Capitalization of local fiscal variables into house prices : theory and new empirical evidences

This thesis analyzes the dynamic of fiscal capitalization across space and over time in the Canton of Zurich. While the literature provides a profusion of evidence on capitalization, there is still a controversy on the extent to which local fiscal variables capitalize into house values and on the appropriate interpretation one must have with respect to the Tiebout hypothesis. The outcome of a significant degree of tax and public expenditures capitalization is often interpreted as a test of the Tiebout hypothesis and associated with a Pareto efficient provision of local public goods. Some authors criticize this view and argue that capitalization of local fiscal variables does not describe a full Tiebout equilibrium. Fiscal capitalization results from scarcity of land that prevents households from moving toward communities with favorable tax-expenditure packages. While it has been raised in the 70's, the question of the persistence of capitalization has not received a definitive answer. This thesis aims to contribute to this debate. Chapter 1 provides a survey of the theoretical literature on the Tiebout model, which is the basis of virtually every empirical capitalization study. Chapter 2 provides a review of the econometric innovations and improvements made by the empirical literature on capitalization. Chapter 3 contributes to the empirical literature on capitalization by providing estimation results of a unique data set from the Canton of Zurich. For our purpose, we use the price of a standardized and comparable single family house in each community. This allows to ignore the control of the structural characteristics in the estimation. We regress this house price variable on a large set of fiscal, neighborhood, environmental and demographic characteristics. Chapter 4 contributes to the debate on the persistence of capitalization and estimates the change of the capitalization rate in response to housing supply differences across space and over time. We use data on land availability as a proxy for housing supply elasticity. We estimate amenity models in which we interact the dummy variable for land availability with local fiscal variables. We also test for a decrease of capitalization over time by interacting local fiscal variables with year and time trend dummies. In addition we compute Spearman correlation to check the presence of a linear trend in the fiscal variables coefficients. Since the time horizon of our sample is relatively small (7 years), we bootstrap the basic capitalization coefficients 10,000 times. Overall, our estimation seems to indicate a persistence of tax and public expenditures capitalization both across space and over time, as most of interaction variables as well as the bootstrapped coefficients are not significant. These results are consistent with the intensive use of zoning regulations at the local level. Fiscal capitalization ; hedonic house price ; amenity models ; housing supply ; property values ; taxes