View by Day
16:30 | Applied Micro Research Seminar
The City University of New York, USA
Author: David Jaeger
Abstract: A large number of papers in the immigration literature rely on geographic variation in the concentration of immigrants to identify the impact of immigration. National flows of immigrants by country of origin are often interacted with the past geographic distribution to create an instrument, in the hopes of breaking the clear endogeneity between labor market conditions and the location choice of immigrants. This “shift-share” instrument is widely used in the spatial correlation approach of estimating the causal impact of immigration. We present evidence that (in the United States) the serial correlation in the countries of origins of the flow immigrants is extremely strong, however, and as a result any estimates based on the conventional “shift-share” instrument are unlikely to identify a causal effect. We propose a “double instrumentation” solution to the problem that produces estimates that are likely to be less biased than those in the previous literature — but only when there is an exogenous event that sufficiently changes the country-of-origin flows. Our results are a cautionary tale for a large body of empirical work (not just on immigration) that uses “shift-share” instruments.
