Speculation

Economic Consequences of Housing
Speculation
Zhenyu Gao
Chinese University of Hong Kong
Downloaded from https://academic.oup.com/rfs/article/33/11/5248/5803087 by Princeton University user on 08 November 2020

Michael Sockin
University of Texas at Austin
Wei Xiong
Princeton University, CUHK Shenzhen, and NBER
By exploiting variation in state capital gains taxation as an instrument, we analyze the

economic consequences of housing speculation during the U.S. housing boom in the
2000s. We find that housing speculation, anchored, in part, on extrapolation of past
housing price changes, led not only to greater price appreciation, economic expansions,
and housing construction during the boom in 2004–2006 but also to more severe economic
downturns during the subsequent bust in 2007–2009. Our analysis supports supply overhang
and local household demand as two key channels for transmitting these adverse
effects. (JEL R21, D84)
Received May 28, 2018; editorial decision August 21, 2019 by Editor Itay Goldstein.
Authors have furnished an Internet Appendix, which is available on the Oxford University
Press Web site next to the link to the final published paper online.
Economists have long been concerned with the economic consequences of

speculation and asset bubbles. A growing strand of the literature, including
work by Shiller (2009), Haughwout et al. (2011), Barlevy and Fisher (2011),
Mayer (2011), Case, Shiller, and Thompson (2012, 2015), Bayer et al.
(2015), Chinco and Mayer (2016), Albanesi, Giorgi, and Nosal (2017),
This paper supersedes an earlier draft, which was circulated under the title “Housing Speculation and Supply
Overhang.” We thank Itay Goldstein (the editor) and two anonymous referees for helpful comments. We are
grateful to Gene Amromin, Barney Hartman-Glaser, Andrew Haughwout, Zhiguo He, Tim Landvoigt, Alvin
Murphy, Charlie Nathanson, and Monika Piazzesi and seminar participants at the Federal Reserve Bank of New
York, the University of Texas at Austin, the Wharton School, the American Economic Association Meetings, the
American Finance Association Meetings, the CICF meetings, the FRB Atlanta, GSU Real Estate Finance, and
the Texas Finance Festival Conference for helpful comments and discussion. Supplementary data can be found
on The Review of Financial Studies web site. Send correspondence to Zhenyu Gao, CUHK Business School, The
Chinese University of Hong Kong, Room 1244, 12/F, Cheng Yu Tung Building, 12 Chak Cheung Street, Shatin,
N.T., Hong Kong telephone: 852-3943-1824. E-mail: gaozhenyu@baf.cuhk.edu.hk.
The Review of Financial Studies 33 (2020) 5248–5287

© The Author(s) 2020. Published by Oxford University Press on behalf of The Society for Financial Studies.
All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
doi:10.1093/rfs/hhaa030 Advance Access publication March 11, 2020
[23:07 27/9/2020 RFS-OP-REVF200031.tex] Page: 5248 5248–5287

Economic Consequences of Housing Speculation
DeFusco, Nathanson, and Zwick (2017), Nathanson and Zwick (2017), and
Soo (2018), has highlighted the importance of housing speculation in driving
the recent housing cycle.1 Housing speculation became a national phenomenon
in the low-interest-rate environment of the mid-2000s, with purchases of non-
owner-occupied homes (second and investment homes) contributing up to
30% of all home purchases during the boom in cities such as Las Vegas.
Housing speculation also represented a source of housing demand largely
orthogonal to the credit expansion to subprime households that occurred during
the housing boom, which is widely regarded, for instance by Mian and Sufi

(2009), Keys et al. (2009), and Justiano et al. (2017), as a key driver of the
boom. Interestingly, areas in which speculation became more prevalent during
the boom period experienced local economic expansions, while subprime credit
expansion occurred more in areas that experienced local economic contractions.
In addition, while supply inelasticity and nonrecourse mortgage laws have a
positive and statistically significant correlation with the fraction of subprime
mortgages in a ZIP code, they have an opposite, smaller, and less significant
correlation with housing speculation.2 As such, speculation represents an
important complementary channel in explaining the cross-sectional variation
in the housing and economic cycles across the United States during the 2000s.
An intuitive hypothesis posits that speculation in the housing market can
have important economic consequences. Housing speculation, anchoring on
extrapolative expectations of past housing price growth, can amplify local
economic conditions by contributing a nonfundamental source of demand to
housing markets. When these speculators purchase more non-owner-occupied
homes in an area during a housing boom, their speculation can magnify the
boom and contribute not only to a greater price drop but also to a more severe
economic contraction, during the subsequent housing bust. Despite its intuitive
appeal, this hypothesis remains elusive to test because of the well-appreciated
endogeneity issue with identification; that is, housing speculation may reflect
local housing demand or other unobservable economic conditions, rather than
be a cause of housing and economic cycles.
In this paper, we undertake this challenge to study how housing speculation
contributed to higher housing prices and local economic expansions during the
boom period of 2004–2006 and adversely affected economic activity during the
bust period of 2007–2009. We measure housing speculation during the boom
by the fraction of non-owner-occupied home purchases in a ZIP code, which
conveniently measures the intensity of housing speculation relative to primary
home demand. For identification, we construct a novel instrument for housing
1 Glaeser (2013) provides an eloquent analysis of nine episodes of real estate speculation in American history and
highlights housing speculation as one of several recurring themes within the episodes.
2 This opposite and less significant correlation suggests that studies that instrument the housing price boom with
supply elasticity or variation in nonrecourse mortgage laws are capturing the impact of subprime rather than
speculation.
5249

The Review of Financial Studies / v 33 n 11 2020
speculation that takes advantage of the variation across U.S. states in their
taxation of capital gains. While homeowners can exclude capital gains from
the sale of their primary residence from their income taxes, this exclusion
does not cover capital gains from selling non-owner-occupied homes. As
U.S. states have significant variation in how they tax capital gains, housing
speculation is more intensive in states with either no or low capital gains taxes.
We therefore construct our instrument as the marginal tax rate for the median
income household in each state.
By instrumenting non-owner-occupied home purchases with the capital gains

tax variable, we find that ZIP codes with a greater share of non-owner-occupied
home purchases during the boom not only had more pronounced housing cycles
during the boom and bust but also experienced greater swings in employment,
payroll, per capital income, and the number of business establishments. The
economic magnitude of these effects is substantial: an increase of 9.9% (1
standard deviation across ZIP codes) in the share of non-owner-occupied home
purchases in 2004–2006 led to a housing price increase of 26.5% during the
boom, and a drop of 37.4% during the bust. Similarly, this increase led to an
increase of 13.7% in real payroll, 8.4% in employment, 12.9% in per capita
income, and 6.8% in the number of establishments in 2004–2006. During
2007–2009, in contrast, it contributed to declines of 15.4% in real payroll,
14.6% in employment, 7.8% in income per capita, and 8.7% in the number
of establishments. These results on the impact of housing speculation are
robust to excluding the so-called “sand states” of Arizona, California, Florida,
and Nevada that saw particularly phenomenal housing cycles. Furthermore,
our analysis shows that, among states with lower capital gains taxes, the
share of non-owner-occupied home purchases responds more strongly to past
housing price increases, even after controlling for past changes in local housing
fundamentals. This finding supports extrapolative expectations as a key driver
of housing speculation.
We also examine two transmission mechanisms to understand how housing
speculation during the boom propagated to the real economy during the bust.
The first is the supply overhang channel, explored, for instance, in Rognlie,
Shleifer, and Simsek (2018). By again using the instrumental variable approach,
we find that areas with more intensive housing speculation during the boom
also had a greater increase in housing construction in the same period, which,
in turn, contributed to the subsequent contraction of the construction sector
during the bust. An increase of 1 standard deviation in instrumented housing
speculation in 2004–2006 led to an increase of 4.2% in building permits in
2004–2006, relative to the number of housing units in 2000, and decreases
of 33.8% in construction-sector employment and 12.4% in nonconstruction
sector employment in 2007–2009. These findings confirm the importance of
the supply overhang channel.
To further explain the substantial downturn experienced by nonconstruction
sectors, we also examine a second transmission channel, local household
5250

demand, as suggested by Mian, Rao, and Sufi (2013) and Mian and Sufi (2014),
by analyzing the impact of housing speculation on nontradable sectors—and
the retail and restaurant sectors more narrowly—which primarily rely on local
consumption demand. We find significant real effects through this channel.
An increase of 1 standard deviation in instrumented housing speculation in
2004–2006 led to a decrease of 15.1% in nontradable sectors’ employment
in 2007–2009, and a decline of 15.6% in the retail and restaurant sectors,
specifically.3 In contrast, housing speculation had a more modest effect on
employment in tradable sectors and in industries other than retail and the

restaurant business.
While our tax instrument allows us to establish and quantify the causal link
between housing speculation, housing prices, and real activity, an important
limitation is that it relies on state-level variation. As such, we do not have
a source of within-state exogenous variation in the fraction of non-owner-
occupied home purchases for the first stage of the IV test. While this lack
of granularity is not ideal for our analysis of housing speculation at the ZIP-
code level, this limitation biases our results against finding supportive evidence
for the housing speculation hypothesis. Nevertheless, this instrument provides
sufficient cross-sectional variation to identify the economic consequences of
housing speculation.
Our study contributes to the rapidly growing literature on housing
speculation. Using credit-report data, Haughwout et al. (2011) document a
large increase in the share of housing purchased by real estate investors during
the recent U.S. housing boom, who took on more leverage and had higher
default rates during the bust. Using micro-level data, Chinco and Mayer (2016)
show that speculation by investment-home buyers played an important role
in the dramatic house price boom and bust cycles in 21 cities, including Las
Vegas, Miami, and Phoenix. Nathanson and Zwick (2017) turn to speculation
in the land market and investigate how land investment by homebuilders
shapes the house price boom in areas with elastic housing supply. DeFusco,
Nathanson, and Zwick (2017) investigate the importance of short-term real
estate investors in explaining housing price and volume dynamics in the recent
housing cycle. While most of these studies have explored the relation between
speculation and housing market outcomes, such as house prices and default, we
provide causal evidence that speculation exacerbated the recent U.S. housing
cycle. In addition, we investigate its causal impact on local economic activity,
including establishments, payroll, employment, and per capita income growth,
during both the housing boom and bust. We also highlight the interaction
between state capital gains, speculation, and extrapolation of past housing
price appreciation as a potential explanation for the cross-sectional variation in
3 Kaplan, Mitman, and Violante (2017), through the lens of a quantitative framework, also find that a shift in
households’ expectations of future capital gains on housing investments deepened the Great Recession through
the household balance sheet channel.
5251

housing speculation. Last, we investigate potential transmission mechanisms

through which housing speculation during the boom propagated to the real
economy during the bust.
Our work is also related to the literature on the economic distortions of
housing booms. Chen et al. (2016) show that firms responded to rising real
estate prices in China by diverting resources from their core businesses to real
estate investment. Charles, Hurst, and Notowidigdo (2016a, 2016b) explore
how the housing boom led to distortions in the employment and educational
attainment decisions predominantly among low-skilled, prime-aged laborers

by temporarily expanding the construction and services sectors. Consistent
with their results, we find that construction and local retail and service sectors
contracted during the housing bust.
While our study focuses on the role of housing speculation in explaining
the recent U.S. housing cycle, the housing literature has explored several
other mechanisms. Himmelberg, Mayer, and Sinai (2005) and Mayer and
Sinai (2009), for instance, focus on the role of interest rates, while Favilukis,
Ludvigson, and Van Nieuwerburgh (2017) emphasize the relaxation of
borrowing constraints. Mian and Sufi (2009, 2011) and Favara and Imbs (2015)
link the surge in household debt and house prices during the recent housing
cycle to an expansion in the supply of credit that resulted from securitization and
subprime lending, Landvoigt, Piazzesi, and Schneider (2015) also emphasize
the role of cheap credit for poor households in driving house price dynamics,
and Mian and Sufi (2010) examine the relation between this buildup and the
cross-section of house price booms and busts in the recent cycle. Furthermore,
Keys et al. (2009, 2010), Purnanandam (2011), and Griffin and Maturana
(2015, 2016), among others, highlight that agency issues associated with
securitization may have helped fuel this credit expansion. Along this dimension,
Griffin and Maturana (2015) document the importance of mortgage origination
misreporting by the worst originators in explaining housing price booms and
busts. Similarly, Mian and Sufi (2015) show mortgage fraud to be associated
with low income ZIP codes that exhibited the strongest mortgage credit growth
in 2002 to 2005. The speculation channel that we analyze is distinct from that
of fraud and misreporting, as our results remain robust when we control for
the misreporting measure of Piskorski, Seru, and Witkin (2015). To the extent
that the credit expansion and agency issues in mortgage origination facilitated
participation by optimistic speculators in housing markets during the boom, we
view these channels as being complementary.
1. Empirical Methodology
We aim to examine the economic impact of housing speculation during the
boom and bust cycle of the U.S. housing markets in the 2000s. The literature
has established that this housing cycle was, in part, driven by a credit expansion
to households across the country, which was precipitated by the rapid growth
5252

of securitization and shadow banking in the early 2000s (e.g., Mian and
Sufi 2009; Keys et al. 2010). While this was a national housing cycle, there
was substantial variation across regions. We are interested in testing whether
housing speculation contributed to this cross-region variation and, specifically,
whether housing speculation during the boom affected the housing cycles and
economic performance during the subsequent downturn.4
To facilitate our cross-region analysis during this national cycle, we divide
the housing cycle into three phases: 2001–2003 as the preboom period, 2004–
2006 as the boom period, and 2007–2009 as the bust period.5 We define

2004–2006 as the boom period because housing speculation, as we will show,
was most intensive during this period. Anchoring on this 3-year boom period,
we analyze how housing price growth in the previous 3 years (the preboom
period) stimulated housing speculation during the boom through home buyers’
extrapolative expectations, and how housing speculation during the boom
affected local economic performance both during the boom period and the
subsequent 3 years (the bust period).
We measure housing speculation in an area in a given year by the fraction
of non-owner-occupied homes in all home purchases financed by mortgages.
Non-owner-occupied homes provide less housing service to their owners than
primary homes, so the decision to purchase a non-owner-occupied home is
driven to a greater extent by buyers’ speculative motives than their decision
to purchase a primary home.6 As such, the fraction of non-owner-occupied
home purchases provides a convenient and intuitive measure of the intensity of
housing speculation relative to primary housing demand.
We face the typical issue of endogeneity in testing the impact of housing
speculation. A large fraction of non-owner-occupied home purchases in an
area might reflect local economic conditions rather than be a cause of housing
and economic cycles. To address this challenging identification issue, we need
an instrumental variable that exogenously affects housing speculation in an
area.
4 We focus on the recent U.S. housing cycle of the 2000s, because the data are more complete for this period
than for earlier years, and because the national housing cycle allows us to directly compare the cross-sectional
variation in housing markets and local economic conditions. Such a cross-sectional analysis is not feasible for
the earlier housing cycles of the 1980s and 1990s, as they were asynchronous and experienced by only a few
cities.
5 This definition is largely consistent with the timing convention in the literature. In particular, 2006 is widely
recognized as the turning point of the cycle, as noted by Glaeser (2013). Haughwout et al. (2013) define the
boom period as 2000–2006 and the bust period as 2007–2010. Ferreira and Gyourko (2018) find that the start
of the house price boom was not synchronized across the United States and house prices for each ZIP code also
peaked in different months. Our speculation measure and several economic outcome variables are at the annual
frequency, so we use 2006 as the turning point of housing cycles across regions.
6 However, this measure has limitations. For instance, investors may purchase their houses for vacation purposes.
Non-owner-occupied house purchases also could be affected by demand from renters and new migrants. To
address this issue, we include various local characteristics, such as the fraction of employment in the sectors
of art, entertainment, and recreation, the ratio of renters, as well as the fraction of migrants, as controls in our
analysis.
5253

To construct such an instrument, we take advantage of the heterogeneity

in capital gains taxes imposed across different states. The primary residence
exclusion allows homeowners to exclude up to $250,000 ($500,000 per couple)
of capital gains from the sale of their primary residence, at both the federal and
state levels, defined as a home that they have owned and lived in for at least
2 of the 5 years prior to the sale. As there is no capital gains exclusion for
sales of non-owner-occupied homes during the recent housing boom, buyers of
non-owner-occupied homes are subject to capital gains taxation.
Taxation of capital gains at the state level is similar to that at the federal level,

but different states impose different capital gains tax rates, and nine states (i.e.,
Alaska, Florida, Nevada, New Hampshire, South Dakota, Tennessee, Texas,
Washington, and Wyoming) impose no capital gains taxes at all. The choice
of capital gains tax rates is not driven by shocks to housing markets. In fact,
during the boom period of 2004–2006, all nine states remained without capital
gains taxes, and only the District of Columbia and Ohio slightly changed their
capital gains tax rates.7
Ample evidence in the literature shows that economic agents across the
United States were overly optimistic about the housing market.8 As optimistic
households might choose to buy investment homes, state capital gains taxes
provide a source of exogenous variation in the fraction of non-owner-occupied
home purchases across areas. In particular, optimistic buyers expect to realize
a capital gain rather than a loss on the sale of an investment home. As such,
capital gains taxes would negatively affect investment home purchases. The
magnitude of this impact could be substantial: as reported by the Bureau of
Census, the average sales price for houses sold in 2003 is $244,550. For a
back-of-the-envelope calculation, using the summary statistics in our sample
in panel A of Table 1, the capital gains would be, on average, $68,000 if one
purchased a house in 2003 and sold it in 2006. This sale would incur a tax of
$3,400 for a 5% average state capital gains tax.
That capital gains taxation represents an important margin for home buyers
and sellers can be seen by revealed preference from the passage of the Taxpayer
Relief Act of 1997. This act introduced the exclusion of capital gains from the
sale of primary residences at the federal level. States followed suit in honoring
this exclusion, providing an additional windfall to their residents. Shan (2011)
studies housing market behavior after the passage of the act and finds that the
semiannual housing sales rate increased by 17% to 24% from baseline levels for
homeowners with capital gains between $0 and $500,000, with an elasticity of
-0.1% to -0.2% in home sales for a $10,000 increase in capital gains taxes. That
7 From 2005 to 2006, for a median income household, the marginal capital gains tax rate changed from 9% to
8.7% in the District of Columbia and from 4.983% to 4.764% in Ohio.
8 Even credit rating agencies, such as Moody’s, calculated the credit risk of mortgage-backed securities during the
boom period under the assumption that housing prices would not decline in the near future. In addition, Cheng,
Raina, and Xiong (2014) find that a sample of securitization agents also increased their own exposures to housing
in 2004–2006.
5254

Table 1
Summary statistics and correlations
A. Summary statistics of the key variables
(1) (2) (3)
Variables N Mean SD
Fraction of non-owner-occupied home purchases in 2004–2006 3,975 0.136 0.0987
Real house price change in 2001–2003 4,027 0.191 0.135
Real house price change in 2004–2006 4,027 0.278 0.191
Real house price change in 2007–2009 4,027 −0.413 0.278
Per capita income change in 2003–2006 4,027 0.0521 0.125
Per capita income change in 2007–2009 4,026 −0.113 0.0957

Change in no. of establishments in 2004–2006 3,942 0.0638 0.0870
Change in no. of establishments in 2007–2009 3,924 −0.0382 0.0706
Real payroll change in 2004–2006 3,942 0.0866 0.179
Real payroll change in 2007–2009 3,924 −0.0998 0.191
Employment change in 2004–2006 3,942 0.0707 0.158
Employment change in 2007–2009 3,924 −0.0831 0.148
Saiz’s elasticity 4,027 1.376 0.715
Fraction of subprime mortgages in 2002 3,468 0.0896 0.0725
Fraction of subprime mortgages in 2005 3,975 0.211 0.138
Mortgage denial rate in 2002 3,468 0.111 0.0573
Mortgage denial rate in 2005 3,975 0.139 0.0507
Non-owner-occupied home mortgage denial rate in 2002 3,414 0.115 0.0956
Non-owner-occupied home mortgage denial rate in 2005 3,939 0.130 0.0637
Fraction of GSE mortgages in 2002 3,468 0.376 0.113
Fraction of GSE mortgages in 2005 3,975 0.193 0.103
Fraction of GSE mortgages for non-owner-occupied home in 2002 3,388 0.383 0.170
Fraction of GSE mortgages for non-owner-occupied home in 2005 3,935 0.181 0.114
ln(population in 2000) 4,027 10.32 0.572
Fraction of the college educated in 2000 4,027 28.12 15.63
Fraction of the employed in 2000 4,027 61.14 8.782
Fraction of workforce in 2000 4,027 64.70 8.106
Median household income in 2000 4,027 49,524 17,274
Poverty rate in 2000 4,027 10.77 7.713
Urban rate in 2000 4,027 94.30 14.21
Fraction of white residents in 2000 4,027 71.89 23.05
Number of households in 2000 4,027 12,935 6,235
Fraction of renters in 2000 3,942 0.355 0.964
Fraction of immigrants in past 5 years (2000) 3,942 0.214 0.0703
Fraction of employment in arts entertainment and recreation in 2000 3,942 0.0832 0.0419
(Continued)
some states have extended the capital gains exclusion to secondary homes since
the financial crisis suggests that taxes on capital gains are a relevant margin in
home buyer decisions.
Motivated by this observation, we instrument the fraction of non-owner-
occupied home purchases during the boom period of 2004–2006 with a tax
variable that incorporates the marginal capital gains tax rate across states. This
variable is equal to zero in states with no capital gains taxes and the marginal tax
rate for a median income household in states with capital gains taxes.9 In using
9 Albanesi, Giorgi, and Nosal (2017) provide evidence that real estate investors during the housing boom were
concentrated in the middle and upper echelons of the income distribution. We have also verified that our results
are robust to using the top marginal capital gains tax rate or a dummy indicator variable of whether a state has a
capital gains tax.
5255

Table 1
(Continued)
B. Correlations of non-owner-occupied versus subprime housing purchases with control variables during the
boom period and the p-values of their differences
Fraction of non-owner-occupied Fraction of subprime
Variables home purchases in 2004–2006 mortgages in 2005 p -value
Real house price change in 0.277∗∗∗ 0.383∗∗∗ .000
2004–2006
Per capita income change in 0.206∗∗∗ −0.388∗∗∗ .000
2003–2006
Change in no. of establishments 0.100∗∗∗ −0.066∗∗∗ .000

in 2004–2006
Real payroll change in 2004–2006 0.105∗∗∗ −0.035∗ .000
Employment change in 0.061∗∗∗ −0.041∗∗ .000
2004–2006
Mortgage denial rate in 2005 0.199∗∗∗ 0.704∗∗∗ .000
Fraction of GSE mortgages in −0.015 −0.564∗∗∗ .000
2005
Non-owner-occupied home 0.199∗∗∗ 0.389∗∗∗ .000
mortgage denial rate in 2005
Fraction of GSE mortgages for −0.015 −0.396∗∗∗ .000
non-owner-occupied home in
2005
Population change in 2003–2006 −0.001 0.011 .592
ln(population in 2000) −0.028 0.315∗∗∗ .000
Fraction of the college educated −0.175∗∗∗ −0.553∗∗∗ .000
in 2000
Fraction of the employed in 2000 −0.453∗∗∗ −0.323∗∗∗ .000
Fraction of workforce in 2000 −0.446∗∗∗ −0.224∗∗∗ .000
Median household income in −0.423∗∗∗ −0.404∗∗∗ .428
2000
Poverty rate in 2000 0.393∗∗∗ 0.497∗∗∗ .000
Urban rate in 2000 0.078∗∗∗ 0.182∗∗∗ .000
Fraction of white residents in −0.113∗∗∗ −0.636∗∗∗ .000
2000
Dummy for states with −0.084∗∗∗ 0.201∗∗∗ .000
nonrecourse mortgage law
Dummy for sand states 0.318∗∗∗ 0.308∗∗∗ .648
Saiz’s elasticity 0.034∗ −0.208∗∗∗ .000
Fraction of renters in 2000 0.236∗∗∗ 0.302∗∗∗ .000
Fraction of immigrants in past 5 0.219∗∗∗ −0.157∗∗∗ .000
years (2000)
Fraction of employment in arts 0.580∗∗∗ 0.069∗∗∗ .000
entertainment and recreation in
2000
this instrument, we implicitly assume that the marginal buyer of non-owner-

occupied homes is an in-state resident. According to a survey by the National
Association of Realtors (2015), the typical investment property is 24 miles from
the buyer’s primary residence. This finding suggests that the typical investment
home buyer is likely to be in-state, supporting the relevance requirement of our
instrument. 10 In Section 2, we also conduct a border analysis for states without
10 Out-of-state investment home buyers introduce a nuanced issue. A buyer expects to pay taxes on future capital
gains in both states—the state of residence and the state where the home resides—but may receive tax credits
from the state of residence to offset the double tax incidence. The buyer thus pays the higher tax rate between
the two states. We expect this issue to mostly affect ZIP codes close to the state border. In Section B of the
5256

capital gains taxes to provide evidence of the relevance of state capital gains
taxes as an instrument for housing speculation.
For our instrument to be valid, it also needs to satisfy the exclusion restriction
for causality with respect to the housing boom and the subsequent housing bust
and economic contraction.11 While economic activity in a state might be related
to its treatment of state-level personal taxation, our analysis requires only that, in
absence of omitted variables correlated with both taxes and changes in housing
prices, the relative magnitudes of the housing boom and bust and real outcomes
during the Great Recession were not directly driven by variation in state-level

capital gains taxation during the boom period.12 We believe that this is the case
for several reasons.
First, capital gains taxation is only a part of state taxes. As summarized by
Fajgelbaum et al. (2019), there are several sources of state tax revenue, such as
personal income taxes, corporate income taxes, general sales taxes, selective
sales taxes (e.g., alcohol sales taxes), and license taxes, which accounted for,
respectively, 35%, 7%, 32%, 15%, and 6.2% of total state tax revenue in 2017.
Capital gains taxes are part of personal income taxes, and there is substantial
variation across states in the total state tax revenue, as well as in the rate for
each of these tax components. There is little evidence to suggest that state fiscal
policies and capital gains taxation affect local economic activity. Da et al. (2016)
documents that state fiscal policies have a negligible effect on firm cash flows
and only affect discount rates if a firm has a concentrated investor base. Several
studies, including Walden (2014) and Gale, Krupkin, and Reuben (2015), also
find little relation between the relative size of the public sector (state and local
taxes as a percentage of personal income) and state differences in economic
growth during the recent recovery. In addition, as a small portion of state tax
revenue, capital gains taxes are even unlikely to affect economic performance
directly.
Second, to directly test for a potential relation of state capital gains taxation
with housing prices and local economic activity, we conduct several placebo
tests, available in Section A of the Internet Appendix, for the presample period
Internet Appendix, we repeat our empirical tests after splitting ZIP codes into two subsamples: (1) ZIP codes
within 50 miles of state borders and (2) ZIP codes further than 50 miles from state borders. Consistent with our
relevance assumption, the impact of housing speculation on housing prices and real outcomes is quantitatively
more pronounced in the second subsample, where there is likely less noise from incorrect assignment of the tax
treatment to non-owner-occupied home purchases. Our results are still quantitatively and statistically significant
for the first subsample, despite this classification issue.
11 One concern is that although homeowners can exclude up to $250,000 ($500,000 per couple) of capital gains
from the sale of their primary residence, the capital gains taxes could still affect households’ primary home
demand. As we construct our speculation measure as the fraction of non-owner-occupied homes in all home
purchases, we filter out the potential effect of the primary home demand to mitigate this concern.
12 We use the level of, rather than the change in, state capital gains taxes as our instrument. The level was stable
during the period of 2004–2006. Although the change in taxation might directly shock housing markets and
local economies, the level of state capital gains taxation likely affected only housing speculators’ decisions when
widespread optimism arose during the boom period.
5257

of 1999 to 2001.13 We find an economically and statistically insignificant

link between state capital gains taxation and housing market and economic
outcomes during this period.
Third, we perform several Granger causality tests, available in Section E
of the Internet Appendix, and find that neither contemporaneous, lagged, nor
leading changes in housing prices or economic performance can predict state
capital gains taxation from 1978 to 2010.14 The lack of any evidence directly
linking state capital gains taxation to housing market and economic outcomes
ensures the validity of our instrument.

2. Data Description
Panel A of Table 1 provides summary statistics for a set of variables used in
our analysis.
2.1 Housing speculation

The Home Mortgage Disclosure Act (HMDA) data set includes comprehensive
individual mortgage application and origination data for the United States.
It discloses owner occupancy for each individual mortgage and indicates
whether the mortgage is for a primary residence or a non-owner-occupied
home. We aggregate the HMDA data to the ZIP code level and calculate
the fraction of mortgage originations for non-owner-occupied homes in the
total mortgage origination as our measure of the share of non-owner-occupied
home purchases.15 The fraction of non-owner-occupied home purchases in
2004–2006 has a mean of 13.6% and a standard deviation of 9.9% across ZIP
codes.
Figure 1 depicts the fraction of non-owner-occupied home purchases for
the United States and three cities, New York, Las Vegas, and Charlotte, from
2000 to 2010. Non-owner-occupied home purchases represent a sizable fraction
of mortgage originations, comprising 15.31% of all new originations in the
United States at its peak in 2005. While this measure of non-owner-occupied
home purchases contains both second home and investment home purchases,
13 We report the results for this presample period, because the IRS data start from 1998. The results are also
insignificant for the variables available in the ZIP Code Business Patterns database since 1994.
14 Specifically, we run the state-level panel regressions of marginal state capital gains tax rate for $50,000 in income
(in 2005 USD) on contemporaneous, lagged (1 or 5 years), or leading (1 or 5 years) changes in housing prices,
and economic performance, including wages and salaries, number of establishments, employment, per capita
income, and unemployment rate. We also control for the corporate tax rate, sales tax rate, and year fixed effects.
We find similar results for the $10,000, $25,000, $75,000, and $100,000 income brackets. We also find similar
results if, instead of 1- and 5-year leads and lags, we use 2, 3, or 4 years. These state-level variables come from
Serrato and Zidar (2018) and NBER TAXISM.
15 Haughwout et al. (2011) use the FRBNY Consumer Credit Panel to determine housing investors based on the
number of first-lien mortgage accounts that appear on their credit reports. Their proprietary data are more reliable
than the HMDA data. Chinco and Mayer (2016) identify out-of-town second home buyers by distinguishing
between the property and tax bill mailing addresses in transaction deeds. These data, however, are not as
comprehensive as the HMDA data with which we are able to conduct a nationwide analysis of housing markets.
5258

35%
Share of non-owner-occupied home purchases
30%
25%
20% New York

Las Vegas
15%
Charloe

10% US
5%
0%
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Year
Figure 1
Fraction of non-owner-occupied home purchases
This figure plots the share of non-owner-occupied home purchases for the United States and three cities, New
York, Las Vegas, and Charlotte. The fraction of non-owner-occupied home purchases in each city is computed
from the “Home Mortgage Disclosure Act” data set.
both types of home purchases are at least partially influenced by the motive to
speculate on housing price appreciation, which became a national phenomenon
in the low-interest-rate environment of the mid-2000s. Among the three cities,
Las Vegas had the highest fraction of non-owner-occupied home purchases,
which rose from a level 17.77% in 2000 to 29.41% in 2005, and then dropped
back down to 17.77% in 2008. New York had the lowest fraction, which, while
having a synchronous rise and fall with the other two cities, remained below
7% during this period.
One may be concerned that our measure of speculation has substantial
measurement error, because it does not include investment home purchases
made in cash, and there is evidence of systematic misreporting of owner
occupancy by banks to MBS investors (see, for instance, Piskorski, Seru, and
Witkin 2015). Consistent with the intuition that our measure of speculation
underestimates the true level of speculation in a ZIP code, Table C2 in Section
C of the Internet Appendix confirms that the ordinary least squares (OLS)
estimates of all our coefficients of interest are biased downward compared
to their IV counterparts. To address the potential issue that our results may
be driven by fraud from misreporting, rather than speculation itself, we add
the misreporting measure from Piskorski, Seru, and Witkin (2015), which
measures the mean fraction of loans in a ZIP code with undeclared second
liens or nonowner occupancy status, as a control in all our main regressions in
Section D of the Internet Appendix.16
16 We do not control for it in our main tables because of the loss of ZIP code observations in matching the data.
5259


Figure 2
Distribution of capital gains tax across U.S. states
This figure plots the map of the marginal state tax rates on capital gains for state median income in 2005 across
U.S. states.
2.2 Capital gains instrument

We use the historical state capital gains tax rate as a key instrument for our
analysis of housing speculation. Specifically, we collect state capital gains tax
data from the Tax Foundation and state median income data from the American
Community Survey conducted by the Census Bureau. We construct the measure
of the capital gains tax burden on housing speculation at the state level based
on the historical tax schedule in these states. We exploit variation in the state
capital gains taxation by measuring the marginal capital gains tax burden for the
median-income residents within a state in 2005.Figure 2 maps the distribution
of capital gains taxes at the state level. As shown in this figure, there are nine
states without capital gains tax: Alaska, Florida, Nevada, New Hampshire,
South Dakota, Tennessee, Texas, Washington, and Wyoming. For states with
capital gains taxes, the marginal capital gains tax rates range from 2.1% in
North Dakota to 9% in Oregon, as examples. The mean of the tax burden on
the intensive margin is 4.77% and the standard deviation is 1.27%.17
To demonstrate that state capital gains taxes influenced speculation during the
boom period, we perform a border analysis by examining ZIP codes within 50
miles of the border of states with zero capital gains taxes, excluding Alaska.18
17 Table C1 in Section C of the Internet Appendix reports reduced-form regressions of house price changes and all
our economic outcomes during the cycle on the tax instrument. Although the coefficients are not economically
interpretable in the context of housing speculation, their statistical significance provides evidence of an economic
link between our tax instrument and economic outcomes, which is central for our IV regressions.
18 Mian, Sufi, and Trebbi (2015) used a similar ZIP-code-level analysis around state borders with different
foreclosure laws to justify state judicial requirements as an instrument for foreclosures. They find a jump in
the foreclosure rate at the border between a judicial state and a nonjudicial state.
5260

Panel A of Table 2 tests for a jump across borders in the fraction of non-
owner-occupied home purchases from 2004 to 2006 and in the fraction of
subprime mortgages in 2005 and reveals that the dummy for whether the
state has capital gains taxes significantly negatively predicts our measure of
speculation, but does not affect subprime credit expansion. The fraction of non-
owner-occupied home purchases jumps by 4.9%, and this economic magnitude
is substantial relative to its mean of 13.6% and standard deviation of 9.9%.
Panel A of Figure 3 graphically confirms a discontinuous jump in the fraction
of non-owner-occupied home purchases at the state borders when plotting the

coefficients from the distance regression, whereas panel B demonstrates that
there is no discernable analogous jump in the fraction of subprime mortgages.
To further confirm that our instrument captures only variation in housing
speculation, panel B of Table 2 reports results from border regressions of the
tax dummy on all our control variables and reveals that all coefficients are
statistically insignificant. Our border analysis consequently provides evidence
that our instrument satisfies the relevance condition, and that it instruments for
housing speculation and not for subprime credit expansion.
2.3 House prices

We use ZIP-code-level house price data from the Case-Shiller Home Price
indices, which are constructed from repeated home sales.19 We further deflate
the Case-Shiller Home Price Indices with the Consumer Price Index (CPI) from
the Bureau of Labor Statistics. The real house price change has a mean of 27.8%
in 2004–2006 across the ZIP codes in our sample, and a mean of -41.3% in
2007–2009.
Figure 4 displays the Case-Shiller real house price indices for the United
States and three cities, New York, Las Vegas, and Charlotte, from 2000 to 2010.
The national housing market experienced a significant boom and bust cycle in
the 2000s with the national home price index increasing over 60% from 2000
to 2006, and then falling back to its 2000 level in 2007–2009. New York had
a real housing price appreciation of more than 80% during the boom and then
declined by over 25% during the bust. Charlotte had an almost flat real housing
price level throughout this decade. Interestingly, Las Vegas, which had the most
dramatic rise and fall in non-owner-occupied home purchases, also experienced
the most pronounced housing price expansion—over 120%—during the boom,
and the most dramatic housing price drop—over 50%—during the bust.
2.4 Local economic performance

We collect data on economic performance at the ZIP code level from various
sources. Annual population and annual per capita income at the ZIP code
19 All our results are quantitatively similar and remain significant if we instead use Zillow housing price indices as
our measure of local housing prices. Section F of the Internet Appendix reports these results.
5261

Table 2
ZIP codes near borders of states without capital gains taxes
A.
(1) (2)
Fraction of non-owner-occupied Fraction of subprime
home purchases in 2004–2006 mortgages in 2005
Dummy for states with −0.0487∗∗ 0.00182
capital gains tax (0.0212) (0.0248)
Distance −2.241∗∗ 0.763
(1.039) (0.964)
Distance squared −40.59∗ −37.77∗

(21.32) (19.72)
Distance cubed 1035.5∗∗ −66.09
(397.3) (423.6)
State-border x 10-mile strips Yes Yes
FE
Observations 1,072 1,072
R -squared .347 .185
B.
Dummy for states with Standard
capital gains tax errors
Mortgage denial rate in 2005 0.00267 (0.0111)
Fraction of GSE mortgages in −0.0109 (0.0638)
2005
Non-owner-occupied home −0.0173 (0.0306)
mortgage denial rate in
2005
Fraction of GSE mortgages −0.0690 (0.0602)
for non-owner-occupied
home in 2005
Population change in −0.0368 (0.0277)
2003–2006
ln(population in 2000) 0.114 (0.157)
Fraction of the college −3.478 (2.131)
educated in 2000
Fraction of the employed in −1.560 (1.607)
2000
Fraction of workforce in 2000 −2.025 (1.441)
Median household income in −2689.7 (1972.3)
2000
Poverty rate in 2000 0.296 (1.434)
Urban rate in 2000 −0.998 (9.154)
Fraction of white residents in 3.712 (4.773)
2000
Fraction of renters in 2000 −.002964 (0.0336)
Fraction of immigrants in −.0273 (0.0209)
past 5 years (2000)
Fraction of employment in −.00983 (0.0180)
arts entertainment and
recreation in 2000
Panel A of this table presents discontinuity tests for the fraction of non-owner-occupied home purchases in
2004–2006 (Column 1) and the fraction of subprime mortgages in 2005 (Column 2) in ZIP codes near borders
of states with no capital gains taxes. Panel B shows discontinuity tests for other control variables in ZIP codes
near borders of states with no capital gains taxes. Distance from borders is divided by 1,000 and is negative on
the side of neighboring states with capital gains taxes. We control for distance and its squared and cubic terms
in all regressions (their coefficients are omitted to save space). Distance from borders is divided by 1,000 and is
negative on the side of neighboring states with capital gains taxes. Standard errors are clustered at the state-border
level. *p <.1; **p <.05; ***p <.01.
5262

A Non-owner-occupied home purchases
0.2
Fraction of Non-owner-occupied homes
0.15
0.1
in 2004-06
0.05

0
-0.05
-0.1
-50 -30 -10 10 30 50
Miles from border
B Subprime mortgages
0.2
Fraction of subprime mortgages in 2005
0.15
0.1
0.05
-0.05
-0.1
-0.15
-0.2
-50 -30 -10 10 30 50
Miles from border
Figure 3
ZIP codes near borders of states without capital gains taxes
This figure plots the fraction of non-owner-occupied home purchases in 2004–2006 (panel A) and the fraction
of subprime mortgages in 2005 (panel B) in ZIP codes near borders of states without capital gains taxes. The
variables of interests are regressed on dummies indicating each 1-mile distance from the border (dummies are
negative for neighboring states with capital gains taxes) and state-border*10-mile strip fixed effects. The figure
plots the coefficients on the distance dummies.
level are available from the Internal Revenue Service (IRS). The IRS does
not, however, provide data for 2000 and 2003. We therefore use the data
for 2002 and 2006 to calculate the changes during the boom period and the
5263

250
Normaized home pirce index

200
150
New York
Las Vegas
100
Charloe
50 US

0
20002001200220032004200520062007200820092010
Year
Figure 4
Case-Shiller home price indices
This figure plots the Case-Shiller home price indices for the United States and three cities, New York, Las Vegas,
and Charlotte. The price index is deflated by the CPI and normalized to 100 in 2000.
changes from 2001 to 2002 for the preboom period. Data about annual total
employment, annual payroll, and the number of establishments at the ZIP code
level are available from the ZIP Code Business Patterns database. We include
both resident income and annual payroll from employers because, as argued by
Mian and Sufi (2009), residents in a certain area do not necessarily work in the
same place that they live. The change in per capita income has a mean of 5.2%
in 2003–2006 and a mean of -11.3% in 2007–2009, which is consistent with the
dramatic economic expansion and recession during the boom and bust period.
Similarly, the employment change has a mean of 7.1% (-8.3%), the change in
the number of establishments has a mean of 6.4% (-3.8%), and the real payroll
change has a mean of 8.7% (-10.0%) in 2004–2006 (2007–2009).
ZIP Code Business Patterns database also provides employment data by
establishment size and by industry. For our analysis, we are interested in the
construction industry as it is directly related to the supply side of housing
markets. We also follow Mian and Sufi (2014) to identify nontradable industries
because they produce nontradable goods and services, which reflect the strength
of local demand. Alternatively, we examine the retail and restaurant industries,
which rely on local consumption.
2.5 New housing supply

To measure supply-side activities in local housing markets, we use building
permits from the U.S. Census Bureau, which conducts a survey of permit issuing
all over the United States. Compared with other construction-related measures,
such as housing starts and housing completions, building permits are more
detailed and available at the county level. In addition, building permits are
issued before housing starts and can therefore predict price trends in a timely
5264

6%
Building permits in 2004 to 2006 relave
5%
to the housing units in 2000
4%
New York
3%
Las Vegas
2% Charloe

US
1%
0%
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Year
Figure 5
New housing supply
This figure depicts building permits in 2000 to 2010 relative to the housing units in 2000 for the United States
and three cities, New York, Las Vegas, and Charlotte.
manner.20 A potential weakness of this measure, however, is that the Census

Bureau does not provide building permit data at the ZIP code level. Specifically,
using 2000 U.S. census data, we measure new housing supply at the county
level during the boom period by the building permits issued from 2004 to 2006
relative to the existing housing units in 2000.21 This measure has a mean of
5.6% across counties in our sample and a substantial standard deviation of
5.6%.
Figure 5 depicts the annual building permits granted in 2000–2010 relative to
the number of housing units in 2000 for the United States and for three cities,
New York, Las Vegas, and Charlotte. At the national level, annual building
permits had a modest increase from 1.05% in 2000 to 1.45% in 2005 and then
a substantial drop to 0.38% in 2009. New York saw very little increase in its
housing supply, with annual building permits staying at a flat level of less than
0.4% throughout this decade. Charlotte had a larger increase in supply than
New York in the 2000s. Interestingly, Las Vegas had the most dramatic rise and
fall in annual building permits, rising from 2.03% in 2000 to a level above 5%
in 2005 and 2006, and then dropping to 0.50% in 2009, roughly in sync with
20 Authorization to start is a largely irreversible process, with housing starts being only 2.5% lower than building
permits at the aggregate level, according to https://www.census.gov/construction/nrc/nrcdatarelationships.html,
the Web site of the Census Bureau. Moreover, the delay between authorization and housing start is relatively
short, on average less than 1 month, according to https://www.census.gov/construction/nrc/lengthoftime.html.
These facts suggest that building permits are an appropriate measure of new housing supply.
21 Our results for new housing supply are robust to allocating new building permits at the county level to ZIP codes,
according to the fraction of employment in residential construction in 2000.
5265

the rise and fall of non-owner-occupied home purchases and the housing price
cycle.
2.6 Credit conditions

We include several variables on credit conditions at the ZIP code level to control
for the credit expansion during the recent housing boom. We use mortgages
originated for home purchases and link the lender institutions on the HUD
subprime home lender list to the HMDA data to identify the mortgages issued

to the subprime households. As the HUD subprime home lender list ended
in 2005, we use the fraction of subprime mortgage originations in 2005 as
the share of low-quality loans in the ZIP code during the housing cycle. This
fraction has a mean of 21.1% and a standard deviation of 13.8%. The HMDA
data set also indicates whether a mortgage application is denied by the lender,
and whether the originated mortgage is sold to government sponsored entities
(GSEs). Consequently, we can also control for the mortgage denial rate and the
share of mortgages sold to GSEs in 2005 at the ZIP code level.22 The mortgage
denial rate has a mean of 13.9% and the fraction of GSE mortgages has a
mean of 19.3%.23 We also include the denial rate and the fraction of mortgages
sold to GSEs for non-owner-occupied mortgages to control for the possible
heterogeneous credit conditions to housing investors.
2.7 Other controls

For housing supply elasticity, we employ the widely used elasticity measure
constructed by Saiz (2010). This measure reflects geographic constraints in
home building by defining undevelopable land for construction as terrain with
a slope of 15 degrees or more and as areas lost to bodies of water including
seas, lakes, and wetlands. This measure has a lower value if an area is more
geographically restricted.24
We also control for various economic fundamentals at the ZIP code level. We
use information from the Census Bureau in 2000 including population, fraction
of college-educated population, fraction of workforce, median household
income, poverty rate, urban rate, and the fraction of white residents. In addition,
we control for whether a state is one of the so-called “sand states” (Arizona,
California, Florida, and Nevada) and whether the state has nonrecourse
22 We control for these variables only in 2005, because we use the subprime mortgage fraction in 2005. The results
are unaffected if we instead choose these controls in 2004–2006.
23 We acknowledge that misreporting is common in mortgage data, as emphasized in Griffin and Maturana (2015,
2016). For example, recent studies, such as those by Avery et al. (2012), Blackburn and Vermilyea (2012), and
Mian and Sufi (2015), cast doubt on the accuracy of HMDA data and, in particular, find that the income variable
could be overstated by home buyers. For this reason, we use income data from the IRS and control for the
misreporting measure from Piskorski, Seru, and Witkin (2015) for robustness.
24 The Saiz (2010) measure is not, however, without its issues. Davidoff (2015), for instance, argues that the Saiz
measure is a poor instrument for housing prices, because it is correlated with many variables related to housing
demand.
5266

mortgage laws. As highlighted, for instance, by Nathanson and Zwick (2017)

and Choi et al. (2016), the sand states experienced phenomenal housing
cycles in comparison to the rest of the United States in such outcomes as
mortgage origination, defaults, and housing price fluctuations.25 The nature of
the mortgage laws in a given state has been found to be an important predictor
of real outcomes in the housing market (Dobbie and Goldsmith-Pinkham 2014)
and of speculative activity in the housing market (Nam and Oh 2016).
In addition, we control for other potential sources of secondary housing
demand beyond speculation. We collect the fraction of renters in a ZIP code, as

well as the fraction of immigrants in the past 5 years, to control for long-term
trends in local demographics. We also collect the fraction of employment in
recreation and entertainment to proxy for the appeal of a ZIP code as a vacation
destination. We construct these variables from the 2000 Census data. While
it is infeasible to consider all potential confounding factors in our analysis,
we believe that with theses controls, we provide an improvement over existing
measures of housing speculation to systematically investigate its causal effects.
Note that the limitation of our sample is that Case Shiller housing price data
requires a sufficient number of housing transactions within a ZIP code to be able
to construct indices based on repeated sales of the same house. After including
all these controls, our data sample covers 3,935 ZIP codes during the boom
period and 3,904 ZIP codes during the bust period across the United States.
2.8 Speculation versus subprime

Figure 6 shows that there is little correlation between the distribution of housing
speculation and that of subprime mortgages across ZIP codes. Statistically,
the correlation coefficient between the fraction of non-owner-occupied home
purchases in 2004–2006 and the fraction of subprime mortgages in 2005
is only 0.004 and is insignificant. This suggests that housing speculation
is a phenomenon largely independent of the credit expansion to subprime
households. Instead, our measure of housing speculation captures the purchases
of non-owner-occupied homes by relatively wealthier households in booming
areas.
Panel B of Table 1 provides evidence of where non-owner-occupied versus
subprime housing purchases were prevalent during the boom period by
examining their correlations with our controls. Although both occurred in
areas with similar initial fundamentals in 2000, along demographic dimensions
including the poverty rate, median household income, and employment, and
were pronounced in the four sand states and areas that had higher housing
price appreciation in 2003–2006 and mortgage denial rates in 2005, they
differed along several substantive dimensions during the boom. Speculation
occurred in areas that experienced economic expansions in income, payroll,
25 In Section G of the Internet Appendix, we rerun all our regressions for the sample excluding the four sand states
for robustness. It is reassuring that our results are not affected by their exclusion.
5267

1
.8
.6

.4
.2
0
0 .2 .4 .6 .8
Fraction of subprime mortgages in 2005
Figure 6
Speculation and subprime households
This figure plots the fraction of non-owner-occupied home purchases in 2004 to 2006 against the fraction of
subprime mortgages in 2005 at the ZIP code level.
and employment. Subprime credit expansion, in contrast, flourished in areas

that experienced economic contractions, and that had inelastic housing supplies,
nonrecourse mortgage laws, and a lower fraction of GSE loans.
2.9 Regression analysis

To account for the relative importance of different ZIP codes in the recent U.S.
housing cycle, we conduct all our regression analyses by weighting observations
by the number of households within the ZIP code in 2000. All our results are
robust to employing an equal-weighting scheme instead. In addition, because
our instrument varies across states, we cluster standard errors at the state level in
all regressions. Although our tax instrument is at the state level, it nevertheless
provides substantial cross-sectional variation for us to identify the effects of
speculation during the recent U.S. housing cycle.
3. Economic Consequences
In this section, we examine the cross-section of housing speculation during the
boom period of 2004–2006, and its economic consequences during both the
boom period and the subsequent bust period of 2007–2009.
5268

3.1 Housing cycle

We first examine the link between housing speculation and the housing cycle.
Figure 7 provides a scatterplot of the real housing price changes during the
boom period of 2004–2006 (panel A) and the bust period of 2007–2009 (panel
B) against the fraction of non-owner-occupied home purchases during the boom
period of 2004–2006 at the ZIP code level. The plot displays a clear association
between more intensive housing speculation and both greater housing price
increases during the boom, and greater subsequent housing price collapses
during the bust.

Table 3 reports the two-stage instrumental variable approach to formally
analyze this relation by using the variable of the marginal capital gains tax
rate for the median income household within the state as our instrument.
Specifically, we first regress the fraction of non-owner-occupied home
purchases during the boom period of 2004–2006 on the tax instrument:
Speculationi,boom = a +bTax i +Controlsi +i , (1)
where we use a list of controls, including the supply elasticity measure, the
fraction of subprime mortgages in 2005, the mortgage denial rate in 2005,
the fraction of GSE mortgages in 2005, the mortgage denial rate for non-
owner-occupied mortgages in 2005, the fraction of GSE mortgages for non-
owner-occupied mortgages in 2005, per capita income change in 2003–2006,
population change in 2003–2006, the change in the number of establishments
in 2004–2006, real payroll change in 2004–2006, the employment change in
2004–2006, the natural logarithm of the population in 2000, the fraction of the
college educated in 2000, the fraction of the employed in 2000, the fraction
of workforce in 2000, median household income in 2000, the poverty rate in
2000, the urban rate in 2000, the fraction of white residents in 2000, the fraction
of employment in arts, entertainment, and recreation in 2000, the fraction of
renters in 2000, the fraction of immigrants in 2000, a dummy for whether a
state has nonrecourse mortgage laws, and a dummy for whether a state is a sand
state.
Column 1 of Table 3 shows the first-stage result that the tax instrument
has a significant explanatory power for the fraction of non-owner-occupied
home purchases. The F-statistic of 69.93 provides reassurance that the tax rate
variable is a valid instrument, with regard to relevance, for the fraction of
non-owner-occupied home purchases.
Next, we analyze the causal effect of housing speculation on price expansion
during the boom period, and the price contraction during the bust period.
Specifically, we regress cumulative changes in housing price in 2004–2006 and
in 2007–2009 on the predicted fraction of non-owner-occupied home purchases
during the boom period of 2004–2006, instrumented by our tax rate variable,
following the first-stage regressions:
Pricei,boom or bust = a +bSpeculationi,boom +Controlsi +i , (2)
5269

A
.8
.6
.4

.2
0
-.2
0 .2 .4 .6 .8 1
Fraction of Non-Owner-Occupied Home Purchases in 2004-06
B
0
-.5 -1
-1.5
-2
0 .2 .4 .6 .8 1
Figure 7
Speculation and housing price cycle
These figures plot the real housing price change during the boom period of 2004 to 2006 (panel A) and the bust
period of 2007 to 2009 (panel B) against the fraction of non-owner-occupied home purchases in 2004 to 2006 at
the ZIP code level.
5270

Table 3
Speculation and housing price cycles
(1) (2) (3)
Fraction of non-owner- Real house Real house
occupied home purchases price change price change
in 2004–2006 in 2004–2006 in 2007–2009
State capital gains tax rate for −0.677∗∗
median income
(0.269)
Fraction of non-owner-occupied 2.685∗∗ −3.785∗∗∗
home purchases in 2004–2006

(1.078) (0.892)
Saiz’s elasticity 0.0170∗∗∗ −0.114∗∗∗ 0.106∗∗∗
(0.00472) (0.0228) (0.0231)
Fraction of subprime mortgages −0.216∗∗∗ 0.901∗∗∗ −1.691∗∗∗
in 2005
(0.0402) (0.286) (0.256)
Mortgage denial rate in 2005 0.509∗∗∗ −1.938∗∗∗ 1.086
(0.0819) (0.642) (0.780)
Fraction of GSE mortgages in −0.175∗∗∗ 0.620∗∗ −0.682∗∗∗
2005
(0.0362) (0.301) (0.260)
Non-owner-occupied home −0.133∗∗∗ 0.225 −0.623∗∗∗
(0.0323) (0.213) (0.208)
Fraction of GSE mortgages for 0.0177 −0.0616 0.107
2005
(0.0198) (0.0765) (0.164)
Per capita income change in 0.0954∗∗ 0.0599 0.395∗
2003–2006
(0.0375) (0.185) (0.239)
Population change in 2003–2006 −0.0417∗∗ 0.134 −0.349∗∗∗
(0.0193) (0.0894) (0.112)
Change in no. of establishments 0.0481∗ 0.223∗ 0.254
in 2004–2006
(0.0243) (0.124) (0.204)
Real payroll change in 2004–2006 0.0204∗∗∗ 0.0113 0.0959∗
(0.00560) (0.0296) (0.0490)
Employment change in −0.00610 −0.0101 −0.0813∗
2004–2006
(0.0106) (0.0300) (0.0490)
ln(population in 2000) −0.0161∗∗∗ 0.0352 −0.0706∗∗
(0.00535) (0.0288) (0.0348)
Fraction of the college educated 0.00119∗∗∗ −0.00488∗∗∗ 0.00339∗
in 2000
(0.000241) (0.00146) (0.00197)
Fraction of the employed in 2000 −0.000672 −0.00875 −0.00339
(0.00196) (0.00869) (0.0171)
Fraction of workforce in 2000 −0.00140 0.0114 −0.00239
(0.00227) (0.00968) (0.0171)
Median household income in −0.00000234∗∗∗ 0.00000587∗∗ −0.00000811∗∗∗
2000
(0.000000261) (0.00000285) (0.00000289)
Poverty rate in 2000 0.00233∗∗ −0.00657∗ 0.00632
(0.000919) (0.00397) (0.00449)
Urban rate in 2000 0.000827∗∗∗ −0.00204∗ 0.00249∗∗
(0.000205) (0.00120) (0.00113)
Fraction of white residents in −0.000266 0.0000601 −0.00307∗∗∗
2000
(0.000249) (0.000989) (0.000936)
Fraction of renters in 2000 −0.156∗∗∗ 0.511∗∗ −0.268
(0.0253) (0.202) (0.265)
(Continued)
5271

Table 3
(Continued)
(1) (2) (3)
Fraction of immigrants in 2000 0.191∗∗∗ −0.412 0.401
(0.0346) (0.289) (0.387)
Fraction of employment in arts 0.742∗∗∗ −1.983∗∗ 2.339∗∗∗
2000
(0.117) (0.840) (0.798)
Dummy for states with 0.0123 −0.0155 0.0976∗

(0.0176) (0.0409) (0.0523)
Dummy for sand states −0.0227 0.229∗∗∗ −0.383∗∗∗
(0.0190) (0.0525) (0.0658)
Constant 0.408∗∗∗ −0.458 1.532∗∗∗
(0.111) (0.650) (0.517)
Observations 3,935 3,935 3,935
First-stage F-statistic 69.93 na na
R -squared .617 .539 .463
This table reports the two-stage least squares regressions of the real house price change in 2004–2006 and
2007–2009 on the fraction of non-owner-occupied home purchases in 2004–2006 instrumented with the state
capital gains tax rate for the median income. Column 1 shows the coefficients of the first-stage regression of the
fraction of non-owner-occupied home purchases in 2004–2006 on the state capital gains tax rate for the median
income. Columns 2 and 3 show the coefficients of the second-stage regression of the real house price changes in
2004–2006 and 2007–2009, respectively, on the instrumented fraction of non-owner-occupied home purchases
in 2004–2006. The regressions control for the supply elasticity measure, the fraction of subprime mortgages in
2005, the mortgage denial rate in 2005, the fraction of GSE mortgages in 2005, the mortgage denial rate for
non-owner-occupied mortgages in 2005, the fraction of GSE mortgages for non-owner-occupied mortgages in
2005, per capita income change in 2003–2006, population change in 2003–2006, the change in the number of
establishments in 2004–2006, real payroll change in 2004–2006, employment change in 2004–2006, the natural
logarithm of population in 2000, the fraction of the college educated in 2000, the fraction of the employed in
2000, the fraction of workforce in 2000, median household income in 2000, poverty rate in 2000, urban rate in
2000, the fraction of white residents in 2000, the fraction of employment in arts, entertainment, and recreation in
2000, the fraction of renters in 2000, the fraction of immigrants in 2000, the dummy for states with nonrecourse
mortgage law, and the dummy of sand states. Observations are weighted by the number of households. Standard
errors are clustered at the state level. *p <.1; **p <.05; ***p <.01.
where the left-hand variable is the housing price change during either the boom
period of 2004–2006 or the bust period of 2007–2009, and Speculationi,boom is
the projected housing speculation from the first-stage regression. We also add
the same control variables from the first-stage regression.
Column 2 of Table 3 shows that the IV coefficient estimate of the impact
of housing speculation on housing prices during the boom is significantly
positive, both statistically and in its economic magnitude: a 1-standard-
deviation increase in the fraction of non-owner-occupied home purchases across
ZIP codes causes a substantial price increase of 26.5%. Column 3 of Table 3
shows the IV estimate of the impact of housing speculation on the housing price
contraction during the bust is significantly negative, both statistically and in its
economic magnitude: a 1-standard-deviation increase in the fraction of non-
owner-occupied home purchases across ZIP codes translates to a substantial
price decline of 37.4%. Taken together, our analysis establishes a causal link
between housing speculation during the boom period of 2004–2006 and the
housing boom and bust cycle of 2004–2009.
What drove housing speculation during the boom period? We now examine
extrapolative expectations as a driver of housing speculation. We hypothesize
5272

Table 4
Extrapolation and housing speculation
(1) (2)
Fraction of non-owner-occupied home purchases in 2004–2006
Real house 0.0231 0.229∗∗∗
price change in 2001–2003 (0.0574) (0.0508)
Interaction −5.413∗∗∗
(0.986)
State capital 0.157
gain tax rate for median income (0.288)

Constant 0.283∗∗∗ 0.260∗∗∗
(0.0768) (0.0818)
Controls Yes Yes
This table reports coefficient estimates from regressing the fraction of non-owner-occupied home purchases in
2004–2006 on the house price change in 2001–2003 (Column 1), the state capital gains tax rate for the median
income, and their interaction (Column 2). All regressions control for the host of lagged variables including the
fraction of subprime mortgages in 2002, mortgage denial rate in 2002, the fraction of GSE mortgages in 2002, the
mortgage denial rate for non-owner-occupied mortgages in 2002, the fraction of GSE mortgages for non-owner-
occupied mortgages in 2002, per capita income change in 2001–2002, the population change in 2001–2002, the
change in the number of establishments in 2001–2003, the real payroll change in 2001–2003, the employment
change in 2001–2003, the natural logarithm of population in 2000, the fraction of the college educated in 2000,
the fraction of the employed in 2000, the fraction of workforce in 2000, median household income in 2000,
poverty rate in 2000, urban rate in 2000, the fraction of white residents in 2000, the fraction of employment in
arts, entertainment, and recreation in 2000, the fraction of renters in 2000, and the fraction of immigrants in 2000,
as well as the dummy for states with nonrecourse mortgage law and the dummy of sand states. Observations
are weighted by the number of households. Standard errors are clustered at the state level. *p <.1; **p <.05;
***p <.01.
that non-owner-occupied home purchases in areas with less capital gains

taxation would anchor more strongly on lagged housing price changes, given
that speculators with extrapolative expectations expect to profit more from
housing speculation in these areas. To test this hypothesis, we run the following
regression:
Speculationi,boom
= a +bPricei,pre−boom +cTax i +dPricei,pre−boom ·Tax i +Controlsi +i ,
(3)
where Speculationi,boom is the fraction of non-owner-occupied home purchases
in ZIP code i during the boom period of 2004–2006 and Pricei,pre−boom refers
to the cumulative housing price change during the preboom period of 2001–
2003. We also interact Pricei,pre−boom with the capital gains tax rate Tax i .
Lastly, we also control for a host of local fundamental variables during the
preboom period.
Table 4 displays the results on extrapolation. Without including the state
capital gains tax rate, Column 1 shows positive though statistically insignificant
relationship between housing speculation and the lagged housing price change.
In contrast, when we introduce the state capital gains tax rate into the
regression, as specified in Equation (3). Column 2 shows that, across ZIP codes,
5273

housing price increases during the preboom period significantly predict higher
fractions of non-owner-occupied home purchases during the boom period.
More importantly, housing speculation at the ZIP code level in states with
less capital gains taxes reacted more strongly to the preboom housing price
increase, and this relation is statistically significant at the 1% level.26 This
finding thus provides evidence that housing speculation, anchored on past house
price changes, contributed a nonfundamental source of housing demand during
the boom, especially in areas more prone to speculative behavior.27 This result
also indicates the important role played by state capital gains taxation in shaping

speculators’ extrapolative behavior, validating relevance of this instrument for
our subsequent analysis.
3.2 Economic cycle

By affecting housing prices, housing speculation can also affect local economic
activity. We again apply the two-stage instrumental variable approach to
formally analyze this relation by using the variable of the marginal capital gains
tax rate for the median income household within a state as the instrument. For
the second stage, we run the following regression:
Yi,boom or bust = a +bSpeculationi,boom +Controlsi +i , (4)
where Yi,boom or bust indicates the cumulative change in economic outcomes

(per capita income, the number of establishments, real payroll, and
employment) during either the boom period of 2004–2006 or the bust period
of 2007–2009, and Speculationi,boom is the projected housing speculation from
the first-stage regression specified in Equation (1).
We first examine to what extent housing market speculation contributed
to local economic expansions during the boom period. Table 5, panel A,
reports the results from using the measures of economic activity during the
boom period of 2004–2006 as the dependent variables. Housing speculation
is positively associated with all these measures. Real payroll, as shown in
Column 3, is most heavily affected by local housing speculation during the
boom: a 1-standard-deviation increase in the fraction of non-owner-occupied
home purchases across ZIP codes leads to a substantial increase of 13.7% in
real payroll. Income per capita and employment, as shown in Columns 1 and
26 The Frisch-Waugh theorem offers an alternative interpretation of our results. The coefficient on past housing prices
and its interaction with the tax instrument is equivalent to regressing the residual from regressing the fraction of
non-owner-occupied home purchases on fundamentals on the residuals from regressing past housing prices and
its interaction on fundamentals. The OLS coefficients therefore capture the responsiveness of nonfundamental
housing demand to past housing price growth that is orthogonal to fundamentals and its interaction with the tax
instrument.
27 Consistent with our results, Wheaton and Nechayev (2008) show that a regression forecasting housing price
appreciation systematically underestimates the realized housing price growth between 1998 and 2005 and that
these forecast errors are positively correlated with the percentage of home sales attributed to investors and second
home buyers within an MSA.
5274

Table 5
Real effects of housing speculation
A. The boom period
(1) (2) (3) (4)
Per capita Change in no. Real payroll Employment

income change of establishments change in change in
in 2003–2006 in 2004–2006 2004–2006 2004–2006
Fraction of non-owner-occupied 1.305∗∗∗ 0.687∗∗∗ 1.383∗∗∗ 0.850∗∗∗

(0.347) (0.245) (0.431) (0.301)
−0.0410∗∗∗ −0.0301∗∗∗ −0.0290∗∗∗ −0.0218∗∗∗

Saiz’s elasticity
(0.00887) (0.00605) (0.00502) (0.00408)
Fraction of subprime mortgages in 0.0962 0.128 0.296∗ 0.182∗
2005
(0.0961) (0.0887) (0.152) (0.101)
Mortgage denial rate in 2005 −0.747∗∗∗ −0.390∗∗ −0.869∗∗∗ −0.552∗∗
(0.211) (0.156) (0.296) (0.224)
Fraction of GSE mortgages in 0.0665 0.196∗∗ 0.324∗∗∗ 0.247∗∗∗
2005
(0.0746) (0.0831) (0.106) (0.0888)
Non-owner-occupied home 0.160∗∗ 0.0629 0.153∗ 0.0556
(0.0756) (0.0402) (0.0805) (0.0485)
Fraction of GSE mortgages for −0.0686∗ 0.0410 −0.00600 0.0341
2005
(0.0384) (0.0256) (0.0483) (0.0418)
Population change in 2003–2006 −0.0635 0.324∗∗∗ 0.300∗∗∗ 0.300∗∗∗
(0.0688) (0.0460) (0.0540) (0.0623)
ln(population in 2000) 0.00197 0.0186∗∗∗ 0.0273∗∗∗ 0.0193∗∗∗
(0.0103) (0.00629) (0.00909) (0.00571)
Fraction of the college educated in 0.00167∗∗ −0.000693 −0.00243∗∗∗ −0.00208∗∗∗
2000
(0.000693) (0.000507) (0.000912) (0.000766)
Fraction of the employed in 2000 −0.00263 0.000487 0.00130 0.000343
(0.00324) (0.00201) (0.00383) (0.00333)
Fraction of workforce in 2000 0.00210 0.00184 0.00306 0.00318
(0.00374) (0.00233) (0.00413) (0.00362)
Median household income in 2000 0.00000304∗∗∗ 0.00000104 0.00000394∗∗∗ 0.00000265∗∗∗
(0.000000819) (0.000000665) (0.00000129) (0.000000996)
Poverty rate in 2000 −0.00121 −0.00243∗∗∗ −0.00137 −0.00108
(0.00152) (0.000771) (0.00189) (0.00119)
Urban rate in 2000 −0.00191∗∗∗ −0.000832∗∗ −0.00129∗∗∗ −0.000929∗∗∗
(0.000306) (0.000346) (0.000422) (0.000322)
Fraction of white residents in 2000 0.00109∗∗∗ −0.0000902 0.000475 0.000386
(0.000375) (0.000209) (0.000314) (0.000275)
Fraction of employment in arts −0.792∗∗ −0.397∗ −0.868∗∗ −0.438∗
2000
(0.402) (0.217) (0.365) (0.262)
Fraction of renters in 2000 0.181∗∗∗ −0.0167 0.112 0.0291
(0.0684) (0.0457) (0.102) (0.0700)
Fraction of immigrants in 2000 −0.284∗∗∗ 0.0968 0.0958 0.108∗
(0.0996) (0.0747) (0.0819) (0.0641)
Dummy for states with −0.00485 −0.00965 0.00965 0.00197
(0.0165) (0.0108) (0.0207) (0.0134)
Dummy for sand states 0.0481∗∗ 0.0511∗∗∗ 0.0883∗∗∗ 0.0684∗∗∗
(0.0234) (0.0149) (0.0233) (0.0175)
Constant −0.0314 −0.284∗ −0.704∗∗∗ −0.497∗∗∗
(0.238) (0.166) (0.242) (0.154)
Observations 3,935 3,935 3,935 3,935
R-squared .211 .251 .102 .082
(Continued)
5275

Table 5
(Continued)
B. The bust period
(1) (2) (3) (4)

2007–2009 in 2007–2009 2007–2009 2007–2009
Fraction of non-owner-occupied −0.794∗∗∗ −0.877∗∗∗ −1.563∗∗∗ −1.475∗∗∗

(0.199) (0.205) (0.356) (0.341)
0.00724∗ 0.0281∗∗∗ 0.0246∗∗∗

Saiz’s elasticity −0.00524
(0.00614) (0.00393) (0.00596) (0.00611)
Fraction of subprime mortgages in −0.284∗∗∗ −0.232∗∗∗ −0.378∗∗∗ −0.369∗∗∗
2005
(0.0544) (0.0489) (0.108) (0.0910)
Mortgage denial rate in 2005 0.0191 0.325∗∗∗ 0.730∗∗∗ 0.688∗∗∗
(0.164) (0.115) (0.225) (0.195)
Fraction of GSE mortgages in −0.141∗∗∗ −0.112∗∗ −0.265∗∗∗ −0.230∗∗∗
2005
(0.0480) (0.0503) (0.0997) (0.0891)
Non-owner-occupied home −0.0828∗∗ −0.147∗∗∗ −0.277∗∗∗ −0.202∗∗
(0.0378) (0.0410) (0.0893) (0.0792)
Fraction of GSE mortgages for 0.0387 0.0205 0.0363 0.0123
2005
(0.0416) (0.0152) (0.0437) (0.0314)
Per capita income change in −0.184∗∗ 0.134∗∗∗ 0.261∗∗∗ 0.195∗∗∗
2003–2006
(0.0836) (0.0370) (0.0692) (0.0740)
Population change in 2003–2006 −0.159∗∗∗ 0.164∗∗∗ 0.119∗∗ 0.115∗∗
(0.0514) (0.0238) (0.0505) (0.0476)
Change in no. of establishments in 0.182∗∗∗ 0.191∗∗∗ 0.479∗∗∗ 0.523∗∗∗
2004–2006
(0.0450) (0.0516) (0.0996) (0.0964)
Real payroll change in 2004–2006 0.0300 0.0326∗∗ −0.107∗∗ 0.162∗∗∗
(0.0194) (0.0126) (0.0503) (0.0327)
Employment change in 2004–2006 −0.0113 −0.0134 −0.0133 −0.308∗∗∗
(0.0157) (0.0126) (0.0357) (0.0290)
ln(population in 2000) −0.0217∗∗∗ −0.0144∗∗∗ −0.0249∗∗ −0.0279∗∗∗
(0.00579) (0.00537) (0.0109) (0.00872)
Fraction of the college educated in −0.000532 0.00115∗∗∗ 0.00379∗∗∗ 0.00277∗∗∗
2000
(0.000473) (0.000349) (0.000561) (0.000407)
Fraction of the employed in 2000 −0.00271 −0.00450∗∗ −0.0110∗∗∗ −0.00777∗∗∗
(0.00275) (0.00223) (0.00395) (0.00281)
Fraction of workforce in 2000 0.00259 0.00295 0.00809∗∗ 0.00526
(0.00277) (0.00239) (0.00401) (0.00324)
Median household income in 2000 −0.00000261∗∗∗ −0.00000178∗∗∗ −0.00000479∗∗∗ −0.00000376∗∗∗
(0.000000690) (0.000000616) (0.00000111) (0.000000854)
Poverty rate in 2000 0.00161 0.00182 0.00331 0.00325∗
(0.00115) (0.00122) (0.00204) (0.00190)
Urban rate in 2000 0.0000585 0.000813∗∗∗ 0.00190∗∗∗ 0.00151∗∗∗
(0.000221) (0.000210) (0.000413) (0.000408)
Fraction of white residents in 2000 −0.00143∗∗∗ −0.000621∗∗∗ −0.000887∗ −0.000764∗∗
(0.000278) (0.000191) (0.000460) (0.000342)
Fraction of employment in arts 0.252∗ 0.642∗∗∗ 1.115∗∗∗ 1.091∗∗∗
2000
(0.152) (0.165) (0.297) (0.333)
Fraction of renters in 2000 −0.120∗∗ −0.113∗∗ −0.277∗∗∗ −0.239∗∗∗
(0.0570) (0.0482) (0.0950) (0.0739)
Fraction of immigrants in 2000 −0.0908∗ 0.0276 0.0368 0.0523
(0.0545) (0.0587) (0.107) (0.0905)
(Continued)
5276

Table 5
(Continued)
(1) (2) (3) (4)

in 2003–2006 in 2004–2006 2004–2006 2004–2006
Dummy for states with −0.000547 0.0126∗ 0.0349∗∗ 0.0366∗∗∗

(0.0115) (0.00749) (0.0144) (0.0141)
Dummy for sand states −0.0332∗ −0.0324∗∗∗ −0.0883∗∗∗ −0.0877∗∗∗
(0.0175) (0.0119) (0.0231) (0.0249)

Constant 0.591∗∗∗ 0.305∗∗∗ 0.489∗∗∗ 0.493∗∗∗
(0.0921) (0.116) (0.181) (0.175)
Observations 3,934 3,904 3,904 3,904
R-squared .237 .241 .095 .074
This table reports the two-stage least squares regressions of economic outcomes in 2004–2006 (panel A) and in
2007–2009 (panel B) on the fraction of non-owner-occupied home purchases in 2004–2006 instrumented with
the state capital gains tax rate for the median income. All regressions control for the supply elasticity measure,
the fraction of subprime mortgages in 2005, the mortgage denial rate in 2005, the fraction of GSE mortgages in
2005, the mortgage denial rate for non-owner-occupied mortgages in 2005, the fraction of GSE mortgages for
non-owner-occupied mortgages in 2005, population change in 2003–2006, per capita income change in 2003–
2006, the change in the number of establishments in 2004–2006, real payroll change in 2004–2006, employment
change in 2004–2006, the natural logarithm of population in 2000, the fraction of the college educated in 2000,
the fraction of the employed in 2000, the fraction of workforce in 2000, median household income in 2000,
poverty rate in 2000, urban rate in 2000, the fraction of white residents in 2000, the fraction of employment in
arts, entertainment, and recreation in 2000, the fraction of renters in 2000, the fraction of immigrants in 2000,
the dummy for states with nonrecourse mortgage law, and the dummy of sand states. Observations are weighted
by the number of households. Standard errors are clustered at the state level. *p <.1; **p <.05; ***p <.01.
4, also increase by 12.9% and 8.4%, respectively. Finally, the change in the
number of establishments, shown in Column 2, is the most modest, although
the effect is still economically meaningful: a 1-standard-deviation increase in
the fraction of non-owner-occupied home purchases across ZIP codes translates
to an increase of 6.8% in the number of establishments.
Panel B of Table 5 reports the results of regressing our measures of economic
activity in the bust period of 2007–2009 on the fraction of non-owner-occupied
home purchases during the boom period of 2004–2006, instrumented by our
tax rate variable. Housing speculation is negatively associated with all four
measures of economic consequences at the 1% significance level during the
bust. Among these measures, real payroll, which is shown in Column 3, is most
heavily affected by local housing speculation during the boom: a 1-standard-
deviation increase in the fraction of non-owner-occupied home purchases across
ZIP codes corresponds to a substantial drop of 15.4% in real payroll. The same
increase in housing speculation also corresponds to substantial drops of 7.8%
in income per capita, 8.7% in the number of establishments, and 14.6% in
employment, as shown in Columns 1, 2, and 4, respectively. The variation
across ZIP codes in their economic responses, consequently, reflects not only
differences in the deterioration of local fundamentals and firm adjustment costs
of employment, wages, and establishments but also differences in exposure to
housing speculation during the boom.
As shown in Tables 3 and 5, some of these control variables are also
highly significant. In particular, the fraction of subprime mortgages in 2005
5277

is significantly correlated with the magnitudes of the housing price boom and
bust, as well as our four measures of the local economic downturn during the
bust period, consistent with the findings of Mian and Sufi (2009, 2014).
In addition to the results presented here, we also report robustness analyses
in the Internet Appendix. As two of the four sand states, Florida and Nevada,
have no capital gains taxes, this raises a potential concern that the effect of
housing speculation on the price increase during the boom, and decline during
the bust, might be driven by these two states. Section G of the Internet Appendix
invalidates this concern by repeating Tables 3 and 5 but excluding the four

sand states. Section F illustrates that our results are quantitatively similar,
and remain significant, if we instead use Zillow housing price data. Section
H shows that our results are robust after excluding control variables related
to local economic performance during the boom period that are potentially
correlated with speculation and endogenous to the housing cycle. Finally, Table
C2 in Section C reports the OLS estimates of Tables 3 and 5. Our IV analyses
reveal a consistent downward bias in the OLS estimates. This is consistent with
investment home buyers reducing their demand as house prices increase and,
consequently, having a smaller impact on economic outcomes, as well as with
a downward bias in our measure of speculation because of cash deals and bank
misreporting.
4. Transmission Mechanisms
Having demonstrated a causal relationship between housing speculation during
the boom period and the decline in local economic activity during the bust, we
now investigate several potential transmission mechanisms by which housing
speculation propagated to the real economy during 2007–2009.
4.1 Supply overhang

By driving up housing demand, housing speculation may have boosted the
supply side of the housing market during the boom. The increased housing
supply would then overhang on the housing market and the local economy
during the bust, as argued by Rognlie, Shleifer, and Simsek (2018). This effect
implies that housing speculation during the boom has a stronger negative
predictive power for economic outcomes in the construction sector than for
the nonconstruction sector. To examine this supply overhang effect, we first
examine the impact of housing speculation on housing supply. Given that the
Census Bureau provides building permit data only at the county level, we
carry out the analysis by aggregating non-owner-occupied home purchases
and all other controls to the county level. Figure 8 provides a scatterplot of
the building permits in 2004–2006 relative to the number of housing units in
2000—our measure of new housing supply—against the fraction of non-owner-
occupied home purchases in the same period. The plot vividly illustrates a
positive relation between housing speculation and new housing supply.
5278

.5
.4
.3

.2
.1
0
0 .2 .4 .6 .8
Figure 8
Speculation and new housing supply
This figure plots building permits in 2004 to 2006 relative to the number of housing units in 2000 against the
fraction of non-owner-occupied home purchases in 2004 to 2006 at the county level.
Table 6 then demonstrates a causal link by regressing the new housing

supply measure on the fraction of non-owner-occupied home purchases in
2004–2006, instrumented by the state capital gains tax rate variable. We report
the two-stage results in Columns 1 and 2, from the regressions specified in
Equations (1) and (2), respectively. A difference from our previous analyses is
that the observations in Table 6 are at the county level. We weight observations
by the total number of households at the county level in 2000, and cluster
standard errors at the state level. As shown in Column 1, the tax instrument
also has significant explanatory power for the fraction of non-owner-occupied
home purchases at the county level. The F-statistic of 19.75 from the first
stage suggests that the instrument is statistically strong for this county-level
test. Column 2 reports the second-stage results. The IV coefficient estimate
of the impact of housing speculation on the new supply during the boom is
significantly positive, establishing a causal link between them. Specifically,
a 1-standard-deviation increase in the fraction of non-owner-occupied home
purchases across counties corresponds to a substantial increase of 4.2% in new
housing supply between 2004 and 2006.
We also expect stronger explanatory power for economic outcomes in the
construction sector than other sectors during the boom period. We examine
this effect by returning to a ZIP code level, and panel A of Table 7 confirms
this hypothesis. Housing speculation drove the increase in employment in the
5279

Table 6
Speculation and new housing supply
(1) (2)
Fraction of non-owner- Building permits in 2004–2006
occupied home purchases relative to the housing
in 2004–2006 units in 2000
State capital gains tax rate for −0.822∗∗∗
median income (0.138)
Fraction of non-owner-occupied 0.428∗∗∗
home purchases in 2004–2006 (0.105)
Constant 0.656∗∗∗ −0.263∗∗

(0.104) (0.114)
Controls Yes Yes
Observations 309 309
First-stage F-statistic 19.75 na
This table reports the two-stage least squares regressions of building permits in 2004–2006 relative to the housing
units in 2000 on the fraction of non-owner-occupied home purchases in 2004–2006 instrumented with the state
capital gains tax rate for the median income. Column 1 shows the coefficients of the first-stage regression of
the fraction of non-owner-occupied home purchases in 2004–2006 on the state capital gains tax rate for the
median income. Column 2 shows the coefficients of the second-stage regression of building permits in 2004–
2006 relative to the housing units in 2000 on the instrumented fraction of non-owner-occupied home purchases
in 2004–2006. All regressions control for the supply elasticity measure, the fraction of subprime mortgages in
2005, per capita income change in 2003–2006, population change in 2003–2006, the change in the number of
mortgage law, and the dummy of sand states. Observations are weighted by the number of households at the
county level. Standard errors are clustered at the state level. *p <.1; **p <.05; ***p <.01.
construction sector more dramatically than that in other sectors: a 1-standard-

deviation increase in the fraction of non-owner-occupied home purchases
across ZIP codes corresponds to an increase of 25.1% in employment in the
construction sector (Column 1) and of 6.1% in other industries (Column 2).
The statistically and economically significant impact of housing speculation on
industries other than the construction sector, however, suggests that additional
transmission mechanisms are needed to explain the overall economic effect of
housing speculation.
Supply overhang can both exacerbate the subsequent housing price bust and
reduce demand for new housing, leading to a large decline in construction
activity during the recession. The impact could be even more pronounced
than that in the boom because construction is irreversible. We examine this
effect by returning to a ZIP-code-level analysis of economic performance of
different sectors during the bust period using the same two-stage regressions
as in panel B of Table 7. Column 1 shows that housing speculation leads
to a severe reduction in employment in the construction sector. Consistent
with the supply overhang channel, the IV coefficient estimate shows that the
impact of housing speculation on the construction sector is more than twice
5280

Table 7
Effects of housing speculation on construction and nonconstruction sectors
A. The boom period
(1) (2)
Construction employment Nonconstruction employment
change in 2004–2006 change in 2004–2006
Fraction of non-owner-occupied 2.543∗∗∗ 0.624∗∗∗
home purchases in 2004–2006 (0.811) (0.239)
Constant −0.676 −0.477∗∗∗
(0.513) (0.123)
Controls Yes Yes

B. The bust period
(1) (2)
Construction employment Nonconstruction employment
change in 2007–2009 change in 2007–2009
Fraction of non-owner-occupied −3.422∗∗∗ −1.253∗∗∗
home purchases in 2004–2006 (0.720) (0.330)
Constant 1.034∗∗∗ 0.459∗∗∗
(0.345) (0.167)
Controls Yes Yes
This table reports the two-stage least squares regressions of the employment change in the construction (Column
1) and nonconstruction sectors (Column 2) in 2004–2006 (panel A) and in 2007–2009 (panel B) on the fraction
of non-owner-occupied home purchases in 2004–2006 instrumented with the state capital gains tax rate for the
median income. All regressions control for the supply elasticity measure, the fraction of subprime mortgages
in 2005, the mortgage denial rate in 2005, the fraction of GSE mortgages in 2005, the mortgage denial rate for
2005, population change in 2003–2006, per capita income change in 2003–2006, the change in the number of
as great as that on total employment (reported in panel B of Table 5)—

a 1-standard-deviation increase in the fraction of non-owner-occupied home
purchases across ZIP codes during the boom is associated with a decrease of
33.8% in construction-sector employment during the bust.
In Column 2, we also examine the change in employment in all industries,
except the construction sector. The result is still both statistically and
economically significant: a 1-standard-deviation increase in the fraction of non-
owner-occupied home purchases across ZIP codes causes a decrease of 12.4%
in nonconstruction sector employment. This result suggests that the economic
effects of housing speculation are not restricted to the construction sector.
4.2 Local demand

We also examine an alternative channel for housing speculation to affect the real
economy through local demand. Housing speculation may have exacerbated
the fluctuations in household wealth during the housing boom and bust, which
5281

may affect their consumption. As suggested by Mian, Rao, and Sufi (2013)
and Mian and Sufi (2014), the shock to household consumption would, in
turn, drive the demand for local services. Thus, housing speculation during the
boom may also lead to an economic cycle through this local demand channel.
As such, we expect housing speculation during the boom to have stronger
explanatory power for economic outcomes in industries that are driven by local
demand.
To examine this local demand channel, we use the classification of
nontradable and tradable industries from Mian and Sufi (2014), 28 who define

these sectors based on an industry’s geographical concentration. Nontradable
sectors service local demand within a region, so their locations tend to be
dispersed geographically. In contrast, tradable sectors supply goods to meet
national demand and are less exposed to local economic conditions, and
therefore they should be more concentrated spatially to take advantage of
economic scale and specific resources. As an alternative, we also examine
the restaurant and retail sectors more narrowly, which mainly rely on local
demand.
Table 8 reports the coefficient estimates from the regression of the fraction
of non-owner-occupied home purchases during the boom period on the change
in employment in the nontradable sectors in Column 1, and the retail and
restaurant sectors in Column 3, during both the boom (panel A) and the
bust (panel B) periods using our IV method. Housing speculation has an
economically profound impact on employment in these sectors at the 1%
significance level: an increase of 1-standard-deviation in the share of non-
owner-occupied home purchases in 2004–2006 is associated with an increase
of 8.8% (a decrease of 15.1%) in the employment of nontradable sectors, and
of 8.9% (15.6%) in the employment of retail and restaurant sectors in 2004–
2006 (2007–2009). These economic magnitudes are similar to those for the
change in overall employment, reported in Column 4 of Table 5, and for the
change in nonconstruction employment, reported in Column 2 of Table 7. This
sizable effect on the nontradable sectors, whether broadly or narrowly defined,
indicates that housing speculation during the housing boom had a substantially
adverse effect on local demand during the housing bust.
For comparison, we also include the estimates for the employment change in
tradable industries in Column 2 and the employment change in industries other
than retail and the restaurant business in Column 4. Housing speculation has an
insignificant effect on the employment of tradable industries and on industries
other than retail and the restaurant business during the boom period. During the
bust period, we also find an insignificant impact on the employment of tradable
sectors and that the impact of housing speculation on the retail and restaurant
business was stronger than that on the other sectors (15.6% versus 11.2% from
28 For the detailed classification, refer to appendix Table 1 of Mian and Sufi (2014).
5282

Table 8
Effect of housing speculation: Demand channel
A. The boom period
(1) (2) (3) (4)
Employment Employment Retail and Employment change
change change restaurant in industries other
in nontradable in tradable employment than retail and
industries in industries in change in restaurant in
2004–2006 2004–2006 2004–2006 2004–2006
Fraction of non- owner-occupied home 0.889∗∗∗ 0.558 0.904∗∗∗ 0.448
purchases in 2004–2006 (0.265) (0.685) (0.261) (0.274)

Constant −0.404∗∗ −0.143 −0.469∗∗∗ −0.483∗∗∗
(0.163) (0.327) (0.155) (0.142)
Controls Yes Yes Yes Yes
Observations 3,969 3,898 3,969 3,931
R -squared .078 .015 .075 .060
B. The bust period
(1) (2) (3) (4)
Employment Employment Retail and Employment
change change restaurant change in
in nontradable in tradable employment industries other than
industries in industries in change in retail and restaurant
2007–2009 2007–2009 2007–2009 in 2007–2009
Fraction of non-owner-occupied home −1.534∗∗∗ −0.297 −1.585∗∗∗ −1.131∗∗∗
purchases in 2004–2006 (0.408) (0.680) (0.590) (0.417)
Constant 0.577∗∗ −0.467 0.569∗ 0.381∗∗
(0.227) (0.376) (0.314) (0.184)
Controls Yes Yes Yes Yes
Observations 3,934 3,851 3,935 3,899
R -squared .067 .005 .037 .010
This table reports the two-stage least squares regressions of the employment change in nontradable and tradable
sectors in 2004–2006 (panel A) and in 2007–2009 (panel B) on the fraction of non-owner-occupied home
purchases in 2004–2006 instrumented with the state capital gains tax rate for the median income. Columns 1 and
2, respectively, present the coefficients for nontradable and tradable industries defined by Mian and Sufi (2014).
Columns 3 and 4 present the results for retail and restaurant sectors and industries other than these two sectors,
respectively. All regressions control for the supply elasticity measure, the fraction of subprime mortgages in
2005, population change in 2003–2006, per capita income change in 2003–2006, the change in the number of
a one-standard-deviation increase in housing speculation). As employment in

tradable sectors relies more on national demand, the adverse effects of local
housing speculation are weaker for these industries.
Taken together, our analysis provides evidence that housing speculation
affected real economic outcomes during the Great Recession through the
supply overhang and the local demand channels. Because employment in
residential construction contributes to local demand, these two channels are
5283

likely complementary, and we are reassured at finding that both are significant
in contributing to the severity of the local recessions during the bust.29
5. Conclusion
In this paper, we provide evidence that housing speculation, as measured by the
fraction of non-owner-occupied home purchases, arose from extrapolation by
speculators of past housing price changes. We document how this speculation
during the boom period of 2004–2006 had positive economic consequences

during the boom period, and adverse consequences during the bust period of
2007–2009. We demonstrate this causal relationship by taking advantage of an
instrument based on variation in state capital gains taxation. Our results suggest
that housing speculation had real economic consequences during the boom, by
increasing housing prices and fueling local economic expansions, and during
the recession, by depressing residential construction employment, as a result of
supply overhang, and by reducing local household demand. Taken together, our
analysis reveals that speculation in housing markets, partly driven by behavioral
biases, affected the real economy, both during and in the aftermath of the recent
U.S. housing cycle.
References
Albanesi, S., G. D. Giorgi, and J. Nosal. 2017. Credit growth and the financial crisis: A new narrative. Mimeo,
University of Pittsburgh, Geneva School of Economics and Management, and Boston College.
Adelino, M., A. Schoar, and F. Severino. 2015. House prices, collateral, and self-employment. Journal of
Financial Economics 117:288–306.
Avery, R. B., N. Bhutta, K. Brevoort, and G. B. Canner. 2012. The mortgage market in 2011: Highlights from
data reported under the Home Mortgage Disclosure Act. Federal Reserve Bulletin 98:1–46.
Barlevy, G., and J. D. M. Fisher. 2011. Mortgage choice and housing speculation. Mimeo, Federal Reserve Bank
of Chicago.
Bayer, P., C. Geissler, K. Mangum, and J. W. Roberts. 2015. Speculators and middlemen: The strategy and
performance of investors in the housing market. Mimeo, Duke University, Georgia State University, ISO New
England, and NBER.
Blackburn, M. K. L., and T. Vermilyea. 2012. The prevalence and impact of misstated incomes on mortgage loan
applications. Journal of Housing Economics 21:151–68.
Case, K. E., and R. J. Shiller. 2003. Is there a bubble in the housing market? Brookings Papers on Economic
Activity 2003:299–362.
29 Although we focus on only two channels, the literature has identified several other potential transmission
mechanisms. By reducing the collateral value of housing (a widely used collateral for firms to raise debt
financing), housing speculation may have affected firms’ access to credit during the housing bust, as studied
in Adelino, Schoar, and Severino (2015) and Schmalz, Sraer, and Thesmar (2017). Another channel is through
the impairment of intermediary balance sheets of local banks during the bust, which may have prevented them
from lending to local firms, as highlighted in Gan (2007) and He and Krishnamurthy (2013). Housing speculation
may also affect local economic activity through a crowding out effect. By encumbering mortgage financing from
local banks, housing speculation may have crowded out limited bank financing to nonhousing investment, as
suggested by Chakraborty, Goldstein, and MacKinlay (2018). We leave it to future research to analyze these
potentially important channels.
5284

Case, K. E., and R. J. Shiller, and A. Thompson. 2012. What have they been thinking? Homebuyer behavior in
hot and cold markets. Brookings Papers on Economic Activity 45:265–315.
———. 2015. What have they been thinking? Homebuyer behavior in hot and cold markets — A 2014 update.
Discussion Paper, Cowles Foundation.
Chakraborty, I., I. Goldstein, and A. MacKinlay. 2018. Housing price booms and crowding-out effects in bank
lending. Review of Financial Studies 31:2806–53.
Charles, K. K., E. Hurst, and M. J. Notowidigdo. 2016a. Housing booms and busts, labor market opportunities,
and college attendance. Mimeo, University of Chicago and Northwestern.

———. 2016b. Housing booms, manufacturing decline, and labor market outcomes. Mimeo, University of
Chicago and Northwestern.
Chen, T., L. X. Liu, W. Xiong, and L. Zhou. 2016. The speculation channel and crowding out channel: Real
estate shocks and corporate investment in China. Mimeo, Princeton University.
Cheng, I. H., S. Raina, and W. Xiong. 2014. Wall Street and the housing bubble. American Economic Review
104:2797–829.
Chinco, A., and C. Mayer. 2016. Misinformed speculators and mispricing in the housing market. Review of
Financial Studies 29:486–522.
Choi, H. S., H. G. Hong, J. D. Kubik, and J. P. Thompson. 2016. Sand states and the U.S. housing crisis. Mimeo,
Singapore School of Management, Columbia University, Syracuse University, and Federal Reserve Board.
Da, Z., M. Warachka, and H. Yun. 2016. Fiscal policy, consumption risk, and stock returns: Evidence from U.S.
states. Mimeo, University of Notre Dame, University of San Diego, and Michigan State University.
Davidoff, T. 2015. Supply constraints are not valid instrumental variables for home prices because they are
correlated with many demand factors. Mimeo, UBC.
DeFusco, A. A., C. G. Nathanson, and E. Zwick. 2017. Speculative dynamics of prices and volume. Mimeo,
Kellogg School of Management and University of Chicago Booth.
Dobbie, W., and P. Goldsmith-Pinkham. 2014. Debtor protections and the Great Recession. Working Paper.
Fajgelbaum, P. D., E. Morales, J. C. S. Serrato, and O. M. Zidar. 2019. State taxes and spatial misallocation.
Review of Economic Studies 86:333–76.
Favara, G., and J. Imbs. 2015. Credit supply and the price of housing. American Economic Review 105:958–92.
Favilukis, J., S. Ludvigson, and S. V. Nieuwerburgh. 2017. Macroeconomic implications of housing wealth,
housing finance, and limited risk-sharing in general equilibrium. Journal of Political Economy 125:140–223.
Ferreira, F., and J. Gyourko. 2018. Anatomy of the beginning of the housing boom: US neighborhoods and
metropolitan areas. Mimeo, NBER.
Gale, W. G., A. Krupkin, and K. S. Reuben. 2015. The relationship between taxes and growth at the state level:
New evidence. Mimeo, Brookings Institute.
Gan, J. 2007. The real effects of asset market bubbles: Loan- and firm-level evidence of a lending channel. Review
of Financial Studies 20:1941–73.
Glaeser, E. L. 2013. A nation of gamblers: Real estate speculation and American history. American Economic
Review Papers and Proceedings 103:1–42.
Glaeser, E. L., J. Gyourko, and A. Saiz. 2008. Housing supply and housing bubbles. Journal of Urban Economics
64:198–217.
Glaeser, E. L., and C. G. Nathanson. 2017. An extrapolative model of house price dynamics. Journal of Financial
Economics 126:147–70.
Griffin, J. M., and G. Maturana. 2016. Who facilitated misreporting in securitized loans? Review of Financial
Studies 29:384–419.
5285

———. 2016. Did dubious mortgage origination practices distort house prices? Review of Financial Studies
29:1671–708.
Haughwout, A., D. Lee, J. S. Tracy, and W. van der Klaauw. 2011. Real estate investors, the leverage cycle, and
the housing market crisis. Federal Reserve Bank of New York Staff Reports.
Haughwout, A., R. W. Peach, J. Sporn, and J. Tracy. 2013. The supply side of the housing boom and bust of
the 2000s. In Housing and the financial crisis, eds. E. L. Glaeser, and T. Sinai, 69–104. Chicago: University of
Chicago Press.
He, Z., and A. Krishnamurthy. 2013. Intermediary asset pricing. American Economic Review 103:732–70.

Himmelberg, C., C. Mayer, and T. Sinai. 2005. Assessing high house prices: Bubbles, fundamentals, and
misperceptions. Journal of Economic Perspectives 19:67–92.
Hurst, E., B. J. Keys, A. Seru, and J. S. Vavra. 2016. Regional redistribution through the US mortgage market.
American Economic Review 106:2982–3028.
Justiano, A., G. Primiceri, and A. Tambolatti. 2017. Credit supply and the housing boom. Mimeo, Federal Reserve
Bank of Chicago and Northwestern University.
Kaplan, G., K. Mitman, and G. L. Violante. 2017. The housing boom and bust: Model meets evidence. Mimeo,
University of Chicago, Stockholm University, and Princeton University.
Keys, B. J., T. Mukherjee, A. Seru, and V. Vig. 2009. Financial regulation and securitization: Evidence from
subprime mortgage loans. Journal of Monetary Economics 56:700–720.
———. 2010. Did securitization lead to lax screening? Evidence from subprime loans. Quarterly Journal of
Economics 12:307–62.
Landvoigt, T., M. Piazzesi, and M. Schneider. 2015. The housing market(s) of San Diego. American Economic
Review 105:1371–407.
Mayer, C. 2011. Housing bubbles: A survey. Annual Review of Economics 3:559–77.
Mayer, C., and T. Sinai. 2009. U.S. House price dynamics and behavioral finance. In Policy making insights from
behavioral economics, eds. C. L. Foote, L. Goette, and S. Meier, 261–308. Boston: Federal Reserve Bank of
Boston.
Mian, A., K. Rao, and A. Sufi. 2013. Household balance sheets, consumption, and the economic slump. Quarterly
Journal of Economics 128:1687–726.
Mian, A., and A. Sufi. 2009. The consequences of mortgage credit expansion: Evidence from the U.S. mortgage
default crisis. Quarterly Journal of Economics 124:1449–96.
———. 2010. Household leverage and the recession of 2007-09. IMF Economic Review 58:74–117.
———. 2011. House prices, home equity-based borrowing, and the US household leverage crisis. American
Economic Review 101:2132–56.
———. 2014. What explains the 2007–2009 drop in employment? Econometrica 82:2197–223.
———. 2015. Fraudulent income oversight on mortgage applications during the credit expansion of 2002 to
2005. Mimeo, University of Chicago and Princeton University.
Mian, A., A. Sufi, and F. Trebbi. 2015. Foreclosures, house prices, and the real economy. Journal of Finance
70:2587–634.
Nathanson, C. G., and E. Zwick. 2017. Arrested development: Theory and evidence of supply-side speculation
in the housing market. Mimeo, University of Chicago and Northwestern University.
Nam, T. Y., and S. Oh. 2016. Non-recourse mortgage law and housing speculation. Mimeo, OCC and Peking
University.
National Association of Realtors. 2015. Investment and vacation home buyers survey 2015.
5286

Piazzesi, M., and M. Schneider. 2009. Momentum traders in the housing market: Survey evidence and a search
model. American Economic Review Papers and Proceedings 99:406–11.
Piskorski, T., A. Seru, and J. Witkin. 2015. Asset quality misrepresentation by financial intermediaries: Evidence
from the RMBS market. Journal of Finance 70:2635–78.
Purnanandam, A. 2011. Originate-to-distribute model and the subprime mortgage crisis. Review of Financial
Studies 24:1881–915.
Rognlie, M., A. Shleifer, and A. Simsek. 2018. Investment hangover and the great recession. American Economic
Journal: Macroeconomics 10(2): 113-53.

Saiz, A. 2010. The geographic determinants of housing supply. Quarterly Journal of Economics 125:1253–96.
Schmalz, M. C., D. A. Sraer, and D. Thesmar. 2017. Housing collateral and entrepreneurship. Journal of Finance
72:99–132.
Serrato, J. C. S., and O. Zidar. 2018. The structure of state corporate taxation and its impact on state tax revenues
and economic activity. Journal of Public Economics 167:158–76.
Shan, H. 2011. The effect of capital gains taxation on home sales: Evidence from the Taxpayer Relief Act of
1997. Journal of Public Economics 95:177–88.
Shiller, R. J. 2009. Unlearned lessons from the housing bubble. In Economicists’ voice, eds. A. Belke and G.
Schnabl, 1553–3832. Berlin, Germany: Walter de Gruyte.
Soo, C. K. 2018. Quantifying sentiment with news media across local housing markets. Review of Financial
Studies 31:3689–719.
Walden, M. L. 2014. Recovery from the Great Recession: Explaining differences among states. Journal of
Regional and Policy Analysis 44:166–74.
Wheaton, W., and G. Nechayev. 2008. The 1998-2005 housing “bubble” and the current “correction”: What’s
different this time? Journal of Real Estate Research 30:1–26.
5287

Speculation

Uploaded by

Speculation

Uploaded by

Economic Consequences of Housing

Downloaded from https://academic.oup.com/rfs/article/33/11/5248/5803087 by Princeton University user on 08 November 2020

By exploiting variation in state capital gains taxation as an instrument, we analyze the

Economists have long been concerned with the economic consequences of

The Review of Financial Studies 33 (2020) 5248–5287

[23:07 27/9/2020 RFS-OP-REVF200031.tex] Page: 5248 5248–5287

Downloaded from https://academic.oup.com/rfs/article/33/11/5248/5803087 by Princeton University user on 08 November 2020

[23:07 27/9/2020 RFS-OP-REVF200031.tex] Page: 5249 5248–5287

Downloaded from https://academic.oup.com/rfs/article/33/11/5248/5803087 by Princeton University user on 08 November 2020

[23:07 27/9/2020 RFS-OP-REVF200031.tex] Page: 5250 5248–5287

Downloaded from https://academic.oup.com/rfs/article/33/11/5248/5803087 by Princeton University user on 08 November 2020

[23:07 27/9/2020 RFS-OP-REVF200031.tex] Page: 5251 5248–5287

housing speculation. Last, we investigate potential transmission mechanisms

Downloaded from https://academic.oup.com/rfs/article/33/11/5248/5803087 by Princeton University user on 08 November 2020

[23:07 27/9/2020 RFS-OP-REVF200031.tex] Page: 5252 5248–5287

Downloaded from https://academic.oup.com/rfs/article/33/11/5248/5803087 by Princeton University user on 08 November 2020

[23:07 27/9/2020 RFS-OP-REVF200031.tex] Page: 5253 5248–5287

To construct such an instrument, we take advantage of the heterogeneity

Downloaded from https://academic.oup.com/rfs/article/33/11/5248/5803087 by Princeton University user on 08 November 2020

[23:07 27/9/2020 RFS-OP-REVF200031.tex] Page: 5254 5248–5287

Downloaded from https://academic.oup.com/rfs/article/33/11/5248/5803087 by Princeton University user on 08 November 2020

[23:07 27/9/2020 RFS-OP-REVF200031.tex] Page: 5255 5248–5287

Downloaded from https://academic.oup.com/rfs/article/33/11/5248/5803087 by Princeton University user on 08 November 2020

this instrument, we implicitly assume that the marginal buyer of non-owner-

[23:07 27/9/2020 RFS-OP-REVF200031.tex] Page: 5256 5248–5287

Downloaded from https://academic.oup.com/rfs/article/33/11/5248/5803087 by Princeton University user on 08 November 2020

[23:07 27/9/2020 RFS-OP-REVF200031.tex] Page: 5257 5248–5287

of 1999 to 2001.13 We find an economically and statistically insignificant

Downloaded from https://academic.oup.com/rfs/article/33/11/5248/5803087 by Princeton University user on 08 November 2020

2.1 Housing speculation

[23:07 27/9/2020 RFS-OP-REVF200031.tex] Page: 5258 5248–5287

20% New York

Downloaded from https://academic.oup.com/rfs/article/33/11/5248/5803087 by Princeton University user on 08 November 2020

[23:07 27/9/2020 RFS-OP-REVF200031.tex] Page: 5259 5248–5287

Downloaded from https://academic.oup.com/rfs/article/33/11/5248/5803087 by Princeton University user on 08 November 2020

2.2 Capital gains instrument

[23:07 27/9/2020 RFS-OP-REVF200031.tex] Page: 5260 5248–5287

Downloaded from https://academic.oup.com/rfs/article/33/11/5248/5803087 by Princeton University user on 08 November 2020

2.3 House prices

2.4 Local economic performance

[23:07 27/9/2020 RFS-OP-REVF200031.tex] Page: 5261 5248–5287

Downloaded from https://academic.oup.com/rfs/article/33/11/5248/5803087 by Princeton University user on 08 November 2020

[23:07 27/9/2020 RFS-OP-REVF200031.tex] Page: 5262 5248–5287

A Non-owner-occupied home purchases

Downloaded from https://academic.oup.com/rfs/article/33/11/5248/5803087 by Princeton University user on 08 November 2020

[23:07 27/9/2020 RFS-OP-REVF200031.tex] Page: 5263 5248–5287

Normaized home pirce index

Downloaded from https://academic.oup.com/rfs/article/33/11/5248/5803087 by Princeton University user on 08 November 2020

2.5 New housing supply

[23:07 27/9/2020 RFS-OP-REVF200031.tex] Page: 5264 5248–5287

Downloaded from https://academic.oup.com/rfs/article/33/11/5248/5803087 by Princeton University user on 08 November 2020

manner.20 A potential weakness of this measure, however, is that the Census

[23:07 27/9/2020 RFS-OP-REVF200031.tex] Page: 5265 5248–5287

2.6 Credit conditions

Downloaded from https://academic.oup.com/rfs/article/33/11/5248/5803087 by Princeton University user on 08 November 2020

2.7 Other controls

[23:07 27/9/2020 RFS-OP-REVF200031.tex] Page: 5266 5248–5287

mortgage laws. As highlighted, for instance, by Nathanson and Zwick (2017)

Downloaded from https://academic.oup.com/rfs/article/33/11/5248/5803087 by Princeton University user on 08 November 2020

2.8 Speculation versus subprime

[23:07 27/9/2020 RFS-OP-REVF200031.tex] Page: 5267 5248–5287

Downloaded from https://academic.oup.com/rfs/article/33/11/5248/5803087 by Princeton University user on 08 November 2020

and employment. Subprime credit expansion, in contrast, flourished in areas

2.9 Regression analysis

[23:07 27/9/2020 RFS-OP-REVF200031.tex] Page: 5268 5248–5287

3.1 Housing cycle

Downloaded from https://academic.oup.com/rfs/article/33/11/5248/5803087 by Princeton University user on 08 November 2020

[23:07 27/9/2020 RFS-OP-REVF200031.tex] Page: 5269 5248–5287

Downloaded from https://academic.oup.com/rfs/article/33/11/5248/5803087 by Princeton University user on 08 November 2020

[23:07 27/9/2020 RFS-OP-REVF200031.tex] Page: 5270 5248–5287

Downloaded from https://academic.oup.com/rfs/article/33/11/5248/5803087 by Princeton University user on 08 November 2020

[23:07 27/9/2020 RFS-OP-REVF200031.tex] Page: 5271 5248–5287

Downloaded from https://academic.oup.com/rfs/article/33/11/5248/5803087 by Princeton University user on 08 November 2020

[23:07 27/9/2020 RFS-OP-REVF200031.tex] Page: 5272 5248–5287

Downloaded from https://academic.oup.com/rfs/article/33/11/5248/5803087 by Princeton University user on 08 November 2020

that non-owner-occupied home purchases in areas with less capital gains