Are riskier cities more compact? An empirical study of the 11 largest census metropolitan areas in Canada, 2016

Abstract

Why are some cities more compact than others? We argue that riskier housing markets have a costlier real option; developers would require greater compensation to build now—thus, developers are being incentivized to delay, giving rise to more compact cities. We test this hypothesis cross-sectionally for Canada's 11 largest census metropolitan areas. We interpret satellite imageries and estimate a hierarchical spatial autoregressive model to account for both the hierarchical and the spatial structure of our data. Our results show that, on average, a one-percentage-point increase in price risk reduces urban land coverage in the fringe by about 0.7 percentage point.