On the Slope of the Beveridge Curve in the Housing Market

Abstract

The co-movement of buyers and vacancies, i.e. the Beveridge Curve, is a key determinant of the cyclical properties of the housing market. It determines the sign of the correlation between prices and key measures of liquidity such as vacancies (i.e. houses for sale), sales, and time-to-sell. As recent work has shown, to account for the core stylized facts of the housing market, search and matching models must be consistent with a positively correlated co-movement of buyers and vacancies—the Beveridge Curve must be upward-sloping. This paper provides evidence that buyers and vacancies are indeed positively correlated along the housing cycle, i.e. the Beveridge Curve on the housing market is upward sloping. Using data on vacancies and time-to-sell, we construct a series for buyers and estimate the slope of the Beveridge Curve. This approach requires only one minimal structural assumption: the existence of a matching function. The regression results confirm the positive relationship between buyers and vacancies over the business cycle. In addition, we provide an estimate of the elasticity of vacancies with respect to buyers. A one percent increase in vacancies is associated with around a two percent increase in buyers, confirming recent findings that buyers are more volatile than houses for sale. We hope this estimate will help future researchers in this area.