20 THE NEW” HOUSING AND MORTGAGE MARKET SPRING 2016The .docx

20 THE “NEW” HOUSING AND MORTGAGE MARKET
SPRING 2016
The New Housing
and Mortgage Market
DOUGLAS DUNCAN
DOUGLAS DUNCAN
is chief economist and
a senior vice president
at Fannie Mae in
Washington, DC.
[email protected]
com
O
ne hears various individuals
ask whether the housing and
mortgage markets are back to
“normal,” or perhaps they con-
jecture that the markets are, in fact, back to
“normal.” Of course, that question implies an
understanding of what constitutes “normal.”
Others suggest there is a “new normal,”
which indicates a view that what was, is no
longer, and that the market has somehow
permanently changed. We will explore that
dichotomy of views in this brief article.
Our primary interests in this article
are in the production and delivery of and

investment in mortgage-related assets as well
as exploring what has changed and what the
future looks like in this market. Because the
number and volume of those assets are deriv-
ative of the underlying real estate, we will
also brief ly describe the U.S. demographic
profile that will drive demand for places to
live. People live in residences that they own
or rent and both are f inanced, so we will
comment on both types of property and what
brings people to live in one or the other.
Finally, we will offer a perspective on what
this means for mortgage asset volumes.
The next subject we will comment
upon is the organization of firms that make
mortgage loans to consumers in the primary
market. A number of post-crisis economic
and policy forces have been acting on these
f irms and changing the opportunities and
constraints they face. The environment has
altered the product set they offer. We offer
a view of how the demographic factors and
the implied potential mortgage-related asset
volumes might look going forward and how
they are likely to impact the number and type
of firms operating in the primary market.
The number and nature of firms oper-
ating in the secondary market have changed
significantly, as well. From a policy perspec-
tive, however, this is the area of least progress.
Irrespective of the lack of legislated change,
there are changes taking place in the sec-
ondary market under the direction of the

conservator.1 The primary market has seen
a shift of volume between traditional f irm
types, but the secondary market awaits poten-
tially greater structural change. This change
includes the mix of investors who ultimately
hold the mortgage assets as well as the types
of assets available to be held.
Much of the change to be discussed is
a result of the policy reaction to the housing
recession. The policy changes were both
monetary and fiscal. The drivers of change
also include what might be called the evo-
lutionary aspects of any market, perhaps
enabled in this case by technologic advance-
ment. We will not discuss the causes of the
recession but rather focus on the changes
wrought by the policy response to it. Not
all changes have been determined as of this
writing, and institutions and markets are still
THE JOURNAL OF STRUCTURED FINANCE 21SPRING
2016
reacting to the initial set of policy adjustments and the
ongoing changes driven by technology.
A host of questions can be asked in considering the
combined effect of the changes in each of these subject
areas. What volume of mortgage assets will be pro-
duced? Who will produce them? What will their origins
be? What form will they take? Who will hold them?
What type of returns will be expected? This article is
too brief to answer all these questions exhaustively or

perhaps even many of them. It will suggest, however,
a number of things researchers should explore and that
policymakers, market participants, or observers should
take into account as they assess the opportunities and
risks.
DEMOGRAPHICS DRIVE HOUSING
AND MORTGAGE DESTINY
Traditional housing and mortgage drivers are
reemerging post-crisis; population, household formation,
and lifecycle included. People have always lived in a struc-
ture built upon land somewhere in proximity to where
they worked, and it will always be thus. Some rent and
some own.2 The determinants of which choice a household
makes include stage of life, personal preference, financial
capacity and performance, tax considerations, supply of
property by type, and relative cost, among others. Being
married and having a child are strongly correlated with
becoming a home owner. Steady employment and earn-
ings growth combined with good credit management are
general preconditions for qualifying for mortgage credit.
Some people who are able to own choose to rent. So,
what are the demographic prospects?
The U.S. demographic profile suggests significant
growth for housing and mortgage assets as the genera-
tion reaching adulthood in the early 2000s, the Millen-
nials, age, as seen in Exhibit 1. Millennials are greater
in number than Baby Boomers; see Exhibit 2.3 Survey
data indicate over 90% of Millennials have a desire to
own. Homeownership varies significantly by age, with
the group of 30- to 34-year-olds being prime first-time
homebuyers and the homeownership rate peaking when
households are in their mid-60s.

Currently, the 30–34 group’s homeownership rate
lags prior cohorts, very likely as a result of the weakness
of the economic recovery because their real incomes are
still well below that of the preceding cohort at the same
age a decade earlier (see Exhibit 3).
E X H I B I T 1
Millennials Are a Large Wave of Potential Home Owners
Source: U.S. Census Bureau: Decennial Census.
SPRING 2016
Many Millennials who are forming house-
holds are renting single-family homes, which
suggests they will align their housing tenure with
their expressed interest when they are financially
capable.4 Inability to get a mortgage is only the
fourth-ranked reason for renting now.5 There is
a clear financial conservatism among younger
households, driven partly by their observations of
the effects of the severe recession and partly by the
slow pace of employment and income growth in
the recovery. Among Millennials, those age 25–34
have always said that the lifestyle benefits are the
best reason to buy a home rather than the financial
benefits, but there is some indication that lifestyle
benefits have been trending down while finan-
cial benefits have been trending up. This trend
is paired with expectations of price appreciation
becoming more aligned with long-term trends, as
illustrated in Exhibit 5.6

Baby Boomers are not driving demand for
rentals at this point, but they are so numerous
that many are, nevertheless, renters.7 Boomers
express a desire to age in place and are remod-
eling their existing homes contrary to expecta-
tions that they would sell their current home and
move to a smaller place after the children exit.8
It remains to be seen how long this stays true as
they age. Disability increases sixfold in the age
range of 65–74 and for those 75 and over. Second-
home purchases have risen, and speculation is that
eventually these households will sell their current
primary home and move their residence to the
smaller second home.
The bottom line on demographics for the
short and intermediate term is that Millennials will
supplant Baby Boomers as the largest age cohort.
Baby Boomers are aging in place, meaning there
will need to be maintenance of existing structures
in addition to increasing the housing stock. The
balance between owning and renting as tenure
choice has returned to a long-term relationship.
THE SUPPLY RESPONSE LAGS
Single-family (1–4 units in the building)
rentals account for 53% of all renter-occupied
units, up from 51% prior to the recession. The
remainder of rentals are 31% in buildings with
E X H I B I T 3
Millennials Have Lower Household Incomes than GenXers
Had at the Same Age

E X H I B I T 2
Millennials Are the Largest Generation in History
Source: U.S. Census Bureau: Decennial Census—Population
Estimates, and Popu-
lation Projections.
Sources: U.S. Census Bureau, 2000 Census, and 2013 American
Community Survey.
2016
5–49 units, 11% in buildings with 50 units or more,
and 5% in manufactured homes and other less common
types of structures.9
The presence of institutional investors in the
single-family rental business is an unusual feature of the
current housing market born from the large
excess supply revealed by the crisis and the
subsequent large price decline. It is unknown
what the ultimate implications are of institu-
tional investors’ participation, but the market
segment seems to have staying power at least
into the intermediate term. Technolog y
seems to be having an inf luence in reducing
costs of managing geographically dispersed
properties.
As seen in Exhibit 6, the number of
multifamily starts per 1,000 households has
expanded at a good clip, running at about its

pre-crisis levels, in response to very strong
rental demand. Despite a positive overall
supply response, affordability in rental housing
remains a concern because much of the new
construction is in Class A properties that
require higher rents. There is potential for
over-building of Class A properties in some
local submarkets existing simultaneously with
a lack of Class B and Class C properties with
more affordable rents.
The supply of single-family homes for sale is lag-
ging and causing real house price appreciation in the
presence of rising demand as employment and income
grow. Construction is running at a pace well below
E X H I B I T 4
Millennials Have Always Seen Lifestyle Benefits as the Best
Reason to Buy a House
Source: Fannie Mae National Housing Survey.
E X H I B I T 5
Average 12-Month Home Price Change Expectations Have
Declined from Their Recent Peak
Source: Fannie Mae National Housing Survey.
SPRING 2016
long-term levels (see Exhibit 7). As in the rental market,
supply is particularly lagging in the lower price home
categories (see Exhibit 8).10 This is evident in the pace of

price appreciation by house price tier nationally as well
as in selected markets (see Exhibits 9 and 10).
The cause of the weak response in single-family
construction is not completely understood. Contrib-
uting factors include the lack of skilled workers;
reduced availability of acquisition, development,
and construction (ADC) credit; reduced supply
of developed lots; and high cost of developing
lots, which puts prof itable home building at
price points that don’t fit traditional “affordable”
income levels.
Home price growth and rent growth
vary by locality. On the national level, we can
gauge their relative growth rates by looking
at the price-to-rent ratio. Since around mid-
2012, home price appreciation has outpaced rent
growth, which is ref lected in the increase in
the price-to-rent ratio (see Exhibit 11).11 Income
growth trailing home price appreciation hurts
home purchase affordability, and strong rent
growth also makes it harder for households to
accumulate the down payments required to pur-
chase a home. As noted earlier, single-family
construction for sale-to-own properties still
lags the level suggested by demographics, and it
would seem that, in the absence of a recession,
it will take approximately three years to achieve
that level.12, 13
THE PRIMARY MORTGAGE MARKET
SHIFTS
One demographic factor already starting

to have an impact on the real estate and mort-
gage finance business is consumer attitudes about
the application of technology to the search pro-
cess. Survey data show that consumers who had
deployed online shopping practices are strongly
interested in shifting that to mobile technology
applications.14 This demand is showing up in
the financial technology (FinTech) investments
being made around the globe.15 Several competi-
tors have emerged in the real estate listing and
search business, and many more are building
tools in the consumer finance space. There is an
interesting dichotomy at present in the mortgage com-
ponent in that, while consumers are focused on search
and comparison capability improvement for both real
estate and its f inancing, existing lenders cite process
efficiency as the basis for their technology investment,
as seen in Exhibit 12.
Source: U.S Census Bureau.
E X H I B I T 7
Single-Family Housing Supply Still behind the Curve
Source: U.S Census Bureau
E X H I B I T 6
Multifamily Construction Picks Up the Pace Post-Crisis
2016
Mortgage lenders face a series of challenges,

particularly on the single-family-home side of
the business. These challenges include adopting
technology tools to meet changes in consumer
behavior as well as a changed regulatory envi-
ronment that has increased the costs of compli-
ance and the end of a policy-induced refinance
driven market. As illustrated in Exhibit 13, data
from the Mortgage Bankers Association show
a clear increase in the compliance component
of operations costs in both loan production and
servicing subsequent to the passage of the Dodd–
Frank Act and the related regulatory changes.
The expectation is that if mortgage volume
falls, there will be firms exiting the business because
the base operating cost has raised the minimum
size at which a firm can successfully operate.
Thus, the recession, housing, and mortgage
market downturn, and related financial market
crisis led initially to consolidation in the industry,
but as the legislative and regulatory response took
shape, the industry has migrated toward a decon-
solidation. Mergers and consolidation among large
depository institutions increased the market shares
of banks initially. As capital rules shifted, legal set-
tlement costs accumulated and regulatory burden
increased, as Exhibit 14 shows, volumes started to
shift toward smaller non-depository lenders.
This migration has taken place in the pres-
ence of a shift in product type and purpose. Mon-
etary policy has been focused, in part, on lowering
nominal interest rates for the purpose of allowing
households to refinance their existing mortgages
and improve household financial stability. Addi-

tionally, the low rates brought buyers, particularly
at higher income levels, into the market to put a
f loor under falling house prices and preserve any
wealth effect related to housing equity wealth.
Monetary policy supported very high levels
of refinance volumes as did the distressed housing
policy initiatives, Home Affordable Modification
Program (HAMP), and Home Affordable Refi-
nance Program (HARP). See Exhibit 15.16
Although the modification programs have
reset provisions that will allow for loan rates to rise
if market rates rise, there are caps on the adjust-
ment that should keep rates at low levels histori-
cally. As these programs were progressing and now
Note: Tier 1: 0–75% of median; Tier 2: 75% –100% of median;
Tier 3:
100% –125% of median; Tier 4: 125% + of median.
Source: CoreLogic.
E X H I B I T 8
Lack of More Affordable Properties
E X H I B I T 9
Continuing to See Faster Home Price Appreciation among
Moderately Priced Homes
Note: Tier 1: 0–75% of median; Tier 2: 75% –100% of median;
Tier 3:
100% –125% of median; Tier 4: 125% + of median.
Source: CoreLogic.

SPRING 2016
approach their end, the underlying home
purchase mortgage volumes have picked
up, although not enough to offset the
decline in refinance activity.
In addition, the product mix
between government, Federal Housing
Administration (FHA) and Veterans
Administration (VA), and conventional,
all non-government loans, changed. The
changes were driven by several factors,
including relative prices of mortgages in
the two components of the market as FHA
reduced its up-front insurance premium.
There have also been changes in the mix
of borrowers, particularly the entry into
the market of large numbers of military
veterans from the first and second Gulf
Wars, thus growing the VA component
of government loans. Because there are no
hard limits on loan size for VA loans, quali-
fied borrowers can refinance from one VA
loan to another VA loan or purchase and
finance a move-up home as well.17 Mean-
while, the FHA appears to be seeing some
volume increases from borrowers who lost
a home previously and are returning to the
market through the FHA’s less-stringent
loan qualification standards.
Stabilized and subsequently rising
home prices meant the staunching of
declines and, ultimately, restoration of

increases in housing equity wealth. The
number of households that owe more on
their home than it is currently worth has
fallen steadily, and the number of house-
holds that have housing equity wealth
available has been increasing. The
expectation is that having low f ixed-
rate, f irst-lien mortgages in a market
expecting rate increases will enhance
the prospects for home equity loan prod-
ucts, but increased conservatism among
owning households regarding the sta-
bility of that equity may imply lower
take-up rates for move-up buying and
equity products. For households with
significant housing equity but low levels
of non-housing equity wealth to draw
E X H I B I T 1 0
Most Metro Areas See Faster Price Appreciation for More
Modest Homes
Source: S&P/Case-Shiller.
Sources: U.S. Bureau of Labor Statistics, FHFA.
E X H I B I T 1 1
Home Prices Rising Faster than Rents
2016
Sources: Fannie Mae Mortgage Lender Sentiment Survey,
National Housing Survey.

E X H I B I T 1 2
Lender and Borrower Mobile Priorities Differ
Source: Mortgage Bankers Association: Quarterly Mortgage
Bankers Performance Report, Servicing Operations Study and
Forum.
E X H I B I T 1 3
Compliance and Servicing Costs Have Grown Since the Dodd–
Frank Act
SPRING 2016
on, however, the potential for growth in the reverse
mortgage product line seems strong.
The level of single-family mortgage debt out-
standing has only recently begun to rise after a sig-
nificant period of moderate decline due to foreclosures
and household deleveraging; see Exhibit 16.18 Although
mortgage origination volumes were high for sev-
eral years, the refinancing volume that composed the
majority of production for several years simply ref lected
churn in the portfolio and, in fact, enhanced the poten-
tial for shortening the maturity of loans and accelerated
extinguishment of the debt altogether.
Foreclosure levels have fallen back to pre-crisis
levels in most states, although the states that have judi-
cial foreclosure laws are still experiencing elevated
but declining levels of distressed loans. This has been
aided by the rise in home prices, which has reduced the

number of home owners who owe more on their home
than it is currently worth.19
The apartment loan market has seen steady volume
growth post-crisis as overall employment has recovered
and builders have expanded production to meet the rise
in apartment demand accompanying the increase in
household formation. Multifamily annual loan volume
has risen steadily as construction has increased, reaching
$199 billion in 2015 after falling to a reces-
sion low of $49 billion in 2009. The largest
single sources of funding have been the gov-
ernment-sponsored enterprises (GSEs), as
Fannie Mae financed $42 billion and Freddie
Mac financed $47 billion in 2015. The FHA
has also been a key funding source, providing
$18.5 billion in 2015.
Overall, the primary market is set to
see growth in home purchase mortgages
and declining refinance activity as mort-
gage interest rates level off or rise from cur-
rent levels. Costs of doing that business have
risen, and while technological improvements
may produce some compliance efficiencies,
the cost increase suggests that the minimum
profitable loan size will be somewhat higher
in the future. As many borrowers have locked
in low fixed-rate funds, the growth of equity
suggests home equity or home equity lines of
credit may see some growth. Within aging
households that have housing equity but low
income, the use of reverse mortgages is likely
to rise. Multifamily debt growth will likely slow over the

midterm, with some potential for a decline in loan per-
formance as overbuilding in some segments and in local
markets is a possibility.
THE SECONDARY MORTGAGE MARKET
ALSO SHIFTS
The secondary market for whole loans and mort-
gage-related securitized products has seen both institu-
tional structural change and investor changes, although
the extent to which one could argue the transformation
is cyclical will play out against the backdrop of these
changes. Key components of the institutional structural
changes are the disappearance of private-label mortgage
security (PLS) issuers and associated securities, the rise of
Ginnie Mae from a volume perspective relative to the two
GSEs, the issuance of credit risk transfer (CRT) securities
by the GSEs, and a shift in how depository institutions
manage whole loan portfolios. See Exhibit 17.
Key components of the investor changes, in addition
to increased whole loan retention at depository institutions,
have been the mandatory declines in the GSE portfolio
holdings, the increase in the U.S. Federal Reserve port-
folio holdings, the support of private investors for the CRT
Sources: Inside Mortgage Finance, Fannie Mae, Freddie Mac,
Ginnie Mae, HMDA, Mar-
ketrac, SNL Financial.
E X H I B I T 1 4
Total Originations—Institution Type Share Shifts to Smaller
Independent Lenders

2016
Source: Treasury Department.
E X H I B I T 1 5
High Levels of Refinancing and Modifications through HARP
and HAMP
SPRING 2016
securities issued by the GSEs, and the return to health of
the private mortgage insurance companies as risk-sharing
entities in the GSE market space (see Exhibit 18).
Private-label mortgage security issuance has been
negligible as legacy securities amortize, with unfavor-
able market conditions leading to a reduction in supply
and liquidity. Concurrently, many investors remain on
the sidelines as unresolved issues in this sector prevent
an accurate pricing of the risk–return tradeoff, and thus,
overall demand has weakened.
At the same time, the market share of Ginnie
Mae increased dramatically, although total MBS issu-
ance declined post-2007 compared with the 2002–2007
time period. The decline in issuance in 2008 was during
the most intense period of the crisis, and the subsequent
rise in issuance from 2009–2013 was the period of most
significant direct policy interventions through specific
mortgage programs at the Federal level and of central
bank interventions to drive rates down and support
house price stability. The period from 2014 through

E X H I B I T 1 6
Mortgage Debt Outstanding and Originations
Sources: U.S. Federal Reserve, Fannie Mae estimates.
2016
2015 represents the slowdown from the monetary policy
support for refinancing and the increasing strength of
the home purchase market.
Portfolio whole loan holdings have risen largely
as a result of income and wealth dynamics post-crisis.
High-income households saw increasingly rapid
growth rate for incomes and faster wealth
accumulation as a result of monetary and
fiscal policies leading depository institu-
tions with wealth management motives to
incorporate mortgage-related debt instru-
ments in their cross-sell product offerings.
This component of the investor base may
be nearing capacit y, and commercial
banks as a group have a long history of
holding whole loan mortgage-related
assets in a narrow band as a share of total
outstanding mortgage debt.20
CRT securities are a market innova-
tion of recent vintage. They are intended
as a vehicle to reduce risk for the GSEs
by sharing it with private investors, thus

reducing the potential taxpayer contingent
liabilities inherent in the conservatorship
status of the GSEs. This market is small
but growing as the market considers the
attributes and performance of the instru-
ments (see Exhibit 19). The securities have
not performed across a full economic cycle,
so the data on cyclical performance are yet
to be acquired, and therefore, pricing is
immature in that sense.
Throughout, the multifamily compo-
nent of the commercial mortgage-backed
secur it y (CM BS ) market per for med
steadily. Total issuance followed the pat-
tern of consumers moving to homeown-
ership for the decade through 2005 and
then, post-crisis, the shift to rebalance
between homeownership and renting.
Volumes have risen steadily post-crisis,
reaching more than $210 billion in 2015.
Expectations for 2016 are that it will be
the strongest year on record and with per-
haps another two or three years of growth
before leveling off.21 One component of
the issuance, the rollover of maturing
loans held by nonbanks, should accelerate
through the 2016–2017 period before roughly f lat-
tening out for the early 2020s, as shown in Exhibit 20.
Somewhere in that time period, there is a possibility
of a recession, given that during 2016, the economy
will be in the fourth longest economic expansion since
World War II.
Source: Inside MBS and ABS.

E X H I B I T 1 8
Agency MBS Investor Breakdown Shows Federal Reserve
Dominance
E X H I B I T 1 7
Mortgage-Related Securities Issuance Has Trended Down
Sources: Fannie Mae, NYSE, Inside Mortgage Finance.
SPRING 2016
considerations are 1) the decisions of the
Federal Reserve regarding the conduct
of monetary policy, including both the
“normalization” of interest rates and the
effects of its decisions regarding its hold-
ings of mortgage-related securities, and 2)
the reform of the secondary market insti-
tutional structure, including the GSEs.
The current mortgage-related assets
component of the Fed’s portfolio is larger
than the combined decline in the portfo-
lios of the GSEs to date. This is impor-
tant for at least two reasons. First, the GSE
portfolios are still in decline and will be
capped at a maximum of $250 billion in
2018.22 Therefore, under policy scenarios
involving the run-off of the Fed’s port-
folio, the GSEs will not be an acquiring
investor. Thus, it raises questions regarding
who will be the investors that are likely to

take the Fed’s place. Second, given that the
Fed purchased MBS for monetary policy
objectives rather than economic returns,
it is unknown how the Fed’s exit would
change private investors’ views on MBS
volume and spreads. These unknown
variables will affect mortgage rates to bor-
rowers in the primary market as well as the
subsequent quantity of credit demand.
The Federal Reserve is also gradu-
ally moving toward a more “normal”
posture for monetary policy, having insti-
tuted its first Federal funds rate increase in
nine years in December 2015. This casts
U.S. monetary policy in juxtaposition to
global central banks that have instituted
negative short-term nominal interest rates
policies. The implications of negative
rates over any timeframe are unknown,
being an historical anomaly. While the
U.S. central bank is resistant to this policy
choice, it must be considered by domestic
and global market participants as it will, by definition,
alter the information contained in market prices.
Questions also surround the secondary market
institutional structure. While the conservatorship of the
GSEs continues, there are potential market structural
shifts under construction in addition to the credit risk
POLICY CONSIDERATIONS DRIVE
THE NEW IN THE MARKET
Finally, there are a number of policy considerations

to take into account for their potential impacts on volume,
composition, rates, and spreads of mortgage-related assets
in the near and far future. The two most important
Note: The 2015 originations estimate is subject to HMDA
revisions.
Source: Fannie Mae.
E X H I B I T 1 9
Credit Risk Share of Originations
Source: Mortgage Bankers Association.
E X H I B I T 2 0
Non-Bank Multifamily Loan Maturities by Investor Type
2016
transfer initiatives. Changes in Federal Reserve policy
will have an impact on the performance of the CRT
market, which the GSEs now support. Presumably, the
potential reform will consider the existence of the CRT
market and the implications of any reform regarding the
potential for stranding that market component. The addi-
tional mechanisms under construction are the Common
Securitization Platform and the Single Security structure.
A great deal has been written about the nature and poten-
tial of these innovations, and we will not address them
directly here. We would note, however, that they do hold
the potential for changing the competitive structure of
the secondary market with implications for the nature of
mortgage-related assets and, correspondingly, the rates

and spreads inherent in the new instruments.
This in turn raises the question of the ultimate fate
of …
NBER WORKING PAPER SERIES
CONCENTRATION IN MORTGAGE LENDING,
REFINANCING ACTIVITY AND
MORTGAGE RATES
David S. Scharfstein
Adi Sunderam
Working Paper 19156
http://www.nber.org/papers/w19156
NATIONAL BUREAU OF ECONOMIC RESEARCH
1050 Massachusetts Avenue
Cambridge, MA 02138
June 2013
We are grateful to Zahi Ben-David, Scott Frame, Andreas
Fuster, Ed Golding, David Lucca, Amit
Seru, Jeremy Stein, Amir Sufi, and seminar participants at the
Federal Reserve Bank of New York,
Harvard University, the NBER Corporate Finance Spring
Meetings, and the UCLA/FRB - San Francisco
Conference on Ho using and the Macroeconomy for helpful
comments and suggestions. We thank
Freddie Mac for data and Toomas Laarits for excellent research
assistance. We also thank the Harvard
Business School Division of Research for financial support. The

views expressed herein are those
of the authors and do not necessarily reflect the views of the
National Bureau of Economic Research.
NBER working papers are circulated for discussion and
comment purposes. They have not been peer-
reviewed or been subject to the review by the NBER Board of
Directors that accompanies official
NBER publications.
© 2013 by David S. Scharfstein and Adi Sunderam. All rights
reserved. Short sections of text, not
to exceed two paragraphs, may be quoted without explicit
permission provided that full credit, including
© notice, is given to the source.
Concentration in Mortgage Lending, Refinancing Activity and
Mortgage Rates
David S. Scharfstein and Adi Sunderam
NBER Working Paper No. 19156
June 2013
JEL No. E44,E52,G21,G23,L85
ABSTRACT
We present evidence that high concentration in local mortgage
lending reduces the sensitivity of mortgage
rates and refinancing activity to mortgage-backed security
(MBS) yields. A decrease in MBS yields
is typically associated with greater refinancing activity and
lower rates on new mortgages. However,
this effect is dampened in counties with concentrated mortgage
markets. We isolate the direct effect
of mortgage market concentration and rule out alternative

explanations based on borrower, loan, and
collateral characteristics in two ways. First, we use a matching
procedure to compare high- and low-concentration
counties that are very similar on observable characteristics and
find similar results. Second, we examine
counties where concentration in mortgage lending is increased
by bank mergers. We show that within
a given county, sensitivities to MBS yields decrease after a
concentration-increasing merger. Our results
suggest that the strength of the housing channel of monetary
policy transmission varies in both the
time series and the cross section. In the cross section,
increasing concentration by one standard deviation
reduces the overall impact of a decline in MBS yields by
approximately 50%. In the time series, a
decrease in MBS yields today has a 40% smaller effect on the
average county than it would have had
in the 1990s because of higher concentration today.
David S. Scharfstein
Harvard Business School
Baker 239
Soldiers Field
Boston, MA 02163
and NBER
[email protected]
Adi Sunderam
Baker Library 245
Harvard Business School
Boston, MA 02163
[email protected]
1

I. Introduction
Housing is a critical channel for the transmission of monetary
policy to the real economy.
As shown by Bernanke and Gertler (1995), residential
investment is the component of GDP that
responds most strongly and immediately to monetary policy
shocks. In addition, housing is an
important channel through which monetary policy affects
consumption. An easing of monetary
policy allows households to refinance their mortgages at lower
rates, reducing payments from
borrowers to lenders. If borrowers have higher marginal
propensities to consume than lenders, as
would be the case if borrowers are more liquidity constrained,
then refinancing should boost
aggregate consumption in the presence of frictions. Indeed,
refinancing is probably the most
direct way in which monetary policy increases the disposable
cash flow of liquidity-constrained
households (Hurst and Stafford 2004).
Using monetary policy to support housing credit has been an
increasing focus of the
Federal Reserve in recent years. In particular, the Federal
Reserve’s purchases of mortgage-

backed securities (MBS) in successive rounds of quantitative
easing have had the explicit goal of
supporting the housing market. One of the aims of quantitative
easing was to lower mortgage
rates by reducing financing costs for mortgage lenders
(Bernanke 2009, 2012). However, it has
been argued that the efficacy of this policy has been hampered
by the high indebtedness of many
households (Eggertson and Krugman, 2012; Mian, Rao, and
Sufi, 2012). “Underwater”
households whose mortgage balances exceed the values of their
homes have been unable to
refinance, potentially reducing the impact of low interest rates
on the economy. Others have
noted that the reduction in MBS yields from quantitative easing
has only been partially passed
through to borrowers, leading to historically high values of the
so-called “primary-secondary
spread” – the spread between mortgage rates and MBS yields
(Dudley, 2012). Fuster, et al.
(2012) consider a number of explanations for the increase in
spreads, including greater costs of
originating mortgages, capacity constraints, and market
concentration, but conclude that the

increase remains a puzzle.
In this paper, we explore in more detail whether market power
in mortgage lending can
explain a significant amount of the increase in the primary-
secondary spread and thereby impede
the transmission of monetary policy to the housing sector. We
build on the literature in industrial
organization that argues that cost “pass-through” is lower in
concentrated markets than in
2
competitive markets – when production costs fall, prices fall
less in concentrated markets than
they do in competitive markets because producers use their
market power to capture larger
profits (e.g., Rotemberg and Saloner, 1987). In the context of
mortgage lending, this suggests
that when the Federal Reserve lowers interest rates, mortgage
rates will fall less in concentrated
mortgage markets than in competitive mortgage markets. This
could dampen the effects of
monetary policy in such markets.

Evidence from the aggregate time series is broadly consistent
with the idea that
concentration in mortgage lending impacts mortgage rates. As
shown in Figure 1, concentration
in the mortgage lending industry increased substantially
between 1994 and 2011. Figure 2 shows
the average primary-secondary spread calculated as the
difference between the mortgage rate
paid by borrowers and the yield on MBS for conforming loans
guaranteed by the government-
sponsored entity (GSE) Freddie Mac. 1 The yield on Freddie
Mac MBS is the amount paid to
investors in the securities, which are used to finance the
mortgages. Thus, the spread is a
measure of the revenue going to mortgage originators and
servicers. The spread rose
substantially from 1994 to 2011. Moreover, as shown in Figure
3, the spread is highly correlated
with mortgage market concentration. The correlation is 66% in
levels and 59% in changes, so the
correlation does not simply reflect the fact that both series have
a positive time trend.
Recent trends are one reason that market power has not received
much scrutiny as an

explanation for rising primary-secondary spreads. Most
recently, the spread spiked in 2011-2012
though concentration in mortgage lending has not increased
since 2010 (Avery, et al., 2012 and
Fuster, et al., 2012).2 These recent trends are misleading for
two reasons. First, they focus on the
market share of the top ten lenders at the national level.
However, evidence suggests that a
significant part of competition in mortgage lending takes place
at the local level, and at the local
level concentration is rising due to increased geographic
segmentation of mortgage lending.3
1 Specifically, Figure 2 shows the time series of the borrowing
rate reported in Freddie Mac’s Weekly Primary
Mortgage Market Survey minus the yield on current coupon
Freddie Mac MBS minus the average guarantee fee
charged by Freddie Mac on its loans.
2 Fuster, et. al. (2012) also argue that the higher fees charged
by the GSEs for their guarantees cannot account for the
rise in spreads.
3 To see this, suppose there are two identical counties where
two lenders each have a 50% market share. Then the
average county market share and the aggregate share of each
lender is 50%. However, if each lender concentrates in
a different county, the average county-level share can go to
100% while their aggregate shares remain at 50%.

3
Second, as we discuss below, in the presence of capacity
constraints, the effects of increased
concentration would be most clearly revealed when MBS yields
fall. Thus, the time series
correlation between spreads and concentration may understate
the true relationship. In this paper,
we use panel data to examine the effects of mortgage market
concentration at the county level.
Rather than focus on the level of the spread between mortgage
rates and MBS yields, we instead
study the relationship between concentration and the pass-
through from MBS yields to mortgage
rates. We provide evidence that increases in mortgage market
concentration are associated with
decreased pass-through at the county level.
Using the yield on GSE-guaranteed MBS as a proxy for the
costs of mortgage financing,
we find that mortgage rates are less sensitive to costs in
concentrated mortgage markets. A
decrease in MBS yields that reduces mortgage rates by 100
basis points (bps) in the mean county

reduces rates only 73 bps in a county with concentration one
standard deviation (18%) above the
mean. Moreover, when MBS yields fall, the quantity of
refinancing increases in the aggregate.
However, the quantity of refinancing increases 35% less in the
high-concentration county
relative to the average county. The effects on mortgage rates
and the quantity of refinancing
compound each other. In a high-concentration county, fewer
borrowers refinance, meaning that
fewer households see their mortgage rates reduced at all. And of
the borrowers that do refinance,
the rates they are paying fall less on average. The magnitude of
the combined effect is
substantial: monetary policy transmission through the mortgage
market has approximately half
the impact in the high-concentration county relative to the
average county.
Our estimates also suggest that increases in the concentration of
mortgage lending can
explain a substantial fraction of the rise in the primary-
secondary spread. Extrapolating from our
results, the 250 bps decline in MBS yields since the onset of the
financial crisis should translate

into a 150 bps reduction in mortgage rates given the current
level of concentration. This implies
that the decline in MBS yields should be associated with an
approximately 100 bps increase in
the primary-secondary spread – roughly the magnitude of the
increase observed by Fuster, et al.
(2012). Our estimates suggest that if the concentration of
mortgage lending were instead at the
4
lower levels observed in the 1990s, the same decline in MBS
yields would have resulted in a
40% smaller increase in the spread – an increase in the spread
of 60 bps rather than 100 bps.4
Of course, mortgage market concentration is not randomly
assigned, so it is difficult to
ascribe causality to these results. We attempt to address
endogeneity concerns in a variety of
ways. First, our basic results are robust to a battery of controls
including county and time fixed
effects, population, wages, house prices, and mortgage
characteristics. Moreover, we control for
the interaction of changes in MBS yields with these
characteristics. Thus, our results show that

market concentration reduces the sensitivity of mortgage rates
to MBS yields even after
controlling for the possibility that this sensitivity can vary with
county characteristics. Second,
we use a matching procedure to ensure that the counties we
study are similar on observable
dimensions. This does not affect the results.
Third, we use bank mergers as an instrument for mortgage
market concentration.
Specifically, we examine a sample of counties where mortgage
lending concentration is
increased by bank mergers, but the counties in the sample were
not the key motivation for the
merger. In particular, we focus on counties where the banks
involved in a merger are important,
but the county itself makes up only a small fraction of the
banks’ operations. Mergers increase
the concentration of mortgage lending in such counties.
However, because the county makes up a
small fraction of each of the bank’s operations, it is unlikely
that the county was an important
driver of the merger. In this sample of counties, we show that
the sensitivity of refinancing and

mortgage rates to MBS yields falls after the merger, consistent
with the idea that increased
concentration causes less pass-through. The exclusion
restriction here is that bank mergers affect
the sensitivity of refinancing and mortgage rates to MBS yields
within a county only through
their effect on market concentration in that county. For the
exclusion restriction to be violated, it
would have to be the case that bank mergers are anticipating
changing county characteristics that
explain our results, which seems unlikely.
Finally, using data on bank profits and employment, we provide
evidence consistent with
the market power mechanism being behind the lower pass-
through of MBS yields into mortgage
rates. Interest and fee income from real estate loans, reported in
the Call Reports banks file with
4 Guarantee fees charged by the GSEs have also increased in
recent years, but Fuster et al. (2012) argue that this
accounts for a relatively small part of the increase in the
primary-secondary spread.
5

the Federal Reserve, is typically positively correlated with MBS
yields because interest income
falls when yields fall. However, we show that interest and fee
income is less sensitive to MBS
yields in high-concentration counties. This suggests that banks
in concentrated mortgage markets
are able to use their market power to protect their profits when
MBS yields fall. Similarly,
employment in real estate credit is typically negatively
correlated with MBS yields; as MBS
yields fall originators hire more workers to process mortgage
applications, or there is entry in
mortgage origination. However, the sensitivity is less negative
(i.e., lower in absolute terms) in
high-concentration counties, meaning that in such counties
originators expand hiring less
aggressively in response to a decline in MBS yields, or there is
less entry. Thus, while it is true
that capacity constraints limit mortgage origination, these
capacity constraints are endogenous to
the degree of competition in the market. In all, the evidence is
consistent with the idea that
mortgage market concentration decreases the transmission of
monetary policy to the housing

sector.
Our results have both time series and the cross-sectional
implications for the
effectiveness of monetary policy. Specifically, the impact of
monetary policy could be
decreasing over time due to the increase in average mortgage
market concentration documented
in Figure 1. In addition, even in the absence of a time series
trend, monetary policy could have
different impacts across counties due to cross sectional
variation in mortgage concentration
across counties.
The remainder of this paper is organized as follows. Section II
gives some relevant
background on the mortgage market, and Section III presents a
brief model to motivate our
empirics. Section IV describes the data, and Section V presents
the main results. Section VI
concludes.
II. Background
A. The Conforming Mortgage Market
We begin with a brief review of the structure of the mortgage
market. Our analysis

focuses on prime, conforming loans, which are eligible for
credit guarantees from the
government-sponsored enterprises (GSEs), Fannie Mae and
Freddie Mac. Such mortgages may
be put into MBS pools guaranteed by the GSEs. The GSEs
guarantee investors in these MBS that
6
they will not suffer credit losses. If a mortgage in a GSE-
guaranteed pool defaults, the GSE
immediately purchases the mortgage out of the pool at par,
paying MBS investors the
outstanding balance of the mortgage. Thus, investors in GSE
MBS bear no credit risk. In return
for their guarantee, the GSE charges investors a guarantee fee.
In addition to the fees charged by
the GSEs, borrowers also pay mortgage lenders origination and
servicing fees (see Figure 4 for a
graphical depiction).
Conforming mortgages must meet certain qualifying
characteristics. For instance, their
sizes must be below the so-called conforming loan limit, which
is set by the Federal Housing

Finance Agency. In addition, borrowers eligible for conforming
mortgages must have credit
(FICO) scores above 620 and the mortgages must meet basic
GSE guidelines in terms of loan-to-
value ratios (LTVs) and documentation.
An important fact for our empirical analysis is that GSE
guarantee fees do not vary
geographically. Indeed, until 2008 the GSEs charged a given
lender the same guarantee fee for
any loan they guaranteed, regardless of borrower (e.g., income,
FICO), mortgage (e.g., LTV,
loan type), and collateral (e.g., home value) characteristics.5 In
2008 the GSEs began to charge
fees that vary by FICO score, LTV, and loan type, but do not
vary by geography or any other
borrower characteristics.6 Thus, for the loans we focus on in
our analysis, the only two
dimensions of credit quality that should materially affect rates
on GSE-guaranteed mortgages are
FICO and LTV.7,8
5 However, there is some relative minor variation in fees
charged across lenders.

6 Fannie Mae publishes their guarantee fee matrix online at:
https://www.fanniemae.com/content/pricing/llpa-
matrix.pdf
7 Loan type does not affect our analysis of mortgage rates
because we restrict our sample to 30-year fixed rate, full
documentation loans.
8 Other determinants of credit quality may have a small effect
on the rates of GSE-guaranteed mortgages due to
prepayment risk. When a GSE-guaranteed mortgage defaults,
the GSEs immediately pay investors the remaining
principal and accrued interest. From an investor’s perspective,
it is as though the loan prepays. If defaults correlate
with the stochastic discount factor, which is likely, this risk will
be priced by investors. However, since prepayments
induced by default are much smaller than prepayments induced
by falling mortgage rates, this effect will be very
small.
7
B. Definition of the Local Mortgage Market
A key assumption underlying our empirical analysis is that
competition in the mortgage
market is local. Specifically, we are assuming that county-level
measures of concentration are
good proxies for the degree of competition in a local mortgage
market. The advent of Internet-

based search platforms like Bankrate.com and LendingTree.com
has certainly improved the
ability of borrowers to search for the best mortgage terms.
However, there is substantial evidence
that many borrowers still shop locally for their mortgages.
Analyzing data from the Survey of
Consumer Finances, Amel, Kennickell, and Moore (2008) find
that the median household lived
within four miles of its primary financial institution in 2004.
They find that 25% of households
obtained mortgages from this primary financial institution,
while over 50% of households
obtained mortgages from an institution less than 25 miles away.
Moreover, borrowers report that they exert little effort in
shopping around for lower
mortgage rates. According to Lacko and Pappalardo (2007), in a
survey conducted by the Federal
Trade Commission, the average borrower considered only two
loans while shopping.9 Thus, it is
likely that local competition has effects on the local mortgage
market. Competition could affect
loan terms like rates and points charged upfront, but could also
manifest itself in other ways. For
instance, lenders may advertise more in more competitive

markets, leading to greater borrower
awareness of lower mortgage rates and increased refinancing
activity. Indeed, Gurun, Matvos,
and Seru (2013) find evidence that local advertising affects
consumer mortgage choices,
suggesting that local competition is important.
III. Model
We now briefly present a simple model of mortgage market
competition. The model is meant
to motivate our empirical analysis, and to show that many of the
results we find in the data can
be obtained in a simple model where differences in market
competition are the driving force. The
model features Cournot competition with capacity constraints
and delivers three main results.
First, the pass-through of MBS yields to mortgage rates is larger
in markets with more competing
lenders. Second, pass-through is asymmetric; mortgage rates
fall less when MBS yields fall than
9 Aubusel (1990) documents the impact of this kind of
consumer behavior on the effective level of competition in
the credit card market.

8
they rise when MBS yields rise. Third, this asymmetry
disappears as there are more competing
lenders in the market.
We assume linear demand for mortgages so that
( )p Q a bQ= −
where p(Q) is the mortgage rate corresponding to demand of Q
in the local area given this rate.
The linear demand assumption can be motivated by assuming
that there are fixed costs to
refinancing and pre-existing mortgage rates are uniformly
distributed.10 Each mortgage
originator is assumed to have pre-existing production capacity q
. When production is below the
pre-existing capacity, the only costs of mortgage production are
the costs of funding the loan,
given by the MBS yield, r. Thus, we are effectively
normalizing other production costs
associated with mortgage origination to zero given that
production is below pre-existing
capacity. However, if a lender wishes to produce more than its
pre-existing capacity, it has
increasing, convex production costs, which capture the idea that

it is costly to produce above
some capacity. For instance, one could think of these convex
costs as capturing loan officer
overtime, strain on back-office capabilities, and other short-run
costs of very high production.
Formally, production costs are given by
( )
( )
2
if
1
if
2
rq q q
C q
rq c q q q q
⎧
⎪
= ⎨
+ − >⎪
⎩
We assume Cournot competition,11 so firms solve the following

maximization problem
( ) ( )maxq p Q q C q− .
10 In particular, suppose that borrowers have existing
mortgages and that the rates on their mortgages, p0, are
uniformly distributed on the interval [x-Δ/2,x+Δ/2].
Refinancing is desirable if the new rate, p, plus transaction
costs, k, are less than the old rate, p0. Thus, the quantity of
refinancing, Q, is equal to M[1-(p+k)/ Δ], where M is a
measure of the size of the market (e.g. population). We can
therefore write the demand function, p(Q) = a – bQ,
where a=Δ-k and b = Δ/M.
11 While it is more natural to model mortgage market
competition as Bertrand, as argued by Kreps and Scheinkman
(1983), Bertrand competition with capacity constraints is
similar to Cournot competition under certain conditions.
Furthermore, the model is merely meant to be illustrative, and
Cournot competition simplifies the analysis
considerably.
9
We solve for the symmetric Nash equilibrium, labeling optimal
production of individual lenders
q* and total equilibrium production Q* = nq*.
Proposition 1. Total equilibrium production depends on the
MBS yield r and is given by

( )
( )
( )
*
*
*
if
,
if
low
high
Q r r r
Q r Q r r r
Q r r r
⎧
⎪ ⎪ ⎡ ⎤ ∈ ⎨ ⎣ ⎦
⎪
<⎪ ⎩
where

Q
low
* (r)=
a−r( )N
b N +1( )
, Q Nq= , Qhigh
* (r)=
a−r( )N
b N +1( )+c
and
( )( )1r a q b N c= − + + , ( )1r a qb N= − + .
Proof. All proofs are given in the Appendix.
The equilibrium depends on the MBS yield r. When the MBS
yield is high, the demand for
loans will be low and can be met using existing capacity. In
contrast, if MBS yields are low,
demand will be high, and lenders will add capacity to meet this
demand. For intermediate values
of MBS yields, the increase in marginal cost associated with
adding capacity is too large and
firms operate exactly at capacity.
We can now study pass-through, the sensitivity of prices and
quantities to changes in MBS

yields, in each region of the equilibrium. Since we are
interested in the behavior of pass-through
as the number of competing lenders changes, it is useful to
normalize pre-existing capacity so
that it is fixed at the industry level. Specifically, let /q Q N=
where Q is aggregate industry
capacity. Thus, as we vary N, aggregate industry capacity is
fixed but is distributed among a
larger number of lenders. Note that this normalization implies
that both r and r approach a bQ−
as N grows large; as the industry becomes very competitive, the
range of MBS yields where
lenders operate exactly at capacity vanishes.
10
The following proposition describes the aggregate sensitivities
of quantities and prices to
changes in MBS yields.
Proposition 2. Mortgage quantities rise when MBS yields fall: *
/ 0Q r∂ ∂ < . In addition,
mortgage rates fall when MBS yields fall: ( )* / 0P Q r∂ ∂ > .
Finally, these sensitivities are larger

in magnitude when there are more lenders: 2 * / 0Q r N∂ ∂ ∂ < ,
( )2 * / 0P Q r N∂ ∂ ∂ > .
When MBS yields fall, the marginal cost of lending falls.
Therefore, lenders produce more
mortgages, and the market clearing price is lower. This is true
even in the region of the
parameter space where lenders must add more capacity. If MBS
yields are low enough, the
demand for mortgages will be high enough that it is worthwhile
for lenders to add capacity. As
the number of lenders increases, each has less effective market
power, so more of the benefit of
low MBS yields is passed on to borrowers. 12
Finally, the model delivers asymmetric pass through, as the
following proposition describes.
Proposition 3. Pass-through is asymmetric. Mortgage rates are
more sensitive to MBS yields
when yields are high: ( ) ( )* */ /low highP Q r P Q r∂ ∂ >∂ ∂ .
Similarly, quantities are more sensitive to
MBS yields when yields are high: * */ /low highQ r Q r∂ ∂ > ∂ ∂
. This difference vanishes as the
number of lenders grows large.
The pass-through of changes in MBS yields is larger when
yields are high and pre-existing

capacity can be used to satisfy demand. When MBS yields are
lower, additional capacity must be
added to meet demand. The additional costs of adding capacity
mean that mortgage rates do not
fall as much as MBS yields fall. However, with more lenders,
this asymmetry vanishes. Each
lender makes a small capacity adjustment, leading to a large
increase in aggregate capacity.
The model, while simple, serves to motivate our empirical
analysis, and shows that the
intuitive link between pass through and market competition can
be formalized. Moreover, the
model underscores the link between industry capacity
constraints and mortgage market
12 It is worth noting that low pass-through can be a symptom of
high market power, but it need not be (Bulow and
Pfleiderer, 1983). The model is meant for illustrative purposes
and the results are sensitive to functional form
assumptions. Ultimately the relationship between pass-through
and market power is an empirical question.
11
competition. It shows that while capacity constraints may be
related to high spreads, the full

impact of the capacity constraints is related to the degree of
competition. In markets with few
lenders, lenders will be reluctant to add capacity to meet
increased demand for mortgages.
IV. Data
The data in the paper come primarily from two sources. The
first is the loan application
register data required by the Home Mortgage Disclosure Act
(HMDA) of 1975. The data contain
every loan application made in the United States to lenders
above a certain size threshold. Of
primary interest in this paper, the data contain information on
whether the loan application was
for a refinancing or a new home purchase, whether the loan
application was granted, a lender
identifier, as well as loan characteristics including year, county,
dollar amount, and borrower
…
2015 V43 4: pp. 993–1034
DOI: 10.1111/1540-6229.12105
REAL ESTATE

ECONOMICS
A Tale of Two Tensions: Balancing Access
to Credit and Credit Risk in Mortgage
Underwriting
Marsha J. Courchane,* Leonard C. Kiefer** and Peter M.
Zorn***
Over the years 2000–2007, mortgage market underwriting
conditions eased in
response to public policy demands for increased
homeownership. This eas-
ing of acceptable credit risk in order to accommodate increased
access to
credit, when coupled with the unanticipated house price
declines during the
Great Recession, resulted in substantial increases in
delinquencies and fore-
closures. The response to this mortgage market crisis led to
myriad changes in
the industry, including tightened underwriting standards and
new market regu-
lations. The result is a growing concern that credit standards are
now too tight,
restricting the recovery of the housing market. Faced with this
history, policy an-
alysts, regulators and industry participants have been forced to
consider how
best to balance the tension inherent in managing mortgage
credit risk without
unduly restricting access to credit. Our research is unique in
providing explicit
consideration of this trade-off in the context of mortgage
underwriting. Using
recent mortgage market data, we explore whether modern
automated under-

writing systems (AUS) can be used to extend credit to
borrowers responsibly,
with a particular focus on target populations that include
minorities and those
with low and moderate incomes. We find that modern AUS do
offer a potentially
valuable tool for balancing the tensions of extending credit at
acceptable risks,
either by using scorecards that mix through-the-cycle and stress
scorecard ap-
proaches or by adjusting the cutpoint—more relaxed cutpoints
allow for higher
levels of default while providing more access, tighter cutpoints
accept fewer
borrowers while allowing less credit risk.
Introduction
U.S. residential mortgage markets changed dramatically during
the past sev-
eral years. In the early 2000s, public policy focused on
expanding credit
access and homeownership and specifically targeted a reduction
in the home-
ownership gap between minority and non-minority households
and between
*Charles River Associates or [email protected]
**Freddie Mac or [email protected]
***Freddie Mac or [email protected]
C© 2015 American Real Estate and Urban Economics
Association
994 Courchane, Kiefer and Zorn

higher and lower income families.1 Relaxation of underwriting
standards, ac-
companied by a surge in subprime lending and an attendant
proliferation of
new products, resulted in many borrowers who could not meet
traditional un-
derwriting standards being able to obtain home mortgages and
achieve home
ownership.
However, the environment changed with the mortgage market
crisis of 2007
and 2008 when the subprime sector collapsed nearly entirely
and delinquency
and foreclosure rates increased throughout the country. In
response, underwrit-
ing standards tightened and legislation was passed imposing
more stringent
regulations on the mortgage industry, particularly the Dodd-
Frank Act reg-
ulations, which introduced both Qualified Mortgage (“QM”) and
Qualified
Residential Mortgage (“QRM”) standards. While providing
assurance that
the performance of recent mortgage originations will reduce the
likelihood of
another housing crisis, this tightening of standards comes at a
significant cost
in terms of access to credit. Balancing the tension between
access to credit
and the management of credit risk remains an ongoing concern.
The rich history of mortgage performance data over this period
offers an
opportunity to better distinguish mortgage programs and

combinations of
borrower and loan characteristics that perform well in stressful
economic en-
vironments from those that do not. The relaxed underwriting
standards of the
2000s provide plentiful performance information on borrowers
who stretched
for credit, but then experienced the stressful post-origination
environment of
declining house prices and rising unemployment. While many of
these loans
performed poorly, a large number performed well. Our goal is to
identify the
characteristics that distinguish between these two groups.
We specifically examine whether the recent data can be used to
create a mod-
ern automated underwriting scorecard that effectively and
responsibly extends
mortgage credit to the general population, and to underserved or
targeted bor-
rowers who reside in low-income communities, make low down
payments
and have poorer credit histories. Our analysis focuses on
mortgage under-
writing, rather than mortgage pricing. This reflects the two-
stage approach
to mortgage lending broadly practiced in the United States—
originators first
underwrite applications to determine whether they qualify for
origination, and
then price the loans that are originated successfully.
1For example, former United States Department of Housing and
Urban Development
(HUD) Secretary Mel Martinez states in 2002 that “The Bush

Administration is
committed to increasing the number of Americans, particularly
minorities, who own
their own homes.”
A Tale of Two Tensions 995
There are four steps necessary to complete this exercise. First,
we empirically
estimate a mortgage delinquency model. Second, we convert the
estimated
delinquency model to an underwriting scorecard for assessing
risk, where
higher scores signify higher risk. Third, we determine a
scorecard value (a
“cutpoint” or risk threshold) that demarcates the marginal risk
tolerance—
score values equal to or below the cutpoint are viewed as
acceptable risk;
score values above the cutpoint are not. Fourth, we process
borrowers through
this prototype of an automated underwriting system. We then
determine the
proportion of the population of mortgage applicants that is
within acceptable
risk tolerances, and the historic performance of these
“acceptable” loans.
The main data we use for this analysis are loan-level
observations from
CoreLogic on mortgages originated in the conventional (prime
and subprime)
and government sectors from 1999 through 2009. For each of
the three market

sectors, we separately estimate the probability that borrowers
will become
90-days or more delinquent on their loans within the first three
years after
origination. Included in the model are standard controls for
borrower and loan
characteristics, as well as for key macroeconomic factors
affecting mortgage
performance post-origination (specifically, changes in house
prices, interest
rates and unemployment rates).
Underwriting scorecards provide ex ante assessments of
mortgage risk at
origination, so creating scorecards requires appropriate
treatment of the
post-origination variables in our estimated models. Two broad
approaches are
possible. One approach attempts to forecast post-origination
variables across
borrower locations and over time. The other approach sets post-
origination
variables to constant values for all borrowers and all time
periods. We use
the latter approach. Specifically, we create two separate
scorecards. The first
scorecard sets post-origination values of house prices, interest
rates and un-
employment rates to their constant long run average levels (a
“through-the-
cycle” scorecard). The through-the-cycle scorecard is inherently
“optimistic”
with respect to credit risk, and therefore reflects a focus on
access to credit.
The second scorecard sets post-origination values of house
prices, interest

rates and unemployment rates to the varying ex post values
defined by the
Federal Reserve in an adverse scenario (a “stress” scorecard) as
defined in the
2014 supervisory stress test for very large banking
organizations.2 The stress
scorecard focuses on “tail” events that are unlikely to occur and
is meant to
prevent crisis outcomes such as those observed during the Great
Recession.
This scorecard therefore represents a focus on credit risk
management.
2See http://www.federalreserve.gov/bankinforeg/stress-
tests/2014-appendix-a.htm.
The next challenge requires choosing appropriate scorecard
cutpoints for
delimiting loans within acceptable risk tolerances. This, in
combination with
the choice of scorecard, is where much of the tension between
credit access
and credit risk resides. Higher cutpoints provide greater access
at the cost of
increasing credit risk; lower cutpoints limit credit risk but
restrict access.
As the choice of a cutpoint is a complicated policy/business
decision, we
provide results for a variety of possible cutpoints, ranging from
a low of a
5% delinquency rate to a high of a 20% delinquency rate. In an

effort to put
forward a possible compromise between access and credit risk,
we explore in
more detail results for alternative cutpoints that are market-
segment-specific;
5% for prime loans, 15% for subprime loans and 10% for
government loans.
We argue that these values represent reasonable risk tolerances
by approxi-
mating the observed delinquency rates in these segments
between 1999 and
2001.
The combination of scorecards and cutpoints creates working
facsimiles of
modern AUS, and we apply these systems to both the full and
target pop-
ulations.3 For this exercise, our “target” population is defined
as borrowers
with loan-to-value (“LTV”) ratios of 90% or above, with FICO
scores of 720
or below or missing, and who are located in census tracts with
median in-
comes below 80% of area median income. This group is
generally reflective
of “underserved” borrowers for whom there is particular policy
concern.
We find that automated underwriting, with a judicious
combination of score-
card and cutpoint choice, offers a potentially valuable tool for
balancing the
tensions of extending credit at acceptable risks. One approach
entails using
scorecards that mix the through-the-cycle and stress scorecard
approaches

to post-origination values of key economic variables. Moving
closer to a
through-the-cycle scorecard provides more focus on access to
credit. Moving
closer to a stress scorecard provides more focus on the control
of risk. The
second approach is to adjust the cutpoint—more relaxed
cutpoints allow for
higher levels of default while providing more access, tighter
cutpoints have
accept fewer borrowers while allowing less credit risk.
Previous Literature
A considerable body of research has examined outcomes from
the mortgage
market crisis during the past decade. Of particular relevance for
this research
3We weight the data using weights based on the proportion of
the target population in
the Home Mortgage Disclosure data (“HMDA”) to ensure that
the target population
in our data is representative of the target population in HMDA.
This allows us to draw
inferences to the full population.
are studies that examine specific underwriting standards and
products that
may be intended for different segments of the population, or
that address the
balancing of access to credit and credit risk.

A recent paper by Quercia, Ding, and Reid (2012) specifically
addresses the
balancing of credit risk and mortgage access for borrowers—the
two tensions
on which we focus. Their paper narrowly focuses on the
marginal impacts of
setting QRM product standards more stringently than those for
QM.4 They
find that the benefits of reduced foreclosures resulting from the
more stringent
product restrictions on “LTV” ratios, debt-to-income ratios
(“DTI”) and credit
scores do not necessarily outweigh the costs of reducing
borrowers’ access
to mortgages, as borrowers are excluded from the market.
Pennington-Cross and Ho (2010) examine the performance of
hybrid and ad-
justable rate mortgages (ARMs). After controlling for borrower
and location
characteristics, they find that high default risk borrowers do
self-select into
adjustable rate loans and that the type of loan product can have
dramatic im-
pacts on the performance of mortgages. They find that interest
rate increases
over 2005–2006 led to large payment shocks and with house
prices declin-
ing rapidly by 2008, only borrowers with excellent credit
history and large
amounts of equity and wealth could refinance.
Amromin and Paulson (2009) find that while characteristics
such as LTV,
FICO score and interest rate at origination are important

predictors of defaults
for both prime and subprime loans, defaults are principally
explained by house
price declines, and more pessimistic contemporaneous
assumptions about
house prices would not have significantly improved forecasts of
defaults.
Courchane and Zorn (2012) look at changing supply-side
underwriting stan-
dards over time, and their impact on access to credit for target
populations
of borrowers.5 They use data from 2004 through 2009,
specifically focusing
on the access to and pricing of mortgages originated for
African-American
and Hispanic borrowers, and for borrowers living in low-income
and minor-
ity communities. They find that access to mortgage credit
increased between
2004 and 2006 for targeted borrowers, and declined
dramatically thereafter.
The decline in access to credit was driven primarily by the
improving credit
mix of mortgage applicants and secondarily by tighter
underwriting standards
4For details of the QRM, see Federal Housing Finance Agency,
Mortgage Market Note
11-02. For details of the QM, see
http://files.consumerfinance.gov/f/201310_cfpb_qm-
guide-for-lenders.pdf.
5See also Courchane and Zorn (2011, 2014) and Courchane,
Dorolia and Zorn (2014).

associated with the replacement of subprime by FHA as the
dominant mode
of subprime originations.
These studies all highlight the inherent tension between access
to mortgage
credit and credit risk. They also stress the difficulty in finding
the “cor-
rect” balance between the two, and suggest the critical
importance of treat-
ing separately the three mortgage market segments—prime,
subprime and
government-insured (FHA)—because of the different borrowers
they serve
and their differing market interactions. The research also
provides some op-
timism that a careful examination of recent lending patterns will
reveal op-
portunities for responsibly extending credit while balancing
attendant credit
risks.
Data
Our analysis uses CoreLogic data for mortgages originated
between 1999
and 2009. The CoreLogic data identify prime (including Alt-A),
subprime
and government loans serviced by many of the large, national
mortgage
servicers. These loan-level data include information on
borrower and loan
product characteristics at the time of origination, as well as

monthly updates
on loan performance through 2012:Q3. Merged to these data are
annual
house price appreciation rates at a ZIP code level from the
Freddie Mac
Weighted Repeat Sales House Price Index, which allow us to
update borrower
home equity over time.6 We prefer this house price index to the
FHFA’s,
as the latter are not available at the ZIP code level. The
CoreLogic data
do not provide Census tract information, so we use a crosswalk
from ZIP
codes to 2000 Census tracts.7 We also merge in county-level
unemployment
rates from the Bureau of Labor Statistics, which are seasonally
adjusted by
Moody’s Analytics.8 Finally, we include changes in the
conventional mortgage
market’s average 30-year fixed mortgage (“FRM”) rate reported
in Freddie
Mac’s Primary Mortgage Market Survey.9
The CoreLogic data are not created through a random sampling
process and
so are not necessarily representative of the overall population,
or our target
6While these data are not publicly available, the metro/state
indices can be found
which are available at:
http://www.freddiemac.com/finance/fmhpi/.
7Missouri Census Data Center, available at:
http://mcdc.missouri.edu/
websas/geocorr12.html.
8The unemployment rate is from the BLS Local Area

Unemployment Statistics
(http://www.bls.gov/lau/).
9These data are available publicly at:
http://www.freddiemac.com/pmms/pmms30.htm.
population. This is not a problem for estimating our
delinquency model, but it
does create concern for drawing inference with our scorecards.
To address this
potential concern, we apply appropriate postsample weights
based on HMDA
data to enhance the representativeness of our sample. We
develop weights by
dividing both the HMDA and the CoreLogic data into
categories, and then
weight so that the distribution of CoreLogic loans across the
categories is the
same as that for HMDA loans. The categories used for the
weighting are a
function of loan purpose (purchase or refinance), state, year of
origination and
loan amount. Because we rely on a postsample approach and
cannot create
categories that precisely define our target population, our
weighting does not
ensure representativeness of the CoreLogic data for this group.
Nevertheless,
it likely offers a significant improvement over not weighting.
We also construct a holdout sample from our data to use for
inference. This
ensures that our estimated models are not overfitted. The

holdout sample was
constructed by taking a random (unweighted) sample of 20% of
all loans in
our database. All summary statistics and estimation results
(Tables 1 and 2
and Appendix) are reported based on the unweighted 80%
estimation sample.
Consistent with our focus on identifying responsible credit
opportunities, we
restrict our analysis to first lien, purchase money mortgage
loans. Summary
statistics for the continuous variables used in our delinquency
estimation are
found in Table 1. Table 2 contains summary statistics for the
categorical
variables.
As shown in Table 1, the average LTV at origination is 97% for
government
loans. This is considerably higher than for the prime market,
where first
lien loans have LTVs less than 80%, on average.10 We also
observe the
expected differences in FICO scores, with an average FICO
score in the prime
sector of 730, 635 for subprime and 674 for government loans.
The prime
market loan amount (i.e., unpaid principal balance, or UPB, at
origination)
averages $209,000 with the government loan amount the lowest
at a mean of
$152,000. The mean value in the subprime population is below
that for prime
at $180,000. DTI ratios do not differ much between prime and
government
loans, and the DTI for subprime is unavailable in the data. As

DTI is a key
focus in the efforts of legislators to tighten underwriting
standards, we use it
when available for estimation. The equity measures post-
origination reflects
the LTV on the property as house prices change in the area.
All three markets faced significant house price declines, as
captured by the
change in home equity one, two or three years after origination.
For all three
10The mean LTV for subprime mortgages is surprisingly low at
83%, although this
likely reflects the absence of recording second lien loans, which
would lead to a higher
combined LTV.
T
ab
le
1
�
S
um
m
ar
y

st
at
is
ti
cs
fo
r
co
nt
in
uo
us
va
ri
ab
le
s
(8
0%
es
ti
m
at
io

n
sa
m
pl
e)
—
st
at
is
ti
cs
no
t
w
ei
gh
te
d.
A
ll
P
ri
m
e

S
ub
pr
im
e
G
ov
er
nm
en
t
L
oa
ns
(u
nw
ei
gh
te
d)
21
,7
25

,5
28
13
,8
94
,1
40
2,
50
3,
16
4
5,
32
8,
22
4
C
hg
in
M
or
tg
ag

e
R
at
e,
O
ne
Y
ea
r
P
os
t-
O
ri
gi
na
ti
on
−0
.1
4%
−0
.1
2%

0.
04
%
−0
.3
0%
%
M
is
si
ng
0.
00
%
0.
00
%
0.
00
%
0.
00
%
S

t.
D
ev
.
0.
61
%
0.
60
%
0.
58
%
0.
64
%
C
hg
.
in
M
or
tg
ag

e
R
at
e,
T
w
o
Y
ea
rs
P
os
t-
O
ri
gi
na
ti
on
−0
.2
1%
−0
.2

2%
−0
.1
0%
−0
.2
5%
%
M
is
si
ng
0.
00
%
0.
00
%
0.
00
%
0.
00
%

S
t.
D
ev
.
0.
56
%
0.
57
%
0.
54
%
0.
53
%
C
hg
.
in
M
or
tg

ag
e
R
at
e,
T
hr
ee
Y
ea
rs
P
os
t-
O
ri
gi
na
ti
on
−0
.2
4%
−0

.2
3%
−0
.2
7%
−0
.2
9%
%
M
is
si
ng
0.
00
%
0.
00
%
0.
00
%
0.
00

%
S
t.
D
ev
.
0.
55
%
0.
55
%
0.
53
%
0.
55
%
D
eb
t-
to
-I
nc

om
e
R
at
io
(D
T
I)
36
.0
8%
35
.8
6%
0.
00
%
36
.8
4%
%
M
is
si

ng
62
.0
3%
58
.7
7%
99
.9
8%
52
.7
2%
S
t.
D
ev
.
15
.1
0%
15
.2
1%

0.
00
%
14
.6
9%
E
qu
it
y
M
ea
su
re
O
ne
-Y
ea
r
P
os
t-
O
ri

gi
na
ti
on
95
.7
2%
95
.5
3%
93
.8
6%
97
.3
8%
%
M
is
si
ng
3.
04
%

2.
06
%
8.
52
%
3.
02
%
S
t.
D
ev
.
10
.3
5%
10
.8
0%
9.
33
%
9.

15
%
E
qu
it
y
M
ea
su
re
T
w
o-
Y
ea
rs
P
os
t-
O
ri
gi
na
ti

on
94
.7
9%
94
.5
3%
94
.5
3%
95
.7
4%
%
M
is
si
ng
3.
04
%
2.
06
%

8.
52
%
3.
02
%
S
t.
D
ev
.
20
.2
0%
21
.0
2%
22
.1
2%
15
.9
5%
E

qu
it
y
M
ea
su
re
T
hr
ee
-Y
ea
rs
P
os
t-
O
ri
gi
na
ti
on
96
.0

4%
95
.7
3%
10
0.
67
%
94
.3
9%
%
M
is
si
ng
3.
04
%
2.
06
%
8.
52

%
3.
02
%
S
t.
D
ev
.
28
.5
0%
29
.0
5%
34
.6
4%
21
.8
4%

T
ab
le
1
�
C
on
ti
nu
ed
.
A
ll
P
ri
m
e
S
ub
pr
im
e
G
ov

er
nm
en
t
F
IC
O
70
1.
9
72
7.
9
63
5.
1
66
5.
1
%
M
is
si
ng

14
.2
4%
15
.7
7%
9.
08
%
12
.7
0%
S
t.
D
ev
.
69
.8
55
.1
63
.6
68

.8
L
oa
n-
to
-V
al
ue
R
at
io
(L
T
V
),
at
O
ri
g.
82
.9
1%
78
.3

4%
83
.2
2%
96
.9
4%
%
M
is
si
ng
1.
62
%
1.
54
%
2.
43
%
1.
44
%

S
t.
D
ev
.
14
.9
5%
14
.6
4%
11
.0
9%
7.
07
%
L
oa
n
A
m
ou
nt

(O
ri
g.
U
P
B
($
))
$1
90
,5
65
$2
07
,3
70
$1
80
,3
74
$1
44
,5
34

%
M
is
si
ng
0.
00
%
0.
00
%
0.
00
%
0.
00
%
S
t.
D
ev
.
$3
28

,1
41
$3
62
,9
15
$2
67
,0
06
$2
27
,5
22
U
ne
m
pl
oy
m
en
t
R
at

e
O
ne
-Y
ea
r
P
os
t-
O
ri
gi
na
ti
on
5.
58
%
5.
40
%
5.
03
%

6.
47
%
%
M
is
si
ng
0.
11
%
0.
10
%
0.
22
%
0.
08
%
S
t.
D
ev

.
2.
27
%
2.
09
%
1.
53
%
2.
88
%
U
ne
m
pl
oy
m
en
t
R
at
e

T
w
o-
Y
ea
rs
P
os
t-
O
ri
gi
na
ti
on
6.
02
%
5.
94
%
5.
40
%

6.
64
%
%
M
is
si
ng
0.
09
%
0.
09
%
0.
12
%
0.
09
%
S
t.
D
ev

.
2.
56
%
2.
53
%
1.
93
%
2.
85
%
U
ne
m
pl
oy
m
en
t
R
at
e

T
hr
ee
-Y
ea
rs
P
os
t-
O
ri
gi
na
ti
on
6.
57
%
6.
55
%
6.
45
%

6.
69
%
%
M
is
si
ng
0.
08
%
0.
08
%
0.
05
%
0.
09
%
S
t.
D
ev

.
2.
78
%
2.
81
%
2.
72
%
2.
72
%
Table 2 � Summary statistics for categorical (class) variables
(80% estimation
sample)—statistics not weighted.
All Prime Subprime Government
ARM 12.60% 48.50% 4.72% 14.91%
Balloon 0.39% 4.91% 0.05% 0.82%
FRM-15 7.68% 1.61% 1.16% 5.63%
FRM-30 68.25% 22.31% 90.14% 67.79%
FRM-Other 4.48% 1.71% 2.73% 3.81%

Hybrid 6.59% 20.97% 1.20% 7.04%
Other 41.05% 33.13% 43.57% 40.71%
Retail 33.70% 21.20% 22.12% 29.87%
Wholesale 25.25% 45.67% 34.31% 29.42%
Full Documentation 29.83% 49.38% 41.80% 34.52%
Missing 38.89% 18.44% 42.30% 37.35%
Not Full Documentation 31.27% 32.19% 15.90% 28.13%
Owner Occupied 83.43% 85.88% 91.84% 85.48%
Not Owner Occupied 16.57% 14.12% 8.16% 14.52%
Condo 13.82% 7.75% 6.97% 11.70%
Single Family 86.18% 92.25% 93.03% 88.30%
mortgage market segments, post-origination equity measures
(post-origination
estimated LTV) averaged over 90%. Post-Origination
unemployment rates are
highest, on average, in the geographies with government loans,
although the
differentials among market segments fell after three-year post-
origination.
Table 2 presents the summary statistics for the categorical
(class) variables
in our sample. Some expected results emerge. The subprime
segment has the
largest share of loans originated through the wholesale channel
at 45.7%,
while the wholesale share for the prime segment was only
25.2%. Nearly half
(48.5%) of subprime loans were “ARM” loans, while only
22.3% of subprime
loans were the standard 30-year FRM product. In contrast,
69.1% of prime
loans were 30-year FRMs and an additional 7.8% were 15-year
FRMs. Nearly
all of the government loans (91.2%) were 30-year FRMs. The

documentation
figures are somewhat surprising, with nearly half (49.4%) of
subprime loans
indicating full documentation. The low share of full
documentation loans in
the prime sector (about 30%) likely reflects the inclusion of
Alt-A loans,
which are defined to be prime loans in the CoreLogic data.11
In our analyses, we focus on access to credit and credit risk
outcomes for all
borrowers. However, many homeownership and affordable
lending programs
11Historically, Alt-A loans were originated through prime
lenders, offering their more
credit worthy customers a simpler origination process.
focus more narrowly on assessing opportunities for responsibly
extending
mortgage credit to borrowers with low down payments and poor
credit his-
tories, or who are otherwise underserved by the prime market
(“target pop-
ulation”). As a result of long standing public policy objectives
focused on
the value of homeownership, both government insured mortgage
programs
(such as FHA) and the GSEs have long held missions to meet
the needs of
underserved borrowers, including low income, minority and
first-time home-

buyers.12 Programs meeting this mission are tasked with
balancing access to
credit for borrowers with any attendant increases in credit risk.
Therefore, aside from our focus on the opportunities provided to
the full pop-
ulation of borrowers, we also provide an analysis of scorecard
outcomes for a
specific target population. We define this target population as
borrowers who
receive first lien, purchase money mortgages on owner-occupied
properties
located in census tracts with median incomes below 80% of the
area median
income, with FICO scores less than or equal to 720 and with
LTV ratios
greater than or equal to 90%.
Limiting our analysis to borrowers who live in lower income
census tracts is
especially constraining, as many borrowers with high LTVs and
lower FICO
scores live elsewhere. However, our data lack accurate income
measures, and
public policy considerations encourage us to include an income
constraint in
our definition of the target population.13 As a consequence,
loans to target
borrowers account for a small percentage of the total loans
made during our
period of study (roughly 4%). We can be assured, however, that
our target
population is composed of borrowers who are an explicit focus
of public
policy.

Figure 1 provides a graphical illustration of the HMDA-
weighted distribution
of target population loans in our sample across the three market
segments.
The dramatic shift over time in the share going to the
government sector is
obvious, as is the reduction in the number of loans originated to
the target
population by all three segments, combined, post-crisis.
12Both the Community Reinvestment Act (CRA) and the
Federal Housing Enterprises
Financial Safety and Soundness Act of 1992 (the 1992 GSE Act)
encouraged mortgage
market participants to serve the credit needs of low- and
moderate-income borrowers
and areas.
13For example, GSE affordable goals are stated with respect to
low- and moderate-
income borrowers and neighborhoods.
Figure 1 � Target population by year and market segment
(weighted).
Analysis
Our analysis begins with the estimation of mortgage
performance models over
the crisis period. We use loan-level origination data from 1999
through 2009
to estimate models of loans becoming 90-days or more
delinquent in the first

three years after origination. These models include standard
borrower and loan
characteristics at origination, as well as control variables
measuring changes
in house prices, unemployment rates and interest rates post-
origination. They
also include several interaction terms for the borrower, loan and
control
variables.
We then use our estimated delinquency models to specify two
underwrit-
ing scorecards—a through-the-cycle scorecard and a stress
scorecard.14 We
next apply various cutpoints (risk thresholds) to our scorecards
to define
levels of acceptable risk. By definition, loans with risk scores
(delinquency
probabilities) at or below the cutpoint are assumed to be within
appropriate
(acceptable) risk tolerances.
The scorecard and cutpoint combinations provide working
prototypes of an
AUS. Our final step applies these prototypes to the full and
target populations
and assesses the results.
Estimating the Models
We estimate three separate delinquency models based on an
80% sample
of first-lien, purchase money mortgage loans in our data.
Separate models
14Additional scorecards constructed using “perfect foresight”

and macroeconomic
forecasts are available from the authors upon request.
were estimated for prime loans (including Alt-A loans),
subprime loans and
government loans, using an indicator provided in the CoreLogic
data to as-
sign each loan to its appropriate segment.15 We estimate
separate models for
each market sector because we believe that there is clear market
segmenta-
tion in mortgage lending. In the conventional market the
lenders, industry
practices, market dynamics and regulatory oversight have
differed between
the prime and subprime segments.16 A similar distinction exists
between the
conventional and government segments—the latter focuses on
first-time bor-
rowers and lower income households. Moreover, acceptable risk
tolerances
will necessarily vary across segments, as may concerns
regarding access to
credit.
Our process differs from the typical construction of
underwriting …
Macroeconomic Dynamics, 17, 2013, 830–860. Printed in the
United States of America.

doi:10.1017/S1365100511000721
MONETARY POLICY, HOUSING
BOOMS, AND FINANCIAL
(IM)BALANCES
SANDRA EICKMEIER
Deutsche Bundesbank
BORIS HOFMANN
Bank for International Settlements
This paper applies a factor-augmented vector autoregressive
model to U.S. data with the
aim of analyzing monetary transmission via private sector
balance sheets, credit risk
spreads, and house prices and of exploring the role of monetary
policy in the housing and
credit boom prior to the global financial crisis. We find that
monetary policy shocks have a
persistent effect on house prices, real estate wealth, and private
sector debt and a strong
short-lived effect on risk spreads in money and mortgage
markets. Moreover, the results
suggest that monetary policy contributed considerably to the
unsustainable precrisis
developments in housing and credit markets. Although monetary
policy shocks
contributed discernibly at a late stage of the boom, feedback
effects of other
(macroeconomic and financial) shocks via lower policy rates
kicked in earlier and appear
to have been considerable.
Keywords: Monetary Policy, Private Sector Balance Sheets,
Asset Prices, Housing

1. INTRODUCTION
The impact of monetary policy shocks on financial conditions,
i.e., asset prices,
lending terms, and balance sheets, has been one of the most
topical issues in
monetary economics over recent years. Interest in the topic has
recently gained
further impetus from the coincidence of rapid property price
inflation (“housing
The views expressed in this paper do not necessarily reflect the
views of the European Central Bank, the Bank
for International Settlements, or the Deutsche Bundesbank. The
paper was mainly written while the second author
was affiliated with the European Central Bank. We would like
to thank participants in workshops/seminars of the
Bundesbank, the ZEW, the People’s Bank of China, the ECB,
the Reserve Bank of Australia, and the Bank of
Canada, as well as the 5th Conference on Growth and Business
Cycles in Theory and Practice (Manchester), the 15th
International Conference on Panel Data (Bonn), the conference
on Computing in Economics and Finance (Sydney),
the BIS/ECB workshop on Monetary Policy and Financial
Stability (Basel), and the 12th Annual DNB Research
Conference on “Housing and Credit Dynamics: Causes and
Consequences” (Amsterdam), for useful comments. In
particular, helpful comments and suggestions by our discussants
Katrin Assenmacher-Wesche, Guido Bulligan, Math-
ias Drehmann, Gabriele Galati, Gert Peersman, and Timo
Wollmershäuser, as well as by Heinz Herrmann, Wolfgang
Lemke, Roberto Motto, Massimo Rostagno, Harald Uhlig,
Alexander Wolman, and three anonymous referees, are
gratefully acknowledged. All errors are our own responsibility.
Address correspondence to: Boris Hofmann, Bank

for International Settlements, Centralbahnplatz 2, 4002 Basel,
Switzerland; e-mail: [email protected]
c© 2012 Cambridge University Press 1365-1005/12 830
MONETARY POLICY AND FINANCIAL IMBALANCES 831
-12
-8
-4
0
4
8
12
-30
-20
-10
0
10
20
30

1980 1985 1990 1995 2000 2005
Real FHFA/OFHEO house price index (% change y-o-y, LHS)
Real S&P/Case-Shiller house price index (% change y-o-y,
LHS)
Real MIT/CRE commercial property price index (% change y-o-
y, RHS)
-8
-4
0
4
8
12
16
-8
-4
0
4
8
12
16

1980 1985 1990 1995 2000 2005
Real household debt (% change y-o-y)
Real corporate non-financial business sector debt (% change y-
o-y)
Real noncorporate nonfarm business sector debt (% change y-o-
y)
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
0.0
0.5
1.0
1.5
2.0

2.5
3.0
3.5
4.0
1980 1985 1990 1995 2000 2005
3-month commercial paper spread (%age pts. over 3-month T-
bill rate)
3-month Eurodollar deposit spread (%age pts. over 3-month T-
bill rate)
30-year mortgage rate spread (%age pts. over 30-year
government bond yield)
-2
0
2
4
6
8
10
-2
0
2

4
6
8
10
1980 1985 1990 1995 2000 2005
Real Federal Funds rate (%)
(a)
(c) (d)
(b)
FIGURE 1. Property prices (a), private sector debt (b), credit
risk spreads (c), and monetary
policy rate (d). Real property prices and real debt have been
computed by deflating with
the GDP deflator. The real Federal Funds rate is the effective
Federal Funds rate less the
year-on-year change in the GDP deflator. Sources: St. Louis
FRED, OFHEO, Bureau of
the Census, Federal Reserve Board, authors’ calculations.
bubble”), a massive expansion of private sector indebtedness
(“credit bubble”),
and very low risk spreads in credit markets (“underpricing of
risk”) on one side,
and, on the other, exceptionally low levels of policy rates in the
United States prior
to the outbreak of the global financial crisis, i.e., between 2001

and 2006, as shown
in Figure 1. This coincidence has led a number of observers—
most prominently
the Bank for International Settlements (2007, 2008) and Taylor
(2007, 2009)—to
832 SANDRA EICKMEIER AND BORIS HOFMANN
argue that an excessively loose monetary policy stance was one
of the key factors
contributing to the imbalances in housing and credit markets
prior to the crisis.1
The goal of this paper is to contribute to the literature on the
transmission
of monetary policy via financial conditions and to explore the
role of monetary
policy in the buildup of imbalances in property and credit
markets before the
financial crisis. To this end, we employ a factor-augmented
vector autoregressive
(FAVAR) model, a novel empirical tool proposed by Bernanke
et al. (2005). The
model enables us to analyze monetary transmission over a wide
range of financial
variables, i.e., property and stock prices, interest rates, credit
risk spreads, and
nonfinancial private sector assets and liabilities,2 based on a
unified, consistent
modeling framework exploiting the close correlation between
these variables in-
dicated by Figure 1. More specifically, the FAVAR model
developed in this paper
extends a standard macroeconomic vector autoregressive (VAR)

model with a set
of (financial) factors summarizing more than 200 quarterly
financial variables.3
To identify the monetary policy shock, we adopt an
identification scheme that
combines contemporaneous zero restrictions and theoretically
motivated sign re-
strictions on short-term impulse-response functions (see, e.g.,
Peersman 2005 and
Uhlig 2005), allowing for contemporaneous interaction between
the policy rate
and financial factors. This identification scheme further enables
us to disentangle
macroeconomic shocks (which are defined here as shocks to real
growth and
inflation) and shocks to financial factors.
The two main contributions of the paper are the following.
First, we provide
a unified and comprehensive characterization of the
transmission of monetary
policy shocks via financial conditions, covering a broad range
of asset prices,
interest rates, risk spreads and private sector balance sheet
components by means of
impulse-response analysis. This is novel, as the related existing
literature has so far
focused on specific aspects of monetary transmission,4 whereas
a comprehensive
analysis of the transmission of monetary policy shocks via
financial conditions
encompassing all these specific aspects is still missing. The
impulse-response
analysis allows assessment of the relative strength of monetary
transmission via

different asset markets, credit markets, and balance sheets and
sheds light on the
relevance of financial frictions in the transmission process.
Second, we assess the role of monetary policy in the buildup of
the precrisis
imbalances in housing and credit markets. A number of recent
academic studies
have explored the contribution of monetary policy shocks, i.e.,
the deviation of
policy rates from their estimated usual reaction patterns or some
postulated re-
action pattern (Taylor rule) to the housing boom [Del Negro and
Otrok (2007),
Taylor (2007), Iacoviello and Neri (2010), Jarociñski and Smets
(2008)], but with-
out coming to consistent conclusions. In this paper we assess,
based on historical
decompositions, the role of monetary policy shocks in the
housing boom as well as
in the two other precrisis phenomena highlighted in Figure 1—
the excessive debt
accumulation in the private nonfinancial sector and the low risk
spreads in credit
markets, which have so far remained unexplored. In this
context, we also show
that the inconsistencies in the results regarding the role of
monetary policy shocks
MONETARY POLICY AND FINANCIAL IMBALANCES 833
in the housing boom produced by previous studies can be linked
to differences in
sample periods.

Besides assessing the role of monetary policy shocks, we also
explore, based
on counterfactual simulations, the role of systematic monetary
policy, i.e., of the
estimated reaction of the policy rate to shocks to financial
factors and to macroeco-
nomic shocks. Because we allow for contemporaneous
interaction between policy
rates and financial factors, we can explore not only the effects
of monetary policy
shocks on financial variables, but also the effect of innovations
in financial factors
on the path of policy rates over time. Via counterfactual
simulations, we then
explore to what extent the reaction of monetary policy to these
innovations has
fed back to housing and credit markets. In this way, we can
tentatively assess the
widely held view that the monetary easing in reaction to the
bursting of the stock
market bubble after 2000 contributed to the subsequent housing
and debt boom.
The main findings of our analysis are as follows. (i) Monetary
policy shocks
have a highly significant and persistent effect on property
prices, real estate wealth,
and private sector debt, as well as a strong short-lived effect on
risk spreads in the
money and mortgage markets. (ii) Monetary policy contributed
considerably to the
unsustainable developments in housing and credit markets that
were observable
between 2001 and 2006. Although monetary policy shocks
discernibly contributed

at a late stage of the boom, feedback effects of other
(macroeconomic and financial)
shocks via lower policy rates on property and credit markets
probably kicked in
earlier and were considerable.
The remainder of the paper is organized as follows. We present
the data in
Section 2 and explain the methodology in Section 3. In Section
4 we analyze
the dynamic effects of monetary policy shocks on asset prices,
interest rates, and
balance sheets. In Section 5 we assess the role of monetary
policy in the precrisis
financial imbalances. Section 6 concludes.
2. DATA
The quarterly data set used in this study is composed of three
standard macro
variables, real GDP growth, GDP deflator inflation, and the
effective Federal
Funds rate [retrieved from the St. Louis Federal Reserve
Economic Data (FRED)
database], as well as 232 financial variables comprising 69
property prices, 62 stock
market indices, 50 money, capital, and loan interest rates and
spreads, 2 monetary
aggregates, and 49 series from private nonfinancial sector
balance sheets. Stock
prices, property prices, monetary aggregates, and balance sheet
variables were
converted to real units by deflation with the GDP deflator. The
choice of variables
is determined by data availability, as well as the aim to estimate
the financial

20 THE NEW” HOUSING AND MORTGAGE MARKET SPRING 2016The .docx

20 THE NEW” HOUSING AND MORTGAGE MARKET SPRING 2016The .docx

Recommended

Recommended

More Related Content

Similar to 20 THE NEW” HOUSING AND MORTGAGE MARKET SPRING 2016The .docx

Similar to 20 THE NEW” HOUSING AND MORTGAGE MARKET SPRING 2016The .docx (17)

More from novabroom

More from novabroom (20)

Recently uploaded

Recently uploaded (20)

20 THE NEW” HOUSING AND MORTGAGE MARKET SPRING 2016The .docx