Stochastic Frontier Estimation of Housing Supply (with David Genesove)

Stochastic Frontier Estimation of Housing Supply (with David Genesove)

Abstract:  Housing prices are determined in large part by unobserved regulations and the costs of building upwards. The standard approach to estimating these costs has been through a conditional mean regression. Leaving aside endogeneity concerns, these estimates will be of limited use in any counterfactual regulation analysis. Using only observed apartment prices and building heights, we estimate "frontier'' (i.e., unregulated market) costs, and thus separate regulation from other costs. Even when price and quantity are simultaneously determined, we show that variation in both demand and regulation identify frontier average costs in an initial region of increasing returns and that variation in demand in unregulated markets identify frontier marginal costs beyond minimum efficient scale. We apply our methods to data on new construction in the Israeli housing market. We find evidence of increasing returns to scale for low buildings. About 50% of housing value can be ascribed to regulation. The share of regulations is higher in more expensive and denser areas. The effects of measurement error and quality differences on price are small relative to regulations.

link to paper

https://drive.google.com/file/d/1H44bqK9cgOZaj__bpnsrT76pI-s1jX4d/view