Households' residential decision: between choices and constraints.

Summary Residential choice revolves around a number of constraints: whether internal to the household or resulting from market distortion, these constraints fundamentally alter the set of household choices. The first part of the thesis focuses on the constraints that the household faces and that are based on one of its characteristics. The literature has studied in detail the issue of discrimination in access to housing by highlighting limited and tacit access to immigrant households, but no study has examined the evidence of potential residential price discrimination. In this article, we study the actual rent differentials between French and immigrant households, using the decomposition proposed by Oaxaca & Blinder (1973). We try to explain these rent differentials in part due to differences in individual choices and in part due to discrimination in the housing market on data from the 1996 housing survey. Although no clear evidence of price discrimination is found, some housing characteristics appear to be more expensive for immigrants. The second part of this thesis focuses on internal household constraints. The work in this first part is a continuation of the work done on the economics of the family and the economics of decision-making regarding their mobility. The first constraint that we present is that of the dwellings that agents occupy before entering into a household. The literature has studied in depth the residential mobility choices of couples at most stages of the life cycle, without looking at the stage of house formation. This would make it possible to take into account the plurality of decision-makers within the household. This chapter presents a collective model of household formation, measuring the probability that a couple will occupy, together, the dwelling that the man occupies alone, the dwelling that the woman occupies alone, versus refusing them both, based on a multinomial logit estimation model. We work on data from the 2002 national housing survey, which allows us to trace all the mobility paths of the two agents. We choose the couple's first home in the sense of its initial occupation. The development of the theoretical model, based on a collective type model, where the agents operate a negotiation, includes the situation where the two agents may not be led to leave their dwelling at the couple's threat points. We propose an enrichment of the model where we argue that the decision of joint housing may imply inefficiency in the couple's future decisions. Although the man's housing choice appears to be a proven choice, the wage inequality between men and women, as well as the age difference within the couple, appear to significantly influence the couples' choice. Second, the choice of housing for a couple must take into account the daily commuting time. Agents, having a localized job, choose housing that meets their needs in a region with a highly heterogeneous spatial occupation and with a non-uniform distribution of housing supply. The idea of the paper is to measure the impact of individual agent characteristics on the joint household travel time decision, as well as how agents allocate these trips. The data on which we work are those of the 1999 population census, enriched with communal data and agents' travel times. We highlight a distinction between male and female behavior based on attraction to job centers and job stability.