oversight

Impact of the Federal Reserve's Quantitative Easing Programs on Fannie Mae and Freddie Mac

Published by the Federal Housing Finance Agency, Office of Inspector General on 2014-10-23.

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

            Federal Housing Finance Agency
                Office of Inspector General




  Impact of the Federal Reserve’s
  Quantitative Easing Programs on
   Fannie Mae and Freddie Mac




Evaluation Report  EVL-2015-002  October 23, 2014
              Impact of the Federal Reserve’s Quantitative Easing
              Programs on Fannie Mae and Freddie Mac

              Why OIG Did This Report
              Fannie Mae and Freddie Mac (the Enterprises) provide liquidity to the housing
              finance system by purchasing qualifying mortgages from lenders and
              packaging them into mortgage-backed securities (MBS) that are sold to
 At A         investors. In exchange for a fee, the Enterprises guarantee that investors will
Glance        receive the timely payment of principal and interest on their MBS regardless of
              the credit performance of the underlying mortgage collateral.
  ———
              As part of its effort to respond to the recent financial crisis and its aftermath,
October 23,   the Federal Open Market Committee (Federal Reserve) has purchased over
   2014       $2.3 trillion of the Enterprises’ MBS under its three Quantitative Easing (QE)
              programs and related initiatives. The Federal Reserve initiated the QE
              programs to, among other things, lower interest rates and thereby stimulate
              growth in the housing markets and the broader economy.

              In this report we assess the effects of the QE programs on the Enterprises’
              recent financial performance and the potential implications for the Enterprises
              of the Federal Reserve’s December 2013 decision to reduce or “taper” its MBS
              purchases.

              OIG Analysis
              The Enterprises Benefited Financially from a Surge in Mortgage Refinancings
              Associated with the QE Programs and Higher Guarantee Fee Rates

              The Federal Reserve’s substantial MBS purchases likely contributed
              considerably to lower long-term mortgage rates from 2008 through mid-2013.
              The lower rates caused mortgage refinancings to surge from 2009 through mid
              2013.

              As the refinancing boom was occurring, FHFA directed the Enterprises to
              sharply increase their MBS guarantee fee rates. Since 2011 the rates have
              more than doubled.

              In 2012 and 2013, the Enterprises benefited financially from the combination
              of the surge in mortgage refinancings and the sharp increases in their MBS
              guarantee fee rates. The new mortgages were packaged into MBS, which were
              subject to the higher guarantee fee rates. Mortgages subject to lower guarantee
              fee rates were prepaid through the refinancings.
              From 2011 to 2013 the Enterprises realized a $4 billion increase in annual
              guarantee fee revenue from new single-family MBS issuances, most of which
              is attributable to refinanced mortgages purchased in 2012 and 2013. The
              Enterprises should generally expect to benefit from the increased guarantee fee
              revenue over the lifetime of the securities, but are subject to certain risks.

              The Federal Reserve’s Tapering of Its MBS Purchases Appears to Have
              Contributed to a Relative Decline in the Enterprises’ More Recent Financial
 At A         Performance
Glance        Long-term mortgage interest rates began to increase in mid-2013 due, in part,
  ———         to the financial markets’ perception that the Federal Reserve would begin
              tapering later in the year as well as other factors, such as an improving
October 23,   economy. Since then, the rates have generally stabilized above their 2013
   2014       levels. The increase has contributed to significant declines in mortgage
              refinancing activity and Enterprise MBS issuances in 2014. Consequently, the
              Enterprises’ expected guarantee fee revenue on MBS issued in the first half of
              2014 fell about 56% compared to their expected revenue on MBS issued in the
              first half of 2013.

              FHFA, the Enterprises, and the Federal Reserve provided us with technical
              comments that we incorporated in the final report as appropriate.
TABLE OF CONTENTS ................................................................
AT A GLANCE ...............................................................................................................................2

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

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

BACKGROUND .............................................................................................................................8
      The Federal Reserve and Its Traditional Monetary Policies Intended to Promote
      Maximum Employment, Stable Prices, and Moderate Long-Term Interest Rates ...................8
      The Federal Reserve Initiated the QE Programs to Augment Its Efforts to Combat the
      Financial Crisis .........................................................................................................................9
              QE I Involved the Purchase of MBS and Treasury Securities ........................................10
              QE II Focused Only on the Purchase of Treasury Securities .........................................10
             The Federal Reserve Began Its Maturity Extension Program, “Operation Twist,”
             and Its MBS Reinvestment Policy ..................................................................................11
              QE III Focused Directly on Purchases of MBS and Treasury Securities .......................11
              The Federal Reserve Is Tapering Its Purchases of MBS and Treasury Securities .........12

ANALYSIS ....................................................................................................................................14
      The Federal Reserve’s Purchases of MBS Under Its QE Programs Contributed
      Considerably to Lowered Mortgage Rates and Increased Refinances ...................................14
             Economic Research Suggests that the Federal Reserve’s MBS Purchases Helped
             Reduce MBS Yields and Long-Term Mortgage Rates ...................................................14
              Lower Mortgage Rates Contributed to Substantial Refinancing Activity ......................16
      Mortgage Refinance Activity Contributed to Significantly Higher Enterprise
      Guarantee Fee Revenue in 2012 and 2013 .............................................................................18
             The Enterprises Have Purchased Relatively Large Volumes of Refinanced
             Mortgages Since 2008.....................................................................................................18
             FHFA Has Directed the Enterprises to Significantly Increase Their MBS
             Guarantee Fees ................................................................................................................18
              The Enterprises’ Purchases of Refinanced Mortgages Present Some Risks ..................20
      The Federal Reserve’s Decision to Taper Its MBS Purchases Has Contributed to
      Significant Declines in Expected Guarantee Fee Revenue for 2014 MBS Issuances ............20


                                           OIG  EVL-2015-002  October 23, 2014                                                               4
CONCLUSION ..............................................................................................................................23

OBJECTIVE, SCOPE, AND METHODOLOGY .........................................................................24

APPENDIX A ................................................................................................................................25
      Overview of the Quantitative Easing Programs’ Impact on Mortgage Interest Rates
      and Refinancing Activity ........................................................................................................25
             Mechanisms by Which the QE Programs Have Lowered MBS Yields and
             Mortgage Interest Rates ..................................................................................................25
             A Summary of Current Economic Research Into the Effectiveness of the QE
             Programs .........................................................................................................................27
             The QE Programs’ Influence on Mortgage Refinance and Home Purchase
             Activity ...........................................................................................................................29

ADDITIONAL INFORMATION AND COPIES .........................................................................30




                                           OIG  EVL-2015-002  October 23, 2014                                                                 5
ABBREVIATIONS .......................................................................

Enterprises           Fannie Mae and Freddie Mac

Federal Reserve       Federal Open Market Committee

The Board             Board of Governors of the Federal Reserve System

FHFA or Agency        Federal Housing Finance Agency

Ginnie Mae            Government National Mortgage Association

LSAP                  Large-Scale Asset Purchases

MEP                   Maturity Extension Program or “Operation Twist”

MBS                   Mortgage-Backed Securities

OIG                   Federal Housing Finance Agency Office of Inspector General

QE                    Quantitative Easing

Treasury              U.S. Department of the Treasury




                         OIG  EVL-2015-002  October 23, 2014                      6
PREFACE ...................................................................................

The purpose of this evaluation report is to assess the effect of the QE programs on the
Enterprises’ recent financial performance, and the potential implications of the Federal
Reserve’s decision in December 2013 to taper its MBS purchases on the Enterprises’ financial
condition.

This report was prepared by Simon Wu, Chief Economist; Jacob Kennedy, Investigative
Evaluator; David P. Bloch, Senior Counsel for Securitization & Risk Management; and
Wesley M. Phillips, Director of the Division of Oversight and Review. We appreciate the
assistance of officials from FHFA, the Enterprises, the Federal Reserve, and other government
agencies in completing this report.

This 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
Deputy Inspector General for Evaluations




                            OIG  EVL-2015-002  October 23, 2014                               7
BACKGROUND ..........................................................................

The Federal Reserve and Its Traditional Monetary Policies Intended to Promote
Maximum Employment, Stable Prices, and Moderate Long-Term Interest Rates

As depicted in Figure 1, below, the Federal Reserve System is comprised of the Board of
Governors (the Board), which is situated in Washington DC, twelve regional Federal Reserve
Banks, and the member banks. 1 The Board and reserve banks share responsibility for
supervising and regulating certain financial institutions and activities, providing banking
services to depository institutions and the federal government, and ensuring that consumers
receive adequate information and fair treatment in their business interactions with the banking
system.2

                                  FIGURE 1. FEDERAL RESERVE STRUCTURE


                                      BOARD OF GOVERNORS
                                               FEDERAL OPEN
                                             MARKET COMMITTEE

                                   12 FEDERAL RESERVE BANKS


                                          MEMBER BANKS
                         Source: http://www.federalreserveeducation.org/about-the-
                        fed/structure-and-functions/
Source: www.federalreserveeducation.org/about-the-fed/structure-and-functions/




1
 The member banks are national banks and state-charted institutions. Membership is required for the former
and is discretionary for the latter. See 12 CFR 208 (regulation H).
2
  The information in this section is based on The Federal Reserve System Purposes & Function, at
http://www.federalreserve.gov/pf/pdf/pf_complete.pdf.




                                  OIG  EVL-2015-002  October 23, 2014                                      8
Moreover, the Federal Reserve is responsible for
monetary policy, which it traditionally exercises through                    Monetary Policy: Actions
its control over the federal funds rate. The mechanisms                      undertaken by the Federal
by which it exerts its control include:                                      Reserve to promote maximum
                                                                             employment, stable prices, and
        Influencing the demand for, and supply of, these                    moderate long-term interest
                                                                             rates.
         balances through the purchase, sale, borrowing,
         or lending of securities—primarily short-term
         Treasury securities—in the open market;
                                                                             Federal Funds Rate: The
        Adjusting the reserve requirements that must be                     overnight interest rate at which
         held at a Federal Reserve Bank; and                                 a depository institution lends
                                                                             funds maintained at the
        Extending credit to depository institutions.                        Federal Reserve to another
                                                                             depository institution or other
The Federal Reserve Initiated the QE Programs to                             eligible entity.
Augment Its Efforts to Combat the Financial Crisis

As the financial crisis began in 2007, the Federal Reserve sought to spur economic recovery
through traditional means. It reduced the federal funds rate by lowering the overnight
borrowing rates applicable to inter-bank lending.3 However, when the federal funds rate and
other short-term rates approached zero percent at the end of 2008, the Federal Reserve
initiated the first of three QE programs, the last of which—QE III—continues to this day.4

Through the QE programs, the Federal Reserve has, generally speaking, sought to strengthen
the economy and housing markets by purchasing U.S. Treasury securities and MBS in order
to lower interest rates and ease credit conditions.5 Figure 2, below, summarizes the timeline
for the three QE programs and related initiatives; the balance of this section provides further
information about them.




3
 The Federal Reserve implemented myriad programs in response to the financial crisis. In this report we focus
upon the QE programs.
4
 Each of the subsequent QE programs was initiated in response to then-contemporaneous economic
conditions.
5
  For example, on September 13, 2012, the Federal Reserve stated that, “to support a stronger economic
recovery and to help ensure that inflation, over time, is at the rate most consistent with its dual mandate, the
Committee [Federal Reserve] agreed today to increase policy accommodation by purchasing additional agency
mortgage-backed securities at a pace of $40 billion per month.” Board of Governors of the Federal Reserve
System, Press Release (September 13, 2012), at
www.federalreserve.gov/newsevents/press/monetary/20120913a.htm.



                                   OIG  EVL-2015-002  October 23, 2014                                           9
                                                                                          6
                         FIGURE 2. TIMELINE OF FEDERAL RESERVE QE PROGRAMS




Source: www.federalreserve.gov.

    QE I Involved the Purchase of MBS and Treasury Securities

In November 2008, the Federal Reserve announced QE I, which involved the purchase of
Enterprise MBS and debt, MBS guaranteed by the Government National Mortgage
Association (Ginnie Mae), and Treasury securities.7 By the time that QE I ended in March
2010, the Federal Reserve had purchased approximately $1 trillion of Enterprise MBS8 and
$135 billion of their debt.9

    QE II Focused Only on the Purchase of Treasury Securities

In November 2010, with the economy still lagging and the unemployment rate elevated, the
Federal Reserve announced QE II. QE II focused on the purchase of longer-term Treasury
securities rather than other types of assets such as MBS. Specifically, the Federal Reserve
committed to purchasing $600 billion of longer-term Treasury securities by the end of the
second quarter of 2011. QE II ended on June 20, 2011.


6
 In August 2010, the Federal Reserve announced its policy to reinvest principal payments from Enterprise
debt and MBS into Treasuries.
7
 Ginnie Mae guarantees investors that they will receive timely payment of principal and interest payments
on MBS comprised of mortgages guaranteed by the Federal Housing Administration and the Department of
Veterans Affairs.
8
  During QE I, the Federal Reserve purchased $1.25 trillion in total MBS. We note that approximately 80%
of the Federal Reserve’s MBS purchases during all three QE programs were Fannie Mae and Freddie Mac
securities; the balance were MBS guaranteed by Ginnie Mae.
9
  The Enterprises sell their MBS to primary market dealers such as Morgan Stanley. In turn, the dealers sell
the MBS on the secondary market to investors, such as banks and pension funds, as well as the Federal
Reserve.




                                   OIG  EVL-2015-002  October 23, 2014                                       10
     The Federal Reserve Began Its Maturity Extension Program, “Operation Twist,” and
     Its MBS Reinvestment Policy

In September 2011, the Federal Reserve initiated Operation Twist.10 Under it, the Federal
Reserve sought to extend the average maturity of its Treasury security holdings. It did so by
purchasing longer-term Treasury securities and selling equal amounts of Treasury securities
with remaining maturities of three years or less.

Operation Twist did not specifically target Enterprise MBS purchases. However, the Federal
Reserve began purchasing such securities while the operation was ongoing and continued
doing so when it ended. When some of the Enterprise MBS in its portfolio reached maturity
or was pre-paid prior to maturity, the Federal Reserve used the proceeds to purchase
additional Enterprise MBS.11 The Federal Reserve labeled these transactions “reinvestment
purchases.”12

     QE III Focused Directly on Purchases of MBS and Treasury Securities

On September 13, 2012, the Federal Reserve initiated QE III to help support a stronger
economic recovery. In addition to its reinvestment purchases, the Federal Reserve committed
to purchasing new MBS at a pace of $40 billion per month and long-term Treasury securities
at a pace of $45 billion per month, for a total of $85 billion per month. From September 2012
to June 2014 the Federal Reserve purchased over $1.3 trillion in Enterprise MBS.13

As shown in Figure 3, below, since the implementation of QE III in the third quarter of 2012,
the Federal Reserve has purchased an increasing percentage of the Enterprises’ annual MBS
issuances. New issuances rose from about 30% in the third quarter of 2012 to about 55% in
the third quarter of 2013.14 From the end of the third quarter of 2013 to the end of the first
quarter in 2014, the percentage of Enterprise MBS purchased by the Federal Reserve soared
to nearly 85%. Thereafter, the Federal Reserve’s monthly MBS purchases began to decline as


10
  The Federal Reserve’s MBS Reinvestment Policy is distinct from its maturity extension program, i.e.,
Operation Twist.
11
  The Federal Reserve’s MBS reinvestment policy involved the reinvestment of principal payments from its
debt holdings, and was intended to prevent the runoff of its MBS holdings.
12
  See www.newyorkfed.org/markets/agency_agencymbs_faq.html for more information on the Federal
Reserve’s reinvestment purchases.
13
 This figure includes the reinvestment of principal payments from Enterprise debt and MBS and Ginnie Mae
MBS.
14
   During the QE programs, approximately 30% of the Federal Reserve’s MBS purchases were reinvestments
of principal payments from agency debt and MBS.




                                  OIG  EVL-2015-002  October 23, 2014                                    11
a result of its “tapering” schedule.15 At the same time, however, new Enterprise MBS
issuances declined even more dramatically, which accounts for the increased percentage of
MBS purchases in late 2013 and early 2014.

                    FIGURE 3. FEDERAL RESERVE PURCHASES OF ENTERPRISE MBS ISSUANCES,
                            FOURTH QUARTER 2011 THROUGH FIRST QUARTER 2014

                 $400                                                                                  100%

                 $350                                                                                  90%
                                                                                                       80%
                 $300
                                                                                                       70%
                 $250                                                                                  60%
      Billions




                 $200                                                                                  50%

                 $150                                                                                  40%
                                                                                                       30%
                 $100
                                                                                                       20%
                  $50                                                                                  10%
                   $0                                                                                  0%




                            Enterprises' Issuance           % Purchased by Federal Reserve

Source: The Federal Reserve Bank of New York – Domestic Market Operations; FHFA Annual Reports to
Congress; and Fannie Mae and Freddie Mac Monthly Volume Summary reports.

     The Federal Reserve Is Tapering Its Purchases of MBS and Treasury Securities

In December 2013, the Federal Reserve stated that it would begin tapering its asset purchases
under QE III due to improved economic activity and gains in the labor markets.16
Accordingly, and as depicted in Figure 4, below, the Federal Reserve’s monthly MBS and




15
  The percentage of the Federal Reserve’s purchase of Enterprise MBS declined to about 60% in the second
quarter of 2014. This decline appears to be due to the Federal Reserve’s decision to taper QE III asset
purchases.
16
   The Federal Reserve determined that the improved economic activity and the gains in the labor market were,
in part, engendered by the programs.




                                  OIG  EVL-2015-002  October 23, 2014                                         12
Treasury security purchases declined from nearly $100 billion in January 2014 to $61.9
billion in June 2014.17

     FIGURE 4. FEDERAL RESERVE MONTHLY ENTERPRISE MBS AND TREASURY SECURITY PURCHASES
                                JANUARY THROUGH JUNE 2014

                $100
                 $90
                 $80
                 $70
     Billions




                 $60
                 $50
                 $40
                 $30
                 $20
                 $10
                  $0
                       Jan 14        Feb 14          Mar 14         Apr 14          May 14          Jun 14

                                             Agency MBS            US Treasuries


Source: The Federal Reserve Bank of New York – Domestic Market Operations.

On September 17, 2014, the Federal Reserve announced that it would reduce its monthly asset
purchases to $15 billion per month, which includes only $5 billion for MBS and $10 billion
for Treasury securities.18




17
   The $100 billion in MBS and Treasury purchases in January 2014 includes the reinvestment of the payoff
from maturing MBS and debt to fund the purchase of additional MBS, as well as the regular monthly purchase
amount.
18
    We note that although the Federal Reserve purchased more than $2.3 trillion in Enterprise MBS since 2008,
its balance of such MBS on March 31, 2014, was only about $1.3 trillion. The difference reflects the fact that,
prior to October 2011, the Federal Reserve permitted maturing or prepaying MBS to run-off its balance sheet
and thereby reduce the size of its overall MBS portfolio. Since then, however, the Federal Reserve has used
the principal payments from its holdings, agency debt, and agency MBS to fund the purchase of additional
MBS.




                                   OIG  EVL-2015-002  October 23, 2014                                          13
ANALYSIS .................................................................................

The Federal Reserve’s Purchases of MBS Under Its QE Programs Contributed
Considerably to Lowered Mortgage Rates and Increased Refinances

The Federal Reserve’s QE programs, among other factors, contributed considerably to the
significant decline in long-term mortgage interest rates that occurred during the period from
2008 through mid-2013.19 Those declining rates, in turn, contributed to a substantial increase
in mortgage refinance activity during the period.

      Economic Research Suggests that the Federal Reserve’s MBS Purchases Helped
      Reduce MBS Yields and Long-Term Mortgage Rates

Economic research that we reviewed indicates that the Federal Reserve’s substantial
purchases of MBS since 2009 contributed to an increase in their price and a corresponding
decrease in their yields (see
                                              FIGURE 5. YIELD ON ENTERPRISE MBS
Figure 5 for yield).20
                                                        SEPTEMBER 2008 THROUGH JUNE 2014
For example, one research paper            6.0%
concluded that shortly after the
                                           5.0%
announcement of the Federal
Reserve’s QE I MBS purchases,              4.0%
the yields on MBS with a 30-year
maturity declined, on average, by          3.0%

107 basis points.21
                                           2.0%

We note that MBS yields                    1.0%
increased moderately in the
summer of 2013 and stabilized at
these higher levels in 2014. This             Source: Bloomberg (Fannie Mae 30-year current coupon
has been attributed, among other              mortgage yield based on bid quote).


19
  See Appendix A of this report for a discussion of the relationship between the Federal Reserve’s MBS
purchases, MBS yields, long-term mortgage interest rates, and mortgage refinance activity.
20
     Appendix A provides a listing and summary of these economic research papers.
21
   A basis point equals one hundredth of one percent, or 0.01%. Thus, a decrease of 107 basis points would
reduce the annual yield on a MBS with a 30-year maturity from 5.00% to 3.93%. See Arvind Krishnamurthy
and Annette Vissing-Jorgensen, The Ins and Outs of LSAPs, Federal Reserve Bank of Kansas City (September
16, 2013), at www.kansascityfed.org/publicat/sympos/2013/2013Krishnamurthy.pdf. The estimated averages
in the working paper include MBS issued by the Enterprises as well as those guaranteed by Ginnie Mae.




                                   OIG  EVL-2015-002  October 23, 2014                                     14
things, to the perception on the part of financial market participants that the Federal Reserve
would begin to taper its MBS and Treasury security purchases later in the year.22

MBS yields and long-term mortgage interest rates
generally move parallel to each other because an                      Yield: The yield on a security, such
MBS’ yield is, essentially, the prevailing mortgage                   as MBS, is its return to the investor.
interest rate less the compensation paid to the                       There is generally an inverse
Enterprises, underwriters, and servicers.23 See                       relationship between a security’s
                                                                      price and its yield. As the price goes
Figure 6, below. Thus, decreasing MBS yields
                                                                      up the yield goes down; and as the
caused, in part, by the QE programs resulted in                       price goes down the yield goes up.
reductions in long-term mortgage interest rates;
this, in turn, led to the spike in refinancing.

As shown in Figure 6, the decline
in the 30-year mortgage rate                               FIGURE 6. ENTERPRISE MBS YIELDS AND
                                                              30-YEAR FIXED MORTGAGE RATES
generally paralleled the decline
                                                           SEPTEMBER 2008 THROUGH JUNE 2014
in MBS yields. Specifically,                 7.0%

the rate declined from just above            6.0%
6% in October 2008 to a record               5.0%
low (in nominal terms) of about              4.0%
3.5% in July 2013. During the
                                             3.0%
remainder of 2013, long-term
                                             2.0%
mortgage rates increased
                                             1.0%
moderately to more than 4%
and have remained above their
2013 lows in 2014.24                                            30-Year Fixed Rate           MBS Yield

We note that several other factors              Source: Freddie Mac Primary Mortgage Market Survey and
likely contributed to the decline in            Bloomberg (Fannie Mae 30-year current coupon mortgage
                                                yield based on bid quote).
MBS yields and long-term interest
rates. These include:


22
   Some commenters have suggested that these perceptions reduced the demand for, and prices of, these
securities.
23
  The difference between the mortgage rates and MBS yields is called the “primary-secondary spread.” See
Andreas Fuster, Laurie Goodman, David Lucca, Laurel Madar, Linsey Molloy, and Paul Willen, The Rising
Gap Between Primary and Secondary Mortgage Rates, The Federal Reserve Bank of New York Economic
Policy Review, December 2013.
24
   The increase in 2013 has been attributed, in part, to perceptions on the part of financial market participants
that the Federal Reserve would begin to taper its QE III asset purchases in the near term.




                                    OIG  EVL-2015-002  October 23, 2014                                           15
        General weakness in the U.S. economy since 2008, which has likely reduced the
         demand for credit, including mortgage credit.

        A worldwide investor “flight to safety,” which has increased demand for U.S.
         Treasury securities and other impliedly “safe” securities, such as Enterprise-issued
         MBS and their debt. Increased demand for these securities has caused a decrease in
         their yields. Given that long-term mortgage rates run parallel to these yields, those
         rates have also declined.

     Lower Mortgage Rates Contributed to Substantial Refinancing Activity

The significant decline in long-term mortgage interest
rates, which were likely influenced considerably by                        Mortgage Refinancing:
the Federal Reserve’s MBS purchases, helped trigger a                      Replacing an older loan with
substantial increase in mortgage refinancing activity                      a new loan, generally due to
                                                                           lower interest rates.
from early 2009 through mid-2013.25

For example, as shown in Figure 7 below, mortgage refinancing activity increased sharply in
the first half of 2009 shortly after the Federal Reserve initiated QE I. During the latter half
of 2010, there was a temporary increase in 30-year mortgage rates and mortgage refinancing
activity declined substantially in the first quarter of 2011. 26 However, in the second quarter
of 2011, general market mortgage refinancing activity began to increase again as 30-year
mortgage rates began to fall.27 This refinancing activity continued its upward trend in 2012,
and accelerated toward mid-year, shortly before the Federal Reserve initiated QE III in the
third quarter of 2012.28 However, as 30-year mortgage rates increased during late 2013,
mortgage refinancing activity declined markedly.


25
   Both Enterprises have stated that the QE programs contributed directly to the increase in mortgage refinance
activity. See Fannie Mae, Signs of Improvement Emerge but Economic Outlook Remains Uncertain (October
18, 2012), at www.fanniemae.com/portal/about-us/media/financial-news/2012/5867.html; and Freddie Mac,
Economic & Housing Market Outlook – QE3 in Motion (October 23, 2102), at
www.freddiemac.com/finance/pdf/Oct_2012_public_outlook.pdf.
26
   It is not clear why mortgages rates increased in 2010. We note that interest rates can change for a variety of
reasons independent of the Federal Reserve’s monetary policy initiatives.
27
   The Federal Reserve initiated MBS reinvestment purchases separately from, but during the tenure of,
Operation Twist. These purchases may have contributed to declining long-term mortgage rates in 2011 and the
increased mortgage refinance activity.
28
   We note that FHFA-directed modifications to the Home Affordable Refinance Program also contributed to
the substantial increase in refinancing activity during this period. For further information, see FHFA-OIG,
Home Affordable Refinance Program, A Mid-Program Assessment, EVL-2013-006 (August 1, 2013), at
http://fhfaoig.gov/Content/Files/EVL-2013-006.pdf.




                                   OIG  EVL-2015-002  October 23, 2014                                            16
                                 FIGURE 7. VOLUME OF MORTGAGE ORIGINATION AND REFINANCING, 2008 THROUGH 2014

                                 $600                                                                                 6.5


                                 $500                                                                                 6.0
 Origination Volume (Billions)




                                                                                                                            Interest Rate (Percent)
                                                                                                                      5.5
                                 $400
                                                                                                                      5.0
                                 $300
                                                                                                                      4.5
                                 $200
                                                                                                                      4.0

                                 $100                                                                                 3.5

                                   $0                                                                                 3.0
                                        2008        2009        2010       2011          2012      2013      2014

                                               Home Purchases          Home Refinances          30 yr Mortgage Rate

Source: Mortgage Bankers Association Quarterly Mortgage Originations Estimates as of May 2014 and the
Freddie Mac Primary Mortgage Survey, historical monthly data for conventional, conforming 30-year fixed-rate
mortgages.

Home purchase mortgages fluctuated from 2009 through mid-2013 while mortgage
refinancing activity sharply increased during the same period. According to federal officials29
and FHFA and Enterprise executives, mortgage refinance activity is highly sensitive to
declining interest rates. Borrowers focus primarily upon interest rates when determining
whether to refinance – and they tend to do so when rates fall sufficiently to offset closing and
other transaction costs.

On the other hand, individuals seeking to purchase homes tend to consider factors in addition
to interest rates, such as their employment prospects. The potentially weakened financial
situation of many potential home buyers during the period 2009 through 2013 may have
impeded their willingness or ability to take advantage of lower rates then available.




29
  We spoke with officials of Treasury, Congressional Budget Office (CBO), and Office of Management and
Budget (OMB).




                                                           OIG  EVL-2015-002  October 23, 2014                                               17
Mortgage Refinance Activity Contributed to Significantly Higher Enterprise Guarantee
Fee Revenue in 2012 and 2013

Increased mortgage refinancing activity significantly benefitted the Enterprises’ financial
performance in 2012 and 2013. During this period FHFA directed the Enterprises to raise the
guarantee fees on MBS for safety and soundness and policy reasons, as well as to implement
legislation designed to offset temporary cuts in payroll taxes. The Enterprises packaged
refinanced mortgages into MBS subject to the higher guarantee fees and, in doing so, replaced
older mortgages in previous MBS issuances subject to lower guarantee fees. Thus, the
Enterprises realized substantial increases in their guarantee fee revenue from 2011 to 2013.

     The Enterprises Have Purchased Relatively Large Volumes of Refinanced Mortgages
     Since 2008

As shown in Figure 8, below, the Enterprises substantially increased the percentage of
refinanced mortgages that they acquired during the period 2006 through 2013. During the
housing boom era of 2006 and 2007, only about half of the mortgages the Enterprises
purchased were refinanced loans. Since 2009 refinanced loans constitute about 77% of the
Enterprises’ mortgage acquisitions.30

          FIGURE 8. PERCENTAGE OF REFINANCE MORTGAGES ACQUIRED BY THE ENTERPRISES
                                2006 THROUGH 2013 ($BILLIONS)

                                2006      2007       2008      2009      2010       2011      2012    2013
Total Mortgage Purchases         $876     $1,125      $941    $1,176       $994      $879    $1,263   $1,156
Refinance Mortgages              $417       $577      $555      $940       $783      $674    $1,010    $822
Percent Refinances                48%       51%        59%       80%       79%        77%       80%     71%

Source: Fannie Mae and Freddie Mac SEC Filings Form 10K.

     FHFA Has Directed the Enterprises to Significantly Increase Their MBS Guarantee Fees

From 2011 to 2013, the Enterprises nearly doubled their average MBS guarantee fees (g-fees)
from about 28 basis points to 54 basis points. FHFA directed the Enterprises to do so because
their formerly low fees exposed them to significant financial losses during the financial crisis.




30
  The Enterprises’ increased percentage of refinanced mortgage purchases from 2009 through 2013 is
consistent with the general trend toward increased mortgage refinances depicted in Figure 7 above.




                                 OIG  EVL-2015-002  October 23, 2014                                       18
Further, FHFA saw the fee increase as a way to attract private capital to housing finance.31
FHFA also raised guarantee fees by 10 basis points in April 2012 as required by the
Temporary Tax Cut Continuation Act of 2011.32

As a result of the sharp increase in guarantee fees, the Enterprises realized significant annual
increases in their expected guarantee fee revenue on MBS issued subsequent to 2011. See
Figures 9 and 10, below.33 Specifically their expected guarantee fee revenue increased by
about $4 billion on MBS issued in 2013 compared to MBS issued in 2011.34

      FIGURE 9. FANNIE MAE AVERAGE GUARANTEE FEES AND MBS ISSUANCES, 2011 THROUGH 2013

                                                                 2011                2012                2013
     Average G-Fee (basis points)                                       28.8                39.9                57.4
     Single-Family MBS Issuance Volume                         $545 billion        $828 billion         $733 billion
     Expected Annual Revenue                                    $1.6 billion        $3.3 billion         $4.2 billion
Source: Fannie Mae SEC Filings Form 10K. Expected revenue calculations estimated by OIG.

     FIGURE 10. FREDDIE MAC AVERAGE GUARANTEE FEES AND MBS ISSUANCES, 2011 THROUGH 2013

                                                                 2011                2012                2013
                                   35
     Average G-Fee (basis points)                                       26.8                38.3                51.4
     Single-Family MBS Issuance Volume                         $305 billion        $446 billion         $435 billion
     Expected Annual Revenue                                  $817 million          $1.7 billion         $2.2 billion
Source: Freddie Mac SEC Filings Form 10K. Expected revenue calculations estimated by OIG.




31
  For more information about guarantee fees, see FHFA-OIG, FHFA’s Initiative to Reduce the Enterprises’
Dominant Position in the Housing Finance System by Raising Gradually Their Guarantee Fees, EVL-2013-
005 (July 16, 2013), at http://fhfaoig.gov/Content/Files/EVL-2013-005_4.pdf.
32
   This statutorily directed increase was not intended to confer a financial benefit upon the Enterprises. Rather,
it was designed to raise revenue over a period of 10 years and thereby offset the costs associated with
temporary reductions in payroll taxes.
33
  The Enterprises’ combined expected annual guarantee fee revenue for single-family MBS issuances (from
Figures 9 and 10) increased from $2.4 billion in 2011 to $6.4 billion in 2013; an increase of 168%.
34
   We note that the structure of guarantee fees also likely resulted in financial benefits to the Enterprises from
the increased refinancing activity discussed above. Guarantee fees include an upfront fee, which is incurred at
the time a mortgage is originated and acquired by an Enterprise, and an ongoing monthly fee. The increased
collection of upfront fees permitted the Enterprises to derive one-time revenue gains on the large volume of
refinanced mortgages they purchased in 2012 and 2013.
35
  Freddie Mac has historically set its guarantee fees lower than Fannie Mae as a means to strengthen market
demand for its MBS. Fannie Mae receives relatively higher market demand for its MBS.




                                    OIG  EVL-2015-002  October 23, 2014                                               19
Officials from the Enterprises, FHFA, and other federal officials generally agreed that the
Federal Reserve’s QE programs have significantly benefitted the Enterprises’ financial
performance in recent years. However, they cautioned that it is nearly impossible to quantify
the extent to which the QE programs resulted in increased Enterprise revenues and earnings,
as there were many other domestic and international macroeconomic factors impacting the
Enterprises’ performance.

      The Enterprises’ Purchases of Refinanced Mortgages Present Some Risks

We also note that Enterprises’ purchases of refinanced mortgages in recent years have
involved the following risks:

          Prepayment risks associated with the MBS in the Enterprises’ retained mortgage
           portfolios: The Enterprises generally package mortgages that they purchase into MBS
           that they sell to investors. However, the Enterprises also hold some MBS in their
           retained mortgage portfolios.36 When borrowers refinance—and thereby prepay—the
           mortgages that collateralize the MBS in the Enterprises’ retained portfolios, the
           Enterprises are deprived of the level of principal and interest that they expected to
           earn over the natural lives of such mortgages.37

          Counterparty Credit Risk: In a recent report, we noted that both FHFA and Enterprise
           officials believe that some of the small and nonbank lenders that focused on the sale of
           refinanced mortgages to the Enterprises present elevated counterparty credit and other
           risks as compared to traditional banks.38

The Federal Reserve’s Decision to Taper Its MBS Purchases Has Contributed to
Significant Declines in Expected Guarantee Fee Revenue for 2014 MBS Issuances

Although the Federal Reserve’s QE programs benefitted the Enterprises’ financial condition
in 2012 and 2013, its decision, among other factors, in late 2013 to taper its MBS purchases
contributed to an upturn in long-term interest rates.39 This, in turn, has contributed to a

36
     The Enterprises’ transfer prepayment risk when they sell their MBS to investors.
37
  For additional information about the Enterprises’ prepayment risk and efforts to mitigate it, see FHFA-OIG,
The Housing Government-Sponsored Enterprises’ Challenges in Managing Interest Rate Risks, WPR-2013-01
(March 11, 2013), at http://fhfaoig.gov/Content/Files/WPR-2013-01_2.pdf.
38
  For more information see FHFA-OIG, Recent Trends in the Enterprises’ Purchases of Mortgages from
Smaller Lenders and Nonbank Mortgage Companies, EVL-2014-010 (July 17, 2014), at
www.fhfaoig.gov/Content/Files/EVL-2014-010_0.pdf.
39
  Mortgage and other interest rates may have increased for other reasons as well, including generally
improved economic conditions and higher demand for credit.




                                     OIG  EVL-2015-002  October 23, 2014                                      20
significant decline in the Enterprises’ guarantee fee revenues on MBS issued in 2014. It
remains to be seen whether this trend will continue.

As discussed previously, the Federal Reserve purchased more than 55% of the Enterprises’
new MBS issuances in 2013, and nearly 85% of them in the first quarter of 2014. In
operating the QE programs, the Federal Reserve purchased MBS according to program-
specific criteria, including predetermined monthly and annual purchase targets.40 By doing
so, the Federal Reserve increased the demand for, and the price of, the Enterprises’ MBS.
This, in turn, helped drive down considerably their yields as well as the mortgage interest
rates that run in tandem with them.

As the Federal Reserve tapers its monthly MBS purchases, other market participants, such as
banks and investment funds, will decide if they should buy MBS based upon their anticipated
risk-adjusted return relative to other investments.41 These market participants may demand a
more favorable price for MBS than does the Federal Reserve.42 A consequent drop in the
price of Enterprise MBS could cause an increase in their yields – and a corresponding
increase in mortgage interest rates.

Indeed, as discussed earlier, MBS yields and long-term interest rates increased moderately in
late 2013 as a result of anticipation in the market that the Federal Reserve would begin to
taper its MBS purchases under QE III. The rise in interest rates contributed to a substantial
decrease in mortgage refinance and home purchase activity which, in turn, has contributed to
significant declines in MBS issuances and expected guarantee fee revenue in the first half of
2014. See Figure 11, below. Therefore, the Enterprises’ financial performance and earnings
have been adversely affected by the higher mortgage rates associated with tapering.




40
   One of the Federal Reserve’s primary missions is to conduct monetary policy. Its purchases and sales of
assets are not intended to generate a profit, as would be the case for any other market participant. Each QE
program had different targets over varied frequencies. For example, the Federal Reserve announced the total
size of the program as well as the anticipated end date for QE I. For QE III, the Federal Reserve announced a
monthly purchase amount, i.e., the size, pace, and composition of its intended purchases.
41
   An investment’s risk-adjusted return is a measurement of the risk that the investor must bear in order to
achieve the anticipated return. The measurement is generally expressed as a number or rating. Such ratings
are applied to investment portfolios, funds, and individual securities.
42
  The Federal Reserve purchases MBS through competitive auctions; so while it is not profit-driven,
competitive market prices are elicited.




                                   OIG  EVL-2015-002  October 23, 2014                                        21
     FIGURE 11. ENTERPRISE SINGLE-FAMILY MBS ISSUANCES AND EXPECTED REVENUES, 2013 AND 2014


                                    Issuances          Avg G-Fee        G-Fee Revenue        Change from
         (January–June Only)       ($Millions)       (Basis Points)       ($Millions)           2013
        Fannie Mae
           2013                     $428,843               55.7              $2,389
           2014                     $161,068               62.8              $1,012             (57.7%)
        Freddie Mac
           2013                     $269,000               49.9              $1,342
           2014                     $111,000               57.0               $633              (52.9%)
        Total
           2013                     $697,843                                 $3,731
           2014                     $272,068                                 $1,644             (55.9%)

Source: Fannie Mae and Freddie Mac SEC Form 10Q – June 2014.

Reductions in the Federal Reserve’s MBS portfolio, under some scenarios, could also affect
the Enterprises’ future financial performance. Specifically, it could put additional upward
pressure on mortgage interest rates. This, in turn, could reduce refinance and home purchase
activity and thereby diminish Enterprise mortgage acquisitions and MBS issuances.43 We
note, however, that under other scenarios these adverse outcomes could be offset by an
improving economy and rising home prices, which could benefit the Enterprises’ financial
performance.




43
  As noted in Figure 3 above, there was a substantial decrease in Enterprise MBS issuances in the fourth
quarter of 2013 and the first quarter of 2014.




                                  OIG  EVL-2015-002  October 23, 2014                                    22
CONCLUSION ............................................................................

The combination of the Federal Reserve’s QE programs and FHFA’s decision to increase the
Enterprises’ guarantee fees contributed considerably to their financial performance in 2012
and 2013. Some of these contributions will bolster the Enterprises’ financial performance
over time. Specifically, the revenues that the Enterprises realized by packaging large amounts
of refinanced mortgages into MBS subject to substantially increased guarantee fees in 2012
and 2013 will continue over the lifetime of the securities.44

More recently, the Federal Reserve’s decision in late 2013 to taper its MBS purchases appears
to have contributed to higher mortgage rates which, in turn, contributed to significant
reductions in the Enterprises’ guarantee fee revenues on MBS issued in 2014. Continued
tapering by the Federal Reserve and the eventual reduction of its massive MBS portfolio
could have an adverse impact upon the Enterprises’ financial performance. Under other
scenarios, however, an improving economy and higher home prices could be of benefit to the
Enterprises’ financial performance. FHFA has a responsibility to monitor these issues and
risks as well as their implications for the Enterprises.




44
   Of course, if there are substantial borrower defaults on the underlying mortgages, the Enterprises would face
significant financial obligations to honor their MBS guarantees to investors.




                                   OIG  EVL-2015-002  October 23, 2014                                           23
OBJECTIVE, SCOPE, AND METHODOLOGY .................................

The primary objectives of this report were to assess the effect of the QE programs on the
Enterprises’ recent financial performance and to assess the potential implications of the
Federal Reserve’s decision to taper its MBS purchases on the Enterprises’ financial condition.

To address these objectives, we interviewed officials at FHFA, Fannie Mae and Freddie Mac,
Treasury, the CBO, and the OMB. We also conducted an informal discussion with the staff at
the Federal Reserve Bank of New York.

To accomplish our analysis, we obtained published data on the Federal Reserve’s MBS
purchases and holdings. We analyzed the Enterprises’ mortgage purchase volume, refinance
volume projections, MBS issuance, and guarantee fee revenue, as well as the 30-year
mortgage interest rate and MBS yields. In addition, we analyzed and incorporated data from
academic research literature, government and industry research papers, and other federal
agencies. The data used in this report covered the period from 2006 through the second
quarter of 2014, when available. We shared the preliminary results of our analysis with
FHFA and Enterprise officials, who generally agreed with our analysis. However, we did
not independently test the reliability of the Enterprises’ or Federal Reserve’s data.

This study was conducted under the authority of the Inspector General Act and is in
accordance with the Quality Standards for Inspection and Evaluation (January 2012), which
was promulgated by the Council of the Inspectors General on Integrity and Efficiency. These
standards require OIG to plan and perform an evaluation that obtains evidence sufficient to
provide reasonable bases to support its findings and recommendations. We believe this report
meets these standards.

FHFA, the Enterprises, and the Federal Reserve provided us with technical comments on a
draft of this report, which we incorporated in the final draft as appropriate.

The performance period for this evaluation was between February 2014 and September 2014.




                             OIG  EVL-2015-002  October 23, 2014                               24
APPENDIX A .............................................................................

Overview of the Quantitative Easing Programs’ Impact on Mortgage Interest Rates and
Refinancing Activity

This appendix provides information on the means by which the Federal Reserve’s QE
programs are believed to have contributed to lower MBS yields. It also addresses the QE
programs’ impact on long-term mortgage interest rates and summarizes the economic
literature on the programs. Finally, it includes a discussion of the relationship between lower
mortgage rates and mortgage refinancing activity.

     Mechanisms by Which the QE Programs Have Lowered MBS Yields and Mortgage
     Interest Rates

As explained in the main report, in operating the QE programs, the Federal Reserve purchased
MBS according to program-specific criteria, including predetermined monthly and annual
purchase targets. By doing so, the Federal Reserve increased the demand for, and the price
of, the Enterprises’ MBS. This, in turn, helped drive down considerably their yields as well
as the mortgage interest rates that run in tandem with them. Recent studies45 have discussed
several additional mechanisms by which the Federal Reserve’s QE programs are said to have
had an impact upon MBS yields and mortgage interest rates:

        1. Market Signaling Effect of QE Programs

Asset purchases by the Federal Reserve are often interpreted by investors and market
participants as signals about the central bank’s intentions regarding interest rates. Unlike the
Federal Reserve’s usual short-term Treasury securities purchases, long-term asset purchases,
such as those made under the QE programs, may increase the credibility of the Federal



45
  Source: Diana Hancock and Wayne Passmore, How the Federal Reserve’s Large-Scale Asset Purchases
(LSAPs) Influence Mortgage-Backed Securities (MBS) Yields and U.S. Mortgage Rates, Federal Reserve Board
Working Paper, 2014-12 (2014) at www.federalreserve.gov/pubs/feds/2014/201412/201412abs.html; Arvind
Krishnamurthy and Annette Vissing-Jorgensen, The Ins and Outs of LSAPs, Federal Reserve Bank of Kansas
City (September 16, 2013), at www.kansascityfed.org/publicat/sympos/2013/2013Krishnamurthy.pdf;
Johannes Stroebel and John B. Taylor, Estimated Impact of the Federal Reserve’s Mortgage-Backed Securities
Purchase Program, International Journal of Central Banking, Vol. 8, No. 2 (June 2012), at
www.ijcb.org/journal/ijcb12q2a1.pdf; Iryna Kaminska and Gabriele Zinna, Official Demand for U.S. Debt:
Implications for U.S. real interest rates, IMF Working Paper (April 2014), at
www.imf.org/external/pubs/ft/wp/2014/wp1466.pdf; and Saty Patrabansh, William M. Doerner, and Samuel
Asin, The Effects of Monetary Policy on Mortgage Rates, FHFA Working Paper 14-2 (June 2014), at
www.fhfa.gov/PolicyProgramsResearch/Research/Pages/Working-Paper-14-2.aspx.




                                 OIG  EVL-2015-002  October 23, 2014                                       25
Reserve’s commitment to maintain low interest rates for the long-term and possibly even after
the start of an economic recovery.46

         2. Removing Risks from Investors’ Portfolios

The Federal Reserve’s purchases reduce interest rates through what is termed the “portfolio
balance effect.” A rise in the demand for a particular financial asset – in this case, MBS –
will increase the asset price and reduce its yield (asset prices and yields move in opposite
directions).47 Thus, when investors sell MBS to the Federal Reserve, they may rebalance their
portfolios by investing the cash proceeds in other assets. This would have the effect of raising
the prices and lowering the yields of these assets. This “ripple effect” of increased asset
prices across a variety of investments is consistent with the Federal Reserve’s intentions for
QE.

This same argument can also be characterized in terms of reducing duration and convexity
(prepayment) risks in the portfolios of private investors. 48 Duration and prepayment risks
are more prominent for fixed-income securities with longer maturities such as MBS, as these
risks positively correspond with the lifespan of the securities. Investors generally demand an
extra return to bear these risks. Thus, by removing a considerable number of risks, the
Federal Reserve’s purchases could lower MBS yields and thereby reduce primary mortgage
rates.

         3. Liquidity Enhancement – An Effect of the QE Programs

The Federal Reserve’s purchases of MBS created liquidity in the marketplace. Liquidity
refers to an entity’s ability to sell or dispose of its assets for cash. In general, investors are
willing to pay a liquidity premium for a security that remains easy to sell. To a certain extent,
the Federal Reserve’s programmed purchases of MBS served to assure investors that it would
continue to purchase MBS even during distressed times. This, in turn, increased the liquidity
of these securities and caused them to become more valuable as a result.




46
  Frequently, bond yields and prices move in anticipation of upcoming Federal Reserve actions and, at times,
even before such actions are announced by the Federal Reserve.
47
  Conversely, if financial assets are seen as interchangeable, then any price and yield effects from the Federal
Reserve’s purchase would be minimal, as investors would be indifferent about exchanging one asset for
another.
48
  Duration is a measure of the sensitivity of the price of a bond to a change in interest rate. Convexity is a
measure of the curvature in the relationship between bond price and interest rate that demonstrates how the
duration of a bond changes as the interest rate changes.




                                    OIG  EVL-2015-002  October 23, 2014                                          26
         4. Concurrent Targeting of Treasury Securities to Lower Interest Rates

Throughout the QE era the Federal Reserve has continued to purchase long-term Treasury
securities. Due mainly to their perceived risk-free nature, Treasury securities are considered
to be one of the most liquid fixed-income securities in the world.49 Consequently, other bonds
are frequently priced based upon their riskiness relative to Treasury securities; their prices are
quoted as a spread between their yields and those of Treasury securities. The wider the spread
between the security’s yield and that of the comparable Treasury security, the riskier it is
perceived to be. All else equal, when the yields on long-term Treasury securities decline as a
result of the Federal Reserve’s purchases, then the yields on other bond securities, such as
MBS, would be expected to decline as well.

     A Summary of Current Economic Research Into the Effectiveness of the QE Programs

The economic research that we analyzed generally supported the view that the QE programs
achieved their intended impact, i.e., lowering interest rates, including MBS yields and primary
mortgage rates. However, researchers disagreed on the magnitude of these impacts. The
following summarizes this research:

        Hancock and Passmore focused on the effect of the QE programs on MBS yields.
         They showed that MBS yield levels decline when the Federal Reserve holds
         substantial amounts of the Enterprises’ securities. In mid-2013, the Federal Reserve
         held about 24% of all MBS, which was equivalent to about a $1.21 trillion portfolio.
         Hancock’s and Passmore’s analysis suggests that the cumulative effect of the MBS
         purchases alone had lowered the MBS yields by 55 basis points.50 Furthermore, they
         found that an increase in the Federal Reserve’s holdings of available Treasury
         securities also lowered the MBS yield.

        Krishnamurthy and Vissing-Jorgensen examined the entire QE period. They showed
         that the large-scale purchases by the Federal Reserve had lowered the MBS and
         Treasury yields as intended, though the effects were much more pronounced during
         QE I than the later QE programs.51 Figure 12, below, captures the immediate changes


49
  The financial markets perceive the odds of the U.S. Federal government defaulting on Treasury securities to
be nearly zero.
50
  Diana Hancock and Wayne Passmore, How the Federal Reserve’s Large-Scale Asset Purchases Influence
Mortgage-Backed Securities (MBS) Yields and U.S. Mortgage Rates, Federal Reserve Working Paper, 2014-12
(February 2014).
51
   Arvind Krishnamurthy and Annette Vissing-Jorgensen, The Ins and Outs of LSAPs, Federal Reserve Bank
of Kansas City (September 16, 2013).




                                  OIG  EVL-2015-002  October 23, 2014                                         27
           in MBS and 10-year Treasury yields around the announcement dates of the QE
           programs.52

               FIGURE 12. CHANGES IN ASSET YIELDS AROUND EVENT DATES (IN BASIS POINTS)

                                               QE I               QE II             Twist             QE III
         10-Year Treasury                      –107               –18                 –7               –3
         30-Year MBS Yield                     –107               –12                –23              –15
Source: Arvind Krishnamurthy and Annette Vissing-Jorgensen, The Ins and Outs of LSAPs, Federal Reserve
Board Working Paper, at 12 (September 16, 2013).

          On the other hand, Stroebel and Taylor focused only on the purchases of MBS by the
           Federal Reserve during the QE I era and measured mortgage rates in terms of their
           spreads over popular benchmarks, such as the London Interbank Overnight Rate swap
           curve and the Treasury yield curve. They concluded that although the mortgage
           spreads may have declined by a statistically significant amount of 30 basis points
           during QE I, a sizable portion of the decline likely could be attributed to concurrent
           declines of default and prepayment risks, rather than the QE program itself.53

          Kaminska and Zinna analyzed the Federal Reserve’s long-term Treasury securities
           purchase operations from 2008 through November 2012, just after the start of the QE
           III program. They found that the QE program had been effective.54 In particular, they
           estimated that in the absence of Federal Reserve purchases, the 10-year real yields of
           Treasury securities would have been higher by as many as 140 basis points.

          Finally, FHFA’s working paper concluded that, as intended, the QE program affected
           long-term interest rates and mortgage rates, with mortgage rates lower than they would
           have been without the QE intervention and reaching historical lows in the post-crisis
           era.55



52
   The total changes could be even higher than the changes illustrated in Figure 15, as some of movements in
yields could take place prior to the Federal Reserve’s announcements of program initiation due to the
anticipation by the markets.
53
   The position of the Federal Reserve is that the declining default and prepayment risks during the QE period
were not coincidental. Rather, the declining risks were at least partially the result of the Federal Reserve’s own
large-scale purchases of mortgage assets.
54
   Iryna Kaminska and Gabriele Zinna, Official Demand for U.S. Debt: Implications for U.S. real interest
rates, IMF Working Paper (April 2014).
55
  Saty Patrabansh, William M. Doerner and Samuel Asin, The Effects of Monetary Policy on Mortgage Rates,
FHFA Working Paper 14-2 (June 2014).




                                   OIG  EVL-2015-002  October 23, 2014                                             28
    The QE Programs’ Influence on Mortgage Refinance and Home Purchase Activity

The QE programs were intended to support the weak housing market in the United States by
lowering mortgage rates and thereby spurring increased mortgage originations.

Mortgage origination volume is comprised of two types of borrowing activities: new home
purchases and existing home refinances. Lower interest rates typically lead to increases in
both types of borrowing, although the effect on refinances is greater than on new home
purchases.

In summary, this is the case because the prevailing interest rate is the most significant factor
in the decision matrix of a borrower who wants to refinance; but it is only one factor among
many in the matrix of a borrower who wants to purchase a new home. Thus, in the absence of
non-economic external factors, e.g., asset division as a result of a divorce, a borrower will not
refinance unless it results in a lower monthly mortgage payment. On the other hand, the
decision to purchase a new home usually involves many other important economic and non-
economic factors, including the stages of life, overall financial situation, personal preferences,
etc.

There is also empirical support for different levels of effects of interest rate movement on
refinances and new purchases. Using historical data, CBO estimates that every 25 basis point
reduction in the prevailing interest rate results in a 5.2% increase in total mortgage origination
over the course of the following year. See Figure 13 below.

             FIGURE 13. MORTGAGE ORIGINATION VOLUME AND INTEREST RATE CHANGE

                                                                                   Weighted-Average
       Interest Rate Change           New Purchase              Refinance               Total
     10 BPS Reduction                    0.6%                     3.7%                   2.0%
     25 BPS Reduction                    1.3%                     9.8%                   5.2%
     50 BPS Reduction                    2.7%                    19.6%                  10.4%
Source: FHFA-OIG’s derived estimates based on separate estimates provided by the Congressional Budget
Office, Fannie Mae, and Freddie Mac.

When broken down further by types of origination activity, the 5.2% increase represents a
9.8% increase in refinance volume and a weighted-average 1.3% increase in new purchase
volume. Accordingly, the impact on the refinancing activity is more than seven times as
much as the impact on the new purchase activity.

Separate independent sensitivity estimates provided by Enterprise officials are similar in
terms of the relative scale. Numerically, the data confirms that interest rate reduction has
a much greater impact on refinance volume than on new purchase volume.



                                OIG  EVL-2015-002  October 23, 2014                                   29
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 Investigation – Hotline
                400 Seventh Street, S.W.
                Washington, DC 20024




                              OIG  EVL-2015-002  October 23, 2014                        30