From military to market: The economic impact of large-scale housing supply shocks

• Applicants: Professor Dr. Fabian Bald; Professor Felix Weinhardt, Ph.D.
• Subject Area: Economic Policy, Applied Economics
• Term: since 2025
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 573309442

This project aims to use quasi-experimental empirical methods, coupled with a structural model of the housing market, to answer the following research questions: What are the effects on housing prices in locations that experience an unforeseen and highly localized increase in the supply of housing stock, and to what extent does this shock affect neighborhoods surrounding new housing supply? Identifying the causal effect of an increase in housing supply on prices is usually challenging since the location of new housing is highly endogenous. New housing is more likely to be built where prices increase. Identification then requires an exogenous shifter that is unlikely to correlate with local housing price trends. For our analysis and as a supply shifter, we will exploit quasi-experimental shocks to German housing stock originating from the unexpected withdrawal of the British military and their families in 2010 - following a decision by then UK Prime Minister David Cameron - and the subsequent sale of their real estate by the Institute for Federal Real Estate (“Bundesanstalt für Immobilienaufgaben, BlmA”). The corresponding housing supply shocks were highly concentrated, substantial in magnitude and amounted to more than one-fourth of average yearly housing stock increases in several medium-sized cities in the German states of Nordrhein-Westfalen and Niedersachsen. In the first step, we are interested in understanding how these substantial supply shocks influence the housing market of affected cities by comparing their price development to those in similar locations but which did not house British military real estate for historical reasons. To further examine the localized effects on neighborhoods surrounding converted real estate (e.g. via amenity spillovers), our small-scale price and treatment data will enable us also to take a much more granular approach. In the second part of the project, we will therefore study changes in house prices at highly disaggregate levels (street, block, etc.) and measure (spillover) effects with a high degree of precision in terms of timing, location, quantity, and an array of other characteristics. While providing causal estimates of the effect of housing supply expansions on within-city and between-city price differences - which is of intense policy interest in its own right - reduced-form estimates alone cannot account for general equilibrium effects or spatial sorting induced by supply shocks. Consequently, we aim to complement our reduced-form analysis with a quantitative spatial model of connected housing and labor markets. The structural model offers two main advantages: (i) it incorporates effects on other endogenous variables, such as worker sorting, and (ii) it allows us to conduct counterfactual welfare analysis in general equilibrium.

• DFG Programme: Research Grants
• International Connection: USA
• Cooperation Partner: Professor Christian Helmers, Ph.D.