| This dissertation is the first to analyze vacancy behavior by examining the interactions of search between tenants and landlords, and new development by potential builders. Based on the notion that similar levels of vacancy may have very different underlying causes and market implications, we propose a model of vacancy formation with tenants, landlords and potential builders strategically interacting with each other in an uncertain dynamic environment. We not only identify important contributors of vacancy formation, such as uncertainty in demand, search frictions, and supply restrictions, but also demonstrate the interactive effects of these factors. This framework of interactive market players allows us to explore previously ignored factors influencing vacancy behavior. As an illustration of how search and development decisions can affect the vacancy rate interactively, we study the effect of degree of monopoly power held by a developer. We find that monopoly power lowers the equilibrium vacancy rate, a result contrary to that in the parallel literature from labor economics. The main reason for this is seen to be a difference in market structure on the supply side. Our model implications are empirically supported when we test them empirically by use of an MSA-level rental office panel data set from the U.S. Finally, we explore the market implications of vacancy rate dynamics, specifically the relationship between the vacancy rate and rental adjustment. In a comparison of alternative model specifications, a hybrid model using natural vacancy rates estimated using our theoretical construct as the basis for specification achieves the best prediction of rental adjustment. Therefore our study on vacancy behavior helps to achieve a better understanding of real estate market behavior and to enhance real estate decision making. |
