Should I Stay or Should I Go? Dutch Evidence on Tax Induced (Return) Migration

It combines information on all employees receiving this 30 percent salary tax exemption from 2001 to 2019 to administrative population-wide data to estimate migration responses to a radical shortening of its duration that only affected certain groups of individuals. Beneficiaries who had arrived within five years from a destination closer than 150km from the Dutch border were retroactively given only five rather than ten years of preferential tax treatment. Causal estimates of tax induced return migration are obtained by comparing movement patterns of workers coming from across this distance threshold who were affected or not by this change because of time already on the scheme.

Our main finding is that taking away tax breaks from migrants, even after they have been in a country for a relatively long time, strongly increases their likelihood to choose to leave. They stay on average 13 fewer month (on a baseline of 7 years) in the country and were 15% less likely to remain beyond the 5 year end of the preferential tax treatment. Crucially we find that this effect was entirely driven by out-migration from those at the very top of the income distribution with no change in staying on decisions for remaining 95% of workers. This is of high potential policy relevance as it suggests that behavioral responses observed for these very rich and highly mobile individuals may not be generalizable to the rest of the population making tax-induced migration decision.