On the economics of housing markets and urban policy: Three essays in real estat economics

In order to assist governments in ensuring efficient housing markets and providing access to suitable housing conditions for all people, research about the functioning of housing markets is needed. My thesis, which is comprised of three essays that repre-sent the main chapters, contributes to the research on the economic processes that work within housing markets, especially with regard to challenges that arise from ur-ban concentration. In particular, I study the following research questions: • What are causes of vacancies in the housing market? • What are the effects of real estate transfer taxes on activity in the housing market? • What are the differences between counties in the provision of technical infra-structure? 1. Causes of Vacancies We summarize theoretical approaches, which may explain the mechanisms leading to vacancies under the assumptions of the standard market model, search and matching theory and behavioral economics. Concerning the latter, we propose a new frame-work to explain why and under which conditions homeowners are hesitant to sell their dwellings, which may lead to vacancies. In this framework, we highlight the fact that even if buyers and sellers are homogenous individuals, their willingness to pay differs depending on market developments. Under the assumptions of the standard market model, we hypothesize that vacancies only arise if the market price equals the suppliers’ minimum price. Next to that, we identify the following factors that could explain an increase in vacancies: Greater market power, higher heterogeneity between dwellings, low holding costs of dwellings, high list prices, a high share of small dwell-ings, less intermediaries and a shorter mandatory period of notice of tenants. Despite the wide range of models explaining vacancies, there is need for new theoretical frameworks. We evaluate the identified hypotheses in the theoretical literature by comparing them with the results found in the empirical studies. Some hypotheses have either been in-vestigated by the empirical literature only to a limited degree or have not been inves-tigated at all. Next to the positive analysis of the mechanisms that cause vacancies, we need to pose the normative question, if and under which circumstances vacancies should be considered problematic. Vacancies in the short run may be a necessity in a search and matching context to reach an efficient market outcome; i.e., to offer poten-tial buyers a heterogeneous pool of options to choose from and to fit their individual needs. Additionally, with the assumptions of behavioral economics, an owner of a vacant dwelling might find greater pleasure in speculating with the object than a buyer would enjoy living in it. However, vacancies could indicate a welfare loss if caused by a monopoly that artificially reduces the housing supply on a market. A reduction of information asymmetry could reduce vacancies and increase welfare. 2. Effects of Real Estate Transfer Taxes German states can set their own real estate transfer tax rates. To date, the real estate transfer tax rates range between 3.5% and 6.5%. Although the tax rates do not seem to be particularly high, the tax results in a relatively high tax amount to be paid; even small changes in the tax rate may cause buyers to accelerate a planned transaction to pay a lower tax rate. In our empirical analysis of different German states, we estimate that a one-percentage-point increase in the transfer tax produces significant anticipation effects for the month just before a tax is increased (about 43% more transactions before and 44% less after) and yields approximately 6% fewer overall transactions and therefore much lower market activity. We show that in many cases, the former first-best option – to buy or sell a single-family home – is apparently no longer the optimal choice. Thus, we expect ownership rates to decrease as letting apartments becomes more attractive than selling. This questions the wisdom of real estate transfer tax increases when other political measures that attempt to support homeownership creation are in place. Addi-tionally, retired households, which tend to stay in houses that are too large after their children have left, could be discouraged from downsizing to their actual needs. Indi-viduals may forgo better job offers in other regions or accept longer commutes, which can have negative consequences on urban labor markets. 3. Efficiency of County Road Provision When houses in urban areas are expensive, an adequate road network allows residents to locate further away from agglomeration centers. This takes pressure off housing prices and infrastructure systems in cities and supports housing markets in rural areas. I use the example of county roads to study whether counties differ in their efficiency of the provision of infrastructure. Efficiency refers to the use of economic resources (input) in the most technologically efficient manner to produce a certain amount of output. Studies on the efficient provision of roads incorporated the ‘quantity’ of roads (e.g., the length of the road network) and the ‘quality’ of roads (e.g., an index that measures road condition) as the relevant output indicators. I address two major problems of previous studies. First, to measure the quality of roads, I acquired new and improved data on road condition by county governments. Second, I focus on the data referring to a road network of approximately the same age. An old road network asks for more frequent maintenance and thus higher costs. Therefore, it is possible that governments with an older road network in their region could be identified as less efficient than governments with a newer road network, even if they do not actually employ their financial resources in a less efficient manner. The results of my study indicate that there are substantial efficiency differences and efficiency reserves in the provision of roads in eastern German counties: Depending on the specification, in the median county, the same level of outputs (i.e., area of roads and road condition) could be achieved using 48% to 70% fewer inputs (i.e., expendi-tures). I also find that my results differ greatly from existing studies applying proxy data for the quality of roads (e.g., the number of accidents)