In this article, we develop a structural econometric model of household location, dwelling and tenure types in a in a dynamic framework with perfect information. We consider an individual living two periods. At the beginning of each period, s/he has to choose whether to stay or not where s/he lives and, in case s/he moves, choose dwelling type, tenure type and location. S/he then has to choose the optimal quantities of housing and consumption level of an outside composite good, subject to budget and other technical constraints.

Borrowing is allowed at first period, but not at the second one. We assume that the interest rate is higher when borrowing for a dwelling than when saving/borrowing on the money market, and that transaction costs apply to real estate.

The individual may leave a bequest to heirs at the end of his/her lifecycle, be it for strategic or altruistic reasons. The bequest is made of money and/or real estate enjoyed when alive. Real estate is then both an investment and a consumption good.

We choose functional forms for utility at each period so that the resulting theoretical model of inter-temporal utility maximization is analytically tractable. Solving for tenure and dwelling types, and quantities of dwelling and composite good results in an indirect utility function computed for any potential sequence of locations. The model can then be easily formulated as mixed Nested Logit Random Utility Maximization model.

The empirical application uses the 2006 French Housing Survey including retrospective information on several demographic and professional events which happened during the years before the survey. These changes are related to household moves during the same period. We use this information to define the periods considered in the empirical application.