For years, economists have pointed to technology, globalization, and shifting labor markets as primary factors contributing to economic polarization in the U.S. But a new study highlights another driving force that cuts across regions — the housing market.

Since 1980, the cost of buying a home in cities like New York, San Francisco, and Los Angeles has surged faster than rents than in smaller cities, reshaping who can afford to stay in urban centers. Middle-income households — teachers, nurses, technicians, and professionals who form the backbone of urban economies — are being priced and pushed out.

In a study published in the Review of Economic Studies, Andrii Parkhomenko, assistant professor of finance and business economics, explains how soaring housing prices are driving middle-income families to move from expensive cities to more affordable places. As they leave, the “middle” of the income distribution in large cities disappears, leaving behind more high- and low-income residents and intensifying the economic polarization.

Parkhomenko’s analysis of the U.S. Census and American Community Survey data from 1980 to 2019 shows that a doubling of house prices is associated with roughly a one percentage point drop in the share of middle-skilled workers and a two point increase in the Gini coefficient, a standard measure of income inequality. Moreover, his analysis of the Current Population Survey data shows that middle-income households are much more likely to migrate to more affordable locations for housing-related reasons, as compared to low- and high-income households.

A model linking housing and inequality

To interpret these patterns, Parkhomenko builds a spatial equilibrium economic model that integrates the rent versus buy choice with location choice. The model shows that when home prices rise faster than incomes or rents, middle-income households are the first to lose access to homeownership in large cities. They either remain renters or relocate to smaller, cheaper cities where they can afford to buy.

Amplifying the effects of technology

Previous studies have emphasized skill-biased technological change (SBTC) — the idea that new technologies favor high-skilled workers — as the main reason why inequality has grown more in big cities. Parkhomenko’s model confirms this, but adds a crucial twist: When house prices rise faster than wages and rents, SBTC’s effects become much stronger.

Model simulations show that if large cities had maintained their 1980 ratios of house prices to rents and wages, they would have experienced virtually the same change in middle-skilled jobs as smaller cities. The widening gap in income inequality between large and small cities would also have been about 40% smaller.