News & Events

Quick Links


Insight | Capital Spillover, House Prices, and Consumer Spending: Quasi-Experimental Evidence from House Purchase Restrictions

November 15, 2021

The paper exploits a quasi-natural experiment in China where local authorities-imposed restrictions on investment home purchases in 2016 and 2017. While these restrictions were effective in containing the surge in local house prices, they triggered capital flight into nearby, unregulated housing markets. House prices in these unregulated cities rose sharply following the out-of-town home purchases despite any improvement in local housing fundamentals. Consumption spending on automobiles increased following the housing wealth increase.

The paper uses a difference-in-differences design to estimate the spillover effects of house purchase restrictions (HPRs). It splits the unregulated prefectural cities into two similarly sized groups, treated cities and control cities, based on each city's distance to the nearest regulated city. The treated cities are within 250 kilometers of the nearest regulated cities. Since the closeness facilitates occasional visits to acquire information on individual houses and to monitor the status of the houses once purchased, treated cities are more likely to attract investors from the regulated cities, and they are more exposed to the capital flight induced by these restrictions.

House prices in the treated cities increased sharply relative to the control cities after the large cities implemented their HPRs. Volumes of home sales also abnormally increased, with a similar size as the reduction in home sales in the regulated cities. Moreover, the paper also documents that web searches from the large, regulated cities for houses in the treated cities also substantially surge. Consistent with a "capital flow", bank deposits increase in the treated cities, in a similar size to the increase in home sale volumes. Such documented abnormal activities in house prices, home sale volume and web searches in the treated unregulated cities are especially salient in cases when the effect of the HPR felt strongest in the large regulated cities.

A key challenge in the finance academia has been to precisely estimate the causal economic effect of house prices out of correlational data. For example, house prices could correlate with household spending simply because they are both affected by people's expectation of future income growth. But that does not mean that house prices "drive" household spending. The quasi-experiment above solves this challenge. Because rents and other dimensions of local fundamentals are all observed to show no different trends in the treated and control cities, the only causal factor is the house price shock. The paper finds that consumer spending in the treated cities increases abnormally, with a marginal propensity to consume (MPC) on housing wealth of 0.048. However, the positive aggregate spending response consists two countervailing effects on two groups of local citizens. Consumer spending of the richer locally-born residents and homeowners in the treated cities rise because of an increased perception of wealth. However, the less wealthy non-locally born residents and renters in the treated cities had to cut back on spending in face of the increasingly higher house prices. Therefore, unintended effect of the HPR policy in the large cities on the nearby unregulated cities is not only consequential, but importantly also distributional.

Beyond providing a quasi-experimental solution to estimate the causal effect of house prices on consumer spending across the homeownership and wealth distribution, this paper makes two additional contributions to the literature. It empirically verifies a new mechanism that generates spillovers between housing markets in which HPRs in a "hot" local housing market force a capital flight and cause housing booms in nearby previously "cold" markets. The analysis recognizes that policies designed to achieve locally optimal outcomes may generate unintended spillovers (Farhi and Werning, 2017; Rodrik, 2019). It also verifies previous theoretical conjectures (Favilukis and Van Nieuwerburgh, 2021) that out-of-town investors in local housing markets would affect local citizen's economic life through their impact house prices and provides the first detailed estimate on their effect of local citizen's consumer spending.

About the author

Yu Zhang is an Assistant Professor of Finance at Guanghua School of Management, Peking University. He received a Ph.D. degree in economics from Princeton University after studying at the Department of Economics and the Bendheim Center for Finance. His research is published on journals including theReview of Financial Studies, Economic Theory and Economic Letters,and focuses on the intersection of macroeconomics and finance, with applications on housing, household finance, asset pricing, and supply chain finance.