Planning for preference heterogeneity, the case of housing and urban green space

Developing urban green space (UGS) is a widely considered way to mitigate negative effects of climate change and improve the quality of living environments in cities. Since UGS has to compete with other land-use demands, land-use models offer a valuable tool to optimize the spatial allocation of UGS. Current methods to empirically estimate the land-use models, however, do not take into account preference heterogeneity that may exist in a population. In this paper, we apply a latent class model on data from a stated choice experiment to estimate housing preferences related to locational characteristics. Three classes that differ in the utilities attached to accessibility, green and price characteristics of a housing location emerge. The estimates are used to specify a housing land-use allocation model that represents the preferences of the different classes. The result of an application to a housing area development problem shows that taking the preference heterogeneity into account can increase the total housing utility value for residents significantly.