Marginal and average prices of land lots should not be equal: A critique of Glaeser and Gyourko's method for identifying residential price effects of town planning regulations

Glaeser and Gyourko's method of measuring the gap between marginal and average land prices of housing lots has become a popular way of demonstrating the degree to which planning controls, or 'regulatory taxes', increase residential land prices. This has led policy-makers across the globe to focus on town planning as a critical determinant of rising home prices. We show, however, that Glaeser and Gyourko's method shows no such thing. Instead, the price gap is a product of the location premium of land, diminishing returns to buyers of residential land size, and historical city development patterns. Numerous shortcomings are identified in their theoretical model, including that (a) they ignore that development happens over time; (b) their 'regulatory tax' is simply the location value of land; (c) they reason inconsistently about the source of land prices, arguing that land at the margins is scarce but locations for whole housing lots are not; and (d) there are no optimal lot sizes nor subdivision incentives in their model. Standard price-taking models of residential land markets that recognize that locations are scarce contain none of these limitations and provide a better interpretation of land price patterns. Empirically, we show that Glaeser and Gyourko's method finds a high 'regulatory tax' even in the absence of regulatory constraints using both simulated suburb development scenarios and historical land sales data from colonial Australia and ancient Mesopotamia. In short, there is no information regarding the effect of planning controls on the supply of new dwellings in a comparison of marginal and average land prices, and this method should not be relied upon to inform planning and housing policy decisions.