Research

Abstract: This paper examines the impact of a large-scale immigration wave into Austria following the Cold War. Using an instrumental variables approach, I estimate the causal effect of immigration on the labor mobility and earnings of native Austrian workers. The immigration shock—which was primarily comprised of low-income blue-collar workers—caused a reallocation of native workers to-wards white-collar jobs. Panel data on the universe of formal workers allows me to identify the margins underlying this reallocation. Immigration did not increase the rate at which blue-collar workers left employment. Instead, immigration increased the rate blue-collar workers transitioned into white-collar jobs and changed the composition of newly hired workers. Point estimates show suggestive evidence of decreased income growth at the bottom of the income distribution, though results are not significantly different from zero for blue-collar workers. Difference-in-differences estimates, which examine medium-run effects, confirm short-run effects of increased native transitions into white-collar jobs. Though, they fail to find significant effects on earnings. The results are consistent with a labor market model in which workers adjust to relative income changes by moving occupations exposed to fewer immigrants.

Abstract: This paper presents a dynamic spatial model of labor markets that examines the adjustment process of workers in the presence of imperfect mobility. The model builds on recent models in trade and urban economics and adapts them to highlight margins that are especially important in the context of immigration. Simulations from a hypothetical economy show that a labor supply shock will affect the mobility decisions not only of those in the exposed region, but also in those regions connected through mobility networks. These spillovers highlight potential pitfalls in empirical work that are not always accounted for in empirical studies. While small, domestic trade can also play a role in the distribution of effects by worsening the terms of trade in immigrant receiving regions. In a counterfactual economy with reduced mobility costs I find a faster adjustment process and smaller welfare losses, this result presents one potential role for policy in facility adjustment and mitigating distributional effects from unexpected shocks.

Abstract: We describe crime effects attributable to the Mariel Boatlift, the 1980 Cuban refugee crisis that increased Miami's population by nearly 10%. Using synthetic control methods to match Miami with cities that exhibit similar pre-intervention crime patterns, we find the phenomenon comparatively increased property crime and murder rates; we also document weaker but suggestive relative growth in violent crime. Compositional features of the newcomers seemingly drive results; the disproportionately young, male Cubans' characteristics highly correlate with illicit activity. Given the group's unique composition and the absence of rigorous screening, our findings likely constitute the "upper bound'' of migration caused crime.

We use mobile phone usage data to identify the extent and patterns of segregation of Syrian refugees in Turkey. We construct a range of dissimilarity and normalized isolation indices using the hourly phone call volume of refugees and natives at a given cell tower as a proxy for the population distribution. The richness of the data allows us to calculate the indices over time, across different provinces, and by the time of day. Segregation levels show high variation across the country, with significantly lower levels of dissimilarity and isolation in provinces with a higher share of refugees. Dissimilarity seems to be declining while isolation is increasing over time across almost every part of the country. Refugee and native mobility across provinces over time appears to explain at least some of these changes. Finally, we see differences in our segregation indices across the hours of the day, suggesting that residential segregation is higher than labor market segregation.

Understanding the determinants of migration patterns and the selection of migrants is of paramount importance in a world in which over 240 million people have migrated and many more are considering doing so. This paper adds to the literature on this topic by (i) using micro-level wage data for a large number of source and destination countries, and (ii) identifying the impact of bilateral variables, e.g. distance, and source and destination country characteristics using a two-step estimation strategy that allows for estimation over a full matrix of migration corridors and better addresses the issue of multilateral resistance to migration. The results highlight how geography (distances, contiguity and country size) and wage differentials matter for determining migration flows and the differential selection of low and high skilled migrants.