oversight

Evaluation of the Federal Housing Finance Agency's Oversight of Freddie Mac's Repurchase Settlement with Bank of America

Published by the Federal Housing Finance Agency, Office of Inspector General on 2011-09-27.

Below is a raw (and likely hideous) rendition of the original report. (PDF)

          FEDERAL HOUSING FINANCE AGENCY
                          333
            OFFICE OF INSPECTOR GENERAL



      Evaluation of the Federal Housing Finance Agency’s
      Oversight of Freddie Mac’s Repurchase Settlement
                     with Bank of America




EVALUATION REPORT: EVL-2011-006          DATED: September 27, 2011
          EXPLANATION OF REDACTIONS IN THIS REPORT

This report includes redactions requested by the Federal Housing Finance Agency
(FHFA) and the Federal Home Loan Mortgage Corporation (Freddie
Mac). According to FHFA and Freddie Mac, the redactions are intended to
protect from disclosure material that they consider to be confidential financial,
proprietary business, and/or trade secret information, which Freddie Mac claims it
would not ordinarily publicly disclose and, if disclosed, could place it at a
competitive disadvantage.
         Evaluation of FHFA’s Oversight of Freddie Mac’s Repurchase Settlement
                                 with Bank of America

Why FHFA-OIG Did This Evaluation                                What FHFA-OIG Found
     [
In the closing days of 2010, the Federal Housing Finance        FHFA-OIG found that FHFA senior management did not timely
Agency (FHFA or Agency), acting in its capacity as the          address significant concerns raised about the loan review process
conservator of the Federal Home Loan Mortgage Corporation       used by Freddie Mac and its ramifications on underlying the
(Freddie Mac or the Enterprise) and the Federal National        settlement. Specifically, FHFA-OIG makes three findings.
Mortgage Association (Fannie Mae) (collectively the
                                                                First, in mid-2010, prior to the Bank of America settlement, an
Enterprises), approved two agreements totaling $2.87 billion
                                                                FHFA senior examiner raised serious concerns about limitations in
under which the Enterprises settled mortgage repurchase
                                                                Freddie Mac’s existing loan review process for mortgage repurchase
claims asserted against Bank of America.
                                                                claims, which, according to the senior examiner, could potentially
Freddie Mac and Fannie Mae have purchased millions of           cost Freddie Mac a considerable amount of money. Freddie Mac’s
mortgages from loan sellers, such as Bank of America. The       internal auditors independently identified concerns about the process
contracts under which the Enterprises purchased the             at the end of 2010. These concerns merited prompt attention by
mortgages provide them with the right to require the sellers    FHFA because they potentially involve significant recoveries for
to repurchase mortgages that do not meet the underwriting       Freddie Mac and, ultimately, the taxpayers. Further, unless
criteria represented and warranted by them. Freddie Mac’s       examined and addressed, the underlying problems are susceptible to
$1.35 billion settlement with Bank of America could serve as    recurrence.
a precedent for future repurchase settlements.
                                                                Second, FHFA did not timely act on or test the ramifications of these
The FHFA Office of Inspector General (FHFA-OIG) began a         concerns prior to the Bank of America settlement. FHFA-OIG did
review after Members of Congress and others questioned the      not independently validate Freddie Mac’s existing loan review
adequacy of the settlements. During the review, two             process and, therefore, does not reach any final conclusion about it.
individuals independently reported their concerns about the     Nevertheless, by relying on Freddie Mac’s analysis of the settlement
Freddie Mac-Bank of America settlement, and FHFA-OIG            without testing the assumptions underlying Freddie Mac’s existing
commenced this evaluation.                                      loan review process, FHFA senior managers may have inaccurately
                                                                estimated the risk of loss to Freddie Mac.
                                                                Third, following the initiation of FHFA-OIG’s evaluation, FHFA, to
What FHFA-OIG Recommends                                        its credit, suspended future Enterprise mortgage repurchase
                                                                settlements premised on the Freddie Mac loan review process and
FHFA-OIG makes two recommendations. FHFA and its
                                                                set in motion activities to test the assumptions underlying the loan
senior management should promptly: (1) act on the specific
                                                                review process. Additionally, other findings tend to support the
and significant concerns raised by FHFA staff and Freddie Mac
                                                                validity of the concerns about the process. For example, on June 6,
internal auditors about Freddie Mac’s loan review process;
                                                                2011, Freddie Mac’s internal auditors issued an audit opinion that
and (2) initiate reforms to ensure more generally that senior
                                                                the Enterprise’s internal governance controls over this process were
managers are apprised of and timely act on significant
                                                                “Unsatisfactory.” Furthermore, at the end of 2010 and then again in
concerns brought to their attention.
                                                                mid-2011, a Freddie Mac senior manager advised the board of
                                                                directors that the Enterprise could recover more in the future if it
                                                                used a more expansive loan review process.
Evaluation Report: EVL-2011-006                                                               Dated: September 27, 2011
TABLE OF CONTENTS
TABLE OF CONTENTS ................................................................................................................ 4

ABBREVIATIONS ........................................................................................................................ 6

PREFACE ....................................................................................................................................... 7

BACKGROUND .......................................................................................................................... 11

About the Enterprises and FHFA .......................................................................................................... 11

Overview of the Mortgage Repurchase Process ........................................................................................ 11

Changes in Mortgage Lending Practices During the Housing Boom ............................................................. 12

Chronology of Key Events and Associated Analysis ................................................................................. 15
     a.   Nine Months Prior to the Bank of America Settlement, an FHFA Senior Examiner Identifies Changes in
       Housing Foreclosure Patterns ....................................................................................................... 15
       b.   FHFA Senior Examiner Raises Concerns that Freddie Mac Did Not Revise Its Loan Review Process for
       Repurchase Claims to Account for Foreclosure Pattern Changes Among Housing Boom Mortgages .......... 18
       c.   FHFA Senior Examiner Views Freddie Mac’s Continued Use of Its Loan Review Process as Potentially
       Costing Freddie Mac “Billions of Dollars” ...................................................................................... 21
       d.    FHFA Senior Examiner Alerts FHFA Staff, Managers, and Senior Managers to the Concerns About
       Freddie Mac’s Loan Review Process ............................................................................................. 22
       e.   Freddie Mac Reaches a Tentative Repurchase Settlement with Bank of America; Freddie Mac’s Internal
       Auditors Independently Raise Concerns About Freddie Mac’s Loan Review Process .............................. 24
       f.      Freddie Mac Management Responds...................................................................................... 26
       g.      FHFA Staff Reviews and Recommends Approval of the Freddie Mac-Bank of America Settlement ... 28
       h.      FHFA’s Acting Director Suspends All Future Enterprise Repurchase Settlements Pending Further
       Review; Freddie Mac’s Internal Auditors Issue an “Unsatisfactory” Audit Opinion ................................ 29


FINDINGS .................................................................................................................................... 32

1.     An FHFA Senior Examiner Raised Significant Concerns About Freddie Mac’s Loan Review Process for
Mortgage Repurchase Claims ............................................................................................................... 32

2.  FHFA Did Not Timely Act on or Test the Ramifications of the Senior Examiner’s Concerns; Consequently,
FHFA May Have Incorrectly Estimated the Risk of Loss to Freddie Mac Before Approving the Bank of America
Settlement ........................................................................................................................................ 33



            Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                                          4
3.     FHFA’s Decision to Suspend Approval of Additional Repurchase Settlements and Freddie Mac’s Continuing
Efforts to Address the Concerns Are Positive Steps .................................................................................. 34


CONCLUSIONS........................................................................................................................... 35

RECOMMENDATIONS .............................................................................................................. 35

SCOPE AND METHODOLOGY ................................................................................................ 37

APPENDIX A ............................................................................................................................... 38
  FHFA Management Comments ........................................................................................................ 38


APPENDIX B ............................................................................................................................... 41
  FHFA-OIG Responses to FHFA Management Comments ...................................................................... 41


APPENDIX C ............................................................................................................................... 42
  Timeline of Relevant Events ............................................................................................................ 42


APPENDIX D ............................................................................................................................... 43
  Timeline of When FHFA Staff Were Alerted to Concerns ...................................................................... 43


ADDITIONAL INFORMATION AND COPIES ........................................................................ 44




          Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                                     5
ABBREVIATIONS
ARM ....................................................................................................... Adjustable Rate Mortgage
Countrywide .................................................................................................. Countrywide Financial
DER............................................................................................. Division of Enterprise Regulation
Fannie Mae......................................................................... Federal National Mortgage Association
FHFA ........................................................................................... Federal Housing Finance Agency
FHFA-OIG ...................................... Federal Housing Finance Agency Office of Inspector General
Freddie Mac .................................................................. Federal Home Loan Mortgage Corporation
HERA.......................................................................Housing and Economic Recovery Act of 2008
MBS ..................................................................................................... Mortgage-Backed Securities
OCO ...................................................................................... Office of Conservatorship Operations




         Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                                 6
                                    Federal Housing Finance Agency

                                       Office of Inspector General

                                              Washington, DC




                                             PREFACE
FHFA-OIG was established by the Housing and Economic Recovery Act of 2008 (Public Law
No. 110-289) (HERA), which amended the Inspector General Act of 1978 (Public Law No. 95-
452). FHFA-OIG is authorized to conduct audits, investigations, and other activities of the
programs and operations of FHFA; to recommend policies that promote economy and efficiency
in the administration of such programs and operations; and to prevent and detect fraud and abuse
in them. This evaluation is one in a series of audits, evaluations, and special reports published as
part of FHFA-OIG’s oversight responsibilities. It is intended to assess FHFA’s review and
approval of Freddie Mac’s settlement of mortgage repurchase claims with Bank of America.

Fannie Mae and Freddie Mac are government-sponsored enterprises that support the nation’s
housing finance system through the secondary mortgage market. The Enterprises purchase
mortgages from loan sellers, such as banks, which can then use the sales proceeds to originate
additional mortgages. The Enterprises either hold the loans in their investment portfolios or pool
them into mortgage-backed securities (MBS) that they sell to investors. The proceeds of such
sales, in turn, fund additional purchases of loans on the secondary market. In 2010, with the
housing crisis continuing, federal government-supported entities collectively controlled 96% of
                                  1
the secondary mortgage market. The Enterprises alone accounted for 70% of the market.

In September 2008, due to mounting mortgage-related losses, the Enterprises were placed into
conservatorships overseen by FHFA, pursuant to HERA. At the same time, the Department of
the Treasury agreed to provide financial support to the Enterprises and, to date, has invested over
                                                                                         2
$162 billion of public funds in them to offset their losses and prevent their insolvency. As



1
 FHFA, Conservator’s Report on the Enterprises’ Financial Performance: Fourth Quarter 2010, at 5, available at
www.fhfa.gov/webfiles/21169/Conservator’s Report_4Q_4_20_11.pdf. The Government National Mortgage
Association, the other federal government-supported entity, accounted for 26% of the secondary mortgage market.
2
  Federal Housing Finance Agency, “Data as of June 9, 2011, on Treasury and Federal Reserve Purchase Programs
for GSE and Mortgage-Related Securities.”



        Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                       7
conservator, FHFA has assumed responsibility for the conservation and preservation of the assets
of each Enterprise.

When a lender or other entity sells a mortgage to either Enterprise, it promises that the loan
complies with certain representations and warranties – principally, that the eligibility of the
property and the creditworthiness of the borrower are characterized accurately in the loan
documents at the time of origination. If the purchasing Enterprise later discovers that the loan
contains a defect (for instance, that the value of the property securing the loan was materially
lower than described in the loan paperwork, or that the borrower did not have the income stated
on the loan application), then the Enterprise has the contractual right to require the seller to
repurchase the loan at its full face value or to indemnify the Enterprise for losses incurred. The
mortgage repurchase process therefore provides an important means for the Enterprises to
mitigate their credit-related losses on foreclosed mortgages and potentially limit taxpayer
exposure to losses as well. Moreover, because the Enterprises typically do not examine the
mortgages they purchase for such defects prior to purchasing them, their repurchase rights
represent their principal defense against defective loans and the risks they pose.

In late December 2010, FHFA’s Acting Director, in his capacity as the Enterprises’ conservator,
approved two repurchase settlement agreements between the Enterprises and Bank of America
totaling $2.87 billion ($1.35 billion for Freddie Mac and $1.52 billion for Fannie Mae). Freddie
Mac’s settlement resolved most past, present, and (with limited exceptions) future repurchase
issues associated with 787,000 loans sold to the Enterprise by Countrywide Financial
(Countrywide). Bank of America purchased Countrywide in 2008. By contrast, Fannie Mae’s
settlement with Bank of America covered only past and present claims, not future ones. The
Freddie Mac settlement could serve as a precedent for future repurchase settlements involving
large financial institutions that sold significant numbers of loans to the Enterprise.

Although the Enterprises’ mortgage repurchase settlements initially generated positive publicity
for Bank of America, Members of Congress and others soon raised concerns about the
                       3
settlement’s adequacy. Accordingly, FHFA-OIG began to survey the settlements in greater
detail. While the survey was under way, two individuals independently provided FHFA-OIG
with information raising significant concerns about the Freddie Mac-Bank of America
settlement. Based on those concerns, FHFA-OIG prioritized its review and commenced this
evaluation.




3
 For example, on January 7, 2011, four Representatives on the House Financial Services Committee wrote to
FHFA’s Acting Director seeking greater detail on the terms of the settlements.



        Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                       8
FHFA-OIG makes three findings:
                                                                                                         4
        1. In mid-2010, prior to the Bank of America settlement, an FHFA senior examiner
           raised significant concerns about limitations in Freddie Mac’s existing loan review
           process for mortgage repurchase claims, which, according to the senior examiner,
           could potentially cost Freddie Mac “billions of dollars of losses.” Freddie Mac’s
           internal auditors independently identified concerns about the process at the end of
           2010. These concerns merited prompt attention by FHFA because they potentially
           involve considerable recoveries for Freddie Mac and, ultimately, the taxpayers.
           Further, unless examined and addressed, the underlying problems are susceptible to
           recurrence in future settlements.

        2. FHFA did not timely act on or test the ramifications of the senior examiner’s
           concerns prior to the Bank of America settlement. FHFA-OIG did not independently
           validate Freddie Mac’s existing loan review process and, therefore, does not reach
           any final conclusion about it. Nevertheless, by relying on Freddie Mac’s analysis of
           the settlement without testing the assumptions underlying the Enterprise’s existing
           loan review process, FHFA senior managers may have inaccurately estimated the risk
           of loss to Freddie Mac.

        3. After this evaluation began, FHFA, to its credit, suspended future Enterprise
           mortgage repurchase settlements premised on the Freddie Mac loan review process
           and set in motion activities to test the concerns raised about the process. In addition,
           Freddie Mac’s internal auditors continued to review the issue, and on June 6, 2011,
           issued an audit opinion that the Enterprise’s internal corporate governance controls
           over this process were “Unsatisfactory.” Furthermore, at the end of 2010 and then
           again in mid-2011, a Freddie Mac senior manager advised the board of directors that
           the Enterprise could recover additional money in the future through a more expansive
           loan review process. Currently, FHFA and Freddie Mac are analyzing the loan
           review process to determine whether greater recoveries in the future are possible.

FHFA-OIG believes that the recommendations in this report will result in more economical,
effective, and efficient operations. FHFA-OIG appreciates the assistance of all those who
contributed to the preparation of this report.



4
 For the purpose of this evaluation, within FHFA: staffers, examiners, and senior examiners report to managers;
managers report to senior managers; and senior managers report to the FHFA Acting Director. Within Freddie Mac,
senior managers report to the Chief Executive Officer.



        Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                       9
This evaluation was led by David Z. Seide, Director of Special Projects; Timothy Lee, Senior
Financial Advisor; and Bruce McWilliams, Investigative Evaluator. This evaluation report has
been distributed to Congress, the Office of Management and Budget, and others and will be
posted on FHFA-OIG’s website, www.fhfaoig.gov.




Richard Parker
Acting Deputy Inspector General for Evaluations




      Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                   10
BACKGROUND
About the Enterprises and FHFA

To fulfill their obligations to provide liquidity to the mortgage finance system, Fannie Mae and
Freddie Mac support what is commonly known as the secondary mortgage market. The
Enterprises purchase from loan sellers residential mortgages that meet their underwriting criteria.
The loan sellers can then use the sales proceeds to originate additional mortgages. The
Enterprises can hold the mortgages in their portfolios or package them into MBS that are, in turn,
sold to investors. In exchange for a fee, the Enterprises guarantee that investors will receive
timely payment of principal and interest on their investments.

HERA provides FHFA with broad authority as the Enterprises’ conservator to conserve and
preserve Enterprise assets and to control and direct their finances and operations. FHFA has
exercised that authority by, among other things, requiring FHFA pre-approval of certain
categories of Enterprise business operations such as settlements of claims exceeding $50 million.
In this regard, FHFA seeks to ensure that these high-dollar settlements are in the best interests of
the Enterprises and the taxpayers.

For the purpose of this evaluation, two offices within FHFA, which report to FHFA’s Acting
Director, are relevant: the Office of Conservatorship Operations (OCO) and the Division of
Enterprise Regulation (DER). OCO coordinates all activities concerning conservatorship issues.
In this case, it took the lead in coordinating FHFA’s review and approval of the Fannie Mae and
Freddie Mac repurchase settlements with Bank of America. DER is an organizational unit
comprised of FHFA examiners who have in-depth knowledge of Enterprise operations and credit
risk work.

Overview of the Mortgage Repurchase Process

Designed to mitigate potential credit losses, the Enterprises’ underwriting standards for loans
they purchase are established in their federal charters and company policies. Lenders and other
entities that sell mortgages to the Enterprises are contractually required to “represent and
warrant” that, at the time of their origination, the loans they sell comply with the Enterprises’
                         5
underwriting standards.



5
 These representations and warranties are detailed in Freddie Mac’s Single Family Seller/Servicer Guide and Fannie
Mae’s Selling Guide.



        Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                       11
The Enterprises have established ongoing, post-purchase quality review processes to verify that
the loans they purchase conform to their underwriting standards. If an Enterprise determines that
a loan did not conform to its underwriting standards at the time of the loan’s origination, then the
Enterprise may require loan seller to repurchase the loan at full face value or to indemnify the
Enterprise for any losses incurred. For example, the Enterprises review mortgages (the majority
of which have gone into foreclosure) to determine whether the representations and warranties
included in them were correct and in compliance with their underwriting standards. Based on
such analysis, the Enterprises determine whether to request that loan sellers repurchase defective
mortgages.

To date, the Enterprises have recovered billions of dollars through their assertion of repurchase
claims. For instance, as of January 2011 Freddie Mac had received repurchase payments from
loan sellers on about 8% of approximately one million loans that it had purchased that were then
in foreclosure.6 As of June 30, 2011, Freddie Mac had outstanding repurchase claims on loans
                                                           7
with a combined unpaid principal balance of $3.1 billion.

Changes in Mortgage Lending Practices During the Housing Boom

With the unprecedented growth in the United States housing market during the 2005 to 2007
housing boom, the quality of loans originated and sold to the Enterprises deteriorated
              8
substantially. Before the boom, the mortgage market largely consisted of fixed rate, amortizing
loans, such as 30-year fixed rate mortgages requiring equal payments each month over the life of
the loan, and adjustable rate mortgages (ARMs) that incorporated features to protect borrowers
from excessive fluctuations in monthly payments (such as “caps” limiting the amount by which
the mortgage’s interest rate can rise over the life of the loan).

However, from 2005 through 2007 there was a substantial increase in non-traditional mortgage
products. These products had significantly enhanced risk profiles compared to more traditional
mortgage products. First, they often included inherently risky attributes, such as significantly
curtailed verification of borrowers’ incomes and assets. Second, non-traditional loans appear to
have significant percentages of representations and warranties defects.9


6
    Freddie Mac QC Disposition of Foreclosures by Funding Year, dated 1/11/11.
7
 Freddie Mac Update August 2011, at 16, available at www.freddiemac.com/investors/pdffiles/investor-
presentation.pdf.
8
    Financial Crisis Inquiry Commission, Financial Crisis Inquiry Report (FCIC Report), at 178-79 (2011).
9
 Freddie Mac data summarizing housing boom era loans eligible for repurchase claims show that for loans
originated in 2006, 2007, and 2008, 18.4%, 20.6%, and 23.4% respectively were “ineligible,” meaning that Freddie
Mac considered these loans potentially good candidates for repurchase claims. Freddie Mac Document, “NPL QC


          Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                          12
Frequently, the non-traditional loans featured “teaser” rates initially resulting in low payments,
but those payments could increase dramatically two, three, or five years after origination when
the rates reset and/or the repayment of principal began. Although borrowers with limited
incomes and credit histories might be able to afford property purchases using such non-
traditional loans during the teaser rate periods, the potential for defaults increased dramatically
when the monthly payments on these loans subsequently reset at higher levels. Aggravating
these conditions, defaults increased as housing prices began to fall at the end of 2006. The
falling prices left many homeowners “underwater” – that is, with mortgage balances exceeding
the value of the homes securing them.

Figure 1 illustrates the dramatic increase in two of the more commonly used non-traditional loan
types during the housing boom years: Interest Only and Option ARM loans. Interest Only loans
permit the borrower to pay only interest on the loan, not principal, for a specified period; Option
ARMs are adjustable rate mortgages that permit the borrower, for a specified period, to choose
among different payment options each month, ranging from traditional interest and principal
payments, to interest only payments, to payments below the amount of interest owed each
       10
month.




Review Results By Loan Characteristics Loans Funded January 2006-December 2009 QC Results as of Mar 3,
2011.” Moreover, Freddie Mac’s internal auditors, in a June 6, 2011, audit opinion report, cited to repurchase rates
exceeding 10% among Alt-A loans from 2005 that entered foreclosure. June 6, 2011, Freddie Mac Memorandum,
Re: Performing Loans Quality Control and Administration Audit (#2011-010), at 10-11.
10
 Federal Reserve Board, Consumer Handbook on Adjustable Rate Mortgages, available at
www.federalreserve.gov/pubs/arms/arms_english.htm.



        Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                         13
Figure 1: Significant Growth in Interest Only and Option ARM Loan Originations in the
          Overall Mortgage Market During 2005-2007 Housing Boom11

        Percentage of All Originations
            30%


            25%


            20%
                        Option ARM
                        Interest Only
            15%


            10%


             5%


             0%
                          2004                  2005                 2006                  2007



Although some non-traditional mortgages had interest rate resets within two years after
origination, many others reset at a later time. For example, according to Freddie Mac, 80% of its
Interest Only loans that originated in 2005 had their first payment adjustment five years after
             12
origination.

There was also significant growth during the housing boom in higher-risk Alt-A mortgages as an
alternative to lower-risk prime mortgages. Offered to those borrowers with credit profiles
approaching those of prime borrowers, Alt-A mortgages often required limited or no
                                                                                     13
documentation of key borrower credit risk characteristics, such as income and assets. For
example, borrowers might only have to state their annual income rather than provide verifying
documentation, such as W-2 tax forms. Such limited- or no-document loans are also referred to

11
  Source: Inside Mortgage Finance, 2011 Mortgage Market Statistical Annual, “Alternative Mortgage
Originations,” at 32.
12
     Sept. 15, 2010, FHFA Analysis Memorandum, at 2.
13
 Government Accountability Office, Testimony of William B. Shear Before the U.S. Congress Joint Economic
Committee on Home Mortgages, at 1 n.1 (July 28, 2009), available at www.gao.gov/new.items/d09922t.pdf.



          Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                       14
as “stated income” (or, more colloquially, “liar”) loans. These categories of loans are not
mutually exclusive; some Alt-A loans incorporated Interest Only or Option ARM payment
structures.

During the housing boom, the Enterprises purchased large volumes of these non-traditional
mortgages from large lenders, such as Countrywide. Countrywide was one of the most
                                                                                      14
aggressive originators of limited- or no-document Interest Only and Option ARM loans.

In early 2008, with the collapse of the housing market, Bank of America purchased
                                                     15
Countrywide, which was then on the verge of failure. Countrywide loans are the dominant
component of the portfolio included within the Freddie Mac-Bank of America settlement and
account for a significant number of repurchase claims asserted by Freddie Mac. For example,
prior to the Bank of America settlement, Freddie Mac reviewed 58% of all Countrywide loans in
foreclosure and made repurchase claims on 24% of them.
                                                                                16
Chronology of Key Events and Associated Analysis

a. Nine Months Prior to the Bank of America Settlement, an FHFA Senior Examiner Identifies
   Changes in Housing Foreclosure Patterns

In March 2010, an FHFA senior examiner, who is assigned to oversee Freddie Mac, noticed in
Freddie Mac-supplied housing data an unusual pattern among foreclosures of loans originated
during the 2005 to 2007 housing boom years. That pattern, as discussed in detail below, may
have significant financial consequences for Freddie Mac and the taxpayers.

Before the housing boom, when the mortgage market was dominated by more traditional loans,
mortgages that defaulted tended to do so during the first three years following origination.
Further, the rate of defaults declined over time as the loans seasoned. This is reflected in Figure
                                                                                 17
2, showing when loans purchased by Freddie Mac in 2001 entered foreclosure.




14
     FCIC Report at 105.
15
     FCIC Report at 250.
16
     A chart summarizing a timeline of key events is included at Appendix C.
17
     Freddie Mac purchases the vast majority of its loans shortly after origination.



           Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                             15
Figure 2: Loans Purchased in 2001 by Freddie Mac that Entered Foreclosure18

        No. of Foreclosure Starts
          14,000

          12,000

          10,000

           8,000

           6,000

           4,000

           2,000

                0
                           1                2                3                4               5
                                           Loan Age at Foreclosure (in years)

But a different pattern exists among loans that Freddie Mac purchased that were originated
during the housing boom. Rather than foreclosures declining over time, Freddie Mac-supplied
housing data revealed foreclosures increasing, three, four, and five years after purchase, as
reflected in Figure 3. It shows that for Freddie Mac-owned mortgages purchased in 2006 there
were relatively few foreclosures within the first two years after purchase but there were
significantly higher numbers of foreclosures during years three through five.




18
     Source: Freddie Mac QC Disposition of Foreclosures by Funding Year, dated 1/11/11.



          Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                        16
Figure 3: Loans Purchased in 2006 by Freddie Mac that Entered Foreclosure19

        No. of Foreclosure Starts
          60,000


          50,000


          40,000


          30,000


          20,000


          10,000


                0
                           1                 2                 3                  4                  5
                                           Loan Age at Foreclosure (in years)

Figure 3 also shows over 100,000 additional loans in default (as compared to 2001-vintage
loans), likely the result of the collapsed housing market and the onset of the financial crisis.

The FHFA senior examiner attributed the reversed pattern to the end of the teaser rate period for
non-traditional mortgages,20 and he recommended further study of the issue. An FHFA staff
memorandum explained:

           [I]t would be reasonable to assume that many of the borrowers, faced with
           significantly increasing payments in the near term and very little equity in their
           home, made the decision to default before their [payments reset to higher levels].
           It would also be reasonable to assume that the stated income and stated asset


19
     Source: Freddie Mac QC Disposition of Foreclosures by Funding Year, dated 1/11/11.
20
  Freddie Mac staff advised FHFA-OIG that they disagree with the senior examiner’s causation hypothesis.
Alternatively, they attribute the reversed pattern of foreclosures shown in Figure 3 to falling home prices leading to
negative equity or “underwater” mortgages. However, causation is irrelevant to the issue in controversy.
Regardless of the cause of these defaults, the search for representations and warranties defects is the point of the
loan review process; and if the search does not begin, then the defects will not be found.



          Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                          17
           underwriting requirement played a role, but neither assumption can be tested
                                          21
           without a review of the loans.

As discussed in more detail below, FHFA did not test the loan review process to validate the
senior examiner’s concerns prior to its review and approval of the Bank of America settlement.

It should be noted that not all causes of foreclosure will justify a repurchase claim. For example,
foreclosures may result from a borrower’s subsequent loss of a job or health issues. But
repurchase claims are fact-specific and based upon representations and warranties defects, such
as missing or erroneous information regarding the quality of a borrower’s assets or income.

b. FHFA Senior Examiner Raises Concerns that Freddie Mac Did Not Revise Its Loan Review
   Process for Repurchase Claims to Account for Foreclosure Pattern Changes Among Housing
   Boom Mortgages

The FHFA senior examiner also observed that, despite the apparently changed foreclosure
patterns associated with housing boom era mortgages, Freddie Mac had not adjusted its process
for identifying loans that might be candidates for repurchase claims. Freddie Mac reviews
intensively for repurchase claims only those loans that go into foreclosure or experience payment
problems during the first two years following origination. Loans that default thereafter are
reviewed at dramatically lower rates. Freddie Mac senior management believe that loan
underwriting defects such as an undisclosed lien on a property – which may be an indication of a
representations and warranties deficiency – are most likely to appear within the first two years
                        22
following origination. Moreover, Freddie Mac management has advised FHFA-OIG that they
also believe that higher rates of loan defaults in later years do not necessarily equate to higher
defect rates. In their view, loans that had demonstrated a consistent payment history over the
first two years following origination and then defaulted in later years (i.e., years three through
five after origination) likely did so for a reason such as loss of employment, which is unrelated to
a representations and warranties defect.23 Based on these assumptions, Freddie Mac does not
review most loans that go into foreclosure more than two years after origination. It reviews such
loans only if they had already exhibited problems such as missed or late payments during the
initial two years after origination or have potential indications of value discrepancies or any
indication of fraud.


21
     Sept. 15, 2010, FHFA Analysis Memorandum, at 2-3.
22
     November 2, 2010, FHFA Analysis Memorandum, prepared by the FHFA Division of Enterprise Regulation, at 3.
23
  As discussed later in this report, Freddie Mac’s internal auditors requested and Freddie Mac management agreed
to test these assertions. Such testing is currently under way.



          Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                         18
This practice meant that most pre-housing boom loans in foreclosure were reviewed for
                      24
repurchase claims. However, the shift in foreclosure patterns among housing boom loans
(loans foreclosed three through five years after origination) meant most of them were not being
reviewed, regardless of their potential viability for repurchase claims. Yet, later payment resets
common among housing boom loans may have temporarily hidden the impact of representations
and warranties defects (e.g., erroneous information about borrower income may not have come
to light until their loan payment resets if the borrowers had sufficient income to satisfy the
“teaser” rate payments but not the later permanent payments). The FHFA senior examiner
shared his concerns with Freddie Mac management in June 2010 at a meeting attended by three
FHFA examiners and an FHFA manager. A June 9, 2010, FHFA memorandum summarized the
issue as follows:

           It was pointed out to [Freddie Mac] that over 93% of the year-to-date [loan]
           foreclosures [(as of June 2010)] from the 2005 and 2006 [loan] vintages have
           been excluded from [loan repurchase] review, eliminating any chance to put
           ineligible loans back to the lenders from those years.25

Figure 4 demonstrates the extent to which Freddie Mac has not reviewed housing boom era
mortgages that went into foreclosure during the third through fifth years after their origination. It
shows that by choosing to review intensively only those loans that defaulted within two years of
origination, Freddie Mac did not examine close to 100,000 2006 vintage loans.




24
 For example, from 2000 through 2004 Freddie Mac reviewed 62% of the 191,853 loans in foreclosure. Freddie
Mac QC Disposition of Foreclosures by Funding Year, dated 1/11/11.
25
     July 9, 2010, FHFA Meeting Notes, at 2.



          Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                       19
                                                                                                 26
Figure 4: Loans Purchased in 2006 by Freddie Mac that Entered Foreclosure

           No. of Foreclosure Starts
            60,000
                                Approximately 100,000 Loans
            50,000                     Not Reviewed


            40,000


            30,000
                          Not Reviewed
                          Reviewed
            20,000


            10,000


                 0
                            1                2              3                4                  5
                                             Loan Age at Foreclosure (in years)


Freddie Mac data further show that for all Enterprise-owned foreclosed loans originated between
2004 and 2007, Freddie Mac has not reviewed over 300,000 loans for possible repurchase
        27
claims. Those loans that were not reviewed (hereafter referred to as “out-of-sample” loans)
have a combined unpaid principal balance exceeding $50 billion. Many of these loans are likely
not candidates for repurchase. For instance, a portion of the loans not reviewed are lower-risk
prime loans, which probably have a lower incidence of representation and warranty defects. On
the other hand, Freddie Mac’s portfolio of housing boom loans includes a substantial number of
Interest Only and Alt-A mortgages, which have a high incidence of defects.28




26
     Source: Freddie Mac QC Disposition of Foreclosures by Funding Year, dated 1/11/11.
27
     Id.
28
  For example, Freddie Mac’s internal auditors have observed that Interest Only and Alt-A loans respectively
comprise 24% and 35% of all 2006 vintage loans in foreclosure, and 38% and 36% of all 2007 vintage loans in
foreclosure. Freddie Mac 2011-010 PL Quality Control & Administration Audit Draft Audit Report Findings
(05/05/11) (Draft Version 4.0), Fig. 3 and supporting data.



            Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                         20
c. FHFA Senior Examiner Views Freddie Mac’s Continued Use of Its Loan Review Process as
   Potentially Costing Freddie Mac “Billions of Dollars”

Throughout 2010, the FHFA senior examiner discussed with Freddie Mac managers his concerns
about the Enterprise’s continued reliance on its current loan review process. In his view, by not
reviewing intensively the mortgages foreclosed upon more than two years after origination for
repurchase claims, Freddie Mac could potentially lose “billions of dollars” that could be used to
mitigate taxpayer losses.29

On June 9, 2010, during a regular monthly meeting involving four FHFA examination staff
members and Freddie Mac senior managers, referenced above, the concerns about Freddie Mac’s
continuing use of its loan review process were discussed (“It was pointed out … that over 93%
of the year-to-date [loan] foreclosures from the 2005 and 2006 [loan] vintages have been
excluded from [loan repurchase] review.”). A Freddie Mac senior manager said he had analyzed
data on “loans defaulting 3-5 years out and concluded that [repurchase] reviews would not prove
fruitful.” But the manager agreed to conduct testing and “acknowledged that looking at the
actual loan files would improve his analysis and so [he] agreed to call in a sample of those loans”
to review.30

However, Freddie Mac officials ultimately did not review such a sample in 2010 or otherwise
test issues related to the senior examiner’s hypothesis. Moreover, FHFA did not require Freddie
Mac to do so or to conduct independent testing. According to an FHFA examination staff
description of a July 26, 2010, meeting of Freddie Mac’s Credit Risk Subcommittee, a Freddie
Mac manager told FHFA staff that loan repurchase review “was ‘resource constrained’ and
sampling older defaults was ‘not the highest and best use of his limited resources.’”31 Weeks
later, the FHFA senior examiner reported to FHFA senior managers that a Freddie Mac manager
had informed him that another Freddie Mac senior manager was “vehemently against looking at
more loans” but had offered “no cogent argument” explaining his resistance.32

29
  As discussed herein, the senior examiner’s concerns were not confined to the Bank of America settlement, but
covered all loan sellers and all potential future settlements. The issue is currently under review by FHFA and
Freddie Mac.
30
     June 9, 2010, FHFA Meeting Notes, at 2.
31
     Sept. 15, 2010, FHFA Analysis Memorandum, at 3.
32
     Sept. 29, 2010, FHFA e-mail, Re: IO and OA defaults.
In a September 23, 2010, internal e-mail chain, the Freddie Mac senior manager told the Freddie Mac manager,
“[w]e have spent a fair amount of time trying to help sellers forecast loan samples and repurchase request[s]. We
have laid out a pretty clear sampling strategy.” Sept. 23, 2010, Freddie Mac e-mail (11:04 AM), Re: NPL Sample
on Older IO ARMs and Options Arms. Later in the same email chain, the senior manager told the manager, who
suggested a temporary review of additional loans for two to three months, that “given the visibility and sensitivity


          Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                            21
Senior Freddie Mac managers disagreed with the FHFA senior examiner’s concerns, at least
partly because they believed a change to a more aggressive approach to repurchase claims would
adversely affect Freddie Mac’s business relationships with Bank of America and other large loan
sellers. During the course of this evaluation, FHFA-OIG staff interviewed the relevant Freddie
Mac senior managers, who asserted that the existing loan review process was appropriate and
that changing the process could potentially cost Freddie Mac business. One senior manager, who
confirmed that he had recommended against further study of the default-timing anomaly, said he
did not believe Freddie Mac would recover enough from a more expansive loan review process
to offset losses of business from Bank of America and other loan sellers. Another Freddie Mac
senior manager also talked about the potential loss of business and emphasized that he did not
believe that the number of repurchase claims would increase appreciably.

d. FHFA Senior Examiner Alerts FHFA Staff, Managers, and Senior Managers to the Concerns
   About Freddie Mac’s Loan Review Process

Between June and December 2010, approximately one dozen FHFA staffers, managers, and
senior managers were alerted to the FHFA senior examiner’s concerns about Freddie Mac’s loan
review process. See Appendix D for a timeline showing when each staffer, manager, and senior
manager was first alerted. Nonetheless, FHFA did not timely act on or test the data underlying
these concerns prior to approval of the Bank of America settlement. FHFA has advised FHFA-
OIG that the senior examiner did not raise his concerns in the context of the normal FHFA
examination process. However, the record is clear that his concerns were known to FHFA senior
management well in advance of the completion of the settlement.

On September 15, 2010, the FHFA senior examiner prepared and circulated to FHFA managers
an Analysis Memorandum describing the concerns. The memorandum recommended that
Freddie Mac change its loan review process to analyze greater numbers of housing boom loans
in foreclosure for repurchase claims. The memorandum also disputed Freddie Mac’s argument
that limited resources undermined its capacity to review a larger sample of loans and concluded
by noting that the Enterprise was potentially losing out on significant potential mortgage
repurchase recoveries.

         By not taking a good look at these defaulted [Interest Only and Alt-A] loans over
         the next 2-3 years, … with a loss severity rate above 40%, Freddie [M]ac could be
         passively absorbing billions of dollars of losses. Since the savings in credit losses
         would dwarf the incremental expenses incurred in reviewing additional loan files,

around [loan reviews] and repurchases, I view any change, even temporary as material. I would prefer we lay out a
proposal here, with clear goals and objectives, then do at least a rough cost benefit.” Sept. 23, 2010, Freddie Mac e-
mail (11:44 AM), Re: NPL Sample on Older IO ARMs and Options Arms.



        Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                         22
           the fundamental question that Freddie Mac and FHFA should be addressing is
           this: How many of the ineligible loans sold to Freddie Mac in the 2005-2007
           origination years should Freddie Mac accept the loss on? (Emphasis in the
                      33
           original.)

FHFA recipients of the memorandum offered differing responses to its contents. One senior
manager told FHFA-OIG that he never read the memorandum because he had never opened the
e-mail attachment containing it. Two managers (a senior manager and a manager) acknowledged
that they had reviewed the memorandum, but they did not remember that the issue could
potentially involve substantial losses to Freddie Mac. Another recipient noted that “this [issue]
is important” and observed that “[o]ver time, I have consistently been concerned about sampling
size. [Freddie Mac] appears to define sample size by the # of [full time employees] it has or
                                                      34
wants, rather than by the true risk in the portfolio.” The senior examiner, in a reply e-mail that
also copied the senior manager – who never read the memorandum – said:

           [S]taffing [for Freddie Mac] isn’t an issue because [Freddie Mac] can hire or use
           vendors, or both. As I said yesterday, if you hire more underwriters, they will pay
           for themselves in the first week. This all goes away in about 2 years, but $billions
                                            35
           will be lost if nothing is done.

Additional e-mails describing the FHFA senior examiner’s concerns were also sent to other
FHFA staff, managers, and senior managers before FHFA approved the Freddie Mac-Bank of
America settlement on December 29, 2010. In a November 23, 2010, e-mail another FHFA
senior manager was advised by the FHFA senior examiner that the concerns involved “billions of
         36
dollars.” A December 9, 2010, e-mail commenting on the then-proposed Freddie Mac-Bank of
America settlement observed that “if the agreement goes as is, those losses [on loans not
reviewed] will be Freddie’s and the discussion is over,” and concluded that “the settlement
                      37
number is too low ….” And, on the eve of the settlement’s approval, a December 28, 2010,
e-mail from the FHFA senior examiner to an OCO staffer again made the same point. It said that




33
     Sept. 15, 2010, FHFA Analysis Memorandum, at 4.
34
     Sept. 30, 2010, FHFA e-mail (8:12 AM), Re: IO and OA defaults.
35
     Sept. 30, 2010, FHFA e-mail (9:12 AM), Re: IO and OA defaults.
36
 Nov. 23, 2010, FHFA e-mail, Re: FW: FHFA AM NEWS SUMMARY 11 22 10. That senior manager told
FHFA-OIG that he did not recall knowing that the issue potentially concerned billions of dollars of losses.
37
     Dec. 9, 2010, FHFA e-mail, Re: BoA settlement with Freddie.



          Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                        23
Freddie Mac’s continued use of its loan review process was a “huge” error, and the resulting
                                                                                38
losses would be “Freddie’s losses, and of course, yours and mine as taxpayers.”

e. Freddie Mac Reaches a Tentative Repurchase Settlement with Bank of America; Freddie
   Mac’s Internal Auditors Independently Raise Concerns About Freddie Mac’s Loan Review
   Process

In early December 2010, Freddie Mac management agreed to a tentative settlement of repurchase
claim issues with Bank of America. The tentative settlement was subject to approval by Freddie
Mac’s board of directors and FHFA. The settlement, which Bank of America wanted to finalize
before the end of the year, required the bank to pay Freddie Mac $1.35 billion in exchange for
relinquishment (with limited exceptions) of all pending and future repurchase claims related to
787,000 mortgage loans previously sold to Freddie Mac by Bank of America and Countrywide.

Enterprise management advised Freddie Mac’s board of directors that the $1.35 billion figure
was a reasonable settlement amount. The figure was premised on the assumption that Freddie
                                                                      39
Mac would in the “expected case” likely recover about                    in repurchase claims from
                                                                     40
Bank of America from the specified portfolio of mortgage loans. Freddie Mac management
further explained, however, that there was “significant uncertainty” (or significant margin of
error) in this figure and that it could vary positively or negatively by                       . Thus,
according to Freddie Mac management, a reasonable recovery in the expected case could range
from about                                .41 The proposed settlement of $1.35 billion was at the
high end of the expected case range. These calculations incorporated the assumptions underlying
Freddie Mac’s existing loan review process, as well as revisions to a financial model Freddie
Mac developed to estimate repurchase claims exposure.




38
     Dec. 28, 2010, FHFA e-mail (12:35 PM), Re: FYI--CW I/Os.
39
   Red text signifies content that FHFA and Freddie Mac claim is confidential financial, proprietary business, or
trade secret information that is redacted in the publicly available version of this report.
40
  Bank of America Repurchase Settlement Proposal (Dec. 17, 2010), at 3. The precise figure given to the board of
directors was           .
41
  Id. The board was further informed that the possible recovery from Bank of America in a “stress case” was
       , and that a reasonable recovery in the stress case could range from about                          . The
“stress case” assumed, among other things, a worsening economy to a greater extent than the “expected case,”
leading to greater numbers of foreclosed loans and greater losses on repurchase claims.



          Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                         24
Freddie Mac’s board of directors was also told that the settlement had a number of benefits, as
follows: 42

               Because of “uncertainty around estimates,” Freddie Mac stood to recover less money
                if it did not settle and instead continued to pursue repurchase claims;

               The settlement would reduce Freddie Mac’s counterparty exposure to Bank of
                America, which was consistently greater than Freddie Mac’s internal risk
                management policy permitted;

               Lower levels of potential Bank of America counterparty exposure could permit
                Freddie Mac to do more “capital markets” business with Bank of America (such as
                issuing MBS and corporate debt);

               “If the counterparty fails,” Freddie Mac would have already been paid and the
                “benefit of representations and warranties [payments would have been] realized
                before failure;”

               The settlement “[i]mproves [Freddie Mac’s] ongoing relationship with Bank of
                America;”

               The settlement would reduce Freddie Mac’s costs associated with reviewing loans for
                repurchase claims;

               The settlement would be “positive [for Freddie Mac’s] current financial results;” and

               The settlement would reduce Freddie Mac’s “ongoing litigation [expense] risk of a
                loan-by-loan enforcement strategy.”

In late November and early December 2010, Freddie Mac’s internal auditors evaluated the
settlement for reasons related to Freddie Mac’s counterparty exposure to Bank of America and
unrelated to the issues raised by the FHFA senior examiner. During the course of their review,

42
     Id. at 5. The board was also told of four risks or “cons” associated with the settlement:
          “Uncertainty about [the internal] estimates could result in losses beyond [the] settlement amount;”
          The “[t]ransfer of credit risk (beyond [the] settlement amount) from Bank of America to Freddie Mac [on
           settled loans would be] ultimately transferred to the taxpayer;”
          “Low probability of counterparty failure;” and
          Freddie Mac would have to change its internal models to account for the settlement.




           Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                             25
the auditors independently questioned Freddie Mac’s existing loan review process and
documented their questions in a December 14, 2010, memorandum. The memorandum made
two recommendations concerning the effect of the loan review process on loans not being
reviewed for repurchase claims. Specifically, the internal auditors recommended that Freddie
Mac management should:

            1. Provide an overview of [Freddie Mac’s] current sampling methodology, including a
               description of the portion of the portfolio that is not sampled; and

            2. Quantify the potential risk of loss that is not or was not the subject of sampling
                                                                  43
               pursuant to current and past sampling strategies.

f. Freddie Mac Management Responds

In response to the internal auditors, Freddie Mac management prepared a memorandum (also
dated December 14, 2010), which attempted to calculate how much money Freddie Mac would
lose by not pursuing repurchase claims on loans that went into foreclosure three to five years
after funding. In other words, Freddie Mac attempted to calculate how much it would be
“leaving on the table” by not changing its existing loan review process to adjust for the changed
circumstances brought about by the housing boom. Freddie Mac management calculated that
                                                                         44
figure to be in the range of                      in the “expected case.” However, Freddie
Mac’s chief internal auditor observed that a potential              loss, which is at the low end
of that range, left little if any of the                   margin of error cushion associated with
the settlement negotiations discussed above. Any amount greater than                   would
exceed the margin of error.

In making their calculation, Freddie Mac management did not have time to undertake a fresh
study based on a representative sample of the “out-of-sample” loans, as requested by the FHFA
senior examiner in June 2010, given the goal of closing the settlement by year-end. Instead,
management used existing data collected for another purpose. It relied on a sample of about
2,200 loans drawn from all loan seller/servicers from which Freddie Mac purchased mortgages
that had gone through repurchase claim review after having gone into foreclosure more than two



43
     Id. at 3.
44
  Dec. 14, 2010, Memorandum from Freddie Mac Senior Management to Freddie Mac’s Internal Auditors, at 3.
The “expected case” assumed that the economy would worsen slightly. Management further assumed that, in a
“stress case,” Freddie Mac could expect to recover larger amounts, specifically              – more than
double the margin of error.



            Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                         26
years after origination.45 However, as Freddie Mac internal auditors have acknowledged, the
loan sample used by management was not representative.46 Among other things, the loans in the
Freddie Mac management sample were drawn from all loan sellers, not only the loans found
within the Bank of America settlement population. This represents a significant difference
because most of the Bank of America loans in foreclosure were originated by Countrywide,
which was among the most aggressive originators of higher-risk, non-traditional loans and whose
loans had significantly above-average numbers of defects subject to repurchase claims.47

Freddie Mac management also justified its current loan review process under a “business
practices” rationale. Freddie Mac management said that maintaining stable customer
relationships that might lead to additional business with loan sellers like Bank of America
justified the existing loan review process. The December 14 memorandum states:

           [T]he sample size is also impacted by our overall business strategy. Our sampling
           strategy is considering several goals, including put-backs of defective loans that
           create losses for the firm, providing incentives for sellers to produce well-
           underwritten loans, and maintaining stable customer relationships. For the
           settlement negotiations with Bank of America, management made a deliberate
           decision not to consider changes to our sampling procedures. Hence, the model
           was built on the assumption that past sampling practices are the best guide for
           future policies. While there is always the possibility that sampling policies will
           change going forward to be either more or less stringent, we did not adjust for
           these explicitly in evaluating the Bank of America settlement. However, we do
           have assumptions in the model that we believe account for potential risk in our
           valuation, in particular, our capital costs.48

In other words, Freddie Mac management asserted that the need to maintain relationships with
loan sellers such as Bank of America was a factor weighing against implementing more
expansive loan review and repurchase policies.


45
  These loans were purportedly a “proxy” for a random sample. In fact, the loans in question had defaulted three,
four, or five years after origination and had good pay histories in the first two post-origination years. Ordinarily
such loans would not be reviewed using Freddie Mac’s current loan review process. This group had been reviewed
because Freddie Mac suspected that the loans might be defective (insofar as their values significantly exceeded local
averages), but further research had found no evidence of defects.
46
     Freddie Mac notes that this fact was disclosed to its board of directors.
47
 Freddie Mac staff has advised FHFA-OIG that before 2010, Countrywide loans had 50% more representations and
warranties violations than the average.
48
     Dec. 14, 2010, Memorandum from Freddie Mac’s Senior Management to Freddie Mac’s Internal Auditors, at 4.



           Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                             27
Freddie Mac’s board of directors approved the Bank of America settlement on December 14,
2010.

Freddie Mac’s chief internal auditor advised the board of directors that management had
“highlighted and quantified the enumerated key risks.”49 At a December 17, 2010, board
meeting, the chief auditor noted that management’s estimate of                   (which, as
discussed above, was the amount Freddie Mac could lose in the settlement by not changing its
loan review process) was “significant.” Given that the proposed settlement allowed only for a
              margin of error in the “expected case,” or low range, the auditor told the board that
“[f]rom this perspective there was little, if any, cushion, left for model uncertainty, further house
price declines or higher severities.” In other words, the auditor regarded management’s low
estimate to be at or very near the margin of error cushion. Any estimated amount greater than
              would exceed the margin of error.

g. FHFA Staff Reviews and Recommends Approval of the Freddie Mac-Bank of America
   Settlement

Starting in early December 2010, FHFA staffers, managers, and senior managers also began to
review the proposed settlement. FHFA senior management summarized their review in a
December 28, 2010, memorandum to the Acting Director that recommended he approve the
settlement. The memorandum provided significant detail about the settlement and included the
package of materials supplied to the Freddie Mac board of directors prior to their approval of the
settlement. The FHFA memorandum discussed Freddie Mac’s and Bank of America’s
motivations to settle, explained the analysis and corporate governance process conducted by
Freddie Mac management, reviewed risk factors, and compared the settlement to other
repurchase settlements. Additionally, one paragraph in the memorandum identified the FHFA
                                                                        50
senior examiner’s concerns about Freddie Mac’s loan review process. The paragraph described
the process and noted that the Freddie Mac management had estimated the risk associated with
the process to be “quantified in the range of                               in recoveries.” But, as
discussed above, Freddie Mac’s estimate had been premised on an unrepresentative sample of
2,200 loans, and it effectively equaled or offset the settlement’s margin of error.51

49
   Dec. 14, 2010, Memorandum from Freddie Mac’s internal auditor to the board of directors, at 4. FHFA believed
that the auditors had considered Freddie Mac’s current loan review process and found it to be “appropriate and
reasonable.” Dec. 28, 2010, Memorandum to the Acting Director, Re: Bank of America Recommended Settlement,
at 5. However, according to Freddie Mac’s chief internal auditor, the internal auditors did not endorse or disapprove
the terms of the settlement. Rather, they raised concerns about risks associated with the settlement and advised the
board of directors that Enterprise management had “highlighted and quantified the enumerated key risks.”
50
     Dec. 28, 2010, Memorandum to the Acting Director, Re: Bank of America Recommended Settlement, at 5.
51
     Dec. 28, 2010, Memorandum to the Acting Director, Re: Bank of America Recommended Settlement, at 5.


          Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                         28
Prior to conducting the settlement review, FHFA did not test the examiner’s concerns (for
instance, FHFA did not insist that Freddie Mac management follow through on the promise
made in June 2010 to test a representative sample of loans in order to validate the senior
examiner’s concerns). Instead, the Agency relied on Freddie Mac’s loan review process and its
analysis of the settlement.

FHFA staff also faced time limitations in light of the goal of closing the settlement by the end of
the month.52 The short timetable affected what could be accomplished. For instance, FHFA
staff suggested bringing in an outside expert to assist staff in their review, but FHFA senior
management declined to do so because of the goal to finalize the deal by year-end.53

h. FHFA’s Acting Director Suspends All Future Enterprise Repurchase Settlements Pending
   Further Review; Freddie Mac’s Internal Auditors Issue an “Unsatisfactory” Audit Opinion

FHFA’s Acting Director approved the settlement on December 29, 2010. However, after this
evaluation began, and on the basis of concerns raised by FHFA-OIG and others about Freddie
Mac’s loan review process and its impact on repurchase settlements, FHFA suspended, pending
further review, all future Enterprise repurchase settlements affected by the methodology
underlying Freddie Mac’s current loan review process.

Additionally, Freddie Mac’s internal auditors continued to examine Freddie Mac’s loan review
process and, on June 6, 2011, they delivered to Freddie Mac’s senior management an opinion
that the Enterprise’s internal controls associated with its loan review process were
                  54
“Unsatisfactory.” The auditors’ report explained that their opinion was “primarily driven by
deficiencies noted with the governance, business rationale, and objectives of the [loan review
process] and oversight of the … process.”

As part of their work, the internal auditors analyzed Freddie Mac-owned loans that were funded
in 2005 and were in foreclosure and – like the FHFA senior examiner – observed a sharp

52
     For example, a December 24, 2010, e-mail from Freddie Mac to FHFA senior management reiterated:
           BofA wants certainty and we will need your [(FHFA’s)] sign-off so we can proceed to finalize
           everything on Tuesday and sign docs on Tuesday or Wednesday with the settlement, payment and
           disclosure on Friday the 31st.
Dec. 24, 2010, Freddie Mac e-mail to FHFA (18:55), Re: BofA settlement.
53
  One senior manager told FHFA-OIG that he felt no time pressure to complete the review. However, others have
told FHFA-OIG that they believed time pressure had an effect.
54
  June 6, 2011, Freddie Mac Memorandum, Re: Performing Loans Quality Control and Administration Audit
(#2011-010), at 1. The opinion addressed the loan review process in general, not the Bank of America settlement in
particular.



          Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                        29
increase in foreclosures more than two years after origination, along with an equally dramatic
fall-off in loan reviews after the second year, as shown in Figure 5 below.

Figure 5: Freddie Mac Internal Auditors’ Depiction of Default Timing Anomaly 55




This observation led the internal auditors (in a June 2011 presentation to the Freddie Mac board
of directors) to assert that “[o]pportunities for increasing the repurchase benefit justify an
                                                        56
expansion of our sampling approach after year two.”

The auditors recommended and management agreed to put additional emphasis on tying loan
review methodologies to the volume of foreclosures (to examine larger numbers of currently
unreviewed loans) and to “place more emphasis on balancing the customer relationship with the
                               57
ultimate cost to the company.”

Consistent with the internal auditors’ findings, the same Freddie Mac senior manager who
prepared the Freddie Mac management estimate at the end of 2010 informed the Enterprise’s
board of directors that he believed Freddie Mac could recover several billion additional dollars
by changing its current loan review process. On May 26, 2011, the senior manager advised the



55
     Id. at 9, Fig. 2.
56
     June 3, 2011, Presentation to the Freddie Mac Board of Directors, re: “Repurchase Sampling Strategy,” at 3.
57
  June 6, 2011, Freddie Mac Memorandum, Re: Performing Loans Quality Control and Administration Audit
(#2011-010), at 1.



            Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                           30
board that Freddie Mac may be able to recover from                          more in future
                                                                            58
repurchase efforts through the use of a more expansive loan review process.

In addition, at the continued urging of the FHFA senior examiner, Freddie Mac management
initiated a more statistically rigorous “out-of-sample” test in February 2011. Management
agreed to sample approximately 1,000 “out-of-sample” Interest Only foreclosed loans originated
during the housing boom era to estimate potential recoveries if a broader loan review process
were employed. On August 31, 2011, Freddie Mac disclosed to FHFA the draft results from this
study, which indicate that at least 15% of the sample loans – a higher percentage than anticipated
by Freddie Mac management in connection with the Bank of America settlement – contain
apparent representation or warranty defects and therefore are subject to repurchase claim to loan
sellers.59 The figure may fall to the extent that loan sellers ultimately cure the defects identified
in some of these loans. Freddie Mac expects to receive final results from that review in about
three months.




58
  May 26, 2011, Freddie Mac Memorandum to Board of Directors, Re: Single-Family Quality Control Process, at 8.
On that day, the senior manager also informed the board that he believes Freddie Mac could lose from
             in new business were it to adopt a more aggressive loan review procedure. In other words, according to
Freddie Mac’s rationale and as a cost-benefit exercise, the senior manager now believes that after deducting those
possible losses from an estimated                         gain, a change in the loan review strategy would leave
Freddie Mac with $500 million to $1 billion in additional revenue.
59
     August 31, 2011, Freddie Mac Memorandum, Bank of America Settlement Loan Process Assumptions Review, at
6.



          Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                        31
FINDINGS
On the basis of the foregoing record, FHFA-OIG finds that:

   1. An FHFA Senior Examiner Raised Significant Concerns About Freddie
      Mac’s Loan Review Process for Mortgage Repurchase Claims

As early as June 2010, prior to the Bank of America settlement, an FHFA senior examiner began
to raise significant concerns about Freddie Mac’s loan review process. Specifically, he noted
that loans that Freddie Mac purchased that were originated during the housing boom defaulted at
higher than expected rates during the third through fifth years after origination. However,
Freddie Mac reviewed intensively only those loans that went into foreclosure or experienced
payment problems during the first and second years following origination. As a result, Freddie
Mac did not review over 300,000 loans for possible repurchase claims. According to the senior
examiner, this could be costing Freddie Mac “billions of dollars of losses.” These concerns
merited further review of the loan review process in 2010, which was not forthcoming. In
support of this finding, FHFA-OIG makes two initial observations.

          First, the concerns raised came from an FHFA senior examiner who had been
           reviewing Freddie Mac’s financial and operational soundness for an extended period
           and continues to do so. Similar concerns were later independently raised by Freddie
           Mac’s internal auditors.

          Second, the concerns relate to a significant risk (potentially involving substantial
           monetary losses) that is susceptible to recurrence in the event the Enterprise enters
           into future repurchase settlements.

FHFA-OIG further notes that the FHFA senior examiner’s concerns were consistent with
Enterprise data provided to FHFA, both before and after the Bank of America settlement.
Specifically, as shown at Figures 2, 3, and 4 above, data indicate a significant shift in the
mortgage default patterns on which the Enterprise’s traditional loan review process was
premised. That is, rather than foreclosures declining two years following their origination,
mortgages originated during the housing boom era showed increasing rates of foreclosure during
the third through fifth years after origination. In other words, the trend data upon which Freddie
Mac’s loan review process is premised appear to be at odds with actual foreclosure patterns
associated with the 2005 to 2007 vintage loans included in the settlement.

These trends could be unrelated to the higher incidence of mortgage origination defects that
might support repurchase claims if, for example, rising unemployment rates related to the


       Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                    32
lingering recession caused more borrowers to default on their prime loans and led to increased
home foreclosure rates. On the other hand, data demonstrate that many of the foreclosures of
loans originated during the housing boom era appear to involve non-traditional loans, which
appear to contain significant percentages of underwriting defects supporting repurchase claims.
In any event, FHFA did not test issues related to the senior examiner’s concerns prior to
approving the Freddie Mac-Bank of America settlement.

Freddie Mac’s internal auditors independently raised concerns in late 2010. In late November
and early December 2010, Freddie Mac’s internal auditors evaluated the Bank of America
settlement for reasons unrelated to the senior examiner’s actions, and, in connection with their
evaluation, they too raised questions about the loan review process.

   2. FHFA Did Not Timely Act on or Test the Ramifications of the Senior
      Examiner’s Concerns; Consequently, FHFA May Have Incorrectly
      Estimated the Risk of Loss to Freddie Mac Before Approving the Bank
      of America Settlement

FHFA, acting as the conservator of the Enterprises, has established a procedure under which it
reviews all Enterprise settlements of more than $50 million to ensure that they preserve and
conserve Enterprise assets and are in the best interests of taxpayers. FHFA-OIG finds that senior
FHFA management did not timely act on or test the ramifications of the FHFA senior examiner’s
concerns prior to approving the settlement, even though one dozen FHFA staffers, managers, and
senior managers were aware of the concerns over a six-month period, as detailed below and as
reflected in Appendix D. FHFA has advised FHFA-OIG that the senior examiner did not raise
his concerns in the context of the normal FHFA examination process. However, the record is
clear that his concerns were known to FHFA management and senior management well in
advance of the completion of the settlement. For example:

          The FHFA senior examiner repeatedly raised concerns about Freddie Mac’s loan
           review process with his direct supervisors (two managers who report to a senior
           manager) within DER in regular meetings throughout 2010. These direct supervisors
           did not follow up on or provide organizational support to substantiate these concerns.

          The FHFA senior examiner alerted two FHFA senior managers to the inaction of his
           direct supervisors.

          Two managers (a senior manager and a manager) acknowledged that they had
           reviewed the September 15, 2011, Analysis Memorandum, but they did not remember
           that the issue could potentially involve substantial losses to Freddie Mac.


       Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                    33
FHFA-OIG did not independently validate Freddie Mac’s existing loan review process and
therefore does not reach any final conclusion about it. Nevertheless, by relying on Freddie
Mac’s analysis of the settlement without testing the assumptions underlying Freddie Mac’s
existing loan review process, FHFA senior managers may have inaccurately estimated the risk of
loss to Freddie Mac. FHFA relied on a Freddie Mac management estimate that the Enterprise
was forgoing no more than                                   by continuing to employ its current loan
review process. That estimate was open to question because, among other reasons – and as
Freddie Mac’s internal auditors acknowledged, the                    projected loss, which was at the
low end of that estimate, left little if any cushion or margin of error, and the estimate itself was
based on an unrepresentative sample of loans.

   3. FHFA’s Decision to Suspend Approval of Additional Repurchase
      Settlements and Freddie Mac’s Continuing Efforts to Address the
      Concerns Are Positive Steps

After FHFA-OIG initiated this evaluation, FHFA suspended further Enterprise mortgage
repurchase settlements that are premised on Freddie Mac’s current loan review process. That is a
positive step, and it may help FHFA better assure that any future repurchase claim settlements
benefit the Enterprises and taxpayers.

In addition, since the close of the Bank of America settlement, Freddie Mac’s internal auditors
have continued to examine the matter and on June 6, 2011, issued an “Unsatisfactory” audit
opinion concerning the internal corporate governance controls involving the loan review process.
In response to that opinion, Freddie Mac management agreed to perform “out-of-sample” testing
of loans not currently reviewed for repurchase claims. Freddie Mac management commenced
such testing before the opinion was issued. In February 2011, at the urging of the FHFA senior
examiner, management agreed to review a sample of 1,000 Interest Only loans originated during
the housing boom that went into foreclosure more than two years after origination. The draft
results from that sample were disclosed to FHFA on August 31, 2011, and they revealed that at
least 15% of such loans – a higher percentage than anticipated by Freddie Mac management in
connection with the Bank of America settlement – include representations and warranties defects
and are subject to repurchase claims to loan sellers. However, the final repurchase rate may be
lower. Final results are expected in about three months.

Moreover, as discussed in footnote 58 and accompanying text, on May 26, 2011, a Freddie Mac
senior manager – who provided management estimates to the Freddie Mac board of directors in
late 2010 – advised the board of directors that the Enterprise could recover from $500 million to
$1 billion net in additional revenue through the use of a more expansive loan review process.



       Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                    34
CONCLUSIONS
FHFA-OIG encourages FHFA and Freddie Mac to continue their efforts to gauge the impact of
the default anomaly associated with housing boom loans and to take remedial actions to address
problems identified. This evaluation reveals a lack of independent action by FHFA senior
management, which may have led and could lead to significant losses by Freddie Mac. Had
FHFA senior management required testing of the concerns raised by an FHFA senior examiner,
FHFA may have been in a better position to evaluate Freddie Mac’s repurchase claim settlement
with Bank of America.

In the aftermath of the settlement, FHFA has suspended approving similar Enterprise repurchase
claim settlements pending further review. Moreover, Freddie Mac’s internal auditors continue to
assess the issue, and Freddie Mac management has agreed to actions to resolve the concerns.




RECOMMENDATIONS
FHFA-OIG makes two recommendations:

1. FHFA and its senior management must promptly act on the significant concerns raised
   about the loan review process.

To ensure that Freddie Mac is maximizing its repurchase claim recoveries:

          FHFA should continue to withhold approval of Freddie Mac repurchase settlements
           until such time as it is confident that the concerns about the Enterprise’s loan review
           process have been resolved.

          FHFA senior management should ensure that Freddie Mac management resolves the
           concerns that prompted their internal auditors to issue an “Unsatisfactory” audit
           opinion.

          FHFA senior management should oversee Freddie Mac’s “out-of-sample” loan
           testing and consider independently validating the testing.

          FHFA should evaluate whether Fannie Mae and Freddie Mac should adopt consistent
           review practices for repurchase claims.




      Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                   35
          FHFA senior management should initiate an independent assessment of Enterprise
           repurchase practices in order to ensure that they are maximizing their repurchase
           claim recoveries.

          FHFA should issue internal guidance regarding its handling of future repurchase
           settlements, should they arise.

2. FHFA must promptly initiate management reforms to ensure more generally that
   senior management is apprised of and timely acts on significant concerns brought to its
   attention.

FHFA senior management must immediately initiate reforms to avoid the kind of management
process shortcomings identified in this evaluation. In particular:

          Direct supervisors must properly and timely address and act upon significant
           concerns brought to their attention (i.e., resolve or elevate issues that pose significant
           potential risks or document decisions not to do so).

          Senior managers, regardless of their position within FHFA, must timely address and
           act on significant concerns, particularly when they receive reports that the normal
           reporting and supervisory process is not working properly.

FHFA’s Acting Director must establish appropriate goals, principles, and procedures at the top
of the FHFA organization to guarantee that significant concerns are properly and timely
addressed and acted upon.




      Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                   36
SCOPE AND METHODOLOGY
To conduct this evaluation FHFA-OIG staff requested and reviewed FHFA and Freddie Mac
documents, including e-mails associated with Freddie Mac’s settlement with Bank of America.
In addition, FHFA-OIG interviewed FHFA senior management and staff, as well as current and
former Freddie Mac senior managers.

FHFA-OIG reviewed HERA, FHFA regulations, and internal policies. FHFA-OIG also obtained
and reviewed publicly available data.

This evaluation was conducted under the authority of the Inspector General Act of 1978, as
amended, and in accordance with the Quality Standards for Inspection and Evaluation (January
2011), which have been promulgated by the Council of Inspectors General on Integrity and
Efficiency. These standards, which are generally adopted by federal agencies, require FHFA-
OIG to plan and perform evaluations so as to obtain evidence sufficient to provide reasonable
bases to support findings and conclusions.

The performance period for this evaluation was from January 1, 2011, to August 30, 2011.

FHFA-OIG provided the Acting Director and FHFA senior management with briefings on this
evaluation, as well as the opportunity to comment officially on the draft version of this report.

FHFA-OIG appreciates the efforts of FHFA and Freddie Mac management and staff in providing
the information necessary to complete this evaluation.




       Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                    37
APPENDIX A
FHFA Management Comments




    Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                 38
Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                             39
Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                             40
APPENDIX B
FHFA-OIG Responses to FHFA Management Comments

FHFA-OIG is pleased that FHFA has agreed to its recommendations and is already taking
actions to address them.

With respect to the first recommendation on the loan review process, although FHFA accepts it
in principle, it does not agree with each of the specific action steps outlined in the report. At the
same time, FHFA has not proposed a specific action plan of its own. Under the circumstances,
FHFA-OIG will continue to monitor the issues discussed in this report and the actions that FHFA
is taking.




       Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                    41
APPENDIX C
Timeline of Relevant Events




   March: FHFA senior examiner
   notices shifts in foreclosure patterns
   among 2005-2007 vintage home loans


                                                          June: FHFA examination staff discuss
                                                          shifts in foreclosure patterns with
   July: Citing resource constraints and                  Freddie Mac managers
   senior management opposition, Freddie
   Mac managers decline to review their
   methodology for selecting loans to
   examine for repurchase claims
                                                          September: FHFA senior manager
                                                          details concerns in a four-page memo
                                                          and circulates to FHFA managers and
                                                          senior managers
   December: Freddie Mac and Bank
   of America agree upon terms of
   repurchase settlement; Freddie Mac
   internal auditors raise concerns about                 December: Additional FHFA staff raise
   loan review process; in response,                      loan review process concerns; FHFA
   Freddie Mac management provides                        Acting Director approves settlement
   justification for existing process




   January: Settlement announced;
   FHFA-OIG begins review                                 June: Freddie Mac internal auditors
                                                          deliver opinion that the Enterprises’
                                                          corporate governance controls are
                                                          “Unsatisfactory” concerning the loan
                                                          review process




      Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                   42
APPENDIX D
Timeline of When FHFA Staff Were Alerted to Concerns61




61
  For the purpose of this timeline and evaluation, FHFA staffers and senior examiners report to managers; managers
report to senior managers; and senior managers report to the FHFA Acting Director.



        Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                       43
ADDITIONAL INFORMATION AND COPIES


For additional copies of this report:

          Call the Office of Inspector General (OIG) at: 202-408-2544

          Fax your request to: 202-445-2075

          Visit the OIG website at: www.fhfaoig.gov



To report alleged fraud, waste, abuse, mismanagement, or any other kind of criminal or
noncriminal misconduct relative to FHFA’s programs or operations:

          Call our Hotline at: 1-800-793-7724

          Fax the complaint directly to: 202-445-2075

          E-mail us at: oighotline@fhfa.gov

          Write to us at: FHFA Office of Inspector General
                      Attn: Office of Investigation – Hotline
                      1625 Eye Street, NW
                      Washington, DC 20006-4001




       Federal Housing Finance Agency Office of Inspector General • EVL-2011-006 • September 27, 2011
                                                    44