Supply Constraints Are Not Valid Instrumental Variables for Home Prices Because They Are Correlated with Many Demand Factors
Sauder School of Business, UBC
Download Document| Date Published | 2014 |
| Version | |
| Primary Author | Thomas Davidoff |
| Other Authors | |
| Theme | Housing Supply |
| Country | United States |
Abstract
Economists sometimes wish to estimate the eff ect of housing prices on outcomes such as wages or investment, ceteris paribus. Such estimates require sources of variation in housing prices uncorrelated with unobserved local amenity or productivity. Such instruments are hard to nd, because households are mobile. Natural and man-made constraints to housing supply are superficially attractive instruments correlated with high and rising home prices. However, in the US, constraints are not significant ly negatively correlated with housing unit growth or the sensitivity of supply growth to productivity shocks. This asymmetry appears to arise because demand pressure causes regulations and because the growing concentration of wealth among individuals with high human capital has increased the price premium attached to amenities such as proximity to mountains and bodies of water. Supply constraints do not appear to have caused home cycle amplitude during the last major US home price cycle; there is mixed evidence that they did in the 1980s.