From shortly after the Civil War until perhaps three decades ago, the United States passed through a remarkable period of economic convergence. During this time, counties became more similar to other counties and states became more alike other states, at least in terms of GDP, per capita income and other measures of economic performance. The way we measure this is simply whether or not poor places grow faster than rich places over a prolonged time.

Economic theory from the 1950s argued that this convergence would occur because investment in machinery would flow to the locations where they were less abundant. The scarcity of capital equipment in poor places would make them more valuable there, and yield a higher return on investment for their owners. In short, this is the mechanism of capitalism. 

In the 1980s, it became clear that this trend was not materializing between nations. However, by including differences in educational attainment among regions, economists could account for differences in convergence. In other words, education, or human capital, began to replace physical capital and equipment as a factor explaining economic growth. The mechanism of capitalism had to take into account more fully the role of human capital than was previously necessary. 

Of course, this model simplifies the economic growth argument. Most economists believe institutions matter as well. Institutions are a fancy word for the rule of law, absence of corruption and reasonably well functioning civil and civic capacity. Writing this column from my office in Muncie, it’d be hard to conclude otherwise. After all, the simple fact that economic growth in Muncie averaged 1.5 percent less than predicted in this 21st century cannot be attributed solely to educational differences. A resident of Muncie need not read the research to conclude something else is at work beggaring the region. While that is for another column, suffice it to say that education also plays a key role in the success of institutions. 

For most of the past 150 years in the U.S., regional differences in education have also been shrinking. Public educational became universal, the GI Bill caused an explosion of college attendance after World War II and household location decisions increasingly favored places with better schools. Taken together, these trends might also have set the stage for the end of regional convergence, and perhaps propelled us into a period where rich places are becoming richer, and poor places poorer. 

Using new county-level data on gross domestic product released last December, I tested whether regional economies were converging towards similar levels of GDP or diverging. It is now clear the nation is in a period of regional divergence. Places that were poor in 2001 grew more slowly through 2018 than did more affluent places. So, places are becoming much more unequal. 

This problem is akin to that of household inequality that has dominated the public debate for more than a decade. But, unlike growing household inequality, two important aspects differ between the two. First, much of the growth in household inequality can be explained by changes in family structure. We don’t have many simple policies to change family formation dynamics, even if we were daft enough to want to become involved. Second, we can readily equalize income inequality through taxes and subsidies. There are good arguments about the efficacy of such policies, but not much argument that we can better equalize how much a family can spend on goods and services. 

For growing regional inequality, the problem is different. It isn’t clear precisely why it is happening now. Part of the explanation is clearly the huge productivity boost cities provide to well-educated workers. The formula is pretty clear. If you want to earn a lot of money, get a good college degree and move to any large city. Plenty of folks are doing just that, but the economic forces that pushed physical capital away from rich places to poor places work differently for human capital. Economists call this force ‘agglomeration.’ This means that the rate of return of being around other educated people rises as the more abundant they become. This is just the opposite of the effect with machinery. 

The best evidence suggests the effects of agglomeration are growing, not subsiding. This naturally concerns many Americans, as few of us think it good for a nation to be transforming into increasingly rich and increasingly poor places. This concern needn’t be supported by a sophisticated economic argument. Most of us love people who live in both in big cities and small towns, and wish them all the chance to prosper. Most of us also don’t wish to see people who live elsewhere as political enemies, a condition which will surely be exacerbated by regional inequality. 

Growing regional inequality is a challenge of our age. Yet, we have few, if any effective policies that can reverse this trend. That is the topic of the next column.