The Limited Relevance of Policy to Growth

With regard to yesterday’s California vs Texas dust-up, the Economist’s Free Exchange blogger argues that state-level comparisons are a bad idea:
One is that the state level is a very poor place to be doing these kinds of comparisons. New York state is home to Manhattan…and Buffalo. Texas has the remarkably successful metropolitan areas of the “Texas T-bone”…and 17 of the poorest 100 counties in the country. California is home to both San Mateo county, home of Silicon Valley and among the richest in the nation, and Imperial county, largely agricultural and poor. State policy isn’t entirely irrelevant, but metropolitan policy and regional geography (and the history of that geography) are far more important in determining economic success.
I think that’s right. But at the same time, I think that’s part of what makes the case for paying some attention to state-level phenomena. The point being made here is that states are not meaningful economic units, metropolitan areas are. But states most certainly are meaningful policymaking units. New York and Buffalo exist in different economic worlds, but they have the same state taxes, taxes which are very different from the taxes in Florida. And I think that what looking at economic performance on the state level does is mostly highlight how little difference these kind of policy issues makes.
Indeed, Charles Kenny argues that you see something similar on an international basis:
There is a high level of persistence of growth within the rich countries of the Organization for Economic Cooperation and Development as a whole. Lant Pritchett notes that, between 1870 and 1989, two thirds of the present high income industrialized countries had per annum GDP/capita growth rates within 0.2% of the US rate. In the post-war period, trade openness, investment, average years of education and literacy rates have all risen in these countries. According to theories of growth that suggest these are significant determinants of economic performance, then, the Organization for Economic Cooperation and Development should have seen increasing growth rates. And yet what minor change there has been in these rates has been downward. Furthermore, rich countries have followed a range of different policies at any one time. French economic policies look very different from US policies across a range of measures from labor law to industrial policy. And yet growth rates in the two countries have tracked remarkably closely.
People talk a lot about the economic impact of public policy choices because it’s not clear what else we would talk about. But the evidence seems to suggest that policy choices don’t impact growth as much as people generally assume.