Four essays in housing economics

Abstract: This essay consists of four self-contained essays in housing economics.Essay I: A dynamic partial equilibrium asset model is used to study the capitalisation effects in themarket for owner-occupied housing that resulted from the Swedish 1991 tax reform. Simulations ofanticipated as well as unanticipated implementations of the tax reform suggest that the 1991 tax reformcaused a substantial decline in real housing prices and housing investments. Further, we find that thepost reform tax system is far less sensitive to inflation compared to the pre-reform tax system.However, the post-reform tax system is still a nominal tax system and therefore inflation will still causelarge and arbitrary tax subsidies to owner-occupiers.Essay II: Housing tenure choice is studied in a two-period model with stochastic rents and houseprices. We show that a risk-averse household chooses the tenure mode with lowest variance in the costof housing. The tenure mode chosen depends on the variance in the prices of apartment buildings, thevariance in the prices of owner-occupied housing and the correlation between these two prices. Whenthe correlation is positive and above a critical level, a risk-averse household prefers ownership as aninsurance against future uncertain housing costs.Essay III: Housing tenure choice is investigated in a numerical two-period model where. both rents andprices on owner-occupied housing are stochastic. With a positive correlation between prices onapartment buildings and prices on private homes, owner occupation serves as an insurance against thefuture uncertainty in housing costs. Because of this insurance mechanism, increased uncertainty in thecost of housing has an ambiguous effect on saving. In addition, non-durable consumption may either bean increasing or a decreasing function of unrealised capital gains from owner-occupied housing.Essay IV: Pooled and longitudinal data for the years 1989 and 1992 were used to study the housingdecisions of young Swedish adults before and after the 1991 Swedish tax reform. Although thehousehold formation and tenure choice decisions of young adults were found to be simultaneouslydetermined, neglecting the cross equation correlation between these two decisions had no major impacton the estimated coefficients. Demographic factors were found to significantly affect both thehousehold formation and tenure choice decisions. Economic factors were also found to significantlyaffect young adults' choice of tenure mode. This is, however, not the case in the household formationdecision. Young adults' decision whether to form a household was found to be rather insensitive toeconomic factors. Furthermore, the impact of the relative cost of owned versus rented housing wasfound to be significantly lower after the Swedish 1991 tax reform. A possible explanation to accountfor this phenomenon is the smaller variation in relative cost of owned versus rented housing betweenhouseholds after the 1991 Swedish tax reform. Finally, neglecting household specific heterogeneity inthe tenure choice estimation causes a downward bias in the parameter estimates.