Modelling the aggregate effects of housing supply policies

Abstract

What are the aggregate effects of housing supply-side policies, such as zoning reforms? In structural models, the answer involves characterising the equilibrium housing price function. I show that a housing price function should separately characterise how policies affect: 1) the response of house prices to new demand ("Elasticity Effect''); 2) the cost of satisfying existing housing demand ("Baseline Effect''). While the former can be calibrated to match estimates of price-demand elasticities such as Saiz (2010), the latter requires a separate calibration. However, popular models in Urban Economics and Economic Geography do not separately characterise and calibrate the Baseline and Elasticity Effects, introducing potential biases in the estimation of long-run policy effects. I propose a characterisation that makes such biases explicit, nests most popular characterisations, and allows to separately characterise and estimate the two effects. Calibrating the Baseline Effect to conservative empirical estimates from the literature, I find housing supply policy effects up to one order of magnitude larger than other characterisations applied to the same model.