Stimulating Housing Markets

• Date: 01 February 2020
• Author: ERIC ZWICK,NICHOLAS TURNER,DAVID BERGER
• Published date: 01 February 2020
• DOI: http://doi.org/10.1111/jofi.12847

THE JOURNAL OF FINANCE •VOL. LXXV, NO. 1 •FEBRUARY 2020

Stimulating Housing Markets

DAVID BERGER, NICHOLAS TURNER, and ERIC ZWICK∗

ABSTRACT

We study temporary fiscal stimulus designed to support distressed housing mar-

kets by inducing demand from buyers in the private market. Using difference-in-

differences and regression kink research designs, we find that the First-Time Home-

buyer Credit increased home sales by 490,000 (9.8%), median home prices by $2,400

(1.1%) per standard deviation increase in program exposure, and the transition rate

into homeownership by 53%. The policy response did not reverse immediately.Instead,

demand comes from several years in the future: induced buyers were three years

younger in 2009 than typical first-time buyers. The program’s market-stabilizing

benefits likely exceeded its direct stimulus effects.

IN THE AFTERMATH OF THE GREAT RECESSION, the U.S. housing market suffered

extraordinary distress (Figure 1). As house price growth slowed, a shortage

of prospective homebuyers caused vacancies to rise and housing inventory to

double from 2004 to 2006 and remain at historic levels through 2008. The boom

coincided with a rapid, widespread rise in household debt secured by real estate

(Mian and Sufi (2015)). When house prices began to fall, defaults, foreclosures,

and further downward pressure on prices ensued (Campbell, Giglio, and Pathak

(2011), Mian, Sufi, and Trebbi (2015), Guren and McQuade (2015)). By mid-

2008, the composition of home sales had shifted dramatically, with nearly 40%

classified as distressed or foreclosure sales.

∗David Berger is at Duke University and the National Bureau of Economic Research (NBER).

Nicholas Turner is at the Federal Reserve Board. Eric Zwick is at Chicago Booth and NBER.

We thank Andrew Abel, Gene Amromin, Michael Best, Jediphi Cabal, Anthony DeFusco, Paul

Goldsmith-Pinkham, Adam Guren, Erik Hurst, Anil Kashyap, Amir Kermani, Ben Keys, Henrik

Kleven, Pat Langetieg, Adam Looney, Janet McCubbin, Matt Notowidigdo, Christopher Palmer,

Jonathan Parker, Amit Seru, Isaac Sorkin, Johannes Stroebel, Amir Sufi, Joe Vavra, Rob Vishny,

Owen Zidar, and seminar and conference participants for comments, ideas, and help with data.

Tianfang Cui, Prab Upadrashta, Iris Song, and Caleb Wroblewski provided excellent research

assistance. The views expressed here are ours and do not necessarily reflect those of the U.S.

Treasury Office of Tax Analysis, the Internal Revenue Service (IRS) Office of Research, Analysis

and Statistics, or the Federal Reserve Board. We all have no relevant or material financial interests

that relate to the research described in this paper. The underlying individual-level tax data were

accessed while Turner worked as staff economist in the Office of TaxAnalysis in the U.S. Treasury.

To comply with Internal Revenue Code (IRC) 6103(j) that defines permissible uses of tax return

data, Treasury civil servants had the right to review preliminary drafts. These reviews focused

on protecting taxpayers from risk of disclosure and did not influence the structure or content of

the paper in a material way. Zwick gratefully acknowledges financial support from the Neubauer

Family Foundation, Initiative on Global Markets, and Booth School of Business at the University

of Chicago.

DOI: 10.1111/jofi.12847

C2019 the American Finance Association

277

278 The Journal of Finance R

Figure 1. The state of the housing market. Panel A plots seasonally adjusted housing inven-

tory, defined as the number of homes listed for sale, from the National Association of Realtors

(NAR). The vertical markers correspond to the FTHC loan program (V1), the start of the FTHC

grant program (V2), the scheduled expiration of the FTHC grant program, and the actual expi-

ration of the FTHC grant program (V3), respectively. Panel B plots the month-by-month share

of existing home sales in DataQuick in each of three categories: nondistress resales, short sales,

and institution-owned sales (REO) or foreclosures. (Color figure can be viewed at wileyonlineli-

brary.com)

Stimulating Housing Markets 279

The debt-induced overhang in the housing market prompted many policy

responses, including debt renegotiation programs to repair household balance

sheets, government asset purchases to support financial markets, and mone-

tary and fiscal policy to spur demand growth.1Yet these policies do not directly

target the problem of housing supply overhang, nor do they promote realloca-

tion when houses are vacant or no longer held by high-value users.

This paper evaluates a complementary policy, the First-Time Homebuyer

Credit (FTHC), which was a $20 billion stimulus program designed to support

U.S. housing markets with a temporary tax incentive for new homebuyers

between 2008 and 2010. We combine data from administrative tax records

with transaction deeds data to measure program exposure and housing market

outcomes for approximately 9,000 ZIP codes, which account for 69% of the

U.S. population. We use difference-in-differences and regression kink research

designs to estimate the effect of the policy on home sales, homeownership, and

the housing market more broadly.

We present five main findings. First, the policy proved effective at spurring

home sales. We estimate that the FTHC increased the number of home sales

during the policy period by 490,000 units nationally. Second, the surge in home

sales did not reverse immediately in the year following the policy period. In-

stead, demand appears to have come from several years in the future. Third,

the policy induced transitions into homeownership. We estimate that receiving

the FTHC increased the likelihood of being a first-time homebuyer by over 50%.

Fourth, the policy response came mainly via existing home sales, implying that

the direct stimulative effects of the program were small. Fifth, the health of the

housing market, as reflected in house prices, improved. A back-of-the-envelope

calculation suggests that the consumption response to the increase in house

prices was likely larger than the policy’s direct stimulus effect.

We first document the effect of the FTHC on home sales. Our difference-

in-differences design compares ZIP codes at the same point in time whose

exposure to the program differs. We define program exposure based on the

number of potential first-time homebuyers in a ZIP code. ZIP codes with few

potential first-time homebuyers serve as a “control group” because the policy

does not induce many people to buy in these places. We measure exposure as

the year-2000 share of people in a ZIP code who are first-time homebuyers.

The key threat to this design is the possibility that time-varying, ZIP-specific

shocks are correlated with our exposure measure. We assess this threat in four

ways. First, we present graphical evidence of parallel pre-policy trends, clear

breaks during the policy period, and spikes at policy expiration. Second, we

show that the results are robust to including city-by-time fixed effects, to us-

ing varying weighting schemes and sample definitions, and to adding explicit

controls for exposure to the subprime bust and to all major contemporaneous

1Diamond and Rajan (2011), French et al. (2010), Shleifer and Vishny (2010a), Hanson,

Kashyap, and Stein (2011), and Eberly and Krishnamurthy (2014) discuss potential policy so-

lutions. A recent empirical literature evaluates some of the programs to address debt overhang

during the Great Recession (Agarwal et al. (2017a,2017b)).