Rural Hospital Closures and Local Economic Decline

This paper studies the local economic impacts of rural hospital closures in the United States. The analysis begins with a difference-in-differences approach using county-by-year panel data on all hospital closures from 2003 through the first half of 2017. The results indicate that closures adversely affect employment, income, labor force participation, establishments, population, rents, and the unemployment rate. Estimated effect sizes grow over time and are explained largely by rural counties that lose their only hospital and in counties where hospitals occupy a large share of the local labor market. While there is little or no evidence of pre-trends, I estimate a range of robustness checks designed to further address endogeneity concerns, such as forward-looking behavior among hospital owners. The results are consistent across these specifications. I also document spillovers, as evidenced by a 1.8 percent decrease in non-hospital employment, an effect that explains 40 percent of the total employment loss. To characterize the significance of the adverse effects, I combine the reduced-form estimates with a spatial equilibrium model of various agents in a local economy. Analysis of the model indicates that rural hospital closures significantly harm welfare, an outcome that is internalized by workers, older residents no longer in the labor force, and landowners