A household life cycle model for residential relocation behaviour

Abstract

In this paper a dynamic household model in connection with a comprehensive urban housing market model for Amsterdam is presented. Household dynamics is modelled here by using the methodology of multidimensional demography. A consistent analysis of the components of a dynamic spatial housing market system incorporating various submodules, multiple factors, spatial aggregation levels and degrees of precision of available data requires the use of an accounting framework. The paper shows the analytical structure and empirical estimation of an accounting model developed for investigating the dynamics on an urban housing market, with a particular focus on changes in household compositions caused by the household life cycle phenomenon. The linkage between the life cycle concept and the demand on the housing market is established by using an explanatory nested logit model. Apart from the few exceptions in a more limited context, this is a rather unique attempt at modelling, in a consistent way, the dynamic allocation of household demand for dwellings in a zonal urban system based on a large-scale socio-economic data set concerning spatial residential relocation behaviour in Amsterdam. On the basis of the empirical outcomes obtained, of various model testing procedures, and of a comparison of observed and predicted outcomes, this housing market model appears to reproduce housing market developments in the past satisfactorily, so that it can likely be used as a tool for predicting future paths of urban housing market demand.