Download this article
- 3,026 article downloads
- 4,350 complete issue downloads
- Total: 7,376
Two empirical papers study US housing markets and treat the impacts of both natural geographic constraints, such as water bodies and steep slopes, and regulatory restrictions. Albert Saiz (2010) concentrates on identifying the impact of natural geographic constraints. Saiz creates measures of developable land using an invariant 50-kilometer radius about an urban focal point. I criticize Saiz’s geographical constraint on the basis that its invariant radius renders comparisons between urban areas of differing sizes of questionable value. Moreover, the inter-relation between the two kinds of constraints—the natural and the regulatory—is highly particular to the locale; for example, the effect of a natural geographical constraint would be neutralized by an interior regulatory constraint such as an urban growth boundary. Haifang Huang and Yao Tang (2010) concentrate on the impact of regulatory constraints on booms and busts in local housing markets, but they also make use of the natural-constraint data furnished by Saiz. Both studies also use the Wharton Land Use Regulatory Index, about which I raise a number of questions. I believe that the procedures employed by the two papers have the tendency to understate, at the aggregate level, the pernicious impact of regulatory restrictions.
Response to this article by Haifang Huang and Yao Tang: Dropping the Geographic-Constraints Variable Makes Only a Minor Difference: Reply to Cox (EJW, January 2011).