Abstract
International migration has become a highly relevant issue over the last decade. In this paper, we study the homeownership and living conditions of the immigrant population in Spain in a period of economic growth. In particular, we analyse the decisions for tenure, quality of housing and the overcrowding rate. Additionally, we test the hypothesis that the behaviour connected to these decisions is divergent among immigrants according to their country of origin. Using data from the National Immigrant Survey, we found that economic factors and immigration-specific variables are strong predictors of the tenure and quality of housing. Individual characteristics and family structure are the main determinants of overcrowding. The results showed differences in immigrant behaviour by country of origin.
Notes
1. A dwelling is considered low quality if it has three or more defects: exterior noise, foul odours, moisture or leaks, construction deficiencies, poor communication, little street cleaning, crime or vandalism.
2. An overcrowded condition is defined as having more than one person per room in the occupied home.
3. As reported by Eurostat, in 2007, Spain is among the countries with higher rates of homeownership (83%), compared to the EU average of 65%, and with rental stock representing 15% of primary residences, which is well below the European average (30%).
4. The overcrowding rate is defined as the ratio of the number of people in crowded conditions to the total population.
5. A detailed description of the superiority of NIS on periodic surveys conducted by the INE can be found in Reher and Requena (Citation2009). The data-set is available under the title Encuesta Nacional de Inmigrantes, 2007 at http://www.ine.es/inebmenu/mnu_dinamicapob.htm.
6. This survey is conducted annually by the INE but has the disadvantage of containing little information on the immigrant population. The data-set is available under the title Encuesta de Presupuestos Familiares, 2007 (base 2006) at http://www.ine.es/inebmenu/indice.htm#142ipc.
7. The information is available at http://www.fomento.es/.
8. To ensure parameter identification, the coefficients associated with the explanatory variables that do not vary across alternatives are assumed to be equal to zero for one of the alternatives. In this case, we have used the low-quality rental alternative as the reference alternative.
9. The coefficients from the multinomial logit model indicate the difference in the likelihood of selecting an alternative relative to the reference alternative. To facilitate the interpretation of the effect that each explanatory variable has on the probabilities of selecting the different alternatives, we calculated the corresponding marginal effects, Note that the sign of this partial derivative can differ from the sign of the estimated coefficient (Green Citation2002, pp 722).