Graph of business migration reveals no new insights

1/10/2014 02:09:00 PM
Tweetable
Breaking: boring map causes boring blog post.
Business Insider ran this map today, showing which states businesses are moving to and from in 2013. I will be honest, I really wanted to make this post about politics. But unfortunately, my brain couldn't make that happen: as far as I can tell, there are no policy similarities that differentiate emmigration states to immigration states.

With all the interesting hypotheses disproven, that leaves us with just the boring one: the map looks like an echo of the listing of states by real per capita GDP, implying that businesses are just leaving high-wage states in search of lower-wage states, exactly as economics says they should do at the margin. And that is broadly speaking a good thing--we ought to want Texas to share in the same high incomes enjoyed by people in New York, and if you do the math, it's clear that they are in the process of catching up...fast.

Now, correlation does not always imply causation (ahem)... The fact that immigration is correlated in the predicted way with wages does not imply that businesses are moving solely in search of lower wages. If we suppose that for whatever reason businesses hate New York and love Texas, then we'd expect that at the margin, firms would probably move their least productive New York workers to Texas. The movement of low productivity workers to Texas from New York means that--all else equal--average productivity in New York rises while average productivity in Texas falls. But I think my original story is more plausible since high immigration states also have higher wage growth rates. And besides, the fact that immigration is uncorrelated with any discernable policies implies that wages themselves are also uncorrelated with policies, so it's hard to tell a story about why businesses would intrinsically hate New York and Illinois but love Texas and California.

As a parting note, let me criticize the graph on Business Insider. The calculation they did to make the graph was the ratio of immigrants to total migration in or out of a state. That's why all the states have massive figures above 40% even though the vast majority of the states had essentially no immigration or emmigration. United Van Lines has a much more appropriately drawn graph, though it doesn't really change any of my conclusions at all.