oversight

Audit of FHFA's Oversight of Fannie Mae's Compliance with the Required Risk Mitigants of Automated Underwriting, Mortgage Insurance, and Homeownership Education for its Purchases of Mortgages with a 97% LTV

Published by the Federal Housing Finance Agency, Office of Inspector General on 2018-03-08.

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

                            REDACTED

                    Federal Housing Finance Agency
                        Office of Inspector General



     Audit of FHFA’s Oversight of
   Fannie Mae’s Compliance with the
       Required Risk Mitigants of
   Automated Underwriting, Mortgage
    Insurance, and Homeownership
     Education for its Purchases of
      Mortgages with a 97% LTV




This report contains redactions of information that is privileged or confidential.


        Audit Report • AUD-2018-003 • March 8, 2018
                Executive Summary
                For more than 20 years, successive administrations agreed that a barrier to
                homeownership for low- and moderate-income people was a significant down
                payment, and they promoted solutions to reduce that barrier to increase
                accessibility to homeownership. Numerous studies have found that saving
                enough cash for a down payment and other up-front closing costs is the
AUD-2018-003    greatest barrier that low-income and minority families face when considering
                homeownership.
March 8, 2018
                The Housing and Economic Recovery Act of 2008 (HERA) established the
                Federal Housing Finance Agency (FHFA or Agency), an independent agency
                responsible for the supervision and regulation of Fannie Mae and Freddie Mac
                (collectively, the Enterprises) and the Federal Home Loan Banks. Using its
                powers under HERA, FHFA placed the Enterprises into conservatorships on
                September 6, 2008.

                As conservator, FHFA issued an expectation to the Enterprises in May 2014
                to “Work to increase access to mortgage credit for creditworthy borrowers,
                consistent with the full extent of applicable credit requirements and risk-
                management practices.” Later that year, in October 2014, the FHFA Director
                announced that FHFA was working with the Enterprises to develop sensible
                and responsible guidelines for mortgages with loan-to-value (LTV) ratios
                between 95% and 97% (high LTV mortgages) to increase access for
                creditworthy but lower-wealth borrowers.

                After reviewing proposals received from the Enterprises, FHFA staff prepared
                a memorandum in early December 2014 (Staff Memorandum) recommending
                that the FHFA Director approve the high LTV mortgage programs proposed
                by the Enterprises. The Staff Memorandum acknowledged that “historical
                performance demonstrates that higher LTV loans can have higher risks than
                lower LTV loans and can have higher loss severities,” but asserted that these
                higher risks can be safely offset by thoughtful compensating factors and risk
                mitigants, including automated underwriting, private mortgage insurance, and
                pre-purchase homeownership education. The Staff Memorandum identified an
                additional control: FHFA’s ongoing oversight of Enterprise purchases of high
                LTV mortgages. The FHFA Director accepted the staff recommendation and
                approved the programs on December 3, 2014.

                We performed this audit to assess FHFA’s oversight of Fannie Mae’s
                implementation of the 97% LTV mortgage program. As part of assessing
                FHFA’s oversight, we obtained (through FHFA) and analyzed Fannie Mae
                data on the mortgages purchased by Fannie Mae under the program and
                whether those mortgages conformed to three FHFA-required credit terms:
                (1) automated underwriting, (2) mortgage insurance, and (3) homeownership
                education.

                Based on our inquiries to FHFA and Fannie Mae, and our analysis of the data
                provided by the Enterprise, we found that Fannie Mae purchased 74,700
                mortgages from December 3, 2014, to December 31, 2016, under the 97%
                LTV mortgage program. Of those mortgages purchased, all were underwritten
AUD-2018-003    using an automated underwriting system and all but two loans utilized
                mortgage insurance or another credit enhancement. With regard to
March 8, 2018   homeownership education, which was required for only about a fourth of
                the 97% LTV mortgage purchases, we found that Fannie Mae relied on the
                lenders’ representations and warranties to determine whether this requirement
                was met. Fannie Mae quality control reviews of purchased loans found 20
                exceptions with homeownership education, representing 3% of loans sampled
                where homeownership education was a requirement. During our field work,
                Fannie Mae advised us that in March 2017, it implemented a “fatal” rule in
                its Loan Delivery system, requiring lenders to confirm that pre-purchase
                homeownership education has been completed, when required, or the
                mortgage will be rejected.

                Our audit also reviewed FHFA’s ongoing oversight of Fannie Mae purchases
                of high LTV mortgages. We found that FHFA engaged in the following
                oversight activities:

                   •   A briefing by FHFA’s Division of Housing Mission and Goals
                       (DHMG), which drafted the Staff Memorandum, to FHFA’s Division
                       of Enterprise Regulation (DER) on the programs’ parameters, to
                       facilitate DER’s ability to conduct supervisory activities.

                   •   DHMG-prepared periodic reporting based on Fannie Mae 97% LTV
                       mortgage data, such as average credit scores and debt-to-income (DTI)
                       ratios of borrowers, loan volume, and delinquency rates, which
                       according to DHMG, are used to assess whether the goals and
                       objectives of the 97% LTV mortgage programs are being met and to
                       flag potential concerns or issues for FHFA leadership.

                   •   Supervisory activities conducted by DER on Fannie Mae’s 97% LTV
                       mortgage program that consisted of an ongoing monitoring activity
                       during the 2016 examination cycle and a targeted examination during
                       the 2017 examination cycle.

                These oversight activities focused on Fannie Mae’s credit risk management
                and have not directly addressed compliance with the three risk mitigants that
                were the scope of this audit.
                We make no recommendations in this report.

                We are also issuing today the results of our audit of FHFA’s oversight of
                Freddie Mac’s 97% LTV mortgage program, which was also approved by
                FHFA’s December 2014 Staff Memorandum. See Audit of FHFA’s Oversight
                of Freddie Mac’s Compliance with the Required Risk Mitigants of Automated
                Underwriting, Mortgage Insurance, and Homeownership Education for its
AUD-2018-003    Purchases of Mortgages with a 97% LTV (AUD-2018-004) (available online
                at www.fhfaoig.gov/reports/auditsandevaluations).
March 8, 2018
                This report was prepared by: Tara Lewis, Audit Director; Pamela L. Williams,
                Auditor-in-Charge; Andrew Gegor, Auditor; and Terese Blanchard, Auditor;
                with the assistance of Bob Taylor, Assistant Inspector General for Audits. We
                appreciate the cooperation of FHFA staff, as well as the assistance of all those
                who contributed to the preparation of this report.

                This report has been distributed to Congress, the Office of Management and
                Budget, and others and will be posted on our website, www.fhfaoig.gov.

                Marla A. Freedman, Deputy Inspector General for Audits /s/
TABLE OF CONTENTS ................................................................
EXECUTIVE SUMMARY .............................................................................................................2

ABBREVIATIONS .........................................................................................................................7

BACKGROUND .............................................................................................................................8
      Significant Down Payments: An Acknowledged Barrier to Affordable
      Homeownership ........................................................................................................................8
      Historic Efforts by the Enterprises to Overcome the Down Payment Obstacle .......................9
      FHFA’s May 2014 Direction to the Enterprises: Increase Access to Mortgage Credit .........10
      FHFA’s Approval of Fannie Mae’s Proposed 97% LTV Mortgage Program........................12
      Subsequent Policy Changes to Fannie Mae’s 97% LTV Mortgage Program ........................15
             Fannie Mae’s Request for Policy Change to Redesign MCM ........................................16
      Subsequent Variances Reported by Fannie Mae to its 97% LTV Mortgage Program ...........17

FACTS AND ANALYSIS.............................................................................................................18
      The Required Risk Mitigants of Automated Underwriting and Mortgage Insurance
      Were Utilized, With Few Exceptions Reported, in Connection with Fannie Mae’s
      Purchases of 97% LTV Mortgages and, for the Risk Mitigant of Homeownership
      Education, Fannie Mae Relied on the Lenders’ Representations and Warranties that
      the Credit Term Was Met .......................................................................................................18
             Data from Fannie Mae Showed that All 74,700 of the 97% LTV Mortgages
             Were Underwritten Using an Automated Underwriting System ....................................19
             Data from Fannie Mae Showed that All But 2 of the 74,700 Mortgages
             Contained Mortgage Insurance or Another Credit Enhancement ...................................19
             Fannie Mae Relied on the Lenders’ Representations and Warranties that the
             Pre-Purchase Homeownership Education Requirement Was Met..................................20
      FHFA’s Oversight of Fannie Mae’s 97% LTV Mortgage Program .......................................21

CONCLUSION ..............................................................................................................................23

FHFA COMMENTS AND OIG RESPONSE ...............................................................................24

OBJECTIVE, SCOPE, AND METHODOLOGY .........................................................................25

APPENDIX: FHFA MANAGEMENT RESPONSE ....................................................................27


                                           OIG • AUD-2018-003 • March 8, 2018                                                              5
ADDITIONAL INFORMATION AND COPIES .........................................................................28




                                 OIG • AUD-2018-003 • March 8, 2018                                        6
ABBREVIATIONS .......................................................................

DER                   Division of Enterprise Regulation

DHMG                  Division of Housing Mission and Goals

DTI                   Debt-to-Income

DU                    Desktop Underwriter

Enterprises           Fannie Mae and Freddie Mac

Fannie Mae            Federal National Mortgage Association

FHFA or Agency        Federal Housing Finance Agency

Freddie Mac           Federal Home Loan Mortgage Corporation

HERA                  Housing and Economic Recovery Act of 2008

HUD                   U.S. Department of Housing and Urban Development

LTV                   Loan-to-Value

MCM                   MyCommunityMortgage

MIs                   Mortgage Insurance Providers

OIG                   Federal Housing Finance Agency Office of Inspector General

Staff Memorandum      Federal Housing Finance Agency Staff Analysis Memorandum

Strategy              The National Homeownership Strategy: Partners in the American
                      Dream

UMDP                  Uniform Mortgage Data Program

UPB                   Unpaid Principal Balance




                          OIG • AUD-2018-003 • March 8, 2018                          7
BACKGROUND ..........................................................................

Significant Down Payments: An Acknowledged Barrier to Affordable Homeownership

Congress passed the Federal Housing Enterprises Financial Safety and Soundness Act of
1992, which President George H.W. Bush signed into law on October 28, 1992. In that
statute, Congress defined the mission of the Enterprises and amended the Enterprise charters
to impose “an affirmative obligation to facilitate the financing of affordable housing for low-
and moderate-income families in a manner consistent with their overall public purposes, while
maintaining a strong financial condition and a reasonable economic return.” 1 At the same
time, Congress directed the U.S. Department of Housing and Urban Development (HUD) to
set and enforce affordable housing goals for the Enterprises to ensure that they met their
affirmative obligation. 2

For more than 20 years, successive administrations agreed that a barrier to homeownership for
low- and moderate-income people was a significant down payment, and they promoted
solutions to reduce that barrier to increase accessibility to homeownership. In 1994, President
Clinton directed the then-HUD Secretary to develop, with leaders of the housing industry,
nonprofit organizations, and government leaders, a strategy to increase homeownership. The
resulting report, The National Homeownership Strategy: Partners in the American Dream
(Strategy), recognized that an impediment to homeownership was that “[l]ow- and moderate-
income families often cannot become homeowners because they are unable to come up with
the required downpayment and closing costs. In many instances, these prospective first-time
homebuyers find that developing the proper savings patterns to accumulate sufficient cash for
the downpayment is difficult.” 3 The Strategy identified specific actions to be taken to lower
barriers to homeownership, including working collaboratively to reduce homebuyer down
payment requirements. 4 Eight years later, in October 2002, President George W. Bush
recognized that barriers to access persisted. In signing into law the American Dream
Downpayment Act the following year, 5 he recognized that “[o]ne of the biggest hurdles to
homeownership is getting money for a down payment” and explained that the statute would

1
    12 U.S.C. § 4501.
2
    12 U.S.C. § 4561(a); 12 U.S.C. § 4566(a)(1).
3
  U.S. Department of Housing and Urban Development, The National Homeownership Strategy: Partners in
the American Dream, at 4-4 (May 1995) (online at
www.globalurban.org/National_Homeownership_Strategy.pdf).
4
    Id. at 4-5.
5
    42 U.S.C. § 12701.




                                     OIG • AUD-2018-003 • March 8, 2018                                8
“help many low-income buyers to overcome that hurdle, and to achieve an important part of
the American Dream.” 6 That statute authorized $200 million per year in down payment
assistance to at least 40,000 low-income families.

Numerous studies over time have found that “saving enough cash for a down payment and for
up-front closing costs is the greatest barrier that low-income and minority families face when
considering homeownership.” 7

Historic Efforts by the Enterprises to Overcome the Down Payment Obstacle

The Enterprises’ amended charters authorize them to purchase a non-federally insured
mortgage for more than 80% of a property’s value, provided that the mortgage includes a
form of credit enhancement. The charters identify three specific forms of credit
enhancements:

      •    The seller retains a participation of not less than 10% in the mortgage;

      •    The seller agrees to repurchase or replace a defaulted mortgage upon demand by an
           Enterprise, for such period and under such circumstances as the Enterprise may
           require; or

      •    The portion of the mortgage amount in excess of 80% LTV is guaranteed or insured
           by a qualified insurer as determined by each Enterprise. 8

One of a number of initiatives implemented by the Enterprises to meet their affordable
housing goals and to reduce the recognized down payment barriers was high LTV mortgage
programs for more than 80% of a property’s value, with credit enhancements. Fannie Mae
announced a pilot program in 1994 to buy conventional home mortgages with only a 3%




6
  Press Release, President Bush Signs American Dream Downpayment Act of 2003 (Dec. 16, 2003) (online at
http://georgewbush-whitehouse.archives.gov/news/releases/2003/12/20031216-9.html).
7
  U.S. Department of Housing and Urban Development, HUD’s Housing Goals for the Federal National
Mortgage Association (Fannie Mae) and the Federal Home Loan Mortgage Corporation (Freddie Mac) for
the Years 2005-2008 and Amendments to HUD’s Regulation of Fannie Mae and Freddie Mac, 69 Fed. Reg.
63655 (at text accompanying note 105) (Nov. 2, 2004) (online at www.gpo.gov/fdsys/pkg/FR-2004-11-
02/pdf/FR-2004-11-02.pdf.
8
    12 U.S.C. § 1454(a)(2); 12 U.S.C. § 1717(b)(2).




                                     OIG • AUD-2018-003 • March 8, 2018                                   9
down payment 9 and that program expanded over time. 10 Freddie Mac introduced a 97% LTV
mortgage program in 1998.

The LTV ratio is defined by Fannie Mae as the ratio between the original amount of the first
mortgage and the lower of the property’s appraised value or sales price. For instance, if a
borrower seeks to purchase a house worth $150,000 and is only able to put down $4,500, the
borrower will need a mortgage of $145,500. The LTV ratio for that mortgage is calculated at
97% ($145,500/$150,000).

FHFA’s May 2014 Direction to the Enterprises: Increase Access to Mortgage Credit

HERA established FHFA, an independent agency responsible for the supervision and
regulation of the Enterprises and the Federal Home Loan Banks. 11 Using its powers under
HERA, FHFA placed the Enterprises into conservatorships on September 6, 2008. As
conservator, FHFA’s stated goal is to “[h]elp restore confidence, enhance capacity to fulfill
mission, and mitigate systemic risk that contributed directly to instability in financial
markets.” 12

Beginning in 2012, FHFA has developed and published formal strategic plans that establish
strategic goals for the Enterprises.

FHFA has also issued annual conservatorship scorecards to set specific expectations for
each strategic plan goal, which enable FHFA and the Enterprises to track progress toward
achieving the goals. Each annual scorecard maps to the strategic plan in place at the time the




9
  Fannie Mae, Information Statement, at 5 (Mar. 31, 1995) (online at
www.fanniemae.com/resources/file/ir/pdf/info-statements/infostmtmar1995.pdf; see also J. Linn Allen,
Lenders Try 3% Home Down Payments: 750,000 Could Qualify for Loans, Chicago Tribune (Feb. 17, 1994)
(online at http://articles.chicagotribune.com/1994-02-17/business/9402170318_1_mortgage-insurance-
gecapital-fannie-mae).
10
   U.S. Department of Housing and Urban Development, HUD’s Housing Goals for the Federal National
Mortgage Association (Fannie Mae) and the Federal Home Loan Mortgage Corporation (Freddie Mac) for
the Years 2005-2008 and Amendments to HUD’s Regulation of Fannie Mae and Freddie Mac; 69 Fed. Reg.
63655 and 63743 (Nov. 2, 2004) (online at www.gpo.gov/fdsys/pkg/FR-2004-11-02/pdf/FR-2004-11-02.pdf).
11
     12 U.S.C. § 4511(a).
12
   Congress vested FHFA with sweeping powers as conservator: FHFA possesses all rights and powers of any
stockholder, officer, or director of the Enterprises. See 12 U.S.C. § 4617(b)(2)(A)(i). For reasons of efficiency,
concordant goals with the Enterprises, and operational savings, FHFA has determined to (1) delegate authority
for general corporate governance and day-to-day matters to the Enterprises’ boards of directors and executive
management and (2) retain authority for certain significant decisions. FHFA requires the Enterprises to consult
with and obtain approval from FHFA, as conservator, on critical matters.




                                     OIG • AUD-2018-003 • March 8, 2018                                              10
scorecard is issued and describes the activities that further each strategic goal. 13 In May 2014,
FHFA issued its 2014 scorecard to the Enterprises. FHFA identified four expectations under
the strategic plan goal titled “Maintain in a safe and sound manner, foreclosure prevention
activities and credit availability for new and refinanced mortgages to foster liquid, efficient,
competitive and resilient national housing finance markets.” One of those expectations was to
“Work to increase access to mortgage credit for creditworthy borrowers, consistent with the
full extent of applicable credit requirements and risk-management practices.” In June 2014,
FHFA issued guidance to the Enterprises to meet this expectation:

     •   Complete an analysis of existing or historical offering of mortgages with a 97% LTV
         ratio limit;

     •   Survey market landscape to identify market and opportunity for 97% LTV lending;

     •   Submit written proposals to offer 97% LTV ratio, first-time homebuyer programs.
         Such offerings must include appropriate risk controls and risk management practices,
         guidelines to ensure approval of creditworthy borrowers, mitigating factors that offset
         incremental risks to lower down payment (e.g., pricing, mortgage insurance, etc.), any
         changes needed for combined loan-to-value standards, and any other factors that may
         help create a sustainable mortgage product; and

     •   Draft conceptual offerings with term sheet.

In October 2014, the FHFA Director announced that FHFA was working with the Enterprises
to develop sensible and responsible guidelines for mortgages with LTV ratios between 95%
and 97% to increase access for creditworthy but lower-wealth borrowers. 14 In written
testimony presented to the Senate Banking Committee on November 19, 2014, the FHFA
Director explained the reasons for the low-down payment initiative:

         Part of the Enterprises’ mission is promoting access to mortgage credit for
         creditworthy borrowers across all market segments. We know that in today’s
         market, there are creditworthy borrowers who have the income to afford monthly
         mortgage payments but do not have the money to make a large down payment and

13
   The 2014 Scorecard reflected the change in the conservator’s goals and priorities outlined in the 2014
Strategic Plan. See FHFA, 2014 Scorecard for Fannie Mae, Freddie Mac and Common Securitization
Solutions (May 13, 2014) (online at
www.fhfa.gov/AboutUs/Reports/ReportDocuments/2014Scorecard051314FINAL.pdf).
14
  Melvin L. Watt, FHFA Director, Prepared Remarks at the Mortgage Bankers Association Annual
Convention, Las Vegas, Nevada (Oct. 20, 2014) (online at www.fhfa.gov/Media/PublicAffairs/Pages/Prepared-
Remarks-of-Melvin-L-Watt,-Director,-Federal-Housing-Finance-Agency-at-the-MBA-Annual-
Convention.aspx).




                                    OIG • AUD-2018-003 • March 8, 2018                                      11
           pay closing costs. As a result, Fannie Mae and Freddie Mac will shortly announce
           purchase guidelines that allow for 3 to 5 percent down payments, which will
           improve opportunities for access to credit for some of these borrowers. 15

FHFA’s Approval of Fannie Mae’s Proposed 97% LTV Mortgage Program

Pursuant to FHFA’s June 2014 guidance, Fannie Mae submitted its proposal to purchase
mortgages with an LTV up to 97% to FHFA on September 30, 2014, and, based on feedback
received from FHFA, submitted a revised proposal on November 5, 2014. FHFA staff
reviewed Fannie Mae’s proposal and prepared a Staff Memorandum in early December 2014
recommending that the FHFA Director approve it. The Staff Memorandum acknowledged
that “historical performance demonstrates that higher LTV loans can have higher risks than
lower LTV loans and can have higher loss severities,” but asserted that these higher risks can
be safely offset by thoughtful compensating factors and risk mitigants. 16 According to that
Staff Memorandum, those controls were:

      •    Strong borrower eligibility requirements and low volume;

      •    Automated underwriting decisions with maximum or minimum parameters for credit
           terms such as debt-to-income (DTI) ratios, reserves, and credit scores;

      •    Private mortgage insurance;

      •    Homebuyer education and counseling (hereafter homeownership education); and

      •    FHFA supervision of Enterprise implementation of approved high LTV programs.

The Staff Memorandum stated that high LTV programs with these controls would provide, in
a safe and sound manner, “access to credit and homeownership opportunities for creditworthy
borrowers who have sufficient income and an ability to pay a mortgage but lack the family
and household wealth to put down a large down payment and pay closing costs.” 17



15
  Senate Committee on Banking, Housing, and Urban Affairs, Statement of Melvin L. Watt, FHFA Director,
An Update from the Federal Housing Finance Agency on Fannie Mae, Freddie Mac, and the Federal Home
Loan Banks (Nov. 19, 2014), (online at www.fhfa.gov/Media/PublicAffairs/Pages/Statement-of-Melvin-L-
Watt-Before-the-US-Senate-Banking-Committee-11-19-2014.aspx).
16
     The Staff Memorandum also emphasizes the relative small scale of the programs.
17
   In FHFA’s Progress Report on its 2014 Scorecard, the Agency noted that the Enterprises’ purchase
guidelines emphasize strong underwriting standards and do not allow the kind of risk layering that occurred in
the years leading up to the housing crisis. See FHFA, Progress Report on the Implementation of FHFA’s
Strategic Plan for the Conservatorships of Fannie Mae and Freddie Mac, at 7 (Mar. 16, 2015) (online at
www.fhfa.gov/AboutUs/Reports/ReportDocuments/SPEC2014ProgressReport3162015.pdf).



                                     OIG • AUD-2018-003 • March 8, 2018                                          12
The Staff Memorandum explained each of these controls.

Strong Borrower Eligibility Requirements and Low Volume. The Staff Memorandum
explained that Fannie Mae’s proposal offered three 97% LTV mortgage products: an
affordable lending product limited to first-time homebuyers through its
MyCommunityMortgage (MCM) program; a “Standard 97” (non-MCM) product limited to
first-time homebuyers; and limited cash-out refinances of existing Fannie Mae mortgages up
to 97% LTV. It explained that the volume of 97% LTV mortgages proposed by Fannie Mae
(and Freddie Mac) represented “a small portion of their annual single-family 30 year flow
business (approximately 1.0% and 1.5%) and an even smaller proportion of their combined
credit guarantee books of business (approximately 0.2% to 0.3%)” based upon “FHFA’s
assumption of $1 trillion in combined 30-year flow over two years ($400 billion and $600
billion for Freddie Mac and Fannie Mae[,] respectively) and $4.5 trillion in existing combined
total guaranteed portfolios.”

The Staff Memorandum reported that Fannie Mae estimated its volume for purchases of 97%
LTV mortgages and refinancings in 2015 to be $5.6 billion but provided no projection after
2015. It explained that there were several factors contributing to Fannie Mae’s relatively low
projected volume in 2015, including roughly three quarters of purchasing activity in 2015 and
its limitation on purchases of 97% LTV mortgages to first-time homebuyers.

Automated Underwriting. The Staff Memorandum endorsed Fannie Mae’s proposal to only
purchase 97% LTV mortgages processed through its automated underwriting system, Desktop
Underwriter (DU), with a minimum credit score of 620 and a maximum DTI index of 45%.
According to the memorandum, DU evaluates components of borrower risk profiles such as
credit history, delinquent accounts, borrower equity, liquid reserves, DTI, and LTV.
Additionally, the DU evaluation considers risk layering; a borrower would be ineligible for a
high LTV mortgage if multiple risk factors were present or layered on top of another. 18

Private Mortgage Insurance. The Enterprises’ charters require mortgage-level credit
enhancement for residential mortgages where the unpaid principal balance (UPB) exceeds
80%. Mortgage insurance is a “credit enhancement” that provides first loss protection to the
Enterprises. 19 According to the Promontory Financial Group, LLC, the “vast majority of loans
over 80% LTV purchased” by the Enterprises use private mortgage insurance as the credit


18
   Fannie Mae defines risk layering as combining multiple higher risk elements on a single mortgage, such as
a low credit score with a high debt ratio. According to FHFA, in the period prior to pre-2009, underwriting
standards became progressively less rigorous resulting in significant risk layering, which led to significantly
increased risk for high LTV programs instead of mitigating that risk.
19
     12 U.S.C. § 1454(a)(2); 12 U.S.C. § 1717(b)(2).




                                     OIG • AUD-2018-003 • March 8, 2018                                           13
enhancement. 20 The Staff Memorandum reported that the final proposals from each Enterprise
required private mortgage insurance for the 97% mortgages that they purchased, as reflected
on the final Term Sheets submitted by the Enterprises. It stated that implementation of
eligibility requirements for Mortgage Insurance providers (MIs) “should reduce the
counterparty risk of the MIs and ensure the availability of the credit enhancement. Further, as
an industry, the MIs are currently stronger than they have been in the past.”

Pre-Purchase Homeownership Education. The Staff Memorandum reported that research
indicated that pre-purchase homeownership education can improve borrower performance and
that “housing counseling and education for first-time homebuyers participating in the
programs [will] serve as an important risk mitigant for the proposed 97% LTV products.”
Pursuant to the Staff Memorandum, “[t]he lender must retain a record of the independent
counseling in the loan file” but the Staff Memorandum does not require the Enterprise to
obtain and retain records of such counseling. 21

Fannie Mae’s final proposed Term Sheet, which was included in the Staff Memorandum,
proposed to require pre-purchase homeownership education only for first-time homebuyers in
the MCM program, not for first-time homebuyers with loans delivered through the
Standard 97 (non-MCM) program or for borrowers refinancing an existing Fannie Mae
mortgage. The Staff Memorandum, which recommends approval of Fannie Mae’s proposal,
provides no explanation for the reasons why the “important risk mitigant” of pre-purchase
homeownership education was not required for all first-time homebuyers.

FHFA Supervision. The Staff Memorandum identified one additional control not proposed
by the Enterprises: FHFA’s ongoing oversight of Enterprise purchases of mortgages with 97%
LTVs. It explained:

         FHFA’s ongoing monitoring of the implementation and performance of
         Enterprise initiatives, in addition to Enterprise quality control findings, is an
         important oversight control. The Enterprises will provide regular reports to FHFA
         on loan delivery volumes, loan performance, and average credit parameters. In
         addition to serving as a monitoring tool, these reports will help FHFA develop
         future policy adjustments, as needed.




20
  Promontory Financial Group LLC, The Role of Private Mortgage Insurance in the U.S. Housing Finance
System, at 8 (Jan. 2011) (online at www.usmi.org/wp-content/uploads/2014/11/Promontory-study-I-Role-of-
PMI.pdf).
21
   For programs requiring homeownership education, Fannie Mae’s Selling Guide requires that such education
be completed prior to closing, and retention of documentation, such as a certificate of course completion, be in
the loan file.



                                    OIG • AUD-2018-003 • March 8, 2018                                             14
The Staff Memorandum did not define the timing or contents of the Enterprises’ required
reports.

According to the Staff Memorandum, FHFA would review information in these regular
reports and “take steps as appropriate as part of the agency’s ongoing oversight and
conservatorship responsibilities.” It recommended that DHMG, which drafted the Staff
Memorandum, brief DER on the programs’ parameters, to facilitate DER’s ability to conduct
appropriate supervisory activities.

                                             *****

The Staff Memorandum opined that 97% LTV mortgage programs with these controls would
provide, in a safe and sound manner, “access to credit and homeownership opportunities for
creditworthy borrowers who have sufficient income and an ability to pay a mortgage but lack
the family and household wealth to put down a large down payment and pay closing costs”
and recommended approval of the Enterprises’ 97% LTV mortgage programs with these
controls. The FHFA Director accepted the staff recommendation and approved the programs
on December 3, 2014. 22

On December 8, 2014, the FHFA Director released a written statement that the Enterprises’
97% LTV mortgage programs provide a responsible approach to improving access to credit
for creditworthy borrowers who can afford a mortgage but lack the resources to pay a
substantial down payment plus closing costs, while also ensuring safe and sound lending
practices.

Subsequent Policy Changes to Fannie Mae’s 97% LTV Mortgage Program

Where the Enterprises seek significant or material changes to existing selling and servicing
guides, they must follow a process established by FHFA to provide notice of the proposed
change so that FHFA has sufficient time to analyze the proposal. Pursuant to this process, an
Enterprise seeking a significant/material policy change submits its proposal to FHFA, and the
Office of Housing and Regulatory Policy, an office within DHMG: researches the policy
topic; coordinates with subject matter experts, as needed; and when its review is complete,
informs the Enterprise of the policy determination (i.e., no further questions or comments,
objection to the implementation of the policy and the grounds, or key comments or concerns.)




22
  The Staff Memorandum also discussed Freddie Mac’s proposed 97% LTV mortgage program, which the
FHFA Director approved on the same date.



                                OIG • AUD-2018-003 • March 8, 2018                                 15
      Fannie Mae’s Request for Policy Change to Redesign MCM

On July 22, 2015, Fannie Mae submitted its proposal to change its MCM program to FHFA.
According to Fannie Mae’s notice of proposed change, the MCM program had been its
affordable lending product since 2001, but that over time, most of the underwriting and
eligibility flexibilities had been removed or transitioned into standard policy. To support
access to credit for creditworthy borrowers, Fannie Mae proposed to redesign MCM with a
new name and change its underwriting and eligibility guidelines, pricing, and homeownership
education and counseling requirements. These proposed changes would modify existing
underwriting criteria for 97% LTV MCM mortgages in the Director-approved December 2014
Staff Memorandum.

As discussed earlier, Fannie Mae’s final proposed Term Sheet for its 97% LTV mortgage
program, which was included in the Staff Memorandum, required pre-purchase
homeownership education only for first-time homebuyers in the MCM program. Its July 2015
notice proposed, in part, two changes: (1) to eliminate the first-time homebuyer requirement
for the MCM redesigned program; and (2) to require pre-purchase homeownership education
for at least one borrower in all MCM redesign transactions. Fannie Mae proposed that its
policy changes would become effective in December 2015.

DHMG staff reviewed Fannie Mae’s proposal and prepared a “Review Memo” dated
August 18, 2015, in which it analyzed the proposed changes and their impact and
recommended that no objection be interposed to the proposed policy changes. The following
day, the FHFA Director and FHFA’s Acting Deputy Director, Division of Conservatorship,
met with Fannie Mae’s Chief Executive Officer and other Fannie Mae executives to review
the proposed policy changes. Fannie Mae representatives briefed the Director on the proposed
policy changes using a PowerPoint entitled “Access to Credit and Affordable Housing,”
which explained proposed changes to underwriting criteria. On August 20, 2015, DHMG
informed Fannie Mae that FHFA had no objections to the policy changes sought in its MCM
redesign request. 23




23
     On September 29, 2015, Fannie Mae announced the rebranded MCM program as HomeReady.




                                  OIG • AUD-2018-003 • March 8, 2018                           16
Subsequent Variances Reported by Fannie Mae to its 97% LTV Mortgage Program

Variances are negotiated terms with individual lenders that are exceptions to the Enterprises’
selling and servicing guides contained within each individual lender’s Master Agreement. 24

For variances within individual Master Agreements, the Enterprises are not required to follow
the same process they must follow for proposed significant or material policy changes to
selling and servicing guides. Specifically, FHFA requires the Enterprises to submit written
reports on a regular basis that show their variance activities. 25 To implement this requirement,
Fannie Mae submits monthly management reports to FHFA detailing variances within
individual Master Agreements.

In its June 2016 monthly variance report to FHFA, Fannie Mae included a variance that
allowed one lender, in partnership with a nonprofit organization, to deliver 97% LTV
mortgages to Fannie Mae with a repurchase obligation that replaced traditional mortgage
insurance. 26 In addition, this variance provided that the nonprofit organization would serve as
the credit guarantor for a portion of the credit losses, thereby acting as an additional credit
enhancement.




24
   A Master Agreement is an “umbrella” document that supplements the general guidelines and requirements
of the Fannie Mae Selling Guide and Servicing Guide and sets forth the additional terms under which Fannie
Mae does business with lenders. Each lender must have a Master Agreement for mortgage-backed securities
and for any loans originated or delivered under negotiated terms.
25
   For a discussion of FHFA’s review of variance activities over the period January 2014 through March 2015,
see OIG, Compliance Review of FHFA’s Implementation of Its Procedures for Overseeing the Enterprises’
Single-Family Mortgage Underwriting Standards and Variances (Dec. 17, 2015) (COM-2016-001) (online at
www.fhfaoig.gov/Content/Files/COM-2016-001_1.pdf).
26
   According to the terms of the variance, Fannie Mae has the right to require the non-profit organization to
repurchase loans under certain circumstances.



                                    OIG • AUD-2018-003 • March 8, 2018                                          17
FACTS AND ANALYSIS ...............................................................

We performed this audit to assess FHFA’s oversight of Fannie Mae’s implementation of the
97% LTV mortgage program. As part of assessing FHFA’s oversight, we obtained, through
FHFA, and analyzed Fannie Mae data on the mortgages purchased by Fannie Mae under the
97% LTV mortgage program approved by FHFA’s December 2014 Staff Memorandum and
whether those mortgages conformed to three FHFA-required credit terms: (1) automated
underwriting; (2) mortgage insurance; and (3) homeownership education. There were other risk
mitigants associated with Fannie Mae’s purchases of high LTV mortgages that were not
included within the scope of this audit, such as maximum or minimum parameters for credit
terms like DTI ratios and credit scores.

In conducting this audit, we obtained and reviewed data reported to Fannie Mae under the
Uniform Mortgage Data Program (UMDP). Established jointly by the Enterprises at the
direction of FHFA, the UMDP requires lenders to report a common set of data elements for
single-family mortgages purchased by the Enterprises. The UMDP was intended to provide
the Enterprises with consistent lender mortgage data and facilitate their ability to reduce risk
(and decrease costs) by focusing on the quality of the mortgage prior to its purchase. For
Fannie Mae, lenders report these data elements through Fannie Mae’s web-based Loan
Delivery system. These data elements include, but are not limited to, automated underwriting,
mortgage insurance, and homeownership education.

The Required Risk Mitigants of Automated Underwriting and Mortgage Insurance
Were Utilized, With Few Exceptions Reported, in Connection with Fannie Mae’s
Purchases of 97% LTV Mortgages and, for the Risk Mitigant of Homeownership
Education, Fannie Mae Relied on the Lenders’ Representations and Warranties that
the Credit Term Was Met

Our audit covered the high LTV mortgages purchased by Fannie Mae from December 3,
2014, to December 31, 2016, (review period) under the program approved by the Director in
the Staff Memorandum, as revised in 2015. 27 In response to our inquiries, Fannie Mae
reported that it purchased 74,700 single-family 97% LTV mortgages with a UPB of




27
  Fannie Mae reported that the first loan delivery for loans included in the scope of this audit was
December 30, 2014.




                                    OIG • AUD-2018-003 • March 8, 2018                                 18
$14.1 billion during this review period. 28 For this universe of mortgages, we asked Fannie
Mae to provide data on the automated underwriting, mortgage insurance, and pre-purchase
homeownership education.

     Data from Fannie Mae Showed that All 74,700 of the 97% LTV Mortgages Were
     Underwritten Using an Automated Underwriting System

Based on our analysis of the data provided by Fannie Mae, we found that all 74,700 of the
97% LTV mortgages purchased by it were underwritten using an automated underwriting
system. 29

Fannie Mae initially reported that its Loan Delivery system showed 5 of the 74,700 mortgages
(0.007%) were underwritten using a proprietary automated underwriting system other than
DU or Loan Product Advisor. Although proprietary automated underwriting was not
permitted under the terms of FHFA’s approval of Fannie Mae’s 97% LTV mortgage program,
Fannie Mae’s Loan Delivery system did not identify these mortgages as ineligible or
otherwise prevent their purchase by Fannie Mae.

In response to our follow-up requests, Fannie Mae reported to us that the lenders for these
5 mortgages represented to Fannie Mae that each of the 5 were underwritten using DU but
had incorrectly recorded a different method of underwriting in the Loan Delivery system.
Fannie Mae also stated to us that, as of November 2016, its Loan Delivery system no longer
allows a lender to report use of a proprietary underwriting system in connection with 97%
LTV mortgages, unless there is a specific lender variance in place at the time of purchase. 30

     Data from Fannie Mae Showed that All But 2 of the 74,700 Mortgages Contained
     Mortgage Insurance or Another Credit Enhancement

Based on our analysis of the data provided by Fannie Mae, of the 74,700 mortgages, we found
that 2 of the 97% LTV mortgages purchased (0.003%) lacked information from the lender
28
   In addition to the 74,700 97% LTV mortgages purchased under the program approved by the Staff
Memorandum that are the scope of this audit, Fannie Mae purchased 84,226 mortgages with LTVs between
95% and 97% and a combined UPB of $14.6 billion between December 3, 2014, and December 31, 2016,
delivered under continuing programs offered by Fannie Mae prior to conservatorship. The mortgage purchases
through these continuing programs are not included in the scope of this audit.
29
  Fannie Mae reported that, during the scope of our audit, lenders delivered 74,674 of the 74,700 97% LTV
mortgages underwritten using DU. Eight lenders delivered, through variances (i.e., negotiated terms), the
remaining 26 loans underwritten using Freddie Mac’s automated underwriting system, Loan Product Advisor.
30
   Fannie Mae also reported that for another 47 mortgages, the lender had recorded in the Loan Delivery
system that both DU and Loan Product Advisor were used to underwrite that loan. In following up on our
inquiry about these loans, Fannie Mae reported that the lenders had incorrectly recorded both underwriting
methods in the Loan Delivery system; for these 47 mortgages, the lenders represented to Fannie Mae that DU
was used, not both.



                                   OIG • AUD-2018-003 • March 8, 2018                                        19
about required mortgage insurance or another credit enhancement. For these two mortgages,
Fannie Mae subsequently reported that one was paid in full and one was repurchased by the
lender.

As part of determining whether the 97% LTV mortgages included mortgage insurance or
another credit enhancement as required, we found that Fannie Mae’s Loan Delivery controls
did not prevent mortgages from being purchased where the lender did not provide information
about mortgage insurance or another credit enhancement. In addition to the two exceptions
explained above, Fannie Mae officials found, in the course of responding to our requests, that
lenders for 24 mortgages did not provide mortgage insurance information upon delivery. For
these mortgages, Fannie Mae contacted the lenders to obtain the mortgage insurance
information. We compared the date Fannie Mae purchased the mortgages to the date the
lender provided the mortgage insurance information and found that the lag in receipt of the
information was up to 28 months.

During our field work, Fannie Mae reported that it implemented a “fatal” rule in its Loan
Delivery system for any loans in which the lender did not provide information about mortgage
insurance or another credit enhancement. This “fatal” rule means that, at the discretion of
Fannie Mae, a mortgage may be rejected if the lender does not provide the mortgage
insurance or other credit enhancement information.

     Fannie Mae Relied on the Lenders’ Representations and Warranties that the Pre-
     Purchase Homeownership Education Requirement Was Met

As discussed earlier, Fannie Mae’s final proposed Term Sheet required homeownership
education for only one category of 97% LTV mortgages – first-time homebuyers in the MCM
program – which the FHFA Director approved. In July 2015, Fannie Mae proposed to expand
the homeownership education requirement to at least one borrower in all MCM redesign
transactions, to which FHFA did not object. Of the 74,700 mortgages, Fannie Mae reported
that 18,769 (25%) of the 97% LTV mortgages purchased were subject to the pre-purchase
homeownership education requirement.

Fannie Mae reported to us that its Loan Delivery system contained fields for lenders to report
at the time of loan delivery: (a) whether pre-purchase homeownership education was required
and (b) whether such education requirement was satisfied; however, Fannie Mae did not use
this information. 31 As a consequence, Fannie Mae did not make use of a front-end
(preventive) control at the time of purchase to determine whether the homeownership
education requirement was met. Instead, Fannie Mae relied on representations and warranties


31
   According to Fannie Mae, when lenders provided this information, it flowed into a system that was not one
from which Fannie Mae ran reports. So, while Fannie Mae had the information, it did not make use of it.



                                   OIG • AUD-2018-003 • March 8, 2018                                          20
by lenders – a back-end (detective) control – that all requirements of the 97% LTV mortgage
program were met, including pre-purchase homeownership education, when required.

Based on our analysis of the data provided by Fannie Mae, we found that, as part of its
quality control program, Fannie Mae completed quality control reviews of 716 of the 18,769
mortgages requiring pre-purchase homeownership education. Those reviews identified 20
exceptions (3% of the sample). 32 For one of the exceptions, Fannie Mae cited a significant
defect and reported that the lender would repurchase the mortgage. For the other 19
exceptions, Fannie Mae cited the lender with a finding for missing homeownership education.

During our field work, Fannie Mae reported that it implemented a “fatal” rule in its Loan
Delivery system in March 2017, requiring lenders to confirm that homebuyers have
completed pre-purchase homeownership education, when required. This “fatal” rule means
that a mortgage will be rejected if the lender does not confirm that homebuyers have
completed pre-purchase homeownership education when it is required. Fannie Mae also
reported that, as of June 15, 2017, the absence of homeownership education documentation
will be considered a significant defect if found during a quality control review.

FHFA’s Oversight of Fannie Mae’s 97% LTV Mortgage Program

As discussed earlier, the Staff Memorandum considered FHFA’s ongoing oversight of
Enterprise implementation of their 97% LTV mortgage program to be a risk mitigant. We
reviewed FHFA’s oversight of Fannie Mae’s implementation of its 97% LTV mortgage
program during the review period and identified the following activities:

     •   DHMG briefed DER on the parameters of the Enterprises’ 97% LTV mortgage
         programs in December 2014.

     •   DHMG staff prepared review reports based on Fannie Mae’s 97% LTV mortgage
         program data from May 2015 to September 2017. According to DHMG, it uses the
         information in these reports to assess whether the goals and objectives of the 97%
         LTV mortgage program were being met and to flag potential concerns or issues for
         FHFA leadership. We reviewed these reports and found that none reported on the
         three credit terms that were the subject of this audit: automated underwriting,
         mortgage insurance, and pre-purchase homeownership education requirements. These

32
   Fannie Mae defines a “significant defect” as “one or more defects that either necessitates a change to the
price on which the loan was acquired or result in the loan being unacceptable for purchase had the true and
accurate information about the loan been known at the time of purchase.” A finding is defined as “one or more
defects that, when considered with other loan features, does not necessitate a change in the price of the loan or
result in the loan being unacceptable even if the true and accurate facts about the loan had been known at the
time Fannie Mae purchased or securitized the loan.”




                                     OIG • AUD-2018-003 • March 8, 2018                                             21
         reports also did not contain information about Fannie Mae’s quality control reviews.
         The reports did include information on homebuyers (such as average credit scores and
         average DTI ratios), loan volume, delinquency rates, and the Enterprises’ return on
         capital/equity. DHMG’s report, as of September 2017, also showed that Fannie Mae
         had acquired a total of 123,753 of the 97% LTV mortgages under its Standard 97
         product. 33 That same DHMG report noted that the 90-day delinquency rate for 97%
         LTV mortgages was 0.22%, 8 basis points higher than its conventional mortgages,
         which had a 90-day delinquency rate of 0.14%.

     •   During the 2016 examination cycle, DER performed an ongoing monitoring activity of
         Fannie Mae’s Single-Family Mortgage Business. 34 The ongoing monitoring activity
         included a procedure to review monthly acquisition metrics for 97% LTV mortgages.
         The 2016 ongoing monitoring activity
         However, DER examiners observed in the closeout memorandum that




     •   During the 2017 examination cycle, DER initiated a targeted examination




                                    The targeted examination was in progress as of March 2018.

The focus of the oversight activities described above was on Fannie Mae’s credit risk
management and did not directly address compliance with the three risk mitigants that were
the scope of this audit.




33
   According to Fannie Mae’s Annual SEC 10-K filings, it reported that it acquired approximately 24,000
97% LTV mortgages during 2015 and that these loans represented 1% of the single-family loans acquired
during the year. For 2016, Fannie Mae reported that it acquired approximately 50,000 97% LTV mortgages,
representing 2% of the single-family loans acquired during the year.
34
   FHFA’s supervisory activities include ongoing monitoring and targeted examinations. According to the
FHFA Examination Manual, the purpose of ongoing monitoring is to analyze real-time information and to use
those analyses to identify Enterprise practices and changes in an Enterprise’s risk profile that may warrant
supervisory attention, while targeted examinations allow for a deep or comprehensive assessment of the area
under review.



                                   OIG • AUD-2018-003 • March 8, 2018                                          22
CONCLUSION ............................................................................

We performed this audit to assess FHFA’s oversight of Fannie Mae’s implementation of the
97% LTV mortgage program. As part of assessing FHFA’s oversight, we obtained (through
FHFA) and analyzed Fannie Mae data on the mortgages purchased by Fannie Mae under the
97% LTV mortgage program approved by FHFA’s December 2014 Staff Memorandum and
whether those mortgages conformed to three FHFA-required credit terms. The FHFA-
required credit terms that we focused on for this audit were: (1) automated underwriting;
(2) mortgage insurance; and (3) homeownership education. Our analysis of data provided by
Fannie Mae, through FHFA, found a high rate of compliance with these credit terms for the
74,700 mortgages purchased by Fannie Mae under its 97% LTV mortgage program.

Fannie Mae reported during our field work that it made changes to its systems and quality
control reviews that, if implemented as described, should enhance controls over all three
credit terms. Specifically, Fannie Mae stated that, as of November 2016, its Loan Delivery
system no longer allows a lender to report use of a proprietary underwriting system in
connection with 97% LTV mortgages, unless there is a specific lender variance in place at the
time of purchase. Fannie Mae also reported that it implemented a “fatal” rule in its Loan
Delivery system for any loans in which the lender did not provide information about mortgage
insurance or another credit enhancement. And, for homeownership education, Fannie Mae
reported that it implemented a “fatal” rule in its Loan Delivery system in March 2017,
requiring lenders to confirm that homebuyers have completed pre-purchase homeownership
education, when required. Lastly, Fannie Mae reported that, as of June 15, 2017, the absence
of homeownership education documentation will be considered a significant defect if found
during a quality control review.

We also found that FHFA conducted oversight of Fannie Mae’s implementation of the 97%
LTV mortgage program. While FHFA’s supervisory activities
         , our review of the workpapers for those activities found that none focused directly on
the three credit terms that were the subject of this audit.

Fannie Mae’s 97% LTV mortgage program approved by FHFA in December 2014 has not
experienced a time of economic stress. As of this writing, the U.S. economy has been stable,
gross domestic product growth has been positive every quarter since the second quarter of
2014, and the unemployment rate currently stands at 4.1%. In the closeout memorandum for a
2016 ongoing monitoring activity, DER examiners observed that

          . In view of the increasing volume of 97% LTV mortgages purchased by Fannie Mae,
it is prudent for FHFA to conduct supervisory activities over Fannie Mae’s 97% LTV



                              OIG • AUD-2018-003 • March 8, 2018                                   23
mortgage program, consistent with the recognition in the Staff Memorandum that such
activities are “an important oversight control.”


FHFA COMMENTS AND OIG RESPONSE .....................................

We provided FHFA an opportunity to respond to a draft of this audit report. FHFA provided
technical comments on the draft report, and those comments were incorporated as appropriate.
FHFA also provided a management response which is reprinted in its entirety in the Appendix
to this report.




                             OIG • AUD-2018-003 • March 8, 2018                                24
OBJECTIVE, SCOPE, AND METHODOLOGY .................................

The objective of our audit was to assess FHFA’s oversight of Fannie Mae’s implementation
of the 97% LTV mortgage program. As part of assessing FHFA’s oversight, we obtained,
through FHFA, and analyzed Fannie Mae data on the mortgages purchased by Fannie Mae
under the 97% LTV mortgage program approved by FHFA’s December 2014 Staff
Memorandum and whether those mortgages conformed to three FHFA-required credit
terms. 35 The FHFA-required credit terms that we focused on for this audit were:
(1) automated underwriting; (2) mortgage insurance; and (3) homeownership education. There
were other risk mitigants associated with Fannie Mae’s purchases of high LTV mortgages that
were not included within the scope of this audit.

To address our objective, we:

     •   Researched and identified applicable laws, regulations, and other guidance that relate
         to FHFA’s oversight of Fannie Mae’s 97% LTV mortgage program;

     •   Obtained and reviewed FHFA and/or Fannie Mae documentation and correspondence
         related to FHFA’s assessment and approval(s) related to Fannie Mae’s 97% LTV
         mortgage program along with changes to the program;

     •   Obtained and analyzed information provided by Fannie Mae related to the universe of
         97% LTV mortgages purchased from December 3, 2014, through December 31, 2016,
         to determine whether Fannie Mae purchased mortgages conforming with the three
         credit terms in the scope of our audit as approved by FHFA’s Staff Memorandum;

     •   Obtained and analyzed documentation regarding FHFA’s oversight of Fannie Mae’s
         97% LTV mortgage program, such as review reports and ongoing monitoring and
         targeted examination results;

     •   Interviewed FHFA and Fannie Mae officials to gain an understanding of Fannie Mae’s
         97% LTV mortgage program (and applicable credit terms);

     •   Observed and reviewed Fannie Mae’s updates to its loan tracking system in response
         to our audit inquiries related to the automated underwriting and mortgage insurance
         for 97% LTV mortgages purchased during the scope of our audit (for example, Fannie



35
   The mortgages with LTVs between 95% and 97% delivered between December 3, 2014, and December 31,
2016, under continuing programs offered by Fannie Mae prior to conservatorship are not included in the scope
of this audit.



                                   OIG • AUD-2018-003 • March 8, 2018                                          25
       Mae updated its loan tracking system after it received mortgage insurance information
       during our audit); and

   •   Obtained a written representation by the Acting Deputy Director, Division of
       Conservatorship, that FHFA, as part of its oversight and conservatorship
       responsibilities, has taken the appropriate actions that provide a reasonable assurance
       of the completeness, accuracy, and reliability of the information (program-related
       data) related to the universe of 97% LTV mortgages purchased by Fannie Mae from
       December 3, 2014, through December 31, 2016, provided for our audit on behalf of
       Fannie Mae.

We conducted this performance audit between March 2017 and March 2018 in accordance
with generally accepted government auditing standards. Those standards require that we plan
and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis
for our findings and conclusions based on our audit objective. We believe that the evidence
obtained provides a reasonable basis for our findings and conclusions based on our audit
objective.




                              OIG • AUD-2018-003 • March 8, 2018                                 26
APPENDIX: FHFA MANAGEMENT RESPONSE .............................




                   OIG • AUD-2018-003 • March 8, 2018         27
ADDITIONAL INFORMATION AND COPIES .................................


For additional copies of this report:

   •   Call: 202-730-0880

   •   Fax: 202-318-0239

   •   Visit: www.fhfaoig.gov



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

   •   Call: 1-800-793-7724

   •   Fax: 202-318-0358

   •   Visit: www.fhfaoig.gov/ReportFraud

   •   Write:

                FHFA Office of Inspector General
                Attn: Office of Investigations – Hotline
                400 Seventh Street SW
                Washington, DC 20219




                               OIG • AUD-2018-003 • March 8, 2018                          28