 19%;
July: 11%;
September: 5%;
Change in return: -32 percentage points.
Auto Leases: All;
May: 11%;
June: 7%;
September: -3%;
Change in return: -14 percentage points.
Auto Leases: A-2;
May: 10%;
June: 5%;
September: -5%;
Change in return: -15 percentage points.
Auto Leases: A-3;
May: 11%;
June: 9%;
September: 0;
Change in return: -11 percentage points.
Auto Leases: A-4;
May: 17%;
June: 8%;
September: 0;
Change in return: -16 percentage points.
Credit Cards: Prime;
March: 14%;
April: 18%;
May: 10%;
June: 7%;
July: 6%;
August: 7%;
September: 5%;
Change in return: -8 percentage points.
Credit Cards: Subprime;
April: 25%;
May: 15%;
June: 15%;
August: 10%;
September: 18%;
Change in return: -7 percentage points.
Private student loans: Lapsed option;
May: 43%;
July: 26%;
August: 18%;
Change in return: -25 percentage points.
Private student loans: Exercise option;
May: 31%;
July: 14%;
August: 13%;
Change in return: -18 percentage points.
Source: GAO analysis of TALF prospectuses, Standard and Poor‘s and
Moody‘s Investors Service reports, and Federal Reserve data.
Note: The percentage point change in return for each asset class is
calculated from the first month of issuance to the most recent month
and may not appear to total correctly due to rounding.
[End of figure]
Most asset classes demonstrated a significant decline in returns, from
a peak of 43 percent among the first student loan ABS accepted under
TALF to lows of -5 percent for 1-year prime auto tranches[Footnote 36]
and 2-year auto leases, suggesting that if investors had used TALF to
finance the purchase of these auto-related TALF eligible securities
they would have locked in losses.[Footnote 37] However, in a sign that
health may be returning to the ABS markets, no TALF borrowings were
needed to finance the purchase of these negative return securities as
the issuances were fully funded by non-TALF investors. The most
dramatic decreases in returns have been in the auto loan tranches with
longer-term maturities and private student loan ABSs. The average
expected return for TALF borrowers that used prime auto loan ABSs as
TALF loan collateral declined by nearly a fifth after March 2009. All
of the TALF-eligible private student loan ABS transactions that were
completed between May 2009 and August 2009 had a unique option feature
that significantly lessened investor's expected returns on equity.
[Footnote 38] Additionally, the September 2009 increase in return for
subprime credit cards was primarily due to the unique structure of an
issuance by a large subprime issuer.[Footnote 39]
The trend of decreasing returns on equity in the overall market for
most asset classes indicates that the returns required to attract
investors have decreased since TALF's implementation, as ABS investors
perceive the assets to be less risky. In addition, the trend
demonstrates that taxpayer subsidies to TALF borrowers have not, over
the course of the program, provided what might be considered excessive
returns. Although returns were high in the beginning, they diminished
once the ABS markets started to revive and overall perceived risk
began to decrease. Moreover, the trend toward negative TALF returns on
equity have coincided with a drop in TALF participation by TALF
borrowers in those asset classes as the rate on the securities' bonds
falls below the loan rate charged by the FRBNY. This occurrence is
consistent with FRBNY's intention for TALF financing to become less
attractive as the ABS market improves and can be viewed as indicating
normalization in the market, since investors can purchase such ABSs
without TALF funding.
New Securitizations Have Generally Been Structured with More Credit
Protections since the Credit Crisis Began:
Certain elements of the securities themselves also reduce the risk of
loss to TALF and the taxpayer. Among these features are credit
enhancements, which have increased since the credit crisis began in
the second half of 2008. Credit enhancements are features in the
structure of a securitization that protect investors from losses due
to defaults on the underlying loans in the securities. Two main forms
of credit enhancement include subordination, which helps ensure that
more highly rated tranches in a security receive priority of payment,
and overcollateralization, which ensures that funds are available if a
borrower stops paying or other credit problems develop with the
underlying loans. For additional details on the various types of
credit enhancement, see appendix V.
Credit enhancements offer several benefits. First, they provide a
cushion against losses, making it less likely that TALF borrowers will
decide not to repay their loans and surrender the collateral because
of credit performance problems in the ABS or CMBS markets. Second,
credit enhancements reduce the probability that TALF LLC will suffer
principal losses on surrendered ABSs and CMBSs from unpaid loans. For
this reason, credit enhancements also provide TALF LLC with an
incentive to decide to hold surrendered collateral to maturity. This
reduces potential losses to TARP funds, which would finance TALF LLC's
purchase of such ABS and CMBS.
We reviewed the credit enhancements on every TALF-eligible ABS issued
between the program's initiation in March 2009 and September 2009 and
compared them with credit enhancements on ABSs by the same issuer
between 2006 and 2008 to identify any changes. As shown in figure 5,
the level of enhancement for every TALF-eligible asset class increased.
Figure 5: Increases in Levels of Credit Enhancement on ABSs Since the
Credit Crisis Began, March 2009 through September 2009:
[Refer to PDF for image: series of horizontal bar graphs]
TALF-eligible ABS:
Prime auto loans:
Percentage of credit enhancement in ABS: TALF: 10.5%;
Percentage of credit enhancement in ABS: Precrisis: 7.9%;
Percentage increase (rounded): 33%.
Subprime auto loans:
Percentage of credit enhancement in ABS: TALF: 44.3%;
Percentage of credit enhancement in ABS: Precrisis: 43.6%;
Percentage increase (rounded): 2%.
Prime credit cards:
Percentage of credit enhancement in ABS: TALF: 18.2%;
Percentage of credit enhancement in ABS: Precrisis: 13.6%;
Percentage increase (rounded): 34%.
Subprime credit cards:
Percentage of credit enhancement in ABS: TALF: 27.5%;
Percentage of credit enhancement in ABS: Precrisis: 22.6%;
Percentage increase (rounded): 22%.
Private student loans:
Percentage of credit enhancement in ABS: TALF: 35.3%;
Percentage of credit enhancement in ABS: Precrisis: 8.9%;
Percentage increase (rounded): 295%.
Auto leases:
Percentage of credit enhancement in ABS: TALF: 25.3%;
Percentage of credit enhancement in ABS: Precrisis: 9.7%;
Percentage increase (rounded): 162%.
Insurance premium finance:
Percentage of credit enhancement in ABS: TALF: 11.0%;
Percentage of credit enhancement in ABS: Precrisis: 9.0%;
Percentage increase (rounded): 19%.
Motorcycle:
Percentage of credit enhancement in ABS: TALF: 23.0%;
Percentage of credit enhancement in ABS: Precrisis: 10.3%;
Percentage increase (rounded): 124%.
Commercial fleet leases:
Percentage of credit enhancement in ABS: TALF: 12.5%;
Percentage of credit enhancement in ABS: Precrisis: 12.1%;
Percentage increase (rounded): 3%.
Equipment:
Percentage of credit enhancement in ABS: TALF: 11.6%;
Percentage of credit enhancement in ABS: Precrisis: 8.2%;
Percentage increase (rounded): 42%.
Mortgage servicing advances:
Percentage of credit enhancement in ABS: TALF: 22.7%;
Percentage of credit enhancement in ABS: Precrisis: 18.9%;
Percentage increase (rounded): 20%.
Nonauto floor plan[A]:
Percentage of credit enhancement in ABS: TALF: 24.0%;
Percentage of credit enhancement in ABS: Precrisis: 9.0%;
Percentage increase (rounded): 167%.
Source: GAO analysis of TALF prospectuses and Moody‘s Investors
Service and Standard and Poor‘s reports.
[A] Nonauto floor plan loans are made to finance nonautomotive
inventory such as recreational vehicles, boats, and industrial
equipment.
[End of figure]
In particular, credit enhancements on ABSs backed by private student
loans, nonauto floor plans, auto leases, and motorcycle loans at least
doubled after the credit crisis.[Footnote 40] In other words, from the
perspective of protecting TARP funds, taxpayers would receive more
than double the amount of credit protection for these particular asset
classes compared with these issuers' ABSs before the financial crisis.
This increase was a combination of market demands and credit rating
agencies' more stringent requirements for achieving AAA ratings on
many ABSs. TALF agents and a TALF issuer confirmed that credit
enhancements had increased since the onset of the financial crisis,
even for non-TALF issuances, and noted that requirements from the
credit rating agencies had contributed to the increases.
TALF Poses Minimal Risks to TARP Even in Adverse Market Conditions;
CMBS Risks are Potentially Higher:
While TARP funds designated for TALF are exposed to potential loss if
TALF LLC uses them to purchase ABSs or CMBSs used as collateral for
unpaid TALF loans, Treasury and FRBNY analyses suggest that the risks
of loss are minimal. According to the Federal Reserve's analysis, the
accumulated excess interest from TALF loans will likely cover any ABS
or CMBS purchases for TALF LLC, and Treasury will not need to provide
any TARP funds for such purchases. Accordingly, the total expenditures
from TARP funds would include only the $100 million placed in TALF
LLC, which is in the form of a loan that would be repaid. According to
the terms of the agreement between FRBNY and Treasury, Treasury will
receive 90 percent of the monies accumulated in TALF LLC when the
program expires. In particular, if TALF LLC has not purchased any
collateral, the excess interest that has accumulated--along with
interest income from investing such money--will go mainly to Treasury;
therefore, Treasury could potentially gain from its commitment to TALF
if losses were minimal.
Treasury Currently Expects to Earn a Profit on TALF:
Treasury hired a contractor to provide an estimate of potential losses
to TARP funds from TALF.[Footnote 41] According to one analysis
conducted by the contractor, any losses to TARP are likely to be far
below the $20 billion that has been set aside for TALF and in fact are
unlikely to exceed about $190 million. Any assets purchased by TALF
LLC would first be paid for with the excess interest and related
interest income that had accumulated in TALF LLC, which totaled almost
$200 million at the end of December 2009.[Footnote 42] According to
the contractor, the analysis of potential loss of TARP funds did not
include this excess interest, so the total estimated losses could be
largely offset by accumulated interest. In fact, the contractor and
Treasury officials said that a subsequent analysis that included
projections for the accumulation of excess interest--and is updated on
a quarterly basis--currently projects more than $1 billion in profit
related to Treasury's TALF exposure.
To assess the reasonableness of Treasury's position that TALF may
actually earn money rather than expose the taxpayer to any losses, we
reviewed the Treasury contractor's model--which is central to
estimating losses for the various asset classes accepted in TALF. We
found this model appeared to incorporate generally reasonable loss
assumptions for the three asset classes that we reviewed and that
comprise the largest portion of the TALF portfolio: credit cards, auto
loans, and student loans. That is, many of the loss estimates were
fairly conservative when compared to recent historical results for
these specific assets. For each asset class, the contractor estimated
expected loss percentages based on its own research and analysis,
which was used to generate total-loss estimates. The model calculates
total losses for each TALF borrower on each asset held in the TALF
portfolio. The portfolio includes current TALF collateral and
projections for future TALF collateral that borrowers will use for
TALF loans, based on information the contractor received from
Treasury. Potential Treasury losses from these various scenarios were
calculated by taking the total loss for the TALF borrower and
subtracting the equity that the TALF borrower holds in ABSs. Any
difference is considered a loss, first to TALF LLC and potentially to
Treasury (through TARP funds) if loans extended to TALF LLC for its
purchase of TALF collateral are not repaid. While some of the
scenarios generated by the model estimated losses from asset classes
that did not have government guarantees, none of the possible
scenarios estimated losses to Treasury from CMBSs.
Uncertainty in the Commercial Real Estate Market Could Create the
Potential for Losses if Conditions Deteriorate:
While Treasury has determined that CMBS-related losses are unlikely
for a number of reasons, our analysis shows that if the commercial
real estate markets were to be affected similarly to real estate
markets in 2008, the potential for loss exists under a worst-case
scenario. Treasury and its contractor said that CMBS losses were
unlikely for a number of reasons ranging from the relatively large
haircuts (at least 15 percent) required on CMBS loans to the level of
credit enhancement associated with CMBSs accepted for TALF. However,
due in part to the recently weak economy, commercial real estate
continues to undergo price deterioration that potentially poses risks
to the TALF legacy CMBS portfolio and could lead to increased
delinquencies and defaults on commercial real estate mortgages. For
example, the potential risks that CMBSs pose to TALF, and thus to
TARP, are underscored by the fact that 63 percent of TALF's CMBS
portfolio was underwritten in 2006 and 2007, when underwriting
standards were at their worst. Moreover, as figure 6 shows, commercial
real estate prices have been falling since early 2008, following the
deterioration in the overall U.S. economy, and shortly thereafter CMBS
delinquencies began to rise sharply. The Federal Reserve and Treasury
have continued to note their ongoing concerns about this segment of
the market.
Figure 6: Commercial Real Estate Prices and CMBS Delinquencies from
January 2004 through October 2009:
[Refer to PDF for image: multiple line graph]
Year: 2004;
January:
Moody's/REAL Price Index: 121.08;
Percentage delinquency: 1.29.
February:
Moody's/REAL Price Index: 123.84;
Percentage delinquency: 1.32.
March:
Moody's/REAL Price Index: 126.04;
Percentage delinquency: 1.33.
April:
Moody's/REAL Price Index: 126.91;
Percentage delinquency: 1.19.
May:
Moody's/REAL Price Index: 127.52;
Percentage delinquency: 1.24.
June:
Moody's/REAL Price Index: 127.78;
Percentage delinquency: 1.22.
July:
Moody's/REAL Price Index: 132.18;
Percentage delinquency: 1.15.
August:
Moody's/REAL Price Index: 133.38;
Percentage delinquency: 1.04.
September:
Moody's/REAL Price Index: 134.66;
Percentage delinquency: 1.01.
October:
Moody's/REAL Price Index: 137.28;
Percentage delinquency: 1.01.
November:
Moody's/REAL Price Index: 138.54;
Percentage delinquency: 1.
December:
Moody's/REAL Price Index: 140.05;
Percentage delinquency: 0.95.
Year: 2005;
January:
Moody's/REAL Price Index: 145.45;
Percentage delinquency: 0.96.
February:
Moody's/REAL Price Index: 146.7;
Percentage delinquency: 0.93.
March:
Moody's/REAL Price Index: 150.36;
Percentage delinquency: 0.9.
April:
Moody's/REAL Price Index: 151.29;
Percentage delinquency: 0.85.
May:
Moody's/REAL Price Index: 154.28;
Percentage delinquency: 0.78.
June:
Moody's/REAL Price Index: 156.26;
Percentage delinquency: 0.77.
July:
Moody's/REAL Price Index: 158.31;
Percentage delinquency: 0.75.
August:
Moody's/REAL Price Index: 161.62;
Percentage delinquency: 0.7.
September:
Moody's/REAL Price Index: 162.76;
Percentage delinquency: 0.68.
October:
Moody's/REAL Price Index: 161.63;
Percentage delinquency: 0.65.
November:
Moody's/REAL Price Index: 162.89;
Percentage delinquency: 0.64.
December:
Moody's/REAL Price Index: 160.64;
Percentage delinquency: 0.56.
Year: 2006;
January:
Moody's/REAL Price Index: 166.57;
Percentage delinquency: 0.55.
February:
Moody's/REAL Price Index: 169.36;
Percentage delinquency: 0.54.
March:
Moody's/REAL Price Index: 169.19;
Percentage delinquency: 0.48.
April:
Moody's/REAL Price Index: 166.1;
Percentage delinquency: 0.49.
May:
Moody's/REAL Price Index: 168.32;
Percentage delinquency: 0.49.
June:
Moody's/REAL Price Index: 170.24;
Percentage delinquency: 0.44.
July:
Moody's/REAL Price Index: 168.1;
Percentage delinquency: 0.41.
August:
Moody's/REAL Price Index: 168.24;
Percentage delinquency: 0.39.
September:
Moody's/REAL Price Index: 168.54;
Percentage delinquency: 0.38.
October:
Moody's/REAL Price Index: 170.52;
Percentage delinquency: 0.37.
November:
Moody's/REAL Price Index: 170.78;
Percentage delinquency: 0.34.
December:
Moody's/REAL Price Index: 174.08;
Percentage delinquency: 0.29.
Year: 2007;
January:
Moody's/REAL Price Index: 179.25;
Percentage delinquency: 0.29;
February:
Moody's/REAL Price Index: 183.47;
Percentage delinquency: 0.28;
March:
Moody's/REAL Price Index: 185.2;
Percentage delinquency: 0.25.
April:
Moody's/REAL Price Index: 186.37;
Percentage delinquency: 0.24;
May:
Moody's/REAL Price Index: 185.59;
Percentage delinquency: 0.23.
June:
Moody's/REAL Price Index: 187.25;
Percentage delinquency: 0.24.
July:
Moody's/REAL Price Index: 188.17;
Percentage delinquency: 0.22.
August:
Moody's/REAL Price Index: 191.11;
Percentage delinquency: 0.24.
September:
Moody's/REAL Price Index: 188.89;
Percentage delinquency: 0.24.
October: Prices peaked in October 2007:
Moody's/REAL Price Index: 191.87;
Percentage delinquency: 0.23.
November:
Moody's/REAL Price Index: 191.4;
Percentage delinquency: 0.27.
December:
Moody's/REAL Price Index: 188.51;
Percentage delinquency: 0.29.
Year: 2008;
January:
Moody's/REAL Price Index: 187.31;
Percentage delinquency: 0.31.
February:
Moody's/REAL Price Index: 191.24;
Percentage delinquency: 0.32.
March:
Moody's/REAL Price Index: 186.92;
Percentage delinquency: 0.35.
April:
Moody's/REAL Price Index: 181.23;
Percentage delinquency: 0.36.
May:
Moody's/REAL Price Index: 174.97;
Percentage delinquency: 0.43.
June:
Moody's/REAL Price Index: 169.28;
Percentage delinquency: 0.46.
July:
Moody's/REAL Price Index: 169.96;
Percentage delinquency: 0.48.
August:
Moody's/REAL Price Index: 169.74;
Percentage delinquency: 0.5.
September:
Moody's/REAL Price Index: 173.92;
Percentage delinquency: 0.54.
October:
Moody's/REAL Price Index: 169.76;
Percentage delinquency: 0.6.
November:
Moody's/REAL Price Index: 164;
Percentage delinquency: 0.75.
December:
Moody's/REAL Price Index: 160.46;
Percentage delinquency: 0.95.
Year: 2009;
January:
Moody's/REAL Price Index: 151.58;
Percentage delinquency: 1.16.
February:
Moody's/REAL Price Index: 150.63;
Percentage delinquency: 1.32.
March:
Moody's/REAL Price Index: 148.07;
Percentage delinquency: 1.59.
April:
Moody's/REAL Price Index: 135.31;
Percentage delinquency: 2.27.
May:
Moody's/REAL Price Index: 125.04;
Percentage delinquency: 2.67.
June:
Moody's/REAL Price Index: 123.82;
Percentage delinquency: 3.02.
July:
Moody's/REAL Price Index: 117.56;
Percentage delinquency: 3.23.
August:
Moody's/REAL Price Index: 114.06;
Percentage delinquency: 3.64.
September:
Moody's/REAL Price Index: 109.6;
Percentage delinquency: 4.01.
October:
Moody's/REAL Price Index: 107.98;
Percentage delinquency: 4.47.
November:
Moody's/REAL Price Index: 109.1;
Percentage delinquency: 4.9.
As of November 2009, prices show a drop of 43% since peak. November
was the first month in which prices increased since September 2008.
Note: Moody‘s/REAL Commercial Property Price Index is a report showing
the change in commercial real estate asset sales. Moody‘s CMBS
Delinquency Tracker is a monthly report showing delinquency rates for
commercial mortgage-backed securities.
Source: GAO presentation of Moody‘s/REAL Commercial Property Price
Index and Moody‘s CMBS Delinquency Tracker data.
[End of figure]
In addition, prices on CMBSs experienced volatility between May 2008
and November 2009. Even the highest credit quality AAA legacy CMBSs
that were used as collateral for TALF loans during the third quarter
of 2009 had dropped by one third, on average, between May and November
2008 during the most severe period of the credit crisis. While prices
have recovered since the expansion of TALF to include CMBSs in the
second quarter of 2009, market observers project that the sector will
continue to perform poorly into 2010.
In light of the ongoing distress in the commercial real estate market,
we analyzed the prices and values of 99 percent of the CMBS collateral
backing loans made by FRBNY during the third quarter of 2009. We
compared the prices at the time the loans were made with the lowest
prices in November 2008, a period of extreme stress for CMBSs. Our
analysis revealed that if legacy CMBSs accepted as TALF collateral as
of September 2009 reached market values equivalent to November 2008
levels, about 88 percent of such collateral would have fallen to
levels at which the TALF borrower's equity would be eliminated.
[Footnote 43] Moreover, more than $3.5 billion owed by TALF borrowers--
or about 85 percent of the total value of TALF legacy CMBS loans--
would have negative equity in this scenario. This extreme market
stress scenario would result in a loss in market value on the part of
these TALF borrowers of nearly $1.2 billion. The haircut investment
for these borrowers totals $665 million, providing significant
economic incentive to walk away, which would result in a worst-case
loss of about $500 million for TALF.
While this worst-case scenario provides useful information for
Treasury to consider as it monitors risk associated with TALF, we
agree with the Federal Reserve and Treasury that there are a number of
factors that affect whether such losses would be realized even in this
adverse scenario. First, as discussed in the ABS analysis, the
accumulated excess interest in TALF LLC would help offset potential
losses. As of December 31, 2009, the fund had almost $200 million in
excess interest, which will continue to increase every month in the
absence of any asset purchases. Second, the risk that all or a large
portion of CMBS assets would be surrendered is significantly mitigated
by an FRBNY requirement that legacy CMBSs prepay a portion of any
returns in excess of certain limits.[Footnote 44] In short, the
requirement helps ensure that the TALF borrower will retain an equity
interest in the underlying CMBSs. The longer the term of the TALF loan
(assuming that the underlying collateral provides a return) the more
equity the borrower holds, and the less likely the borrower is to
surrender the CMBS collateral. Third, as with ABSs, and as indicated
by Treasury officials, TALF LLC could hold the securities acquired at
a discount from par--instead of selling them--and earn interest income
and the equity forfeited by the borrower as long as the underlying
mortgages in the security continue to perform well. Fourth, the
estimated loss represents less than 1 percent of total TALF loans as
of December 2009, and total TALF CMBS loans represent about 14 percent
of all TALF loans. Because CMBSs is a small portion of the portfolio,
it would present a smaller proportion of total losses. However, if
recent TALF borrowing trends hold, CMBS loans are likely to increase
as the percentage of total TALF loans. Finally, only senior credit-
enhanced tranches within each CMBS can be accepted as collateral.
Thus, even if credit losses on the commercial mortgages underlying the
TALF CMBS securitizations are significantly higher than currently
expected in today's stressed commercial real estate environment, these
senior tranches are unlikely to experience principal or interest
payment interruptions. While losses associated with CMBSs currently
appear unlikely, these securities warrant ongoing scrutiny because of
continuing economic uncertainty and the distressed conditions in the
commercial real estate market.
Treasury Worked with the Federal Reserve and FRBNY to Analyze Risks
Related to TALF but Did Not Fully Document Analysis Supporting Final
Decisions:
FRBNY, the Federal Reserve, and Treasury worked in a collaborative
manner to design certain elements of TALF, according to these agency
officials. The Federal Reserve led the initial efforts to determine
collateral eligibility and Treasury recommended one asset class and
assessed the risks of others. As part of this work, Treasury hired a
contractor to conduct independent analyses, and the contractor raised
concerns about accepting certain assets as TALF collateral, the size
of the haircuts that were required, and other program terms. While
Treasury officials said that the contractor's concerns were ultimately
resolved, they could not provide documentation showing how Treasury
resolved the contractor's concerns, or how the contractor's analysis
informed Treasury's final decisions. Treasury also did not document
how they reached major decisions that were made with FRBNY and the
Federal Reserve. The lack of an effective process to make and document
decisions may inhibit transparency and accountability of Treasury's
monitoring of the $20 billion of taxpayer funds at risk through TARP.
The Federal Reserve and Treasury Worked Together to Determine the
Eligibility of Proposed TALF Collateral and Their Potential Risks:
Treasury, the Federal Reserve, and FRBNY officials with whom we spoke
said that the agencies have a positive working relationship when
making decisions on TALF. For example, Federal Reserve and FRBNY
officials said that they consulted with Treasury to select the types
of ABSs to include in TALF as eligible collateral.[Footnote 45] Under
this process, the Federal Reserve identified all eligible collateral
except SBA loan guarantees.
Treasury officials said that one asset class that FRBNY proposed--
insurance premium finance loans--required additional analysis to
assess the risks.[Footnote 46] These loans are not as widely traded or
as well understood as other asset classes. Treasury officials worked
with Federal Reserve officials to better understand the asset class,
including its risks and factors mitigating such risks, and its
importance to small businesses. Treasury's contractor also reviewed
risk information on this class of ABSs. Treasury and Federal Reserve
officials determined that this asset class should be eligible for
TALF. According to Treasury officials, other asset classes were also
reviewed by both agencies and not included in the program because of
their risks.
According to Treasury officials, the following criteria were used by
Treasury to evaluate the eligibility of asset classes for inclusion in
TALF: (1) whether including certain asset classes would have a
significant or beneficial effect on the broader economy, small
businesses, or consumers; and (2) whether assistance was needed
because of a market failure in what were otherwise safe asset classes.
[Footnote 47] Based on these criteria, Treasury recommended to the
Federal Reserve and FRBNY that TALF accept as collateral securities
backed by SBA loan guarantees. SBA has two loan guarantee programs--
section 7(a) and section 504--that support financing for small
businesses.[Footnote 48] Treasury requested that these securities be
included because they would assist in carrying out TARP's goals of
supporting small businesses and the risks were deemed to be low
because of their government guarantees.
As part of the asset class selection process, the Federal Reserve and
Treasury each analyzed the potential for loss that the TALF assets
presented. Treasury hired a contractor to, among other things, conduct
an independent analysis of the credit and other risks of the TALF
asset classes and to determine appropriate haircuts for each of the
asset classes. In its initial reports to Treasury, the contractor
raised some concern about accepting certain asset classes for TALF,
provided suggestions for program changes, and, in a few instances,
disagreed with haircuts that FRBNY suggested. According to Treasury
officials and the contractor, any differences were ultimately
reconciled.
As an example, the contractor's report noted that auto floor plan ABSs
faced risks because of the financial problems that the major domestic
auto manufacturers faced, and the nationally recognized statistical
rating organizations were in many cases unwilling to provide AAA
ratings on these securities. The contractor recommended a higher
haircut to encourage TALF borrowers to conduct thorough due diligence
for this asset class. Treasury questioned the contractor's methodology
and asked the contractor to redo its analysis of auto dealer floor
plan ABSs. After working with Treasury and FRBNY to understand the
differences in methodologies between its analysis and FRBNY's for this
asset class, the contractor agreed with FRBNY's estimates on haircuts
and even suggested a haircut lower than FRBNY's.
Treasury Did Not Fully Document Its Analysis or Basis to Support All
Major Agreements with the Federal Reserve and FRBNY:
Although Treasury told us what its reasons were for not accepting all
of the contractor's recommendations, Treasury officials were unable to
provide documentation showing when the contractor conducted the
analyses or when or how Treasury made decisions based on these
analyses. Further, no documentation was available showing how major
differences were resolved, including those involving program terms,
the eligibility of asset classes, and differences in haircut estimates
among Treasury, Treasury's contractor, the Federal Reserve, and FRBNY.
Additionally, Treasury officials could not provide documentation on
the rationale for major program decisions that Treasury, the Federal
Reserve, and FRBNY officials reached. Treasury officials told us that
FRBNY, the Federal Reserve, and Treasury had a positive working
relationship when making decisions on TALF and described the process
as "fluid;" therefore, there was not always documentation of
discussions and final outcomes. Moreover, Treasury officials said that
they spoke almost daily to Federal Reserve and FRBNY officials. In
some cases, according to Treasury officials, FRBNY and the Federal
Reserve consulted Treasury, although technically no consultation was
required. Treasury and the Federal Reserve did not formally document
their conversations, but the end result of those conversations was
documented and reflected publicly on the TALF Web site administered by
FRBNY.
Finally, some of the early decisions on TALF were made by Treasury
officials who are no longer at the agency. Without documentation,
there are no records to show, for instance, how certain suggestions
made by these officials about asset classes or program terms were
incorporated into policy choices for TALF.
Our Standards for Internal Control in the Federal Government states
that internal control activities help ensure that government
management directives are carried out.[Footnote 49] Such activities
are critical to helping ensure accountability and stewardship for
government resources, and include proper documentation of major
decisions. In the context of the Emergency Economic Stabilization Act
of 2008--and the unprecedented size and scope of government assistance
to support the financial sector--transparency and accountability are
of the utmost importance. As an example, for the largest program in
TARP--the Capital Purchase Program--all major decisions are recorded
in meeting minutes that report who was present, what decisions were
made, and when they were made.[Footnote 50] In past TARP reports, we
recommended that Treasury increase the transparency and accountability
of TARP, in part by documenting and reporting certain processes and
decisions.[Footnote 51]
As we noted in past TARP reports, given the economic environment
surrounding the creation of TARP, and subsequently TALF, during the
fall of 2008, the change in administrations and the lack of staff that
Treasury's administrative office for TARP--the Office of Financial
Stability--faced, Treasury may have initially had difficulty
establishing its decision-making processes for TALF and recording
decisions and important meetings on TALF program terms. At the time
that TALF was created, the Office of Financial Stability had been in
existence for barely a month, and its strategy and overall staffing
needs were not yet in place. The broader context at the onset of the
program--with unprecedented economic challenges and low, impermanent
staffing--may help explain why such processes were not established and
documented when the program was first established.
However, for TALF decision-making processes and the activities of TALF
LLC to be viewed as credible, Treasury needs to ensure that it has
developed an effective process to document the basis for its
decisions. A year has passed since Treasury began rolling out TARP-
related programs, and other larger programs--such as the Capital
Purchase Program--have established systems for documenting decisions
and the rationale for decisions. But Treasury's decisions for TALF
still lack a clear process for tracking how important program
decisions are made and why. Without such documentation, ascertaining
what information has been considered to protect TARP funds committed
to TALF is difficult. Further, the lack of documentation inhibits
transparency and accountability.
Treasury's and FRBNY's Indicators Suggest That Credit Market
Conditions Have Shown Some Improvement, but Treasury Lacks Performance
Indicators in the Event that It Must Purchase TALF Assets:
FRBNY, in conjunction with Treasury, monitors TALF by tracking
indicators--such as securitization volumes, changes in pricing, and
TALF loan volumes--by amount and borrower type to identify any
possible impact from TALF. Our analysis of these and other indicators
suggests that market conditions have begun to improve for some TALF-
eligible asset classes, but that others, such as CMBSs, continue to
show weakness. However, any assessment of the effectiveness of an
individual program presents challenges. As we have reported, no
indicator can provide a definitive measure of TALF's impact because a
myriad of programs have been initiated to stabilize the markets,
including actions taken under the Capital Purchase Program and the
Automotive Industry Financing Program.[Footnote 52] Challenges remain
for some of the TALF-eligible asset classes, and FRBNY and Treasury
monitor the performance of TALF loans and collateral to be aware of
all potential risks to TARP funds. However, according to Treasury
officials, Treasury has not yet developed a plan for tracking assets
that might be surrendered to TALF LLC or for publicly disclosing how
up to $20 billion in TARP funds would be monitored.
Treasury, FRBNY, and the Federal Reserve Collaborate on Monitoring
Market Indicators:
Federal Reserve and Treasury officials said that they collaborated on
monitoring indicators that could help measure TALF's effectiveness in
improving conditions in the securitization markets and, in turn, its
impact on the availability of credit to households and small
businesses. Although the officials have said they do not have specific
benchmarks or targets they hope to achieve for ABS issuance volumes or
volumes of TALF loans, they are monitoring those indicators. Officials
also track interest spreads for TALF-eligible asset classes, the
number of borrowers accessing the facility, and information about TALF
borrowers, such as investor type. FRBNY collects data on these
indicators to monitor TALF's impact. According to Treasury officials,
they review FRBNY's metrics related to TALF, including cash flows from
TALF loans that it receives monthly from FRBNY, and speak daily with
FRBNY officials.
New Issuances in Various Asset Classes Increased after TALF's First
Subscription in March 2009:
To determine the condition of the securitization markets, we also have
been monitoring similar indicators, such as new ABS issuances and
changes in interest rates, types of investors, and spreads--using data
from before and after TALF's implementation.[Footnote 53] In general,
data from the indicators that we have collected show increases in
securitization volumes, little change in the cost of credit, and
declines in perceptions of risk in certain asset classes since TALF
began. ABS issuances in all of the most liquid TALF-eligible sectors
dropped sharply in 2008 from their peak levels in 2006 and 2007. As
figure 7 indicates, new issuance of ABSs had come to a virtual halt in
2008, significantly reducing a major source of credit for consumers
and businesses.[Footnote 54] While securitization volumes increased
since the end of 2008, these increases have not been sustained
throughout 2009.
Figure 7: Fluctuations in ABS Issuances, First Quarter 2005 through
Fourth Quarter 2009:
[Refer to PDF for image: multiple line graph]
2005, Q1;
Credit card: $12.0 billion;
Auto: $9.59 billion;
Student loan: $5.9 billion;
CMBS: $31.4 billion.
2005, Q2;
Credit card: $25.9 billion;
Auto: $11.4 billion;
Student loan: $8.1 billion;
CMBS: $30.4 billion.
2005, Q3;
Credit card: $17.9 billion;
Auto: $14.2 billion;
Student loan: $9.4 billion;
CMBS: $38.1 billion.
2005, Q4;
Credit card: $21.1 billion;
Auto: $17.8 billion;
Student loan: $22.8 billion;
CMBS: $52.3 billion.
2006, Q1;
Credit card: $18.6 billion;
Auto: $17.3 billion;
Student loan: $16.5 billion;
CMBS: $45.6 billion.
2006, Q2;
Credit card: $18.0 billion;
Auto: $15.5 billion;
Student loan: $21.7 billion;
CMBS: $42.5 billion.
2006, Q3;
Credit card: $26.6 billion;
Auto: $16.0 billion;
Student loan: $16.4 billion;
CMBS: $44.0 billion.
2006, Q4;
Credit card: $25.0 billion;
Auto: $11.6 billion;
Student loan: $15.5 billion;
CMBS: $69.3 billion.
2007, Q1;
Credit card: $13.4 billion;
Auto: $25.0 billion;
Student loan: $23.4 billion;
CMBS: $46.0 billion.
2007, Q2;
Credit card: $27.1 billion;
Auto: $23.0 billion;
Student loan: $9.6 billion;
CMBS: $74.4 billion.
2007, Q3;
Credit card: $14.5 billion;
Auto: $22.1 billion;
Student loan: $10.0 billion;
CMBS: $54.2 billion.
2007, Q4;
Credit card: $16.0 billion;
Auto: $24.3 billion;
Student loan: $7.6 billion;
CMBS: $25.4 billion.
2008, Q1;
Credit card: $10.3 billion;
Auto: $28.8 billion;
Student loan: $8.4 billion;
CMBS: $4.4 billion.
2008, Q2;
Credit card: $14.0 billion;
Auto: $28.2 billion;
Student loan: $14.6 billion;
CMBS: $2.0 billion.
2008, Q3;
Credit card: $3.2 billion;
Auto: $10.4 billion;
Student loan: $6.4 billion;
CMBS: $3.1 billion.
2008, Q4;
Credit card: $1.6 billion;
Auto: 0;
Student loan: 0;
CMBS: 0.
2009, Q1;
Credit card: $7.6 billion;
Auto: $6.5 billion;
Student loan: $2.0 billion;
CMBS: 0.
2009, Q2;
Credit card: $15.0 billion;
Auto: $20.4 billion;
Student loan: $8.3 billion;
CMBS: $3.43 billion.
2009, Q3;
Credit card: $19.3 billion;
Auto: $15.5 billion;
Student loan: $6.1 billion;
CMBS: 0.
2009, Q4;
Credit card: $9.4 billion;
Auto: $3.6 billion;
Student loan: $4.8 billion;
CMBS: $2.2 billion.
Source: GAO analysis of Thomson Reuters IFR Markets data.
[End of section]
After having shown little activity since the last quarter of 2008,
issuance of credit card, auto, and student loan ABSs increased after
the initial TALF subscription in March 2009 (shown as first quarter
2009 in figure 7). The majority of ABS issuances in the credit card
and auto sectors have been supported by TALF loans. Specifically, of
the $46 billion in ABSs issued on credit card debt in 2009, $29.7
billion, or about 65 percent, have been eligible for TALF financing.
Similarly, about 88 percent--or $44.9 billion--of the $51.2 billion in
ABSs issued on auto loans in 2009 were TALF-eligible deals. For more
detailed information on securitization volumes, see appendix VII.
By the third quarter of 2009, credit card and student loan issuances
had declined again in dollar terms, while auto issuances continued to
increase. A number of factors--such as the combined effects of the
numerous stimulus programs, changes in consumer demand for credit, and
investor willingness to invest--also may have contributed to the trend
in securitization volumes in these sectors. Federal Reserve officials
suggested that some companies may have been hesitant to issue credit
card ABSs because of uncertainty regarding the continued availability
of the FDIC's "Safe Harbor" rule in light of new accounting rules
effective for annual financial periods beginning after November 15,
2009.[Footnote 55]
Figure 7 also shows that CMBS securitization volumes peaked in mid-
2007 and dropped significantly in late 2007 and early 2008 before
coming to a complete halt by the end of 2008. CMBS volumes did not
pick up again until the second quarter of 2009, but those were the
result of repackaging existing securitizations rather than new CMBS
issuances. FRBNY officials noted that the first new CMBS deal to come
to the market since summer 2008 was a TALF-eligible, single-issuer
deal in the fourth quarter. CMBS issuances have remained relatively
flat nonetheless, and it is unclear if they will increase in the
future. According to FRBNY and Treasury officials, one possible reason
for the small CMBS issuance volumes could be that the time it takes to
complete a CMBS deal is often longer than for other asset classes.
Interest Rates in Most Asset Classes Have Generally Not Decreased
Since TALF Was Announced:
One of TALF's goals is to increase the availability of credit to
consumers and businesses. TALF assistance to the securitization
markets is intended to result in lower loan rates and increased credit
availability to businesses and individual consumers, including the
auto loan, credit card, and student loan sectors that account for the
majority of securitizations. [Footnote 56] Recent increased activity
in the securitization markets has been accompanied by a substantial
decrease in interest rates for loans originated by auto finance
companies. However, there have been few changes in credit card rates
or interest rates for consumers in auto loans originated by commercial
banks. Consumer interest rates remain flat.
Because auto finance companies rely more heavily on securitizations
for funding than commercial banks, the effects of positive changes in
the securitization markets are more likely to be reflected in their
loan rates than in those of commercial banks. As figure 8 shows, auto
loan rates offered by commercial banks remained fairly steady before
and after the implementation of TALF. The average finance company auto
rate has been consistently below commercial bank auto rates, with the
exception of the fourth quarter of 2008, perhaps reflecting the
financial challenges facing the auto industry at that time.[Footnote
57] Since then, rates at auto finance companies have declined from an
average of 7 percent to approximately 3 percent. This reduction
coincides with the launching of TALF but may also reflect assistance
from the numerous government stimulus programs, especially those
focused on the auto industry. While fixed credit card rates have
remained fairly flat in recent years, variable credit card rates have
increased by approximately one percent since TALF's inception. FRBNY
officials attributed the elevated credit card rates to increased
charge-offs, which have raised companies' costs of funds. These rate
changes could also be the result of credit card companies' efforts to
anticipate the implementation of the remaining part of the Credit Card
Accountability Responsibility and Disclosure Act of 2009, which will
take effect in February 2010.[Footnote 58]
FRBNY officials said that it is possible that without TALF interest
rates on loans to consumers and small businesses would be much higher
than they are now. Issuers of TALF-eligible ABSs have told FRBNY that
without TALF they would have made fewer loans and those loans would
have been at higher rates. Data on private student loan rates are
difficult to obtain, but FRBNY officials said that Sallie Mae has
reduced its rates on private student loans over the past few months.
Figure 8: Trends in Average Finance Rates for Credit Cards, Auto Loans
at Banks, and Auto Finance Companies, June 2008 through December 2009:
[Refer to PDF for image: multiple line graph]
Date: June, 2008;
New car rates: 6.99%;
Used car rates: 7.74%;
Variable card rates: 11.88%;
Fixed card rates: 11.35%;
Auto finance company rates: 5.18%.
Date: July, 2008;
New car rates: 7.06%;
Used car rates: 7.74%;
Variable card rates: 11.88%;
Fixed card rates: 11.27%;
Auto finance company rates: 5.05%.
Date: August, 2008;
New car rates: 7.01%;
Used car rates: 7.71%;
Variable card rates: 11.78%;
Fixed card rates: 11.28%;
Auto finance company rates: 4.90%.
Date: September, 2008;
New car rates: 7.04%;
Used car rates: 7.77%;
Variable card rates: 11.88%;
Fixed card rates: 11.34%;
Auto finance company rates: 5.38%.
Date: October, 2008;
New car rates: 7.04%;
Used car rates: 7.8%;
Variable card rates: 12%;
Fixed card rates: 11.31%;
Auto finance company rates: 6.06%.
Date: November, 2008; TALF announcement;
New car rates: 7.18%;
Used car rates: 7.89%;
Variable card rates: 12%;
Fixed card rates: 11.3%;
Auto finance company rates: 6.91%.
Date: December, 2008;
New car rates: 7.09%;
Used car rates: 7.79%;
Variable card rates: 12.06%;
Fixed card rates: 11.08%;
Auto finance company rates: 6.58%.
Date: January, 2009;
New car rates: 6.97%;
Used car rates: 7.71%;
Variable card rates: 12.06%;
Fixed card rates: 10.86v
Auto finance company rates: 5.73%.
Date: February, 2009;
New car rates: 6.88%;
Used car rates: 7.69%;
Variable card rates: 12.09%;
Fixed card rates: 10.71%;
Auto finance company rates: 5.05%.
Date: March, 2009;
New car rates: 6.9%;
Used car rates: 7.69%;
Variable card rates: 12.24%;
Fixed card rates: 10.76%;
Auto finance company rates: 4.5%.
Date: April, 2009;
New car rates: 6.81%;
Used car rates: 7.62%;
Variable card rates: 12.33%;
Fixed card rates: 10.69%;
Auto finance company rates: 4.08%.
Date: May, 2009;
New car rates: 7.05%;
Used car rates: 7.86%;
Variable card rates: 12.33%;
Fixed card rates: 10.78%;
Auto finance company rates: 3.56%.
Date: June, 2009;
New car rates: 6.98%;
Used car rates: 7.82%;
Variable card rates: 12.4%;
Fixed card rates: 10.85%;
Auto finance company rates: 3.44%.
Date: July, 2009;
New car rates: 7.01%;
Used car rates: 7.86%;
Variable card rates: 12.26%;
Fixed card rates: 11.04%;
Auto finance company rates: 3.43%.
Date: August, 2009;
New car rates: 7.02%;
Used car rates: 7.86%;
Variable card rates: 12.38%;
Fixed card rates: 11.2%;
Auto finance company rates: 3.43%.
Date: September, 2009;
New car rates: 7.02%;
Used car rates: 7.84%;
Variable card rates: 12.34%;
Fixed card rates: 11.37%;
Auto finance company rates: 3.43%.
Date: October, 2009;
New car rates: 6.86%;
Used car rates: 7.71%;
Variable card rates: 12.34%;
Fixed card rates: 11.42%;
Auto finance company rates: 3.42%.
Date: November, 2009;
New car rates: 6.71%;
Used car rates: 7.6%;
Variable card rates: 12.31%;
Fixed card rates: 11.68%.
Date: December, 2009;
New car rates: 6.77%;
Used car rates: 7.59%;
Variable card rates: 12.34%;
Fixed card rates: 11.68%.
Source: Bankrate.com and the Federal Reserve.
Note: Auto finance company rates from the Federal Reserve are only
available through October 31, 2009.
[End of figure]
TALF Loan Volumes and the Composition of TALF Participants Have
Changed since the Program Began:
As illustrated in table 3, the volume of TALF loans made since the
inception of the program has fluctuated by month, with loan volumes
peaking in May 2009 and June 2009. As of December 31, 2009, a total of
$61.6 billion in TALF loans had been granted; however, the balance of
loans outstanding at that date was $47.5 billion due to loan
prepayments and principal paydowns. As we reported previously, agency
officials indicated that improvements in securitization and loan
markets had made issuers less dependent on TALF support.[Footnote 59]
However, according to FRBNY officials, other issuers remain more
heavily dependent on investor access to TALF financing. FRBNY
officials noted that there have been a number of prepayments, and
market participants also told us that financing under TALF was now
less favorable because better financing terms could be found in the
private sector for certain asset classes. Notably, the first and only
subscription for new-issue CMBSs occurred in November 2009. FRBNY and
Treasury officials stated that the slow new-issue CMBS activity may be
due to the length of time it takes to complete a deal.
Table 3: TALF Loans Disbursed by Asset Class, March 2009 through
December 2009:
Asset class: Auto;
March: $1,909 million;
April: $797 million;
May: $2,311 million;
June: $2,946 million;
July: $2,831 million;
August: $555 million;
September: $1,160 million;
October: $191 million;
November: 0;
December: 0;
Total by loan type: $12,699.
Asset class: Credit card;
March: $2,804 million;
April: $891 million;
May: $5,515 million;
June: $6,023 million;
July: $1,459 million;
August: $2,554 million;
September: $4,399 million;
October: $224 million;
November: $63 million;
December: $1,529 million;
Total by loan type: $25,461 million.
Asset class: Equipment;
March: 0;
April: 0;
May: $446 million;
June: $590 million;
July: 0;
August: 0;
September: $111 million;
October: $39 million;
November: $57 million;
December: $199 million;
Total by loan type: $1,441 million.
Asset class: Floor plan;
March: 0;
April: 0;
May: 0;
June: 0;
July: 0;
August: $1,005 million;
September: 0;
October: $887 million;
November: $445 million;
December: $172 million;
Total by loan type: $2,510 million.
Asset class: Insurance premium finance;
March: [Empty];
April: [Empty];
May: [Empty];
June: $464 million;
July: 0;
August: 0;
September: $530 million;
October: 0;
November: 0;
December: 0;
Total by loan type: $994 million.
Asset class: Servicing advances;
March: 0;
April: 0;
May: 0;
June: $439 million;
July: $34 million;
August: $108 million;
September: 0;
October: $475 million;
November: 0;
December: $138 million;
Total by loan type: $1,193 million.
Asset class: Student loan;
March: 0;
April: 0;
May: $2,281 million;
June: $227 million;
July: $987 million;
August: $2,445 million;
September: $177 million;
October: $288 million;
November: $85 million;
December: $665 million;
Total by loan type: $7,155 million.
Asset class: Small business;
March: 0;
April: 0;
May: $87 million;
June: $29 million;
July: $62 million;
August: $147 million;
September: $162 million;
October: $262 million;
November: $409 million;
December: $275v;
Total by loan type: $1,433 million.
Asset class: Legacy CMBSs;
March: [Empty];
April: [Empty];
May: [Empty];
June: 0;
July: $636 million;
August: $2,148 million;
September: $1,351 million;
October: $1,931 million;
November: $1,330 million;
December: $1,282 million;
Total by loan type: $8,677 million.
Asset class: New[Empty]issue CMBSs;
March: [Empty];
April: [Empty];
May: [Empty];
June: 0;
July: 0;
August: 0;
September: 0;
October: 0;
November: $72 million;
December: 0;
Total by loan type: $72 million.
Asset class: Total;
March: $4,713 million;
April: $1,688 million;
May: $10,639 million;
June: $10,717 million;
July: $6,009 million;
August: $8,962 million;
September: $7,890 million;
October: $4,297 million;
November: $2,461 million;
December: $4,259 million;
Total by loan type: $61,636 million.
Source: GAO analysis of FRBNY data.
Note: Not all numbers will total due to rounding. Until September 2009
the FRBNY's Web site showed only TALF loans requested, not loans
disbursed. FRBNY provided us with data prior to September 2009. In
addition, in discussions with FRBNY we learned that the program
changed in April 2009 to allow for new interest rates on loans secured
by ABSs with weighted average lives to maturity of less than 2 years
had resulted in the refinancing of some existing TALF loans. As a
result, May 2009 figures may include double counting due to this
refinancing.
[End of table]
As shown in figure 9, approximately 75 percent of TALF loans involved
the purchase of ABSs backed by auto loans and leases, credit card
receivables, and student loans. This activity reflects the historical
trends in auto, credit card, and student loan securitizations, which
represent the majority of the ABS markets.
Figure 9: Composition of TALF Loans Disbursed by Asset Class as of
December 2009:
[Refer to PDF for image: pie-chart]
Credit cards: $25,461 million (41%);
Auto: $12,699 million (21%);
CMBS: $8,677 million, legacy (14%); $72 million, new issue (0%);
Student loans: $7,155 (12%);
Floor plan: $2,510 million (4%);
Equipment: $1,441 million (2%);
Small Business Administration: $1,433 million (2%);
Mortgage servicing advances: $1,193 million (2%);
Insurance premium finance: $994 million (2%).
Source: GAO analysis of FRBNY data.
[End of figure]
According to Treasury and Federal Reserve officials, TALF was designed
to encourage broader investor participation in the securitization
markets, with the goal of reviving consumer lending. These officials
noted that the securitization markets stopped functioning in 2008 when
many investors stopped purchasing these securities. The lack of
securitization market activity disrupted a significant source of
funding for businesses and consumers. Gradually, some of these
investors have returned to the markets, but at a slower rate than
during past market downturns. Specifically, Treasury officials noted
increasing participation in TALF securitization by asset managers,
hedge funds, and traditional institutional investors such as pension
funds and insurance companies. They consider the return of investors
to the securitization markets to be a measure of the program's
success. Hedge funds traditionally have not invested in ABSs because
of the low returns relative to other opportunities. However, FRBNY
officials believe the access to low-cost financing through TALF made
ABS returns attractive to hedge funds. FRBNY also noted participation
by private investors and banks.
Differences in Prices and Benchmarks Have Decreased for Most TALF-
Eligible ABS Collateral:
As we have discussed, one method of measuring market participants'
perceived risk of a security is to compare the difference between the
security's yield and a benchmark yield. The difference is called a
spread, and wide spreads, or large differences, generally indicate
that participants perceive high risk in the market that requires a
high rate of return. As perceived risk declines, differences in such
prices decrease, or narrow. During the fourth quarter of 2008 and
first quarter of 2009, spreads likely reflected high expected costs of
selling securities prior to their maturity, which contributed to low
desirability for those securities. Figure 10 shows the change in
spreads in the following TALF-eligible asset classes: auto loan,
credit card, student loan, and CMBS. A trend of widening spreads in
these asset classes began in mid-2007, indicating negative perceptions
about risk. Although there were fluctuations throughout 2008, spreads
began to narrow in early 2009, indicating a perceived decline in risk
by market participants and potentially improved credit market
conditions.
Figure 10: Trends for AAA ABS and CMBS Spreads from January 2005
through September 2009:
[Refer to PDF for image: multiple line graph]
Date: January 2005:
Auto: 2.1%;
Credit Cards: 1.3%;
FFELP loan: 3%;
CMBS: 9.5%;
Private student loans: 19%.
Date: February 2005:
Auto: 0.9%;
Credit Cards: 0.5%;
FFELP loan: 3%;
CMBS: 7%;
Private student loans: 15.9%.
Date: March 2005:
Auto: 0.4%;
Credit Cards: 0%;
FFELP loan: 3%;
CMBS: 6.5%;
Private student loans: 13.9%.
Date: April 2005:
Auto: 3.9%;
Credit Cards: 0%;
FFELP loan: 1%;
CMBS: 5.5%;
Private student loans: 17.9%.
Date: May 2005:
Auto: 6%;
Credit Cards: 1%;
FFELP loan: 2%;
CMBS: 10%;
Private student loans: 22.7%.
Date: June 2005:
Auto: 6.1%;
Credit Cards: 1.2%;
FFELP loan: 1%;
CMBS: 10.5%;
Private student loans: 26.2%.
Date: July 2005:
Auto: 4.3%;
Credit Cards: 0.7%;
FFELP loan: 2%;
CMBS: 11.5%;
Private student loans: 23.2%.
Date: August 2005:
Auto: 4.3%;
Credit Cards: 0.7%;
FFELP loan: 1%;
CMBS: 11.5%;
Private student loans: 21.4%.
Date: September 2005:
Auto: 3.6%;
Credit Cards: -0.3%;
FFELP loan: 1%;
CMBS: 11.5%;
Private student loans: 22.8%.
Date: October 2005:
Auto: 3.6%;
Credit Cards: -0.3%;
FFELP loan: 1%;
CMBS: 12.5%;
Private student loans: 25.3%.
Date: November 2005:
Auto: 3.3%;
Credit Cards: -0.7%;
FFELP loan: 2%;
CMBS: 12.5%;
Private student loans: 24.8%.
Date: December 2005:
Auto: 3.4%;
Credit Cards: -0.2%;
FFELP loan: 2%;
CMBS: 12.5%;
Private student loans: 27.0%.
Date: January 2006:
Auto: 3.8%;
Credit Cards: 0.5%;
FFELP loan: 2%;
CMBS: 14%;
Private student loans: 24.7%.
Date: February 2006:
Auto: 2.8%;
Credit Cards: 0%;
FFELP loan: 1%;
CMBS: 14%;
Private student loans: 25.1%.
Date: March 2006:
Auto: 1.3%;
Credit Cards: -0.7%;
FFELP loan: 0%;
CMBS: 11.5%;
Private student loans: 22.0%.
Date: April 2006:
Auto: 0.8%;
Credit Cards: -1.7%;
FFELP loan: 0%;
CMBS: 8.5%;
Private student loans: 19%.
Date: May 2006:
Auto: -0.5%;
Credit Cards: -1.8%;
FFELP loan: 0%;
CMBS: 8.5%;
Private student loans: 17.7%.
Date: June 2006:
Auto: -0.9%;
Credit Cards: -2.5%;
FFELP loan: -1%;
CMBS: 7%;
Private student loans: 19.9%.
Date: July 2006:
Auto: 0.4%;
Credit Cards: -1.8%;
FFELP loan: -1%;
CMBS: 7%;
Private student loans: 19.1%.
Date: August 2006:
Auto: -0.3%;
Credit Cards: -2.2%;
FFELP loan: -1%;
CMBS: 7%;
Private student loans: 17.7%.
Date: September 2006:
Auto: -0.6%;
Credit Cards: -2.7%;
FFELP loan: -1%;
CMBS: 7%;
Private student loans: 18.9%.
Date: October 2006:
Auto: 0.4%;
Credit Cards: -2.7%;
FFELP loan: -1%;
CMBS: 7%;
Private student loans: 18.7%.
Date: November 2006:
Auto: 1.1%;
Credit Cards: -2.7%;
FFELP loan: -1%;
CMBS: 7%;
Private student loans: 19.7%.
Date: December 2006:
Auto: 4.3%;
Credit Cards: -2%;
FFELP loan: 0%;
CMBS: 7%;
Private student loans: 23.3%.
Date: January 2007:
Auto: 5%;
Credit Cards: -0.8%;
FFELP loan: 0%;
CMBS: 7%;
Private student loans: 21.5%.
Date: February 2007:
Auto: 4.4%;
Credit Cards: -0.7%;
FFELP loan: 0%;
CMBS: 7%;
Private student loans: 21.2%.
Date: March 2007:
Auto: 4.1%;
Credit Cards: -0.5%;
FFELP loan: 0%;
CMBS: 8.5%;
Private student loans: 23.1%.
Date: April 2007:
Auto: 5%;
Credit Cards: 0.2%;
FFELP loan: 0%;
CMBS: 7.5%;
Private student loans: 24.5%.
Date: May 2007:
Auto: 5.3%;
Credit Cards: 0.7%;
FFELP loan: 1%;
CMBS: 9.5%;
Private student loans: 30.3%.
Date: June 2007:
Auto: 4.9%;
Credit Cards: -0.2%;
FFELP loan: 1%;
CMBS: 8.5%;
Private student loans: 24.7%.
Date: July 2007:
Auto: 5%;
Credit Cards: 0%;
FFELP loan: 2%;
CMBS: 8.5%;
Private student loans: 27.8%.
Date: August 2007:
Auto: 13.8%;
Credit Cards: 7.3%;
FFELP loan: 9%;
CMBS: 19%;
Private student loans: 61.9%.
Date: September 2007:
Auto: 52.3%;
Credit Cards: 28.7%;
FFELP loan: 34%;
CMBS: 57.5%;
Private student loans: 66.6%.
Date: October 2007:
Auto: 33.5%;
Credit Cards: 25.7%;
FFELP loan: 34%;
CMBS: 60%;
Private student loans: 53.4%.
Date: November 2007:
Auto: 34%;
Credit Cards: 21.8%;
FFELP loan: 29%;
CMBS: 57.5%;
Private student loans: 74.5.
Date: December 2007:
Auto: 91.9%;
Credit Cards: 49.2%;
FFELP loan: 48%;
CMBS: 80%;
Private student loans: 126.0%.
Date: January 2008:
Auto: 95.6%;
Credit Cards: 55%;
FFELP loan: 58%;
CMBS: 100%;
Private student loans: 112.5%.
Date: February 2008:
Auto: 125%;
Credit Cards: 66.8%;
FFELP loan: 64%;
CMBS: 125%;
Private student loans: 309.3%.
Date: March 2008:
Auto: 180.6%;
Credit Cards: 95%;
FFELP loan: 93%;
CMBS: 212.5%;
Private student loans: 398.7%.
Date: April 2008:
Auto: 202.5%;
Credit Cards: 115%;
FFELP loan: 105%;
CMBS: 300%;
Private student loans: 289.8%.
Date: May 2008:
Auto: 168.8%;
Credit Cards: 110%;
FFELP loan: 111%;
CMBS: 300%;
Private student loans: 201.0%.
Date: June 2008:
Auto: 131.9%;
Credit Cards: 65%;
FFELP loan: 68%;
CMBS: 300%;
Private student loans: 179.0%.
Date: July 2008:
Auto: 120.6%;
Credit Cards: 74.2%;
FFELP loan: 61%;
CMBS: 325%;
Private student loans: 211.8%.
Date: August 2008:
Auto: 173.1%;
Credit Cards: 106.7%;
FFELP loan: 111%;
CMBS: 375%;
Private student loans: 269.4%.
Date: September 2008:
Auto: 221.3%;
Credit Cards: 132.5%;
FFELP loan: 125%;
CMBS: 387.5%;
Private student loans: 317.4%.
Date: October 2008:
Auto: 347.5%;
Credit Cards: 250.8%;
FFELP loan: 264%;
CMBS: 600%;
Private student loans: 427.4%.
Date: November 2008: TALF announcement;
Auto: 659.4%;
Credit Cards: 479.2%;
FFELP loan: 300%;
CMBS: 675%;
Private student loans: 685.4%.
Date: December 2008:
Auto: 650%;
Credit Cards: 554.2%;
FFELP loan: 338%;
CMBS: 800%;
Private student loans: 1248.9%.
Date: January 2009:
Auto: 643.8%;
Credit Cards: 577.5%;
FFELP loan: 375%;
CMBS: 950%;
Private student loans: 1082%.
Date: February 2009:
Auto: 450%;
Credit Cards: 376.7%;
FFELP loan: 224%;
CMBS: 1000%;
Private student loans: 1203.8%.
Date: March 2009:
Auto: 378.8%;
Credit Cards: 329.2%;
FFELP loan: 188%;
CMBS: 1000%;
Private student loans: 1066.1%.
Date: April 2009: TALF subscription;
Auto: 331.3%;
Credit Cards: 305.8%;
FFELP loan: 164%;
CMBS: 1000%;
Private student loans: 907.3%.
Date: May 2009:
Auto: 250%;
Credit Cards: 200%;
FFELP loan: 116%;
CMBS: 750%;
Private student loans: 771.9%.
Date: June 2009:
Auto: 182.5%;
Credit Cards: 131.7%;
FFELP loan: 106%;
CMBS: 750%;
Private student loans: 707.8%.
Date: July 2009:
Auto: 167.5%;
Credit Cards: 149.2%;
FFELP loan: 101%;
CMBS: 750%;
Private student loans: 664.4%.
Date: August 2009:
Auto: 121.9%;
Credit Cards: 125%;
FFELP loan: 93%;
CMBS: 550%;
Private student loans: 387.7%.
Date: September 2009:
Auto: 100%;
Credit Cards: 92.5%;
FFELP loan: 69%;
CMBS: 550v
Private student loans: 459.2%.
Source: GAO analysis of spread data from multiple banks.
[End of figure]
Treasury Reviews TALF-Related Data from FRBNY's Indicators but Has Not
Developed Indicators to Collect and Disclose Data on Future TALF LLC
Assets:
Treasury reviews the data that FRBNY collects on TALF loan volumes and
borrowers by type, securitization volumes, and changes in pricing.
Treasury officials noted that personnel at both agencies were
responsible for a variety of tasks in tracking TALF-related metrics.
We found that Federal Reserve officials, particularly at FRBNY,
typically took the lead in collecting data and calculating metrics. We
also found that Treasury officials did not have a plan to collect and
analyze information related to assets that might be placed in TALF
LLC--assets to which Treasury would have an exposure. Such information
might include the purchase and sale price of the assets, their current
market value, total outstanding loans by Treasury to TALF LLC for the
ABS purchases, and the rationale behind TALF LLC's possible future
sale of assets. Treasury has not yet developed such a plan because no
TALF collateral has been surrendered thus far, and Treasury believes
it is unlikely that it will have to use TARP funds to finance TALF
LLC's purchase of surrendered collateral. Moreover, Treasury does not
have a plan to publicly communicate such information in the event that
collateral is surrendered and placed there. In previous TARP reports
and in this report, we have discussed the importance of improving the
transparency and accountability of TARP programs. We have also
recommended that Treasury build on existing oversight procedures to
better monitor and report on the use of TARP funds and to better
quantify program results.
Although Treasury is not responsible for implementing or administering
TALF, it has pledged support to TALF LLC with the first $20 billion of
potential loans to allow it to purchase surrendered TALF collateral.
As discussed earlier, commercial real estate continues to show
weakness and could potentially pose greater risks to TARP funds.
Without a system for tracking and reporting on any potential assets
such as CMBSs that are surrendered to TALF LLC, Treasury cannot assure
transparent management of these assets or determine if it is achieving
its goals under CBLI with respect to the use of TARP funds for TALF-
related activities. Further, without properly planning for its role in
managing the collateral should they have to be purchased by TALF LLC,
Treasury may not be able to effectively assess any risks associated
with such assets or exercise appropriately its decision-making
responsibilities regarding the potential sale of any assets.
Conclusions:
TALF is one of several programs created by the Federal Reserve to help
address the recent crisis in the financial sector. Specifically, this
program was designed to restart securitization markets, a critical
part of financial markets. Given the myriad of programs initiated to
stabilize the financial system and increase credit availability, it is
difficult to attribute improvement in markets to any one program.
Nevertheless, according to a variety of indicators, TALF appears to be
contributing to measured improvements in the securitization markets.
As of December 31, 2009, $61.6 billion in loans were made through TALF
and TALF LLC had received $100 million of the $20 billion in TARP
funds committed to the program. In addition to the $20 billion, funds
provided by FRBNY to operate TALF could expose additional risks.
However, because we are statutorily prohibited from auditing the
Federal Reserve's monetary policy activities, we believe our ability
to completely assess and report on taxpayers' exposure to the entire
program or the Federal Reserve's management of the program is limited.
Although the government has taken a number of steps to mitigate the
risk of loss from TALF, in the long term risks remain. For example,
while analyses by the Federal Reserve and a Treasury contractor that
were based on predictions of market performance and other factors
estimated that a loss of a substantial portion of the $20 billion TARP
commitment would be unlikely based on current conditions in the
securitization markets, we found that until the TALF borrowers repay
their loans, TALF still presents risks. While we acknowledge that
overall market conditions have generally improved since 2008, some
asset classes--specifically CMBS--are still performing poorly and may
continue to perform badly for the foreseeable future. Moreover,
markets remain fragile and predicting how the overall ABS markets will
perform in the future and how borrowers might respond to new declines
in the markets is difficult. A return to 2008 conditions could have
adverse impacts on the program, such as significantly reducing the
value of TALF collateral, providing an economic incentive for
borrowers to walk away from their loans, and requiring TARP funds be
used to buy TALF collateral. However, several TALF program features
make this less likely.
Treasury, which worked with FRBNY and the Federal Reserve on certain
decisions related to TALF, was not able to provide documentation on
how these decisions were made. As we noted in past TARP reports,
Treasury has yet to develop systems to ensure the transparency and
accountability for TARP activities by implementing a strong,
transparent strategic framework with appropriate oversight mechanisms.
Among other things, these mechanisms would ensure accountability by
tracking why decisions are made, and whether goals are being achieved.
Documenting the basis for decisions is an important part of the
decision-making process. Moreover, documenting the rationale for major
program decisions would help ensure that the program objectives are
being met and that it is functioning as intended. Unless Treasury
documents the rationale for major program decisions that it made with
the Federal Reserve, it cannot demonstrate accountability for meeting
the goals of TALF and could unnecessarily place TARP funds at risk.
Believing it is highly unlikely that it will have to use TARP funds to
finance ABS purchases by TALF LLC, Treasury has not taken steps to
develop a set of metrics or a plan for tracking and reporting on the
performance of the collateral that could be placed in TALF LLC. While
TARP funds may never be used to finance purchases of ABS or CMBS used
as TALF collateral, Treasury should at least be prepared for the
possibility. Without a plan for collecting and analyzing such data,
Treasury would have to develop one as it is financing or after it has
financed collateral purchases by TALF LLC and risks being ill prepared
to make informed decisions on whether TALF LLC should keep collateral
until the securities mature or sell them. Unlike many other programs
that were developed and implemented in the midst of the crisis,
Treasury has an opportunity to be strategic by developing a plan in
the event that its role in TALF is triggered. In addition, without a
plan Treasury cannot measure TALF's success in meeting its goals under
CBLI with respect to any assets that are placed in TALF LLC. Finally,
without a plan for communicating the findings that result from
tracking and analyzing such metrics, the public will not be aware of
how the assets are managed and financed, undermining Treasury's
efforts to be fully transparent about TARP activities.
Matter for Congressional Consideration:
To enable GAO to audit TARP support for TALF most effectively, we
recommend that Congress provide GAO with audit authority over all
Federal Reserve operational and administrative actions taken with
respect to TALF, together with appropriate access authority.
Recommendations for Executive Action:
To improve transparency of decision making on the use of TARP funds
for TALF and to ensure adequate monitoring of risks related to TALF
collateral, we recommend that the Secretary of the Treasury direct the
Office of Financial Stability to take the following actions:
1. Given the distressed conditions in the commercial real estate
market, as part of its ongoing monitoring of TALF collateral, continue
to give greater attention to reviewing risks posed by CMBSs.
2. Strengthen the process for making major program decisions for TALF
and document how it arrives at final decisions with the Federal
Reserve and FRBNY. Such decisions should include how Treasury
considers expert and contractor recommendations and resolves those
recommendations that differ from those of the Federal Reserve and
FRBNY.
3. Conduct a review of what data to track and metrics to disclose to
the public in the event that TALF LLC purchases surrendered assets
from FRBNY. Such data and metrics should relate to the purchase,
management, and sale of assets in TALF LLC that potentially impact
TARP funds. Metrics related to TALF LLC could include periodic reports
on the date and purchase price of assets; fluctuations in the market
value of assets held; the date, price, and rationale when assets are
sold; and the total amount of loans outstanding to Treasury.
Agency Comments and Our Evaluation:
We provided a draft of this report to Treasury for its review and
comment. We also provided the draft report to the Federal Reserve to
verify the factual information they provided to us about TALF.
Treasury and the Federal Reserve provided written comments that we
have reprinted in appendixes VIII and IX, respectively. Treasury and
the Federal Reserve also provided technical comments that we have
incorporated as appropriate.
In their response Treasury welcomed our recognition that TALF
contributed to improvements in the securitization markets but believed
that the draft report understated the success of the program. In so
doing Treasury reiterated several points that were already underscored
in the draft report. For example, as discussed in the draft report and
Treasury's response, we acknowledged that securitization volumes in
markets had come to a complete halt in 2008, but increased after
TALF's first subscription in March 2009. Moreover, we also noted that
recent TALF subscription levels for the majority of eligible asset
classes have tapered off, which is an indication that investors'
perception of risk has decreased. As we have noted in our previous
TARP reports, any assessment of the effectiveness of TALF is
complicated by the fact that a variety of programs have been
established by the Federal Reserve, Treasury, and others to stabilize
the markets--making it virtually impossible to definitively single out
and measure TALF's impact.
Treasury also stated that it disagreed with our methodology related to
potential losses for CMBSs. As we discussed in the report, the adverse
scenario analysis of the TALF CMBS portfolio was not intended to
project an expected loss amount for this portfolio but to help assess
the possible range of losses in TALF. We used a stress scenario and
selected loss assumptions that were similar to those the Federal
Reserve imposed on the 19 bank holding companies that participated in
the 2009 stress test. Treasury states that it would take a 65 percent
loss on underlying commercial real estate prices to experience losses.
While we agree that this is an unlikely event, commercial real estate
prices have already fallen by an average of 43 percent since prices
peaked. Combined with the fact that CMBSs have much longer time
horizons than other TALF ABS asset classes and hence greater
uncertainty of outcomes, we continue to believe that CMBS warrants
ongoing attention.
Treasury also noted that it appreciates the recommendations GAO makes
in the report to strengthen the documentation of decisions Treasury
made concerning changes to the program. Treasury stated it is
committed to ensuring that not only TALF, but TARP as a whole, is
administered in a way that protects the taxpayer. We believe that
development of a sound decision-making process that includes steps for
formal approval and documentation of the basis of the final decisions
at an appropriate management level will improve transparency and
accountability of the TALF program. As we noted in past TARP reports
and most recently in the October 2009 report, Treasury has yet to
develop systems to ensure the transparency and accountability for TARP
activities by implementing a strong, transparent strategic framework
with the appropriate oversight mechanisms. Among other things, these
mechanisms would ensure accountability by tracking why decisions are
made and whether goals are being achieved.
Finally, regarding our recommendation that Treasury review what data
to track and metrics to disclose to the public in the event that TALF
LLC purchases surrendered assets from FRBNY, Treasury noted that it
will continue to enhance its existing reporting on its investments in
TALF that strikes an appropriate balance between its goal of
transparency and the need to avoid compromising either the competitive
positions of investors or Treasury's ability to recover funds for
taxpayers. We believe that having a plan in place for tracking and
reporting on the performance of any collateral that could be placed in
TALF LLC will help Treasury strike that balance.
In its comments, the Federal Reserve did not agree with our
recommendation that Congress consider providing GAO with authority to
audit the Federal Reserve's TALF operational and administrative
actions because it disagreed that there are limitations on GAO's
authority to audit these Federal Reserve activities. The Federal
Reserve also noted that it fully cooperated in GAO's conduct of this
audit and provided us access to records and personnel.
The Federal Reserve did cooperate and voluntarily provided all access
we requested in this audit of Treasury. We appreciate this
cooperation, which enabled us to factually describe the TALF program
and to evaluate Treasury's involvement in it. However, we believe the
express statutory prohibition in 31 U.S.C. § 714(b) on GAO auditing
the Federal Reserve's monetary policy and discount window activities,
which the Federal Reserve believes include TALF's operation and
administration, prohibits us from auditing the Federal Reserve's TALF
activities, even from the perspective of TARP. We limited the scope
and conduct of this audit accordingly, and thus did not request access
to information to audit the Federal Reserve's performance of these
activities.[Footnote 60] Further, the Federal Reserve's decision to
voluntarily provide requested access in this instance, while helpful,
does not create GAO authority for access to information the agency may
not volunteer, nor GAO authority to audit the Federal Reserve's TALF
operational activities or other performance. In our view, our lack of
authority to audit the Federal Reserve's actions limited our ability
to fully assess the risk to taxpayer funds presented by TALF.
Accordingly, we continue to believe that Congress should provide GAO
with authority to audit the Federal Reserve's operation and
administration of the TALF program. Our detailed response to the
Federal Reserve's comments on these issues is contained in appendix
VIII.
We are sending copies of this report to the Congressional Oversight
Panel, Financial Stability Oversight Board, Special Inspector General
for TARP, interested congressional committees and members, Treasury,
the federal banking regulators, and others. The report also is
available at no charge on the GAO Web site at [hyperlink,
http://www.gao.gov].
If you or your staffs have any questions about this report, please
contact Orice Williams Brown at williamso@gao.gov or (202) 512-8678.
Contact points for our Offices of Congressional Relations and Public
Affairs may be found on the last page of this report. GAO staff who
made major contributions to this report are listed in appendix XI.
Signed by:
Gene L. Dodaro:
Acting Comptroller General of the United States:
List of Committees:
The Honorable Daniel K. Inouye:
Chairman:
The Honorable Thad Cochran:
Vice Chairman:
Committee on Appropriations:
United States Senate:
The Honorable Christopher J. Dodd:
Chairman:
The Honorable Richard C. Shelby:
Ranking Member:
Committee on Banking, Housing, and Urban Affairs:
United States Senate:
The Honorable Kent Conrad:
Chairman:
The Honorable Judd Gregg:
Ranking Member:
Committee on the Budget:
United States Senate:
The Honorable Max Baucus:
Chairman:
The Honorable Charles E. Grassley:
Ranking Member:
Committee on Finance:
United States Senate:
The Honorable David R. Obey:
Chairman:
The Honorable Jerry Lewis:
Ranking Member:
Committee on Appropriations:
House of Representatives:
The Honorable John M. Spratt, Jr.
Chairman:
The Honorable Paul Ryan:
Ranking Member:
Committee on the Budget:
House of Representatives:
The Honorable Barney Frank:
Chairman:
The Honorable Spencer Bachus:
Ranking Member:
Committee on Financial Services:
House of Representatives:
The Honorable Charles B. Rangel:
Chairman:
The Honorable Dave Camp:
Ranking Member:
Committee on Ways and Means:
House of Representatives:
[End of section]
Appendix I: Objectives, Scope, and Methodology:
The objectives of this report are to (1) analyze the risks that the
Term Asset-Backed Securities Loan Facility (TALF) presents to Troubled
Asset Relief Program (TARP) funds and therefore to taxpayers, (2)
evaluate how the Department of the Treasury (Treasury) analyzed the
risk of TALF assets and used this information in making decisions on
TALF with the Board of Governors of the Federal Reserve System
(Federal Reserve) and the Federal Reserve Bank of New York (FRBNY),
and (3) assess the condition of securitization and credit markets
before and after TALF's implementation based on indicators tracked by
Treasury and FRBNY.
GAO has statutory limitations on auditing certain functions of the
Federal Reserve.[Footnote 61] Because of these limitations, the
evaluative content of this report is limited to Treasury's role of
safeguarding TARP funds related to TALF and we did not review or
evaluate any monetary policy actions taken by the Federal Reserve or
FRBNY with respect to TALF. We collected information on Federal
Reserve practices related to TALF, but did not audit those practices.
Specifically, we did not evaluate the sufficiency of how certain TALF
program terms, such as haircuts and interest rates, were arrived
at.[Footnote 62] In addition, we did not assess FRBNY's system of
internal control or the role of TALF participants such as agents,
borrowers, and auditors in certifying and validating compliance with
certain TALF terms. Finally, we did not validate the comments or
background information provided to us by Federal Reserve and FRBNY
officials about TALF.
To address the first objective, we first reviewed publicly available
documentation on the Web sites of the Federal Reserve and FRBNY. We
also interviewed Treasury, FRBNY, and Federal Reserve officials to
understand how TALF fits in to Treasury's Financial Stability Plan and
how risks to the taxpayer were reduced in TALF's design. Next, we
assessed how Treasury reviewed the risks of the various asset classes
considered for TALF eligibility by collecting and analyzing reports
that Treasury requested through a contractor, Bank of New York Mellon,
which in turn subcontracted the work to NSM Structured Credit
Solutions, which has since been acquired and is now known as
RangeMark. We also reviewed the subcontractor's methodology for
assessing the likelihood of loss to TARP funds and interviewed the
subcontractor, contractor, and Treasury officials about the
assumptions in the loss model. We also reviewed other factors that
have an impact on the risk to TARP funds and taxpayers, including the
return on equity for TALF borrowers, credit enhancement of TALF
securities, and the risks of asset-backed securities (ABS) and
commercial mortgage-backed securities (CMBS).
* To assess the changes in return on equity (ROE), we analyzed the
returns based on information collected from prospectuses for TALF-
eligible ABSs on credit cards, auto loans, auto leases, and private
student loans issued between March and September 2009. Some of these
prospectuses were provided by the Federal Reserve. We also used
information collected from reports from Moody's Investors Service and
Standard & Poor's. We calculated returns for fixed-rate bonds by using
the tranche-level[Footnote 63] interest rate paid to the FRBNY. For
floating-rate bonds, we used a spread between the interest rate paid
to the FRBNY and an index, such as the London Interbank Offered Rate.
This is the "coupon" variable in the equation below:
ROE = Coupon - (1 - Haircut%)* Rate _ on _ loan _ paid _ to _ FRBNY.
divided by:
Haircut%.
* To assess the levels of credit enhancement for TALF securities, we
analyzed information collected from prospectuses related to public and
private offerings of TALF-eligible securities issued between March and
September 2009, along with related reports from Standard & Poor's and
Moody's Investors Service. For each security, we compared the level of
credit enhancement for the TALF issuance with that issuer's most
recent securitization prior to TALF, which ranged from 2004 through
2008.
* To understand the recent activity in CMBS markets, we collected
information from Moody's Investors Service on commercial real estate
prices (Moody's/REAL Commercial Property Price Index) and on CMBS
delinquency (Moody's CMBS Delinquency Tracker). We determined that the
data was reliable for our purposes of demonstrating recent trends in
the commercial mortgage sector. In addition, we collected CMBS price
performance data from Thomson Reuters DataScope and determined that
the information on the price, yield, and performance of securities was
reliable for our purposes of understanding trends in CMBS prices and
vintages for the TALF portfolio.
* We interviewed a range of market participants and market observers
about the taxpayer protections and other features of TALF, to include
three dealers that also serve as TALF agents; three issuers (one for
credit cards, one for auto loans, and one for student loans) and an
SBA securities dealer; three industry associations representing the
CMBS market, the hedge fund industry, and small and regional banks; a
buy-side investment firm with interest in TALF; two large auditing
firms that provide auditor attestations for TALF; an attorney with
securitization market expertise; two TALF-qualified credit rating
agencies, or nationally recognized statistical rating organizations;
an academic in banking and securitization at the Massachusetts
Institute of Technology; an analyst from Brookings Institution; an
analyst from a student loan firm; and a representative of a consumer
advocacy organization.
* For details on our methodology for assessing adverse scenario losses
from TALF ABSs and CMBSs, see appendix II.
To address the second objective on Treasury's analysis of the risk
associated with TALF assets and how that analysis was used to make
decisions with the Federal Reserve and FRBNY related to TALF, we
analyzed reports from a subcontractor with Treasury--NSM Structured
Credit Solutions--that provided assessments of various risks of TALF
to TARP funds. In analyzing these reports, we reviewed the asset class
risk assessments, the recommendations made to change TALF program
terms, and the suggested haircuts for each asset class. We also
interviewed Treasury's contractor and the subcontractor for
clarification on the reports and to understand Treasury's interaction
with both. In addition, we interviewed Treasury officials about their
role in reviewing and shaping the terms of TALF, how they considered
the analysis and recommendations of the subcontractor, how they
decided to include certain asset classes, and how they came to agree
on haircuts and other program terms with the FRBNY and Federal
Reserve. We also interviewed officials from FRBNY and the Federal
Reserve about the reasons for differences in haircuts and other TALF
program terms and how they were resolved with Treasury and the
subcontractor.
To address the third objective on changes in the securitization and
credit markets before and after TALF was created, and to understand
how Treasury tracks the impact of TALF and its potential risks to TARP
funds, we collected and analyzed information from a variety of data
sources relevant to the ABS, CMBS, and credit markets. Specifically:
* To review changes in securitization markets for ABSs backed by auto
loans, credit cards, student loans, and commercial mortgages, we
collected data from Thomson Reuters IFR Markets, a database that
collects information on activity in the securitization markets. To
analyze changes in interest rates for auto loans and credit cards, we
reviewed quarterly data from the Federal Reserve's G.19 Consumer
Credit Release, a widely used data source, as well as weekly data
provided by BankRate.com. We selected the auto, credit card, student
loan, and CMBS asset classes because they were the most widely traded
in securitization markets and the latter had recently experienced
significant trading and price volatility. Because reliable interest
rate data for private student loans and commercial mortgages were more
difficult to obtain, we collected and analyzed data only on auto loans
and credit cards. We validated the securitization and interest rate
information against reports and data provided by credit rating
agencies, issuers, and dealers. We determined that the data sources
were sufficient for our purposes of demonstrating trends in the
markets before and after TALF was created.
* To report on the amount of TALF loans settled, we accessed data
publicly available on the FRBNY Web site and also information provided
to us from FRBNY for periods when FRBNY did not publicly report the
settled loan amounts, but only the requested amounts. Because of the
limitations on our audit authority, we did not review the internal
systems that generated this information.
* To analyze spreads for ABSs backed by auto loans, credit cards,
student loans, and CMBSs, we analyzed dealer-provided data from three
dealer banks. Because this spread information is not available from
one data provider, we determined that collecting data from three
dealers--and ensuring that the numbers were within an acceptable range
of each other--would ensure the reliability of such data for our
purposes.
* To determine what information Treasury collects to assess TALF's
impact on securitization and credit markets and the risks TALF poses
to TARP funds, we interviewed officials from the Treasury about what
data they collect and received reports that Treasury's subcontractor
provided on the various risks of TALF activities. We also interviewed
Federal Reserve and FRBNY officials about what type of data they
collect related to TALF's impact on the securitization and credit
markets.
[End of section]
Appendix II: Methodology for Market Value Analysis of ABSs and CMBSs:
To understand the possible range of losses to Troubled Asset Relief
Program (TARP) funds from the Term Asset-Backed Securities Loan
Facility (TALF), we conducted an analysis based on extreme market
value losses, similar to those experienced in the asset-backed
securities (ABS) and commercial mortgage-backed securities (CMBS)
markets in November 2008. This provides an alternative approach to the
Department of the Treasury (Treasury) subcontractor's analysis, and
provides an estimate of how large losses potentially could be in the
event that the markets returned to their November 2008 lows. Selecting
November 2008 as the market low point is generally consistent with the
approach used by the Board of Governors of the Federal Reserve System
(Federal Reserve) in its "stress tests" for determining the capital
that large bank holding companies must maintain. Our scenario provides
a more-adverse than expected loss estimate for our sample of ABSs and
all of the CMBS loans remaining as of September 30, 2009.
Our first analysis focused on ABSs. We conducted a market value
analysis on a sample of the three largest asset classes--ABSs backed
by credit cards, auto loans, and student loans--because they make up
the majority of TALF's portfolio. Of the $42.5 billion in TALF loans
backed by credit card, auto loan, and student loan ABSs that had been
disbursed as of September 30, 2009, we took a sample of $16.5 billion,
or 39 percent. The sample was selected to broadly match the makeup of
these asset classes in this subset of the TALF portfolio (see table 4).
Table 4: GAO Sample Selection for ABS Extreme Market Value Loss
Analysis:
Credit cards;
Collateral: $9.35 billion;
Percent of sample: 57%;
Percent of TALF Portfolio (as of Sept. 30, 2009): 56%.
Auto loans;
Collateral: $4.54 billion;
Percent of sample: 27;
Percent of TALF Portfolio (as of Sept. 30, 2009): 30.
Student loans;
Collateral: $2.64 billion;
Percent of sample: 16;
Percent of TALF Portfolio (as of Sept. 30, 2009): 14.
Total;
Collateral: $16.53 billion;
Percent of sample: 100%;
Percent of TALF Portfolio (as of Sept. 30, 2009): 100%.
Source: GAO.
[End of table]
Within each asset class the sample was selected to include ABSs that
gave the largest sample size on a TALF loan dollar basis; hence larger
deals predominate. In addition, TALF loans were spread across 6 of the
7 TALF ABS subscription months between March and September.
Nevertheless, it is a nonprobability sample and is not necessarily
generalizable to all TALF deals.
We modeled TALF collateral cash flows using assumptions consistent
with the FRBNY's assumptions on the rate at which ABS principal is
paid back to investors. Then we calculated the discount rate that
brought the price back to par as of the issuance date. To this
discount rate, we added the incremental yield (or spreads) that would
be required to deplete the borrowers' entire equity investment and any
excess interest that had built up in TALF LLC as of September 30,
2009, for that specific tranche.[Footnote 64] These incremental
spreads were compared with the widest levels seen in November 2008 for
the appropriate asset class and expected average life. November 2008
spreads were obtained from Wall Street ABS-dealer weekly price data
that are published for the more widely traded ABS classes. If the
spread seen in November 2008 was greater than that required to deplete
the borrower's equity and TALF's excess interest, a stress loss was
calculated. No loss was assumed if the required spread widening was
less than the extremes of November 2008.
Our analysis makes the following assumptions: (1) excess interest has
accumulated as of September 30, 2009 at the tranche level of each TALF
security; (2) borrowers will surrender their TALF ABS collateral to
the FRBNY and stop paying the TALF loan when the ABS market value
falls below the TALF loan balance; (3) TALF will mark-to-market the
surrendered collateral, ignoring any recovery that Treasury might make
if the ABS collateral fully pays all cash flows over the life of the
securities; and (4) the change in market value is strictly based on
mark-to-market, with no assumption about the underlying credit
performance of the ABS.
For the separate analysis on the risks that legacy CMBS collateral may
pose to TARP funds, we compared the prices on the 139 legacy CMBS
CUSIPs that were accepted by FRBNY as of September 30, 2009, with the
exception of 2 for which no price information was available.[Footnote
65] These prices were then compared with the lowest prices that, on
average, most CMBS across the TALF portfolio reached in November 2008.
This CMBS analysis did not include consideration of the excess
interest accumulated in TALF LLC but otherwise made the same
assumptions noted above for the ABS analysis. As discussed earlier,
the $198 million of excess interest that had accumulated in the cash
collateral account as of December 31, 2009, would be available to
absorb the first losses born on surrendered collateral prior to any
outlay by Treasury, and this amount is expected to increase over time.
In conducting this analysis we utilized certain data from Thompson
Reuters Datascope and the Federal Reserve.
[End of section]
Appendix III: The Securitization Process Explained:
Securitization is a process that packages relatively illiquid
individual financial assets--such as loans, leases, or receivables--
and converts them into interest-bearing, asset-backed securities (ABS)
that are marketable to capital market investors. As outlined below,
the market participants in securitization--borrowers, consumer and
small business lenders, investment banks or pool assemblers, credit
rating agencies, and investors--each derive specific benefits from the
transaction. For example, borrowers might gain access to loanable
funds with more favorable terms, such as longer repayment periods and
lower interest rates, than may otherwise be available. Similarly,
securitization offers consumer and small business lenders a funding
source for making new loans, improving balance sheet and capital
management, and diversifying fee or income streams. Securitization
also allows the cash flows from asset pools to be structured to
satisfy the maturity, risk, and return preferences of investors.
The degree to which participants receive these benefits depends, in
large part, on how efficiently the markets for securitized assets are
functioning. With accurate and more comprehensive performance data
regarding financial assets, capital markets can more easily profile
the risk of a pool of similar assets. This risk can be divided and
sold to investors who are willing to purchase it at an acceptable risk-
adjusted return, sometimes called the "investor-required yield." As
the markets for securitized asset classes grow in volume and
liquidity, and as the performance and risk characteristics of those
assets become better understood, investor-required yields on
particular ABSs and transaction costs of securitizing those assets may
decline. Declining investor-required yields and transaction costs can
lower the cost of financing for consumer and small business lenders
and ultimately borrowers. Conversely, with inadequate performance
data, and low volumes of similar financial assets, these benefits may
not sufficiently materialize for securitization to be a viable
financing arrangement for consumer and small business lenders or
borrowers.
Figure 11: The Securitization Process:
[Refer to PDF for image: illustration]
* Investors purchase securities with cash flows that have desirable
risk-return and maturity characteristics. A single pool can often
contain multiple classes, or "tranches," of securities.
* The issuing trust holds the pool of assets which are insulated from
the performance and credit of the underwriter and originating lenders.
The issuing trust sells securities to investors.
* Nationally recognized statistical rating organizations (NRSRO), or
credit rating agencies, assess the performance and expected losses
(credit quality) of the pool of assets, including internal and
external credit enhancements, and provide a credit rating on the
securities to be sold.
* The underwriter structures assets within the ABS trust and
facilitates the sale of securities to investors. Underwriters, as part
of structuring, stratify the credit and payment positions of cash
flows generated from the pool into different classes of securities, or
tranches, based on investor preferences.
* For the Term Asset-Backed Securities Loan Facility (TALF), auditors
either provide an attestation or use agreed upon procedures to certify
certain characteristics of TALF collateral.
* Lenders originate loans that conform to underwriting criteria
acceptable to a pool of loans and ultimately sell the loan into the
pool. Lenders may fund credit enhancements to support the credit
quality of a pool of loans and service the loan by collecting payments
for distribution to the issuing trust, which in turn remits payments
to bondholders.
* Borrowers finance consumer spending using credit cards, auto and
student loans, and then provide specified repayments of principal and
interest to lenders.
Source: GAO.
[End of figure]
[End of section]
Appendix IV: Descriptions of Asset-Backed Securities:
Auto Loan and Lease:
Asset-backed securities (ABS) for the auto market compose the largest
share of ABS issuances. Auto securitizations are collateralized with a
fixed pool of loans. In most cases, these transactions are divided
into at least four senior segments, or tranches, which have the same
payment priority in the event of default but different priorities for
principal repayment (with the exception of the shortest pay securities
or A1 tranche designed to be marketable to money market funds, which
take priority). Tranches are structured so that all scheduled
principal amortization and prepayments of principal are paid back
first to the tranche with the lowest interest rate. This tranche is
generally designated the A1 tranche. Once the principal on the first
tranche is paid off, subsequent principal is paid to the A2, A3, and
A4 bondholders. The sizes of the tranches are designed so that the
expected average life on these securities is generally consistent
within each tranche--for instance, the A1 average life is usually 3
months, the A2 average life 1 year, the A3 average life 2 years, and
the A4 average life approximately 3 years or more. In some deals,
there are also "subordinate tranches," or tranches that receive
ratings below AAA. In many cases, the issuer retains subordinate
tranches rather than selling them to the public. Auto ABSs include the
following subasset classes: prime auto loans, subprime auto loans,
auto leases, and motorcycle loans.
Credit Card:
Credit card ABSs tend to use a master trust structure through which a
credit card issuer collateralizes a series of ABS issuances with
receivables from a large pool of credit card accounts. This pool is
not a static set of account balances but absorbs new receivables as
they are created. New issuance can be used to support an increase in
the size of the receivables collateralizing the securitizations.
Investments in credit card ABSs are usually divided into senior and
subordinated, or junior, tranches where the investors in the senior
tranches are paid first.
Student Loan:
Student loan ABSs can be collateralized with either federally
guaranteed Federal Family Education Loan Program (FFELP) loans or
consumer loans that are not part of a government guarantee program.
Student loan ABSs tend to have longer terms to maturity than other ABS
classes due to the longer repayment terms and the fact that students
do not tend to pay any principal or interest until at least 6 months
after they graduate, thus lenders might not receive cash flows on a
student loan for years after the initial cash disbursement.
Insurance Premium Finance:
Insurance premium finance ABSs are collateralized with loans made to
businesses to finance their property and casualty insurance coverage.
The typical commercial insurance policy requires a down payment, with
equal monthly payments, typically over a time frame shorter than the
term of the insurance policy, which in effect creates
overcollateralization. When a policy is canceled, refunds of unearned
premiums will be used for making payments to the securitization trust
for the remaining term of the loan to protect the ABS holders.
Commercial Mortgage-Backed Securities:
Cash flows on commercial mortgage-backed securities (CMBS) are
generally backed by principal and interest payments on a pool of
commercial mortgage loans. Most commercial mortgage loans are
structured with a 30-year amortization term, but CMBS terms are
generally shorter than the corresponding amortization terms. However,
recent years have seen an increase in the number of loans with
interest-only periods during which the mortgagee pays no principal on
the mortgage. Commercial mortgages are made on a wide variety of
different property types, including rental apartment buildings,
industrial properties, office buildings, hotels, healthcare related
properties, and retail properties such as shopping malls, strip malls,
and freestanding outlets. CMBSs are highly structured and frequently
have more than 20 tranches in their capital structure. The coupon
payment generally is positively correlated with both the expected
average life of the tranche and the risk that the bondholder will not
receive the entire principal amount. Also differentiating CMBSs from
other asset classes is their sensitivity to the underlying commercial
real estate prices and the cash flow generated from the commercial
properties backing the mortgages.
Commercial Fleet Leases:
Unlike other ABS classes for which investors own direct stakes in the
trust assets, commercial fleet lease ABSs are collateralized with
special units of beneficial interest (SUBI) in open-ended leases and
fleet management receivables on a pool of vehicle leases mainly for
commercial trucks, trailers, and equipment. The leases are made on a
per-vehicle basis to large corporate customers with fleets that may
have more than 5,000 vehicle leases with the issuer. Open-ended leases
require the lessee to reimburse any loss in a vehicle's residual value
to the lessor. Commercial fleet ABSs usually are structured out of a
master trust with the ability to issue numerous term securities.
Collateral in the master trust can be replenished with new or renewed
leases as older contracts prepay and expire. The lease SUBI entitles
the ABS holders to receive the monthly lease payments. This SUBI also
includes beneficial interest in all the vehicles that are being leased
or are in the process of being leased but have not completed the
process. The fleet management receivables SUBI includes beneficial
interest in the receipt of management and other fees that the lessees
pay to the lessor.
Mortgage Servicing Advances:
Mortgage servicing advance ABSs are collateralized with receivables
owed to the servicer for servicing advances made by the servicer to
and on behalf of the residential mortgage-backed securitization (RMBS)
trusts. There are three types of advances: principal and interest,
which cover these types of payments on delinquent loans; escrow
advances, which cover expenses related to maintaining ownership of a
mortgaged property, including property taxes and insurance premiums;
and corporate advances, which are costs for the process of
foreclosure, including attorney and other professional fees and
expenses related to maintaining a repossessed home. As there is no
interest paid to the RMBS trusts when advances are paid back out of
either the proceeds of the liquidation of a repossessed property (loan-
level servicing advances) or broader pool level cash flows (pool-level
servicing advances), a discount factor is applied. The discount factor
reflects the estimated time frame for repaying the loan. As a result
of this discount factor, the issuer receives less than the face value
of the servicing advance at the time of securitization.
Dealer Floor Plans:
Floor plan ABSs are collateralized by loans made to finance either
automobiles or nonautomotive durable goods. Nonautomotive floor plan
inventory includes, among other things, recreational vehicles, boats,
motorcycles, industrial equipment and farm equipment, appliances, and
electronics. Automotive dealer floor plan arrangements tend to be
between a single financial entity and a dealer network. Nonauto
dealers can have multiple floor plan arrangements with several capital
providers. Financing could be in the form of revolving or nonrevolving
lines of credit. Once a floor plan agreement is in place, dealers
place orders for inventory from the manufacturer and specify that a
lender will provide the financing. The loan is repaid by proceeds from
inventory sales, or the dealer can arrange to repay the loan in
monthly installments.
Equipment:
Equipment ABSs are collateralized with retail installment sale
contracts, loans and leases secured by new and used agricultural
equipment, construction equipment, industrial equipment, office
equipment, copiers, computer equipment, telecommunications equipment,
and medical equipment, among other things.
SBA-Loan Backed ABS:
The Small Business Administration (SBA) provides guarantees on loans
made to small businesses. The most common SBA loan programs are 7(a)
and 504. In the 7(a) program, SBA guarantees up to 85 percent of the
loan amount made by participating lenders. 7(a) loans are usually made
for general business purposes, including working capital, equipment,
furniture and fixtures, and land and buildings. 504 program loans are
typically long-term, fixed-rate loans for the purpose of expanding or
modernizing a small business. When pooled together for securitization
purposes, underlying loans must have similar terms and features--for
example, similar maturity dates.
[End of section]
Appendix V: Credit Enhancement:
Credit enhancements are features in the structuring of a
securitization that protect investors in the securitization from
losses due to defaults on the underlying loans. The following are some
methods of credit enhancement that have been used on asset-backed
securities (ABS) eligible for the Term Asset-Backed Securities Loan
Facility (TALF).
Subordination: This feature is a method of prioritizing cash flows
from the underlying loan collateral. The senior tranches within a
securitization get priority over subordinate, or junior, tranches in
the event of a default on the underlying collateral.[Footnote 66] All
TALF-eligible securitizations must have a AAA rating from at least two
TALF-eligible nationally recognized statistical rating organizations,
and all of the AAA-rated ABSs have first priority for cash flows.
While most AAA tranches or bonds within an ABS have the same priority
in the event of a default, the sequential nature of the principal
paydown for certain classes of ABSs (for example, auto loans) leads to
higher risk of default for the tranches with weighted average lives
that extend further into the future. This higher risk requires the
issuer to pay a higher interest rate or coupon on longer tranches. Any
losses are applied to the most subordinate tranche first.
Overcollateralization: When the total face value on the loan
collateral underlying an ABS is greater than the face value of the
bonds, the securitization is said to be overcollateralized. These
assets are maintained on the balance sheet of the issuer and are the
first to absorb credit losses on the collateral.
Reserve account: This is a cash account set up at the origination of
an ABS. This account is accessed when the cash flows from the
collateralized loan assets are insufficient to cover the contractual
payments on the bonds, including servicing and other fees.
Excess spread: Excess spread refers to the funds leftover after
payments to bondholders and other contractual obligations have been
met. This can be used to make up for insufficient cash flows if the
underlying borrowers are delinquent or default on the loan.
Yield-supplement overcollateralization: This feature applies to
securitizations with assets in the underlying pool that are paying
interest that is below the coupon rate on the bonds. For example,
borrowers frequently pay very low interest rates on loans within
certain auto loan securitizations. These loans are often extended with
advantageous borrower terms as part of a sales promotion. Generally,
the issuer will set up a yield-spread overcollateralization account to
make up the difference over some portion of the life of the
securitization. The initial balance in this account is set as the
present value of the shortfall on those loans for the life of the
loans.
Mortgage servicing advance discounts: A form of enhancement that is
implicit in the discounted price at which the securitization trust
purchases the servicing advance receivables from the mortgage
servicing company issuing the security. The servicing advances are
segregated into several classifications based on whether the servicing
advances are treated at the pool-or loan-level in order of repayment,
whether the underlying mortgage is located in a state that has a
judicial or nonjudicial foreclosure regime, and the type of cash flow
for which the servicer is advancing payment. The servicing advance is
classified into one of three classes: principal and interest advance,
escrow advances, and corporate advances. Principal and interest
advances are made by mortgage servicers to holders of residential
mortgage-backed securities for mortgages whose underlying borrowers
are delinquent on their monthly payments. Escrow advances are used to
pay the property taxes, insurance premiums, or other property-related
expenses that the borrowers should have paid. Corporate advance costs,
usually in the form of attorneys' and other professional fees, are
also accounted for in the event that the servicer incurs them while
foreclosing on and liquidating repossessed real estate.
The advance discount percentage is calculated based on assumptions
about the length of time it will take to repay that particular type of
advance and the risk that it might not be paid back. Pool-level
servicing advances have the lowest discount percentage, because the
advances can be repaid to the servicer out of the entire pool's
available funds, including principal and interest payments received
for nondelinquent mortgages. Loan-level servicing advances are not
repaid until the borrower repays all the money advanced or from the
proceeds of the sale of the repossessed property (the likely scenario
in a default). Servicing advances on mortgages secured with properties
in judicial foreclosure states have higher discount rates than those
in nonjudicial states, because judicial foreclosures take longer.
Principal and interest servicing advances are:
viewed as the safest instruments and have lower discount rates than
escrow, which in turn has slightly lower discount factors than
corporate advances.
[End of section]
Appendix VI: Additional Information on TALF Compliance:
According to Federal Reserve Bank of New York (FRBNY) officials, FRBNY
has in place a number of compliance measures to (1) ensure that
borrowers and collateral are eligible for the Term Asset-Backed
Securities Loan Facility (TALF); (2) protect FRBNY from fraudulent
activity; (3) reduce the risk of fraud and address conflicts of
interest; (4) ensure that agents have adequate compliance regimes; and
(5) build multiple layers of compliance where possible.
TALF has a certification regime in place for a number of TALF
participants. TALF agents and sponsors must certify that they are
complying with certain TALF requirements, and TALF agents review the
eligibility of TALF borrowers. According to FRBNY officials, TALF
agents are the first line of defense against fraudulent participants
in TALF, as they conduct "Know Your Customer" reviews of potential
TALF borrowers. FRBNY noted that it had antifraud measures in place
and receives referrals for those investors that TALF agents raised
concerns about. Moreover, FRBNY officials stated that they have
developed an inspection program to conduct on-site reviews of TALF
Agent's "Know Your Customer" programs and files. The entire process is
under the management of FRBNY, without the Department of the
Treasury's (Treasury) participation.
Sponsors and issuers must include in any offering document a
certification required by FRBNY. Borrowers must also provide
representations to the TALF agent, who conducts the review of the
borrower. According to FRBNY officials, because the issuers and
sponsors include certification to TALF eligibility and acknowledge
certain responsibilities related to the TALF collateral in the
offering documents, any material misrepresentations would be covered
under relevant securities laws. In addition, the TALF agents and
borrowers also make certain representations on their eligibility and
the eligibility of the collateral. Though this is a certification and
self-disclosure regime, FRBNY officials told us they have established
additional measures to detect and address noncompliance. First, FRBNY
has a 24-hour fraud hotline. Second, it has hired a law firm to assist
in assessing fraud risks associated with the program. Third, it is
cooperating with other government and law enforcement agencies to
gather additional information on potential TALF participants.
In addition to certifications, FRBNY requires auditor attestations for
non-CMBS collateral, which state that the Report on Management
Compliance fairly states compliance with certain TALF program criteria
specified by FRBNY. For CMBS collateral, agreed upon procedures (AUP)
are required to provide more detailed specifications on what to
review.[Footnote 67] FRBNY has published broad guidelines to the
auditors for carrying out these responsibilities, which are paid for
by the issuers of TALF-eligible securities. In addition, FRBNY
requires that the attestation and AUP processes follow standards
issued by the Public Company Accounting Oversight Board and the
American Institute of Certified Public Accountants. Most of the
information that the auditors review is provided by the issuers and is
not verified independently, according to auditors we spoke with. FRBNY
officials added that loan-level testing is required and this includes
a review of original loan files or electronic versions thereof.
According to Treasury officials, Treasury provided some input into the
design of the auditor attestations and AUPs but primarily leaves
oversight of this function to the Board of Governors of the Federal
Reserve System and FRBNY, which are responsible for designing and
implementing TALF. Treasury does not review these documents; however,
should the assets be placed to TALF LLC, Treasury may review them.
[End of section]
Appendix VII: New Securitization Volumes Have Increased since the
Inception of the TALF Program:
Since the Term Asset-Backed Securities Loan Facility's (TALF) March
2009 inception, securitization volumes have increased in some TALF-
eligible sectors. For auto loan securitization, new issuance dropped
off significantly in the third quarter of 2008, bottoming in the
fourth quarter (see table 5). By 2009 issuance began to pick up,
especially in the second and third quarters, when most were TALF-
eligible. There were 45 issuances through December 2, 2009, a marked
contrast to the peak of 85 in 2005.
Table 5: Auto Loan Securitization Volume, 2005 through Fourth Quarter
2009 (dollars in millions):
Year of issuance: 2005;
Deals: 85;
Total issuance volume: $76,912;
Number of TALF securitizations: [Empty];
Number of non-TALF securitizations: [Empty];
TALF Issuance Volume: [Empty];
Non-TALF issuance volume: [Empty].
Year of issuance: 2006;
Deals: 78;
Total issuance volume: $88,114;
Number of TALF securitizations: [Empty];
Number of non-TALF securitizations: [Empty];
TALF Issuance Volume: [Empty];
Non-TALF issuance volume: [Empty].
Year of issuance: 2007;
Deals: 69;
Total issuance volume: $71,015;
Number of TALF securitizations: [Empty];
Number of non-TALF securitizations: [Empty];
TALF Issuance Volume: [Empty];
Non-TALF issuance volume: [Empty].
Year of issuance: 2008;
Deals: 32;
Total issuance volume: $29,113;
Number of TALF securitizations: [Empty];
Number of non-TALF securitizations: [Empty];
TALF Issuance Volume: [Empty];
Non-TALF issuance volume: [Empty].
Year of issuance: 2009;
Deals: 45;
Total issuance volume: $51,192;
Number of TALF securitizations: 35;
Number of non-TALF securitizations: 10;
TALF Issuance Volume: $44,941;
Non-TALF issuance volume: $6,251.
Year of issuance: Q1 2008;
Deals: 10;
Total issuance volume: $10,265;
Number of TALF securitizations: [Empty];
Number of non-TALF securitizations: [Empty];
TALF Issuance Volume: [Empty];
Non-TALF issuance volume: [Empty].
Year of issuance: Q2 2008;
Deals: 14;
Total issuance volume: $14,022;
Number of TALF securitizations: [Empty];
Number of non-TALF securitizations: [Empty];
TALF Issuance Volume: [Empty];
Non-TALF issuance volume: [Empty].
Year of issuance: Q3 2008;
Deals: 5;
Total issuance volume: $3,226;
Number of TALF securitizations: [Empty];
Number of non-TALF securitizations: [Empty];
TALF Issuance Volume: [Empty];
Non-TALF issuance volume: [Empty].
Year of issuance: Q4 2008;
Deals: 3;
Total issuance volume: $1,600;
Number of TALF securitizations: [Empty];
Number of non-TALF securitizations: [Empty];
TALF Issuance Volume: [Empty];
Non-TALF issuance volume: [Empty].
Year of issuance: Q1 2009;
Deals: 5;
Total issuance volume: $7,583;
Number of TALF securitizations: 3;
Number of non-TALF securitizations: 2;
TALF Issuance Volume: $5,213;
Non-TALF issuance volume: $2,370.
Year of issuance: Q2 2009;
Deals: 11;
Total issuance volume: 14,952;
Number of TALF securitizations: 10;
Number of non-TALF securitizations: 1;
TALF Issuance Volume: $13,445;
Non-TALF issuance volume: $1,507.
Year of issuance: Q3 2009;
Deals: 14;
Total issuance volume: 19,294;
Number of TALF securitizations: 12;
Number of non-TALF securitizations: 2;
TALF Issuance Volume: $18,394;
Non-TALF issuance volume: $900.
Year of issuance: Q4 2009;
Deals: 15;
Total issuance volume: $9,363;
Number of TALF securitizations: 10;
Number of non-TALF securitizations: 5;
TALF Issuance Volume: $7,889;
Non-TALF issuance volume: $1,474.
Source: GAO analysis of Thomson Reuters IFR Markets data.
[End of table]
Credit card securitization volumes show a similar pattern (see table
6). The peak in credit card securitizations occurred in 2007, with 112
issuances, a sharp contrast to 2009 when only 35 securitizations were
issued through December 2, 2009. The majority of credit card ABSs
issuances in 2009--about 65 percent--have been TALF supported.
Table 6: Credit Card Receivables Securitization Volume, 2005 through
Fourth Quarter 2009 (Dollars in millions):
Year of issuance: 2005;
Deals: 80;
Dollar volume: $53,019;
Number of TALF securitizations: [Empty];
Number of non-TALF securitizations: [Empty];
TALF issuance volume: [Empty];
Non-TALF issuance volume: [Empty].
Year of issuance: 2006;
Deals: 93;
Dollar volume: $60,374;
Number of TALF securitizations: [Empty];
Number of non-TALF securitizations: [Empty];
TALF issuance volume: [Empty];
Non-TALF issuance volume: [Empty].
Year of issuance: 2007;
Deals: 112;
Dollar volume: $94,539;
Number of TALF securitizations: [Empty];
Number of non-TALF securitizations: [Empty];
TALF issuance volume: [Empty];
Non-TALF issuance volume: [Empty].
Year of issuance: 2008;
Deals: 67;
Dollar volume: $67,319;
Number of TALF securitizations:
[Empty];
Number of non-TALF securitizations: [Empty];
TALF issuance volume: [Empty];
Non-TALF issuance volume: [Empty].
Year of issuance: 2009;
Deals: 35;
Dollar volume: $45,988;
Number of TALF securitizations: 23;
Number of non-TALF securitizations: 12;
TALF issuance volume: $29,713;
Non-TALF issuance volume: $16,275.
Year of issuance: Q1 2008;
Deals: 25;
Dollar volume: $28,785;
Number of TALF securitizations: [Empty];
Number of non-TALF securitizations: [Empty];
TALF issuance volume: [Empty];
Non-TALF issuance volume: [Empty].
Year of issuance: Q2 2008;
Deals: 24;
Dollar volume: $28,180;
Number of TALF securitizations: [Empty];
Number of non-TALF securitizations: [Empty];
TALF issuance volume: [Empty];
Non-TALF issuance volume: [Empty].
Year of issuance: Q3 2008;
Deals: 18;
Dollar volume: $10,354;
Number of TALF securitizations: [Empty];
Number of non-TALF securitizations: [Empty];
TALF issuance volume: [Empty];
Non-TALF issuance volume: [Empty].
Year of issuance: Q4 2008;
Deals: 0;
Dollar volume: 0;
Number of TALF securitizations: [Empty];
Number of non-TALF securitizations: [Empty];
TALF issuance volume: [Empty];
Non-TALF issuance volume: [Empty].
Year of issuance: Q1 2009;
Deals: 2;
Dollar volume: $6,500;
Number of TALF securitizations: 1;
Number of non-TALF securitizations: 1;
TALF issuance volume: $3,000;
Non-TALF issuance volume: $3,500.
Year of issuance: Q2 2009;
Deals: 15;
Dollar volume: $20,385;
Number of TALF securitizations: 8;
Number of non-TALF securitizations: 7;
TALF issuance volume: $13,835;
Non-TALF issuance volume: $6,550.
Year of issuance: Q3 2009;
Deals: 15;
Dollar volume: $15,500;
Number of TALF securitizations: 12;
Number of non-TALF securitizations: 3;
TALF issuance volume: $10,775;
Non-TALF issuance volume: $4,725.
Year of issuance: Q4 2009;
Deals: 3;
Dollar volume: $3,603;
Number of TALF securitizations: 2;
Number of non-TALF securitizations: 1;
TALF issuance volume: $2,103;
Non-TALF issuance volume: $1,500.
Source: GAO analysis of Thomson Reuters IFR Markets data.
[End of table]
Student loan securitization volumes show similar patterns to the auto
and credit card asset classes, with a marked low of no new deals in
the last quarter of 2008 (see table 7). Lenders may be tightening
their lending standards, potentially resulting in fewer loans and
reducing the need to access the securitization markets as frequently
as in the past. Although both Federal Family Education Loan Program
(FFELP) and private loan securitizations are TALF eligible, to date no
FFELP deals have been underwritten to TALF eligibility standards.
There have been five TALF private student loan securitizations since
the program's inception.
Table 7: Student Loan Securitization Volume, 2005 through Fourth
Quarter 2009 (Dollars in millions):
Year of issuance: 2005;
Total student loan deals: 42;
Total dollar volume: $46,277;
FFELP deals: [Empty];
Private deals: [Empty];
FFELP volume: [Empty];
Private dollar volume: [Empty];
Private TALF dollar volume: [Empty];
Private Non-TALF dollar volume: [Empty].
Year of issuance: 2006;
Total student loan deals: 43;
Total dollar volume: $70,058;
FFELP deals: [Empty];
Private deals: [Empty];
FFELP volume: [Empty];
Private dollar volume: [Empty];
Private TALF dollar volume: [Empty];
Private Non-TALF dollar volume: [Empty].
Year of issuance: 2007;
Total student loan deals: 28;
Total dollar volume: $50,672;
FFELP deals: 23;
Private deals: 5;
FFELP volume: $42,663;
Private dollar volume: $8,009;
Private TALF dollar volume: [Empty];
Private Non-TALF dollar volume: [Empty].
Year of issuance: 2008;
Total student loan deals: 22;
Total dollar volume: $29,427;
FFELP deals: 21;
Private deals: 1;
FFELP volume: $29,302;
Private dollar volume: $125;
Private TALF dollar volume: [Empty];
Private Non-TALF dollar volume: [Empty].
Year of issuance: 2009;
Total student loan deals: 20;
Total dollar volume: $21,247;
FFELP deals: 13;
Private deals: 7;
FFELP volume: $12,350;
Private dollar volume: $8,897;
Private TALF dollar volume: $7,367;
Private Non-TALF dollar volume: $1,530.
Year of issuance: Q1 2008;
Total student loan deals: 6;
Total dollar volume: $8,400;
FFELP deals: 6;
Private deals: 0;
FFELP volume: $8,400;
Private dollar volume: 0;
Private TALF dollar volume: [Empty];
Private Non-TALF dollar volume: [Empty].
Year of issuance: Q2 2008;
Total student loan deals: 10;
Total dollar volume: $14,624;
FFELP deals: 10;
Private deals: 0;
FFELP volume: $14,624;
Private dollar volume: 0;
Private TALF dollar volume: [Empty];
Private Non-TALF dollar volume: [Empty].
Year of issuance: Q3 2008;
Total student loan deals: 6;
Total dollar volume: $6,403;
FFELP deals: 5;
Private deals: 1;
FFELP volume: $6,278;
Private dollar volume: $125;
Private TALF dollar volume: [Empty];
Private Non-TALF dollar volume: [Empty].
Year of issuance: Q4 2008;
Total student loan deals: 0;
Total dollar volume: 0;
FFELP deals: 0;
Private deals: 0;
FFELP volume: 0;
Private dollar volume: 0;
Private TALF dollar volume: [Empty];
Private Non-TALF dollar volume: [Empty].
Year of issuance: Q1 2009;
Total student loan deals: 2;
Total dollar volume: $2,047;
FFELP deals: 1;
Private deals: 1;
FFELP volume: $547;
Private dollar volume: $1,500;
Private TALF dollar volume: 0;
Private Non-TALF dollar volume: $1,500.
Year of issuance: Q2 2009;
Total student loan deals: 5;
Total dollar volume: $8,315;
FFELP deals: 4;
Private deals: 1;
FFELP volume: $5,722;
Private dollar volume: $2,593;
Private TALF dollar volume: $2,593;
Private Non-TALF dollar volume: 0.
Year of issuance: Q3 2009;
Total student loan deals: 5;
Total dollar volume: $6,124;
FFELP deals: 1;
Private deals: 4;
FFELP volume: $1,910;
Private dollar volume: $4,214;
Private TALF dollar volume: $4,184;
Private Non-TALF dollar volume: 30.
Year of issuance: Q4 2009;
Total student loan deals: 8;
Total dollar volume: $4,761;
FFELP deals: 7;
Private deals: 1;
FFELP volume: $4,171;
Private dollar volume: $590;
Private TALF dollar volume: $590;
Private Non-TALF dollar volume: 0.
Source: GAO analysis of Thomson Reuters IFR Markets data.
Note: Thomson Reuters IFR Markets data does not break out FFELP and
private securitizations until 2007.
[End of table]
This report discussed the severe disruption in the commercial real
estate sector following the economic downturn. Table 8 shows that
commercial mortgage-backed securities (CMBS) volumes peaked in 2006
with 97 issuances before dropping dramatically to just 7 by 2008.
These sharp declines in part motivated the inclusion of CMBSs as a
TALF-eligible asset class. Three new-issue deals have been offered
since TALF was expanded to CMBSs, and only one used TALF for
financing. Other CMBS deals in 2009 were repackaging of existing
securitizations. Part of the sluggish activity in the CMBS sector
could be attributed to the length of time it takes to put together a
deal, which officials have noted is considerably longer than for the
other asset classes. There could be other reasons as well. As we
discussed in this report, the CMBS sector continues to show signs of
volatility resulting from sharp declines in commercial real estate
prices and increases in CMBS delinquency rates.
Table 8: CMBS Securitization Volume, 2005 Through Fourth Quarter 2009
(Dollars in millions):
Year of issuance: 2005;
Number of deals: 80;
Par value: $152,271;
TALF: [Empty];
Non-TALF: [Empty].
Year of issuance: 2006;
Number of deals: 97;
Par value: $201,419;
TALF: [Empty];
Non-TALF: [Empty].
Year of issuance: 2007;
Number of deals: 73;
Par value: $199,925;
TALF: [Empty];
Non-TALF: [Empty].
Year of issuance: 2008;
Number of deals: 7;
Par value: $9,482;
TALF: [Empty];
Non-TALF: [Empty].
Year of issuance: 2009;
Number of deals: 6;
Par value: $2,500;
TALF: $400;
Non-TALF: $2,100.
Year of issuance: Q1 2007;
Number of deals: 17;
Par value: $45,959;
TALF: [Empty];
Non-TALF: [Empty].
Year of issuance: Q2 2007;
Number of deals: 25;
Par value: $74,360;
TALF: [Empty];
Non-TALF: [Empty].
Year of issuance: Q3 2007;
Number of deals: 19;
Par value: $54,236;
TALF: [Empty];
Non-TALF: [Empty].
Year of issuance: Q4 2007;
Number of deals: 12;
Par value: $25,370;
TALF: [Empty];
Non-TALF: [Empty].
Year of issuance: Q1 2008;
Number of deals: 3;
Par value: $4,387;
TALF: [Empty];
Non-TALF: [Empty].
Year of issuance: Q2 2008;
Number of deals: 2;
Par value: $1,955;
TALF: [Empty];
Non-TALF: [Empty].
Year of issuance: Q3 2008;
Number of deals: 2;
Par value: $3,140;
TALF: [Empty];
Non-TALF: [Empty].
Year of issuance: Q4 2008;
Number of deals: 0;
Par value: 0;
TALF: [Empty];
Non-TALF: [Empty].
Year of issuance: Q1 2009;
Number of deals: 0;
Par value: 0;
TALF: [Empty];
Non-TALF: [Empty].
Year of issuance: Q2 2009;
Number of deals: 2;
Par value: $343;
TALF: [Empty];
Non-TALF: $343.
Year of issuance: Q3 2009;
Number of deals: 0;
Par value: 0;
TALF: [Empty];
Non-TALF: [Empty].
Year of issuance: Q4 2009;
Number of deals: 4;
Par value: $2,157;
TALF: $400;
Non-TALF: $1,757.
Source: GAO analysis of Thomson Reuters IFR Markets data.
[End of table]
[End of section]
Appendix VIII: Comments from the Department of the Treasury:
Department Of The Treasury:
Assistant Secretary:
Washington, DC 20220:
February 1, 2010:
Thomas J. McCool:
Director, Center for Economics Applied Research and Methods:
U.S. Government Accountability Office:
441 G Street, N.W.
Washington, D.C. 20548:
Dear Mr. McCool:
The Department of the Treasury (Treasury) appreciates the opportunity
to review a draft of the GAO's latest report on Treasury's Troubled
Asset Relief Program (TARP) which focuses on the Term Asset-Backed
Securities Loan Facility (TALF). Treasury welcomes the recognition by
the GAO that (1) TALF is contributing to improvements in the
securitization markets and (2) TALF is unlikely to result in a loss to
taxpayers.
Indeed, on the first issue, the draft report understates the success
of the program. The securitization markets have dramatically improved
since the fourth quarter of 2008 as a result of TALF. Before the
launch of TALF, the asset-backed securitization (ABS) market came to a
complete halt. There were no new ABS issuances that supported credit
to consumers and small businesses. There was very little investor
confidence and, as a result, spreads for asset backed securities
widened to historic levels. For instance, spreads on securities backed
by auto loans and credit card receivables widened to 600 basis points
from their historic average of less than 100 basis points. Since the
launch of TALF, investor confidence has improved dramatically:
ABS spreads have been reduced to at or near their historic averages,
which represents a reduction of 70% to 90% depending on the collateral
type. Similarly, issuers of ABS have been able to extend credit to
consumers and small businesses, which is the overall goal of TALF.
On the second issue, as the report notes, Treasury currently projects
a net profit from TALF. The GAO report expresses concern that there
could be losses with respect to commercial mortgage backed securities
(CMBS). We disagree with GAO's methodology in projecting those losses
and note that, for a senior CMBS to experience losses, the underlying
property pool would have to experience losses in excess of 65%,
assuming a 30% AAA subordination requirement and a 20% TALF haircut.
As the report notes, TALF is administered by the Federal Reserve Bank
of New York. Treasury does not run the program or make loans directly
to private parties. Treasury continues to work collaboratively with
the Federal Reserve to structure TALF to minimize taxpayer risk and
achieve the program's purpose of encouraging lending to consumers and
businesses. With respect to the operation of the program, we
appreciate the suggestions the GAO makes in the report to strengthen
the documentation of decisions Treasury made concerning changes to the
program. Treasury is committed to ensuring that not only TAIT, but
TARP as a whole, is administered in a way that protects the taxpayer.
In this regard, we note that the GAO recently completed its audit of
the annual financial statements of the Office of Financial Stability,
which cover the first year of operations of TARP, and rendered an
unqualified opinion. GAO also performed a review of internal controls
and found no material weaknesses (and nor were any issues cited with
respect to TALF). Treasury will also continue to enhance its existing
reporting on its investments in TALF that strikes an appropriate
balance between our goal of transparency and the need to avoid
compromising either the competitive positions of investors or
Treasury's ability to recover funds fOr taxpayers.
We appreciate the time your staff took to discuss the report with us
and to review areas where we disagreed with your characterization of
the program. We look forward to reviewing the final report and
continuing to work with the GAO regarding any remaining areas of
disagreement and Treasury's implementation of programs to stabilize
the financial system.
Sincerely,
Signed by:
Herbert M. Allison, Jr.
Assistant Secretary for Financial Stability:
[End of section]
Appendix IX: Comments from the Board of Governors of the Federal
Reserve System:
Board Of Governors Of The Federal Reserve System:
Scott G. Alvarez, General Counsel:
Washington, D.C. 20551:
January 29, 2010:
Mr. Richard Hillman:
Managing Director:
Financial Markets and Community Investment:
Government Accountability Office:
Washington, D.C. 20548:
Dear Mr. Hillman:
I am writing regarding the draft audit report of the Government
Accountability Office (GAO) on the Treasury Department's
decisionmaking and reporting on the Term Asset-Backed Securities Loan
Facility (TALF) program,[Footnote 74] a joint program of the Federal
Reserve and the Treasury Department (Treasury). Treasury participates
in the TALF under the Troubled Asset Relief Program and the GAO
conducted the audit pursuant to its statutory responsibility to audit
and oversee the Troubled Asset Relief Program.
The draft report indicates that GAO believed it could not conduct an
effective review of the Troubled Asset Relief Program because the
Federal Reserve participates in and administers portions of the TALE.
The Federal Reserve does not agree with this assessment and believes
that the draft report does not reflect the full scope of the GAO's
audit authority with respect to the TALF. For the reasons explained in
the attached memorandum, we believe that the GAO has ample authority
to audit all aspects of the TALF's operations and the Treasury's
participation in that program, including the manner in which the
Federal Reserve administers the program on behalf of the Federal
Reserve and Treasury, within the scope of existing law. Indeed, the
Federal Reserve provided the GAO full access to information about its
administration of the TALF program, including access to all related
personnel and records, in reliance on this view.
The Federal Reserve fully cooperated with the GAO in connection with
its audit of the TALF. We provided all documents and information
requested by the GAO about the TALF program. Federal Reserve staff
maintained an ongoing dialogue with GAO staff in order to provide an
overview of the TALF program and answer detailed questions. As is
clear from the draft report, the Federal Reserve provided the GAO
extensive information concerning the terms, conditions, and operations
of the TALE. as well as the manner in which the Federal Reserve
administers the TALF program on behalf of the Federal Reserve and the
Treasury. For example, the Federal Reserve provided the GAO with
detailed, loan-level data concerning all collateral pledged by
borrowers for each loan extended by the Federal Reserve under TALF
through October 2009, as well as a memorandum explaining the
methodology used by federal Reserve staff in developing the
recommended "haircuts" that the Federal Reserve and the Treasury
should establish for each type of collateral eligible for the TALF. In
addition, as the report recognizes, the Federal Reserve provided the
GAO with information concerning the certifications that issuers and
sponsors must provide to the Federal Reserve regarding collateral
pledged to secure a TALF loan and the due diligence conducted by the
Federal Reserve and its collateral monitors to assess the credit risks
of collateral offered by potential borrowers to secure a TALF loan.
Each of these requirements is an integral component of the protections
built into the TALF to protect taxpayers.
Nevertheless, the GAO has asserted that restrictions in current law
prohibit the GAO from auditing the Federal Reserve's role in
administering the TALF in connection with the GAO's audit of the
Troubled Asset Relief Program. We believe this assertion is belied by
the draft audit report itself, the analysis and recommendations of
which rely heavily on information provided by the Federal Reserve
concerning the Federal Reserve's administration of the TALF. For
example, the GAO's analysis in the report of potential losses on TALF
loans backed by commercial mortgage-backed securities (CMBS) is based
on the loan-level data provided by the Federal Reserve, as well as
other information provided by the Federal Reserve regarding its
operation of the TALF.[Footnote 75]
Moreover, for the reasons discussed in the attached memorandum, we
believe that current law provides the GAO full authority to audit the
operations of the TALF--including the internal systems used by the
Federal Reserve to operate the TALF program and the internal controls
established and used by the Federal Reserve to limit the potential
risks to taxpayers from TALF lending--as part of the GAO's audit of
the Troubled Asset Relief Program. We have strongly encouraged the GAO
to adopt this view, which we believe is consistent with Congress'
intent. To the extent the GAO disagrees with our view of the GAO's
statutory authority to audit the TALF, we believe the cooperation and
access to information provided to the GAO by the Federal Reserve as
part of this review strongly illustrates our voluntary agreement to
allow GAO to conduct a complete and thorough audit of the Treasury's
role, decisionmaking, reporting, and exposure under the Troubled Asset
Relief Program with respect to all aspects of the TALF.
The Federal Reserve stands ready to continue to work with the GAO
concerning the TALF to allow the GAO to fully assess the performance
of and risks to, the Troubled Asset Relief Program. Federal Reserve
staff separately has provided GAO staff with technical and correcting
comments on the draft report. We hope that these comments were helpful.
Sincerely,
Signed by:
Scott G. Alvarez:
Attachment:
Memorandum Attachment:
Background on the TALF:
The TALF is a joint program of the Federal Reserve and the Treasury
Department ("Treasury"), each of which participates in the TALF under
separate legal authorities. The Board of Governors authorized the
Federal Reserve Bank of New York to participate in the TALF under
section 13(3) of the Federal Reserve Act (12 U.S.C. § 343) in order to
assist financial markets in meeting the credit needs of consumers and
businesses of all sizes by facilitating the issuance of asset-backed
securities (ABS) collateralized by a variety of consumer and business
loans. The TALF also is intended to improve market conditions for ABS
more generally by, among other things, restarting the securitization
markets for consumer- and business-related ABS. These markets came to
a halt in mid-2008, cutting off an important source of funds to
support consumer and business lending. In this way, the TALF furthers
the monetary policy objectives of the Federal Reserve as established
by Congress: price stability and maximum sustainable employment.
The Treasury participates in the TALF through the Troubled Asset
Relief Program (TARP) in order to achieve the objectives and purposes
of the TARP as established by the Congress in the Emergency Economic
Stabilization Act of 2008 (EESA),[Footnote 76] These purposes include
restoring liquidity and stability to the financial system of the
United States, promoting jobs and economic growth, and maximizing
overall returns to the taxpayers of the United States.
Under the TALF, the Federal Reserve makes non-recourse loans that are
fully secured by specified types of high-quality consumer- and
business-related ABS, including, for example, ABS backed by student
loans, floorplan loans for automobile dealers, and commercial real
estate loans. Borrowers commit their own risk capital in the form of
haircuts against the collateral, which serve as the borrower's equity
in the transaction and act as a buffer to absorb any decline in the
value of the collateral in the event the borrower does not repay the
loan.
The Federal Reserve Bank of New York operates and administers the TALF
on behalf of the Federal Reserve and the Treasury. However, because of
the joint nature of the program, all material terms governing the
operation of the program--including the types of ABS that are eligible
to be pledged as collateral for a TALF loan and the appropriate
haircut to be applied to each class of eligible collateral--are agreed
to by both the Federal Reserve and the Treasury. Once agreed, these
terms are posted on the public websitc of the Federal Reserve Bank of
New York.[Footnote 77]
An important aspect of the TALF is the credit protection provided by
Treasury under the TARP. Specifically, the Treasury has agreed to
provide up to $20 billion under the TARP to cover any losses that may
ultimately be incurred by the TALF.[Footnote 78] However, the Federal
Reserve transfers to the TALF LLC any excess interest earned by the
Federal Reserve on TALF loans above the Overnight Index Swap (OIS)
rate plus 25 basis points. These amounts, plus any returns earned on
their investment prior to use. are available to absorb first any loss
that may be incurred on TALF loans. As the GAO report acknowledges,
both the Federal Reserve and the Treasury currently project that any
potential losses on TALF loans would be fully covered by this retained
excess interest and the returns achieved through its investment and,
accordingly, the TARP will not incur a net loss on its commitment to
the TALF.
Statutory Framework Governing GAO Audits:
The EESA expressly authorizes the GAO to audit the programs and
activities of the Treasury under the TARP--including the "CALF--for
purposes of conducting, ongoing oversight of the activities and
performance of the TARP, which is a fiscal program operated by the
Treasury. See 12 U.S.C. § 116(a) and (b)(2).
The GAO also has broad authority to audit the activities and programs
of the Federal Reserve. For example, all of the Federal Reserve's
supervisory arid regulatory functions are subject to audit by the GAO
to the same extent as the supervisory and regulatory functions of the
other federal banking agencies. However, for sound public policy
reasons, the Congress has excluded monetary policy deliberations and
operations (including discount window operations) from the scope of
GAO audits. 31 H.S.C. § 714(b). Considerable experience shows that
monetary policy independence--within a framework of legislatively
established objectives and public accountability--tends to yield a
monetary policy that best promotes price stability and economic growth.
The unique, hybrid nature of the TALF initially created questions
concerning the GAO's ability to audit the TALF in conjunction with its
audits and oversight of the TARP.[Footnote 79] In response, the
Congress in May 2009 enacted new statutory provisions specifically to
clarify and ensure the GAO's ability to audit the TALF for purposes of
assessing the performance of the TARP. These provisions, which were
included in an amendment to the EESA offered by Senator Grassley and
adopted as part of the Helping Families Save Their Homes Act,
specifically provide:
Notwithstanding any other provision of law, and for purposes of
reviewing the performance of the TARP, the Comptroller General shall
have access, upon request, to any information, data, schedules, books,
accounts, financial records, reports, files, electronic
communications, or other papers, things, or property belonging to or
in use by the TARP, any entity established by the Secretary under the
Act, any entity that is established by a Federal reserve hank and
receives funding from the TARP [such as the TALF LLC], or any entity
(other than a governmental unit) participating in a program
established under authority of the Act".[Footnote 80]
When these provisions were adopted by the Senate. Senator Grassley,
the sponsor of the amendment, submitted for the record written
material indicating the "actions currently covered by the new
Language."[Footnote 81] This material specifically provides that the
amendment "would expand the GAO's authority to oversee the TARP.
including the joint Federal Reserve-Treasury Term Asset-Backed
Securities Loan Facility (TALF)."[Footnote 82] Importantly, Senator
Grassley also indicated that the amendment was not intended to
"threaten monetary policy independence."[Footnote 83]
Application to the TALF:
The Federal Reserve firmly believes that these statutory provisions
provide the GAO the authority to audit the terms, conditions, and
operations of the TALF--including those aspects of the TALF that are
administered by the Federal Reserve--as necessary to understand and
assess the performance of, and risks to. the TARP. Among the matters
that GAO may audit under this authority are:
* the minimum credit rating standards established for TALF collateral
by the Federal Reserve and Treasury;
* the due diligence requirements imposed by the Federal Reserve's
Master Loan and Security Agreement on broker-dealers that act as
agents in the TALF borrowing process;
* the certifications required of issuers and sponsors of the ABS
pledged as collateral for a TALF loan;
* the minimum haircuts established by the Federal Reserve and the
Treasury for each class of eligible TALF collateral;
* evaluations conducted by the Federal Reserve or the Treasury of the
potential for TALI, loans to result in losses to the program: and;
* the internal controls established and used by the Federal Reserve
for purposes of ensuring that participants in the TALF program and all
collateral pledged in support of TALF loans meet the requirements
established for the program by the Federal Reserve and the Treasury.
Each of these matters affect the risk that ultimately may he borne by
TARP and, thus, are within the scope of the GAO's authority to audit
the TALF in order to assess and evaluate the performance of and the
risk to, the TARP. In addition, information concerning these matters
is "in use" by the TARP because these standards, requirements,
procedures, and controls are integral parts of the measures taken
under the joint TALF program to limit the risks to which the TARP may
be exposed.
The Federal Reserve already has provided extensive information to the
GAO on these matters, as well as all other matters identified by the
GAO, to assist the GAO in fulfilling its audit responsibilities under
the EESA.
We believe this approach not only is fully consistent with the
language of the relevant statutory provisions, but also represents the
best way to balance and achieve the Congressional intent underlying
the directive for GAO to audit the performance of the TARP and the
prohibition on GAO audits of monetary policy deliberations and
operations (including discount window operations). It gives
substantive effect to each of the relevant statutory provisions,
[Footnote 84] ensures that the GAO may fully understand and assess the
TARP, and protects the independence of monetary policy. In particular,
the Federal Reserve believes that this approach does not abridge the
limitations in 31 U.S.C. § 714(b) on the ability of the GAO to audit
monetary policy or discount window deliberations, decisions, and
operations of the Federal Reserve.
[End of section]
Appendix X: Analysis of Legal Comments Submitted by the Federal
Reserve:
As noted, in its comments, the Board of Governors of the Federal
Reserve System (Federal Reserve) did not agree with our recommendation
that Congress consider providing GAO with authority to audit the
Federal Reserve's operational and administrative actions because it
disagreed that there are limitations on GAO's authority to audit these
Federal Reserve activities.
However, we believe the express statutory prohibition in 31 U.S.C. §
714(b) on GAO auditing the Federal Reserve's monetary policy and
discount window lending activities, which the Federal Reserve believes
includes the Term Asset-Backed Securities Loan Facility's (TALF)
operation and administration, prohibits us from auditing the Federal
Reserve's TALF activities, even from the perspective of the Troubled
Asset Relief Program (TARP). We limited the scope and conduct of this
audit accordingly, and thus did not request access to information to
audit the Federal Reserve's performance of these activities.[Footnote
68] Further, the Federal Reserve's decision to voluntarily provide
requested access in this instance, while helpful, does not create GAO
authority to audit the Federal Reserve's TALF operational activities
or other performance. In our view, our lack of authority to audit the
Federal Reserve's actions limited our ability to fully assess the risk
to taxpayer funds presented by TALF.
The basis of the Federal Reserve's view that GAO has authority to
audit its TALF operational and administrative actions is a May 2009
amendment to the Emergency Economic Stabilization Act of 2008 (EESA),
the statute authorizing TARP. While GAO vigorously pursues its audit
and access authority in all appropriate circumstances, the amendment
referenced by the Federal Reserve, section 601 of the May 2009
amendments, provides no authority for GAO to audit the Federal
Reserve.[Footnote 69] Rather, responding to congressional concern that
GAO lacked authority to obtain access to information directly from
banks and other firms receiving TARP funds, the amendment provided GAO
with such access to enable us to more effectively review Treasury's
actions under our existing TARP audit authority. As the Federal
Reserve correctly noted in its comments, section 601 included access
to records of any entity "that is established by a Federal reserve
bank and receives funding from the TARP," thus covering records of
TALF LLC, the special purpose vehicle created by the Federal Reserve
Bank of New York to which Treasury has committed up to $20 billion of
TARP funds. But GAO's authority to obtain access to records of TALF
LLC does not provide GAO access to other TALF program information, nor
GAO authority to audit the Federal Reserve's operation or
administration of TALF.
As support for its view that the May 2009 amendment in section 601
authorized GAO to audit the Federal Reserve's TALF performance, the
agency quoted a portion of remarks made by Senator Grassley, a lead
sponsor of the amendment. As Senator Grassley noted, however, he was
describing the Federal Reserve's position. The Senator provided
material for the record stating in part, "According to Federal Reserve
staff,...[Senate] amendment No. 1020 [section 601] would expand GAO's
authority to oversee TARP, including the joint Federal Reserve-
Treasury Term Asset-Backed Securities Loan Facility (TALF)...."
[Footnote 70] As noted, however, the Federal Reserve was only
partially correct: section 601 provided GAO authority to access
records of the Treasury-funded TALF LLC in order to audit Treasury,
but not authority to audit and evaluate the Federal Reserve's TALF
actions.
That section 601 did not authorize GAO to audit the Federal Reserve is
confirmed by the legislative history of a second amendment authored by
Senator Grassley, section 801 of the May 2009 amendments. An earlier
version of section 801 would have modified GAO's Federal Reserve
authority under 31 U.S.C. § 714 to authorize audit of all Federal
Reserve emergency actions under section 13(3) of the Federal Reserve
Act, as well as Federal Reserve TARP-related actions, thus giving GAO
TALF audit authority. Changes to section 801 shortly before the
committee vote eliminated GAO's TALF audit authority, however, leaving
GAO authorized only to audit Federal Reserve actions taken under
section 13(3) "with respect to a single and specific partnership or
corporation"--that is, AIG, Citigroup, Bank of America, and Bear
Stearns. This language was enacted as a new subsection (e) to 31
U.S.C. § 714.[Footnote 71]
Finally, the Federal Reserve commented that because of TALF's unique
"hybrid" nature--it serves objectives of both monetary policy and
TARP--GAO has "ample authority" to audit TALF operations, Treasury's
participation in them, and the Federal Reserve's administration of
TALF "on behalf of" Treasury, all from the perspective of TARP. In
this regard, the Federal Reserve noted that in practice, it obtains
Treasury's input and agreement on many aspects of TALF. However, the
view that GAO can separately audit the Federal Reserve's TALF
performance as long as the audit is limited to TARP objectives,
without violating the statutory prohibition against GAO auditing
Federal Reserve monetary policy actions, is not supported by either
the language of 31 U.S.C. § 714, its original legislative history, or
the amendments Congress enacted to it in May 2009.
Section 714(b)(2) prohibits GAO from auditing the Federal Reserve with
respect to "deliberations, decisions, or actions on monetary policy
matters, including discount window operations...." In the Federal
Reserve's view, all of its TALF activities are an exercise of its
monetary policy and discount window lending authority, thus
prohibiting GAO from auditing these actions at least from the
perspective of monetary policy. While the Federal Reserve stated it
believes the prohibition does not apply to non-monetary policy aspects
of "hybrid" Federal Reserve actions, permitting GAO to audit the
agency from the perspective of TARP, for example, the debate leading
to enactment of § 714(b) in the Banking Agency Audit Act of 1978
indicates Congress did not intend to create such an exception.
Representative Ashley, the author of the final version of the § 714(b)
prohibitions, made clear that the monetary policy prohibition extends
to all discount window lending actions, even those serving multiple
objectives such as regulation/supervision and monetary policy. Rep.
Ashley explained that all discount window lending is necessarily
"inextricably bound up in monetary policy" and is intended to be
covered by the prohibition.[Footnote 72] Under this reading, GAO is
prohibited from auditing the Federal Reserve's TALF discount window
lending activities even from the perspective of TARP.
The Federal Reserve's position also conflicts with Congress' enactment
of § 714(e) in May 2009, noted above, authorizing GAO to audit Federal
Reserve section 13(3) actions with respect to a single and specific
partnership or corporation. If GAO already could audit TARP aspects of
"hybrid" Treasury and Federal Reserve activity, such additional
authority would have been unnecessary regarding AIG, for example,
because both Treasury and the Federal Reserve already were providing
assistance to AIG. Yet GAO was required, as the Federal Reserve then
agreed, to await enactment of additional authority in order to audit
this joint assistance.[Footnote 73]
In light of these statutory restrictions on GAO's authority to audit
the Federal Reserve's TALF activities, we continue to believe that
Congress should provide GAO with authority to audit the Federal
Reserve's operation and administration of the TALF program.
[End of section]
Appendix XI: GAO Contact and Staff Acknowledgments:
GAO Contact:
Orice Williams Brown, (202) 512-8678 or williamso@gao.gov:
Staff Acknowledgments:
In addition to the contacts named above, Karen Tremba (Assistant
Director), Angela Burriesci, Emily Chalmers, Rudy Chatlos, Joe
Cisewski, Rachel DeMarcus, Mike Hoffman, Robert Lee, Sarah McGrath,
Marc Molino, Tim Mooney, Omyra Ramsingh, Susan Sawtelle, and Cynthia
Taylor made important contributions to this report.
[End of section]
Footnotes:
[1] Securitization is a process where financial assets are brought
together into interest-bearing securities that are sold to investors.
Banks receive financing for future loans from securitizations and may
rely on the ability to securitize in making certain types of loans.
[2] Pub. L. No. 110-343, Div. A., 122 Stat. 3765 (2008) (codified at
12 U.S.C. §§ 5201 et seq.).
[3] The Federal Reserve originally announced that TALF would lend up
to $200 billion, although it subsequently stated that it would be
willing to increase the limit to as much as $1 trillion if needed. As
of June 2009, Federal Reserve officials suggested that an increase to
$1 trillion is unlikely.
[4] A nonrecourse loan essentially means that in the event a borrower
defaults on the loan, the borrower's exposure is limited to the assets
used to secure nonrecourse loans. Accordingly, the lender, in this
case FRBNY, cannot seek repayment of the outstanding debt from any of
a borrower's other assets. However, if a participant in TALF breaches
the loan agreement by, for example, misrepresenting its eligibility
for a TALF loan, then the loan loses its nonrecourse status and the
borrower is personally liable for the outstanding debt.
[5] Section 116 of EESA, 122 Stat. at 3783 (codified at 12 U.S.C. §
5226).
[6] See GAO, Troubled Asset Relief Program: Additional Actions Needed
to Better Ensure Integrity, Accountability, and Transparency,
[hyperlink, http://www.gao.gov/products/GAO-09-161] (Washington, DC:
Dec. 2, 2008); GAO, Troubled Asset Relief Program: Status of Efforts
to Address Transparency and Accountability Issues, [hyperlink,
http://www.gao.gov/products/GAO-09-296] (Washington, DC: Jan. 30,
2009); GAO, Troubled Asset Relief Program: March 2009 Status of
Efforts to Address Transparency and Accountability Issues, [hyperlink,
http://www.gao.gov/products/GAO-09-504] (Washington, DC: Mar. 31,
2009); GAO, Troubled Asset Relief Program: June 2009 Status of Efforts
to Address Transparency and Accountability Issues, [hyperlink,
http://www.gao.gov/products/GAO-09-658] (Washington, DC: June 17,
2009); GAO, Troubled Asset Relief Program: One Year Later, Actions Are
Needed to Address Remaining Transparency and Accountability
Challenges, [hyperlink, http://www.gao.gov/products/GAO-10-16]
(Washington, DC: Oct. 8, 2009); and GAO, Financial Audit: Office of
Financial Stability (Troubled Asset Relief Program) Fiscal Year 2009
Financial Statements, [hyperlink,
http://www.gao.gov/products/GAO-10-301] (Washington, DC: Dec. 9, 2009).
[7] Section 714 of Title 31 of the U.S. Code limits GAO's authority to
audit certain Federal Reserve activities. Specifically, GAO audits of
the Federal Reserve and Federal Reserve banks may not include, among
other things, "deliberations, decisions, or actions on monetary policy
matters, including discount window operations, reserves of member
banks, securities credit, interest on deposits and open market
operations..., or transactions made under the direction of the Federal
Open Market Committee" 31 U.S.C. § 714 (b)(2)-(3). This prohibition
limits GAO's ability to audit the Federal Reserve's actions taken with
respect to TALF. The Helping Families Save Their Homes Act of 2009,
enacted on May 20, 2009, amended Section 714 to provide GAO authority
to audit the Federal Reserve's actions taken under Section 13(3) of
the Federal Reserve Act, 12 U.S.C. § 343, with respect to a single and
specific partnership or corporation. See Pub. L. No. 111-22, 123 Stat.
1632, 1662-63, codified at 31 U.S.C. § 714(e). Among other things,
this amendment provides GAO with authority to audit Federal Reserve
actions taken with respect to three entities also assisted under TARP--
Citigroup, Inc., American International Group, Inc., and Bank of
America Corporation--but does not provide GAO with authority to audit
Federal Reserve monetary policy actions taken with respect to TALF. We
have raised this concern about audit authority in previous testimonies
and reports. See GAO, Troubled Asset Relief Program: Status of Efforts
to Address Transparency and Accountability Issues, [hyperlink,
http://www.gao.gov/products/GAO-09-1048T] (Washington, DC: Sept.
2009); [hyperlink, http://www.gao.gov/products/GAO-10-16]; and
[hyperlink, http://www.gao.gov/products/GAO-09-658].
[8] See appendix VI for more information on the compliance program for
TALF.
[9] Eligible asset classes are those types of ABSs accepted for TALF.
The amount of equity borrowers are required to hold in the securities
pledged as collateral to secure TALF loans is calculated as a
percentage of the pledged collateral's value (either its par value,
market value, or a value determined from FRBNY internal risk
assessments).
[10] For additional information on the securitization process, see
appendix III.
[11] Special purpose vehicles are legal entities, such as limited
liability companies, created to carry out a specific financial purpose
or activity.
[12] For additional information on these asset classes, see appendix
IV.
[13] Nationally recognized statistical rating organizations, which are
registered with the Securities and Exchange Commission (SEC), have
designed credit ratings to provide investors with information about a
security's credit quality. SEC requires the agencies to undergo an
initial application review and subsequent examinations. The rating
scales are comparable, ranging from AAA, or its equivalent
(excellent), to D, or its equivalent (poor). Anything rated BBB or
higher is considered investment grade, and anything below BBB is
considered noninvestment grade or speculative. Credit ratings are
assigned without regard to whether a security is TALF-eligible or not.
The TALF-eligible rating agencies are Fitch Ratings; Moody's Investors
Service; Standard & Poor's; DBRS, Inc.; and Realpoint LLC.
[14] An issuer is an entity that sells ABSs backed by its assets and
can be a special purpose vehicle or similar legal entity. A sponsor is
an entity that facilitates such a transaction. For additional
information on the securitization process, see appendix III.
[15] For more information on the certification process and the role of
auditors, see appendix VI.
[16] Congress authorized SIGTARP as an oversight entity for TARP.
SIGTARP has made numerous recommendations to improve the
implementation of TARP and TALF. See EESA § 121, 122 Stat. at 3788
(codified at 12 U.S.C. § 5231).
[17] The primary dealers are 18 banks and securities broker-dealers
that trade in U.S. government securities with FRBNY. FRBNY had added 4
more broker-dealers to serve as agents for borrowers accessing TALF,
and as of December 31, 2009, there were a total of 22 TALF agents, but
currently not all have executed the Master Loan and Security Agreement.
[18] These documents also indicated that FRBNY created a proprietary
list of such "top-tier" entities for its own internal proposes and has
not shared it with the TALF agents.
[19] The 2001 USA Patriot Act established new and enhanced measures to
prevent, detect, and prosecute money laundering and terrorism. One of
the more important measures for financial institutions was addressed
in section 326, Verification of Identification, more commonly referred
to as "Know Your Customer." Pub. L. No. 107-56, 115 Stat. 272 (Oct.
26, 2001).
[20] CUSIP stands for the Committee on Uniform Security Identification
Procedures. CUSIP numbers are alphanumeric identifiers assigned to
individual securities.
[21] In general, the weighted average life is the average number of
years that unpaid principal is outstanding. The TALF program's
definition varies according to asset class and assumptions that FRBNY
developed about the likelihood of prepayment for the different ABS
classes. The weighted average life for CMBSs is based on the
assumption that each loan amortizes according to schedule and pays in
full on its maturity date, without prepayment. FRBNY's Web site has
further details and the formulas used to compute the weighted average
life of collateral.
[22] Trepp LLC and PIMCO assist FRBNY by providing valuation,
modeling, analytics and reporting, and advising on these matters.
Trepp LLC focuses only on CMBSs. PIMCO analyzes both mortgage-backed
and nonmortgage-backed ABSs.
[23] Administrative fees due on the settlement date are equal to 10
basis points (bps) or 0.10 percent of the loan amount for nonmortgage-
backed ABS collateral, and 20 bps for CMBS collateral.
[24] A sector is a group of securities that are similar with respect
to maturity, type, rating, industry, or coupon.
[25] Generally, the percentage is applied to the lesser of the market
value of the collateral or the par value. In addition, to account for
potential risks in legacy CMBSs, the FRBNY conducts a risk assessment
of pledged legacy CMBS collateral and applies the haircut to the least
of the market-, par-, or assessment-values. For more details, see
[hyperlink, http://www.newyorkfed.org/markets/talf_faq.html].
[26] Under certain circumstances, the excess interest earned on ABS or
CMBS above the TALF loan interest payable will be remitted to the TALF
borrower only up to a proportion of the original haircut amount, with
the remainder of such amount applied to the TALF loan principal. FRBNY
refers to this as a net carry diversion, and it accelerates the
repayment of the TALF loan and ensures that the TALF borrower
maintains a significant equity stake in the ABS or CMBS collateral
over the life of the loan.
[27] A basis point represents .01 percent, or one-hundredth of a
percent. In other words, 100 basis points equals 1 percent. For
example, the difference between 1.00 percent and 1.50 percent would be
expressed as 50 basis points. The OIS rate is a type of interest rate
swap that is based on daily federal funds rates.
[28] TALF loans are nonrecourse, except in cases involving fraud or
other breaches of certain representations.
[29] This amount excludes the $84 million (initial $100 million loan
from Treasury less the $16 million set aside for administrative
expenses), which is held in a separate sub-account. Therefore, total
funds that could potentially be available to purchase surrendered
collateral could be greater, according to Treasury officials, though
additional TALF LLC expenses related to surrendered collateral could
also be taken from this amount. For that reason, we refer to $198
million as the amount in TALF LLC that is dedicated to purchasing
surrendered assets.
[30] As noted in the background, TARP funds can be used to purchase
ABS or CMBS collateral if not enough excess interest has accumulated
in TALF LLC.
[31] According to the terms of FRBNY's Master Loan Security Agreement,
TALF borrowers can choose to surrender collateral in order to stop
making payments on a loan. If they do not voluntarily surrender
collateral, FRBNY has the right to take possession of it. In either
case, TALF LLC is likely to purchase the associated collateral.
[32] See background for additional detail on the rating requirements
for TALF collateral.
[33] For TALF purposes, returns on equity are the earnings on the
amount invested in a particular security or the amount that an
investor can expect from an investment.
[34] Because of our audit limitations, we did not review the Federal
Reserve's process for surveying the market on what returns on equity
would encourage participation for a program like TALF.
[35] Our analysis focused on auto, credit card, and student loan ABSs,
because these securities make up the majority of the market and are
the most commonly traded asset classes.
[36] Some securitizations--such as ABSs backed by auto loans--are
divided into different segments, or tranches. A tranche is a piece of
a securitization that has specified risks and returns. For additional
information on the securitization process and tranches within
securitizations, see appendixes III and IV.
[37] It should be noted that just because an ABS meets the TALF
eligibility requirements, it does not necessarily follow that
investors will use TALF to finance their purchase. As part of the
price discovery process that the ABS underwriters engage in as part of
the marketing of a new issue of ABS, they interact with prospective
buyers to achieve the lowest interest rates on the offered bond
tranches that will still sell the full offering. As the TALF
eligibility verification process must be completed prior to the issue
date, it is possible that bonds will have sufficient demand to drive
the price up or bonds' coupon rates (since ABSs are issued at or very
close to prices of 100 percent of par) down to a level near or below
the TALF finance rates paid to the Federal Reserve. This is why
investors are increasingly unwilling to use TALF to finance the
purchase of TALF eligible securities. In addition, this trend explains
why more issuers are choosing to forego the process of seeking TALF
eligibility as they feel it is no longer necessary to ensure that
their bonds will find buyers.
[38] This option allowed the issuer to purchase back the bonds after a
period of between 30 and 36 months at a significant discount on the
face value. For one issuer, the discounts were 7 percent for May 2009
and 6 percent for both July and August 2009. The August 2009
transaction from a new private student loan issuer had a 10 percent
discounted exercise price. Exercise by the issuer at this discounted
price implicitly reduces the amount of principal that the bondholder
receives and negatively impacts the return. The issuer makes up for
this expected shortfall in principal payback with a higher interest
rate or "coupon" on the bond.
[39] This subprime credit card issuance was unique in that two
tranches had average lives of nearly 5 years, which is longer than
most TALF-eligible credit card ABS collateral. Investors in ABSs with
longer average lives require higher interest rates to compensate for
the greater risk of holding the bond over a longer time period.
However, TALF loans for these ABSs are still 3-year loans with the
same interest rate as would be paid by a borrower with a shorter-lived
ABS. To some extent the increased return is compensation for the risk
the borrower is taking given the mismatch on the maturity of the loan
and that of the ABS.
[40] Although there was a single auto floor plan ABS deal in September
2009, we could not locate a prior issuance from this issuer to use as
a comparison for precrisis levels of credit enhancement; therefore,
this asset class is not included in this analysis.
[41] Treasury asked the Bank of New York Mellon, as the administrator
for TARP, to analyze the risks that TALF posed to TARP funds. In turn,
the bank hired a firm to conduct this analysis, NSM Structured Credit
Solutions, which is now known as RangeMark. The firm's
responsibilities include determining how the various TALF asset
classes perform and the risks they present to TARP. The contractor
made certain assumptions about the composition of the portfolio and
the risks of the various types of collateral, and estimated losses in
an extreme scenario. The results of this analysis contributed to
Treasury's recent financial statement, which anticipates that its
commitment of funds for TALF will actually provide a return. See
[hyperlink, http://www.gao.gov/products/GAO-10-301].
[42] As stated earlier, according to FRBNY officials the amount
increases at approximately $30 million per month, based on the current
loan portfolio.
[43] We compared the prices on all of the CMBS CUSIPs, with the
exception of two, that were accepted by FRBNY as of September 30,
2009. For this analysis, we did not calculate the excess interest that
would have accumulated in TALF LLC.
[44] Specifically, the amount of money that FRBNY distributes to TALF
borrowers that is in excess of the interest due on the TALF loan will
be limited for all 5-year loans and for 3-year loans related to legacy
CMBSs. In the first 3 years of a 5-year loan, the limit is 25 percent
per year of the haircut amount, 10 percent in the fourth loan year,
and 5 percent in the fifth loan year. Any amount above this excess is
applied to the principal on the TALF loan. For 3-year loans for legacy
CMBSs, the limit is 30 percent of the haircut per year. This
requirement ensures that borrowers retain some amount of equity
interest in the securities and reduces the likelihood of nonpayment.
More details can be found on the FRBNY Web site at [hyperlink,
http://www.newyorkfed.org/markets/talf_faq.html].
[45] FRBNY also solicited input from market participants on asset
classes that might benefit from TALF. According to FRBNY officials,
this is part of its regular market monitoring function. Two of these
market participants, which are also TALF agents, noted that officials
from the Federal Reserve requested their feedback about TALF before
the program was announced, seeking information such as how to restart
securitization markets, what the latest developments were in those
markets, and which asset classes should be considered.
[46] Insurance premium finance loans are used to finance property and
casualty insurance premiums.
[47] For more information on the approved asset classes and the dates
they were announced as acceptable collateral for TALF, see appendix IV
and the background.
[48] The 7(a) and 504 programs aim to facilitate the accessibility and
affordability of financing to small businesses. Under the 7(a)
program, SBA generally provides lenders with guarantees on up to 85
percent of the value of loans made to qualifying small businesses in
exchange for fees to help offset the costs of the program. Under the
504 program, which generally applies to small business real estate and
other fixed assets, SBA also provides certified development companies
with a guarantee on up to 40 percent of the financing of the projects'
costs in exchange for fees, while the small business borrowers and
other lenders provide the remaining 60 percent of the financing with
no guarantee. For additional information, see GAO, Small Business
Administration's Implementation of Administrative Provisions in the
American Recovery and Reinvestment Act, [hyperlink,
http://www.gao.gov/products/GAO-10-507R] (Washington, DC: Apr. 16,
2009).
[49] GAO, Standards for Internal Control in the Federal Government,
[hyperlink, http://www.gao.gov/products/GAO/AIMD-00.21.3.1]
(Washington, D.C.: November 1999).
[50] See [hyperlink, http://www.gao.gov/products/GAO-09-161].
[51] See GAO, Troubled Asset Relief Program: Treasury Actions Needed
to Make the Home Affordable Modification Program More Transparent and
Accountable, [hyperlink, http://www.gao.gov/products/GAO-09-837]
(Washington, D.C.: July 23, 2009) and [hyperlink,
http://www.gao.gov/products/GAO-10-16].
[52] See [hyperlink, http://www.gao.gov/products/GAO-10-16] and GAO,
Auto Industry: Summary of Government Efforts and Automakers'
Restructuring to Date, [hyperlink,
http://www.gao.gov/products/GAO-09-553] (Washington, DC: Apr. 23,
2009).
[53] In this instance, spread refers to the difference between a
security's yield and a benchmark yield. For ABSs, the customary
benchmark yield is interest rate swap yields. Interest rate swaps are
contracts in which one party agrees to pay a fixed interest rate to
another party in exchange for a floating rate.
[54] We focused on auto, credit card, and student loan ABSs and CMBSs
because these asset classes represent the largest sectors of the
securitization markets and, in the case of CMBSs, experienced the
greatest levels of disruption that could be measured.
[55] The Federal Deposit Insurance Corporation (FDIC) Rule 360.6
provides that the FDIC will not use its statutory authority as
conservator or receiver to disaffirm or repudiate certain contracts
pertaining to any financial assets transferred by an insured
depository institution to a special purpose entity in connection with
a securitization or participation, provided that such transfer
satisfies all conditions for sales accounting treatment under
generally accepted accounting principles (GAAP). For most insured
depositary institutions, the 2009 GAAP modifications will be effective
for reporting periods after January 1, 2010. However, in November
2009, FDIC amended Rule 360.6 to extend the safe harbor to assets
transferred in connection with securitizations completed prior to
March 31, 2010, so long as those securitizations complied with the
accounting rules as they existed prior to the 2009 GAAP modifications.
See, "Defining Safe Harbor Protection for Treatment by the Federal
Deposit Insurance Corporation as Conservator or Receiver of Financial
Assets Transferred by an Insured Depository Institution in Connection
with a Securitization or Participation," 74 Fed. Reg. 59066 (Nov. 17,
2009) (interim rule and request for comments).
[56] Auto finance companies such as Ford Motor Credit Company, GMAC,
and Chrysler Financial provide financing for consumer automotive and
dealer purchases. We focused on interest rates for credit cards and
automotive loans and did not evaluate changes in interest rates for
student loans because reliable and consistent data were unavailable
for private student loans. We also did not evaluate changes in
interest rates for commercial mortgages, because these rates depend on
factors such as geographic location and property type, making it
difficult to draw broad conclusions about interest rate trends.
[57] See [hyperlink, http://www.gao.gov/products/GAO-09-553].
[58] Provisions in the law will limit credit card companies' ability
to increase interest rates, among other things, so some companies have
raised interest rates in anticipation of the law taking effect. See
GAO, Credit Cards: Rising Interchange Fees Have Increased Costs for
Merchants, But Options for Reducing Fees Pose Challenges, [hyperlink,
http://www.gao.gov/products/GAO-10-45] (Washington, DC: Nov. 19, 2009).
[59] See [hyperlink, http://www.gao.gov/products/GAO-10-16].
[60] A performance audit of the Federal Reserve's TALF operational and
administrative activities would, for example, have involved evaluating
the sufficiency of how certain TALF program terms were arrived at,
such as whether the haircuts or the amount of equity the TALF borrower
holds in the collateral protected TALF from losses. We also would have
evaluated FRBNY's system of internal controls and the role of TALF
participants in certifying and validating compliance with certain
program requirements. In addition, we would have interviewed some of
the entities that helped administer TALF to validate agency
information and to better understand their roles and interactions with
the Federal Reserve and FRBNY. These would have included entities such
as the TALF agents, Bank of New York Mellon, Trepp, or CW Capital.
[61] Section 714 of Title 31 of the U.S. Code limits GAO's authority
to audit certain Federal Reserve activities. Specifically, GAO audits
of the Federal Reserve and Federal Reserve banks "may not include
deliberations, decisions, or actions on monetary policy matters,
including discount window operations, reserves of member banks,
securities credit, interest on deposits and open market operations...,
or transactions made under the direction of the Federal Open Market
Committee" 31 U.S.C. § 714 (b)(2)-(3). This prohibition limits GAO's
ability to audit the Federal Reserve's actions taken with respect to
TALF. The Helping Families Save Their Homes Act of 2009, enacted on
May 20, 2009, amended Section 714 to provide GAO authority to audit
Federal Reserve Board actions taken under Section 13(3) of the Federal
Reserve Act with respect to a single and specific partnership or
corporation. See Pub. L. No. 111-22, 123 Stat. 1632, 1662-63. Among
other things, this amendment provides GAO with authority to audit
Federal Reserve actions taken with respect to three entities also
assisted under TARP--Citigroup, Inc., American International Group,
Inc., and Bank of America Corporation--but does not provide GAO with
authority to audit Federal Reserve monetary policy actions taken with
respect to TALF generally.
[62] Haircuts set the amount of equity the TALF borrower holds in the
collateral and is a percentage assigned by the FRBNY. Haircuts vary by
FRBNY's assessment of market risks for each sector and subsector.
[63] Some securitizations--such as ABSs backed by auto loans--are
divided into different segments, or tranches. A tranche is a piece of
a securitization that has specified risks and returns.
[64] A tranche is a piece of a securitization that has specified risks
and returns.
[65] CUSIP stands for the Committee on Uniform Security Identification
Procedures. CUSIP numbers are alphanumeric identifiers assigned to
individual securities.
[66] Some securitizations--such as ABSs backed by auto loans--are
divided into different segments, or tranches. A tranche is a piece of
a securitization that has specified risks and returns.
[67] According to the guidance provided by the American Institute of
Certified Public Accountants, Public Company Accounting Oversight
Board, and discussions with two auditors involved in the process,
attestations are generally considered a higher form of assurance than
AUPs. This is because in an attestation the auditor must attest to all
of its statements and design a methodology that provides reasonable
assurance of the accuracy of its attestation. In the case of an AUP,
the process is stipulated ahead of time and the auditor simply follows
it, providing a statement as to the outcome of the process.
[68] A performance audit of the Federal Reserve's TALF operational and
administrative activities would, for example, have involved evaluating
the process by which certain TALF program terms were arrived at, such
as whether the haircuts or the amount of equity the TALF borrower
holds in the collateral protected TALF from losses. We also would have
evaluated FRBNY's system of internal controls and the role of TALF
participants in certifying and validating compliance with certain
program requirements. In addition, we would have interviewed some of
the entities that helped administer TALF to validate agency
information and to better understand their roles and interactions with
the Federal Reserve and FRBNY. These would have included entities such
as the TALF agents, Bank of New York Mellon, Trepp, or CW Capital.
[69] See section 601 of the Helping Families Save Their Homes Act of
2009, Pub. L. No. 111-22, 123 Stat. 1632, 1659 (May 20, 2009),
codified at 12 U.S.C. § 5226.
[70] 155 Cong. Rec. S5181 (daily ed. May 6, 2009).
[71] Section 801 (Senate Amendment 1021) as introduced would have
removed all restrictions on GAO's authority to audit the Federal
Reserve in 31 U.S.C. § 714(b), including the prohibition against
auditing monetary policy actions. 155 Cong. Rec. S4991 (daily ed.
April 30, 2009). As detailed by Senator Grassley, a substitute
amendment then narrowed GAO's audit authority to all Federal Reserve
section 13(3) actions and TARP-related actions, 155 Cong. Rec. S5118-
19 (daily ed. May 5, 2009), and the final version narrowed GAO's
authority even further, to Federal Reserve section 13(3) actions "with
respect to a single and specific partnership or corporation." 155
Cong. Rec. S5173-74, S5181-82 (daily ed. May 6, 2009); sec. 801, Pub.
L. No. 111-22, 123 Stat. 1662-63 (May 20, 2009).
[72] 123 Cong. Rec. H11015-20 (daily ed. Oct. 14, 1977).
[73] See GAO, Troubled Asset Relief Program: Status of Government
Assistance Provided to AIG, [hyperlink,
http://www.gao.gov/products/GAO-09-975] (Washington, D.C.: Sept. 21,
2009), p. 2.
[74] [hyperlink, http://www.gao.gov/products/GAO-10-25], as provided
to Federal Reserve staff on January 15, 2010.
[75] For example, the draft report notes that the risk that the TALF
would experience a loss due to the surrender of a substantial amount
of CMBS assets is significantly mitigated by the Federal Reserve's
requirement that borrowers obtaining TALF loans backed by legacy CMBS
collateral prepay a portion of any returns in excess of certain
established limits.
[76] See section 2 of the EESA (12 U.S.C. § 5201).
[77] See [hyperlink,
http://www.newyorkfed.org/markets/talf_terms.html].
[78] As of December 31, 2009, only $100 million of this commitment had
been disbursed under the TARP in support of the TALF.
[79] See Statement of Acting Comptroller Gene L. Dodaro before the
Senate Finance Committee, March 31, 2009.
[80] See Helping Families Save Their Homes Act of 2009, § 601. Pub. L.
111-22, 123 Stat.1632, 1659 (codified at 12 U.S.C. § 5226(a)(2)(B))
(emphasis added). The amendment also provides certain protections for
proprietary and trade secrets information obtained by the GAO.
[81] 155 Cong. Rec. S5181 (Daily ed. May 6, 2009)(emphasis added).
[82] Id.
[83] Id.
[84] The Supreme Court has indicated that separate statutes should he
interpreted, whenever possible, in a manner that gives effect to each.
See Lockheed Corp. v. Spink, 517 U.S. 882, 895 n.6 (1996) (Court
"would be reluctant to infer that [one federal statute] bars conduct
affirmatively sanctioned by other federal statutes" absent clear
evidence that Congress intended such a conflict); County of Yakima v.
Confederated Tribes and Bands of the Yakima Indian Nation, 502 U.S.
251, 265 (1992) ("Judges are not at liberty to pick and choose among
congressional enactments, and when two [or more] statutes are capable
of co-existence, it is the duty of the courts, absent a clearly
expressed congressional intention to the contrary, to regard each as
effective.") (internal quotations and citations omitted).
[End of section]
GAO's Mission:
The Government Accountability Office, the audit, evaluation and
investigative arm of Congress, exists to support Congress in meeting
its constitutional responsibilities and to help improve the performance
and accountability of the federal government for the American people.
GAO examines the use of public funds; evaluates federal programs and
policies; and provides analyses, recommendations, and other assistance
to help Congress make informed oversight, policy, and funding
decisions. GAO's commitment to good government is reflected in its core
values of accountability, integrity, and reliability.
Obtaining Copies of GAO Reports and Testimony:
The fastest and easiest way to obtain copies of GAO documents at no
cost is through GAO's Web site [hyperlink, http://www.gao.gov]. Each
weekday, GAO posts newly released reports, testimony, and
correspondence on its Web site. To have GAO e-mail you a list of newly
posted products every afternoon, go to [hyperlink, http://www.gao.gov]
and select "E-mail Updates."
Order by Phone:
The price of each GAO publication reflects GAO‘s actual cost of
production and distribution and depends on the number of pages in the
publication and whether the publication is printed in color or black and
white. Pricing and ordering information is posted on GAO‘s Web site,
[hyperlink, http://www.gao.gov/ordering.htm].
Place orders by calling (202) 512-6000, toll free (866) 801-7077, or
TDD (202) 512-2537.
Orders may be paid for using American Express, Discover Card,
MasterCard, Visa, check, or money order. Call for additional
information.
To Report Fraud, Waste, and Abuse in Federal Programs:
Contact:
Web site: [hyperlink, http://www.gao.gov/fraudnet/fraudnet.htm]:
E-mail: fraudnet@gao.gov:
Automated answering system: (800) 424-5454 or (202) 512-7470:
Congressional Relations:
Ralph Dawn, Managing Director, dawnr@gao.gov:
(202) 512-4400:
U.S. Government Accountability Office:
441 G Street NW, Room 7125:
Washington, D.C. 20548:
Public Affairs:
Chuck Young, Managing Director, youngc1@gao.gov:
(202) 512-4800:
U.S. Government Accountability Office:
441 G Street NW, Room 7149:
Washington, D.C. 20548:
