One of the greatest challenges facing the United States is the growing economic disparity between its different regions. Over the last forty years, a handful of prosperous areas have pulled away from the rest of the country. This is a remarkable reversal from the prior 100 years, when the poorer parts of the nation were the places that, economically speaking, grew fastest. Today, more than a third of Americans live in metropolitan areas that are significantly richer or poorer than the nation as a whole—almost three times the share that did in 1980.
This growing geographic inequality is creating a host of serious problems. It makes it harder for the federal government to set economic policy, because one interest rate and money supply must meet the needs of rich and poor places at the same time. It may also be contributing to the extensive regional variation in upward mobility rates. More fundamentally, economic divergence between regions of the country undermines the idea that we’re all one nation, in this together. When the average person in a place like Brownsville, Texas makes less than half as much as the average person in Washington D.C., they might wonder what exactly they have in common.
Because the United States apportions political representation on the basis of geography, economic divides between places are deeply consequential for U.S. politics. Hillary Clinton famously won counties that collectively produce almost two-thirds of U.S. gross domestic product, even as she lost the national election. One of the most powerful predictors of Democratic vote share in 2018 was prosperity.
In order to reduce the economic disparities between places, we need to understand their source. Because some places are doing well while others are not, we tend to assume that disparities are largely a local problem. The successful regions are doing something right, while the less-than-successful ones are doing something wrong. As a result, we often propose solutions that are targeted specifically at places that need to ‘catch up.’
But this is, at best, a limited way of thinking about the problem. At worst, it’s downright wrong. As I recently discovered, the national rise in income inequality is, by itself, responsible for more than half of the growing gap between rich and poor places. As the very richest people have taken a larger and larger share of national income, the average incomes in the places where they were already concentrated has gone up too. This—not changes in where different types of people live, or where different types of jobs are found—is the main contributor to our growing regional disparities. Some places were already a little wealthier than others. As national inequality increased, this gulf widened dramatically.
This finding is at odds with the conversation on regional inequality, which is largely focused on the “sorting” of workers into cities by their education or income level. These explanations note that college-educated workers increasingly live in a different set of metro areas from high school-educated workers. Some argue that this is because certain cities have created an urban culture that attracts high-income workers, while others claim that it stems from changes in the geography of labor demand tied to the rise of geographically concentrated “knowledge” industries. As a result, most of the policy prescriptions are focused on things like subsidizing employment in struggling regions or piecemeal efforts to bring major white-collar employers to rust belt cities. One prominent urban policy writer recently recommended doing this by moving parts of the federal government.
But while sorting has certainly occurred, these prescriptions will not make a dramatic difference. According to my research, sorting is responsible for only a quarter of the economic divergence that developed between 1980 and today. It’s the national trend of rising inequality that balkanized America’s regions.
To show this, I conducted two simulations. In the first, I created a timeline where the regional economic demographics of America froze starting in 1980. In this world, the percentage of New Yorkers who fall in the richest 10 percent of the country remains the same today as it did when Ronald Regan first took office. The percentage of people who fall in the bottom 90 percent likewise remains unchanged. Wealthier people didn’t move to New York, and poorer people didn’t leave. But nationwide inequality still grew by the same amount. In other words, the income of America’s top 10 percent grew at the same rate between 1980 and today as it did in real life.
How unequal would America’s different regions be in this scenario? As it turns out, inequality between America’s different areas would have grown by 54 percent as much as it did in the real world. Again, that’s with no movement of people whatsoever. Indeed, according to my findings, America’s growing regional inequalities aren’t even really driven by the merely rich. They’re driven by the very rich. Half of the growth in disparities between regions comes from the richest 1 percent of Americans alone.
In the second simulation, I assumed that the opposite was true. There was no rise in nationwide inequality between 1980 and today, but people and jobs moved between regions just as they did in real life. This world would have only produced 23% as much economic divergence between regions.
This makes it unlikely that local jobs programs, federal aid, and better urban amenities will make a serious dent in regional inequality. It also means it is incorrect, as well as callous and politically unwise, to say that the struggles of places like St. Louis are their own fault. These struggles are primarily the result of national economic shifts. Fixing them will almost certainly require action at the national level.
To do this, we need to reverse or redo the many changes to economic policies made during the 1970s and 1980s that bolstered national inequality. These types of policies—lowering taxes on the wealthy, cutting the minimum wage, relaxing antitrust enforcement, and reducing trade barriers—aren’t typically thought of in spatial terms. But because different types of people, industries, and jobs are unevenly distributed across places, they have a spatial impact. It’s time we recognize this fact.
