Research

This paper measures the effect of job loss on the subsequent labor market outcomes of immigrants as a function of the time spent in the host country at the time of displacement. The evidence comes from yearly employer-employee administrative data from Canadian taxes (2001-2019), linked to immigration records. I look at immigrants displaced during mass layoffs, which provide plausibly exogenous job separations. I estimate the impact of displacement in two distinct ways. First, through an event study approach. Second, through a regression-based approach that allows me to quantify how differences in the composition of pre-displacement characteristics contribute to the heterogeneous treatment effects and how the heterogeneity in earnings loss is linked to specific differences in post-displacement outcomes. I find that recent immigrants experience smaller and less persistent earnings losses from displacement, with a 21% decrease in earnings one year after displacement, compared to 26% for those who have been in the host country longer. Recent immigrants also display better post-displacement outcomes in other dimensions, such as lower time spent nonemployed and higher geographic mobility. I show that differences in pre-displacement characteristics account for 50% of the heterogeneous treatment effects on earnings: each additional year in Canada results in 0.8 percentage points larger earnings losses, but only 0.4 percentage points when controlling for pre-displacement characteristics. Age at displacement alone explains half of this difference. Differences in post-displacement outcomes account for an additional 40% of the heterogeneity in earnings losses, with time spent nonemployed being the most important mechanism.