The Hidden Relationship Between Housing, Migration, and Inequality

In the November issue of the Journal of Urban Economics, Peter Ganong of Chicago’s Harris School of Public Policy and Daniel Shoag of the Harvard Kennedy School investigate a troubling question: “Why has regional income convergence in the U.S. declined?“

In economics, convergence—or the “catch-up effect”—is the hypothesis that per capita incomes in poor and rich economies eventually converge because incomes in poor economies grow more quickly than incomes in rich economies. For a hundred years, from 1880 to 1980, this hypothesis held true in the United States: Per capita incomes in poorer states indeed rose more quickly than per capita incomes in richer states, resulting in a shrinking gap between high-income and low-income states. Since the 1980s, however, the rate of convergence slowed dramatically. To explain this phenomenon, most research has focused on the demand side of the labor market; by contrast, Ganong and Shoag explore the supply side. Specifically, they investigate how housing costs and migration patterns impact the labor supply.

Historically, both high- and low-skilled workers migrated from low-wage states to high-wage states, slowing wage growth in high-wage states and accelerating wage growth in low-wage states. But since the 1980s, both the level of migration and the rate of income convergence between high- and low-wage states declined. Ganong and Shoag show that differences in housing costs contributed to this effect. Increased land use regulation since 1965 made it more difficult for developers to build new housing in response to population growth. Over time, this decreased housing supply relative to demand, pushing up housing costs in areas with more regulations.

Ganong and Shoag use data from the American Community Survey to track interstate movement and data from State Economic Areas and Public Use Microdata Areas to construct time series measuring wages, rents/housing prices, and skill levels. Motivated by the observation that low-skilled workers were migrating away from high-wage areas while those areas experienced a disproportionate rise in housing costs, they build a corresponding model of the relationship between wages, housing costs, skill level, and region.

The authors then measure land use regulation intensity by counting all incidents of the phrase “land use” in state appeals and Supreme Court rulings, under the assumption that any impactful regulation would have been contested. They find that before 1980 there was a strong relationship between income growth and current income, with lower-income states growing faster than high-income states. But since 1980, states with a less constrained supply of housing experienced greater income convergence.

High-skilled workers are less sensitive to higher housing costs than low-skilled workers because they generally spend a smaller share of their incomes on housing. Over time, this variation in sensitivity to housing costs produced a change in migration patterns among high- and low-skilled workers. In the past, both high- and low-skilled workers were relatively free to move from state to state in search of higher wages, driving income convergence across the country. However, rising housing costs since the 1980s constrained the movement of low-skilled workers, forcing low-skilled workers away from higher-wage states since 1995. This is because income after housing costs is actually lower for some low-skilled workers in high-wage regions than in low-wage regions. For example, after housing costs are factored in, a janitor in NYC may earn less than a janitor with a comparable wage in Mississippi. Over time, this “regional sorting” results in a concentration of high-skilled workers in high-wage, high-cost states and low-skilled workers in low-wage, low-cost states.

This research offers a partial explanation for increasing inequality since the 1980s. Much talk of addressing inequality is oriented at the federal level with proposals for more progressive tax structures and income redistribution. However, this research reveals an opportunity for policymakers at the state and local levels to make significant, if gradual, impacts on inequality by using the tools at their disposal to increase the housing stock and so reduce barriers to migration. The authors estimate that if interstate income convergence had continued from the 1980s at the pre-1980 rate, the increase in hourly wage inequality from 1980 to 2010 would have been 8 percent smaller. Going forward, more research is needed to figure out exactly how large a role housing costs played in the post-1980 decline in income convergence, relative to other causes.

Article source: Ganong, Peter & Daniel Shoag. “Why has regional income convergence in the U.S. declined?” Journal of Urban Economics. 102 (2017): 76–90

Featured photo: cc/(dabldy, photo ID: 477193287, from iStock by Getty Images)