The price of residential land varies widely. Land price differentials at the edge of the city are often used as a criterion for the transformation of agricultural land into city area.
The presence of agglomeration externalities makes full decentralization of these decisions unlikely to yield a first best outcome. This market failure has been recognized by policy makers by implementing zoning policies. Having a better understanding of the variety in land prices and the future evolution of these prices is crucial for the adequate design of these zoning policies and for investment decisions in housing and public infrastructure. This issue is therefore at the core of the research interest of the Keynes Fund, as stated in its statutory goals.
Answering these questions is the aim of the proposed project. Data are available on more than a million transactions of residential real estate and on the population disaggregated by 3000 ZIP codes for the Netherlands from 1996 and 2012. Econometric techniques allow a decomposition of these transaction prices into the value of the land and the building on top of it. Hence, we can construct the land prices and the marginal value of a unit of construction by ZIP code and over time. This has been done in previous projects.
The aim of the proposed project is to explain:
- The level of land prices from the spatial structure of economic activities, using insights from the literature on urban structure and detailed data on commuting patterns.
- The growth rate of these prices from the (expected) growth of each city.
The challenging aspect is that differences in the expected growth affect the level of land prices via the option value of future increases in the rental rate for that land. These option values play an important role in the decision when to transform agricultural land into city area and what density of construction to apply in this transformation. An analysis of the data on the population by ZIP code done for another project has shown that there is in fact substantial persistence in the growth rates of cities. This persistence yields variation in the expected growth rate, which is an input in the calculation of the option value component in land prices.
The current proposal asks for a post doc who can do the work on data selection, the exploratory data analysis, the programming the formulas for option values, and the econometric analysis. The methodology of this study will have wider applicability. Contacts with the Departments of Architecture and Land Economy reveal that similar data are available for the UK.
