Saturday, July 4, 2015

Refugee resettlement as a matching problem

There are a lot of displaced people in the world today, both outside their country of origin and within. The conflict in Syria is a big contributor. The poverty in Africa is another. Here's a recent NY Times story, about a UN report, whose headline summarizes the story well:
60 Million People Fleeing Chaotic Lands, U.N. Says

The international refugee accords place most responsibility for resettlement on the "country of first asylum."  If you were smuggling yourself out of Africa, or Syria, you'd have good reason, therefore, to try to get to Sweden before declaring yourself a refugee, but it's a lot easier to get to Turkey or Greece or Italy.  However some classes of refugee can seek resettlement elsewhere, and the U.S. takes a small number of these (around 70,000/year).

American policy is to try to settle refugees across the country, the idea being that this might ease assimilation, and avoid overburdening particular cities and towns. But, of course, once refugees get to the U.S., they are completely free to move around. So there's a matching problem of refugees and cities.

The case of Somali refugees makes this clear: although they've been resettled around the country, many of them quickly move to join the growing community in and around Lewiston, Maine. (Here's a nice story dated 2007...
Letter From Maine: New in Town--Somali refugees began arriving in Lewiston, Maine (pop. 36,000) six years ago. Word spread that Lewiston had good schools, a low crime rate and cheap housing — and the Somalis began arriving in droves.

And here's a Wikipedia page: History of the Somalis in Maine

The point of all this is that people aren't passive, you can't keep them where they are sent if they don't want to stay there (even if moving means giving up various kinds of refugee assistance).

Hillel Rapoport of the Paris School of Economics has been thinking of this in a European context, in which one of the questions is to which countries should refugees be resettled?  How a tradable refugee-admission quota system could help solve the EU’s migration crisis.  Even in Europe, I'm not sure how well refugees can be resettled in the countries to which they are assigned, but the barriers to moving are probably substantially higher than for moves in the U.S.

The EU is thinking about moving refugees, maybe in directions they want to go (although this isn't clear): see e.g. this recent story. EU leaders agree to relocate 40,000 migrants. "EU leaders holding late-night talks in Brussels have agreed to relocate tens of thousands of migrants who have arrived in Italy and Greece." But it's hard, and they aren't really reaching agreement: In Testy Debate, E.U. Leaders Fail to Agree on Quotas to Spread Migrants Across Bloc

So, we have a matching problem here. How to resettle refugees to places that they are willing to stay in, while meeting the other goals that we'd like to achieve?

It's not a bad question to ponder on the 4th of July, for a nation made up of immigrants, many of whom escaped from somewhere to come to the USA.

No comments: