The Immigration Game

Imagine there are two countries, Mexico and the United States. The U.S. government is worried about illegal immigration from Mexico, and the Mexican government is trying to avoid the political pain that comes from fixing its badly mismanaged economy. Here is how we might imagine that problem playing out:


               Mexico

U.S.	Reform	Don’t Reform
Fence	(-a,-b-d)	(-a,-d)
No Fence	(-d,c-b)	(-d,c)

a is the cost to the U.S. of building a fence to keep migrants out. b is the domestic political cost to the Mexican government of economic reform. c is the gain to the Mexican government when illegal immigrants send money back from the U.S. And d is the cost to each country of providing public services to those who migrate if there is a fence. Under each combination of actions, the table thus shows the payoffs to each government.

The way economists solve this problem – to find out which combination of actions by each government is rational – is to look for the Nash equilibria, the combination of actions where each party is taking the best action given the other party’s action. There are potentially two: if a is less than d then the U.S. builds a fence and Mexico doesn’t reform. If a exceeds d then the U.S. doesn’t build a fence and Mexico doesn’t reform.

In other words, it never pays Mexico to reform. The political costs of reform never make it worthwhile, whether the immigrants are there or not. And thus we have the U.S./Mexico immigration problem in a nutshell: the presence of the U.S. labor marker gives the Mexicans no incentive to take the painful steps to make their economy more efficient, preferring instead that that economy’s victims go north and send money home. I have recently argued that the costs to the U.S. of “fixing” the illegal immigration problem would be prohibitive – not just the lost productivity, but the gradual evolution into a police state with its hands deeply involved in every business in the land.

From the Mexican government’s point of view the current situation is about as good as can be expected. It would much rather have all those surplus, unemployable workers working hard up north then back home causing trouble. And they seem to know that the American anger over illegal immigration is a thing that comes and goes. Unless the Mexican people can be given a stake in economic reform back home, so that the Mexican government will have an incentive to overcome the high but short-lived political costs of major change, it will never pay the Mexican government to take tough steps.

How might that happen? One way would be if workers and businessmen in Mexico suddenly found that they too had a greater stake in domestic reform. And that in turn might happen if the two economies become more closely connected – in particular, if Mexicans living here and there could be induced to set up more cross-border commerce, which would put everyone – workers and managers, Mexican and American – on the side of better governance. And the best way to make that happen is probably to free the restraints on cross-border movements of factors, including labor, as we did already for goods. While the assumption is that this would cause Mexicans to come here by the tens of millions and stay permanently, in fact once they were free to come and go as they pleased they could work here and return home and speed up the reform of the society to which they are most attached, their own.

In other words, a North American version of the European Union round about 1990, without the ambition-destroying welfare/regulatory state and the demographic death spiral, would be just the thing. Mexico would become like Spain or Ireland, a nation that egested immigrants for decades before they made enough money elsewhere and, thanks to free labor flows, came back home and revitalized their own societies. I am hard-pressed to think of anything else that would get two nations out of a very costly equilibrium.