Assaf Razin and Efraim Sadka have a new paper up on NBER called Migration into the Welfare State, in which they survey the literature on the tax burden of migration and work with available migration and tax policy data to make some assertions about how taxation impacts inward and outward migration. From the abstract in the paper:
We develop a stylised EU-type model of rich capital-abundant (and productive) countries and poor capital-scarce countries in order to explain a key feature of tax policies and inter- and intra-migration flows. We examine how this model can explain the differences in the tax rates and the generosity of the welfare state, on the one hand, and migration flows, on the other hand, between rich and poor countries, within a union and from the rest of the world. An upward-slopping supply of migrants from outside the union and the relatively low endowment of capital of these migrants gives rise to a fiscal externality.The jargon-heavy style of this abstract continues through the paper and I cannot say that I really understand some of the arguments. Razin and Sadka seem to conclude that under conditions of tax competition (for example, involving migration from countries outside of the EU), rich countries let in too many immigrants, offer them too many social benefits, and levy too high a tax on capital (which immigrants don't typically pay) to pay for it all, while coordination fixes things (although confusingly the papers seems to suggest that coordination brings down capital gains taxes but intensifies migration flows, so I am not sure). Unfortunately this paper requires too much deciphering and parsing work by the reader (an all-too-common phenomenon with academic papers) and at this time of year there is little hope for that.
A plea to academic authors who write papers on tax topics especially (and a reminder to myself when I look at my own writing): readers need a clear and succinct summary, written in plain, jargon-free language, to help them summon the desire and the mental energy to work through a paper. The more complex and data-filled the paper, the more a return to Zinsser is to be recommended.