Summary of Home ownership, job duration, and wages

In this paper, the authors first hypothesized that employed home owners, compared to renters, are less likely to be unemployed, have longer employment spell lengths, and earn higher wages. Then the authors tested the hypotheses empirically based on a rich Danish micro data set. The data set they used covered one percent of the Danish population from years 1993 to 2001 and contained detailed information of housing, labor market experiences and demographic characteristics.   The pattern of the data showed that on average homeowners have more favorable labor market outcomes than renters.

To investigate the impact of home ownership on job duration and wages, the authors specified a competing risks duration model for job spells, used a mixed proportional hazard model for the labor market transitions to capture unobservable work characteristics, and estimated a standard human capital wage equation.

To identify the causal linkage between homeownership and hypothesis, the authors modelled the selection process into homeownership and used two different identification strategies to check the robustness of their results. The first identification related to the fact that some people change their home ownership status during job spells. This identification required authors to observe job spells both when the individual is a home owner and when he is a renter. Conditional on heterogeneity, it helped authors to see the relation between employment outcomes and home ownership status. The second identification was about exclusion restrictions, which helped tease out variables that affect home ownership decision but have no direct impact on labor market outcomes. The authors considered regional home ownership rate, home owner status of the parents, and home ownership rate in the region of birth as variables that only affected home ownership status. The whole model endogenized home ownership status.

The results of the paper were conform to the hypotheses. Home owners are less likely to transit into new jobs, and they earn higher wages. The paper suggests that there are labor market gains for homeownership. The problem of the model is that to make it more tractable, the authors restricted the correlation structure.