We normally think of income inequality as a function of differences in class or socioeconomic status. But much more than generally realized, geographic differences are a major source of inequality as well, argue Brink Lindsey and Steven Teles in this extract from their new book The Captured Economy.


Editors’ note: In advance of this week’s panel discussion “Radical Markets and the Captured Economy” at the Stigler Center, we offer this extract adapted from panelists Lindsey and Teles’s recent book The Captured Economy: How the Powerful Enrich Themselves, Slow Down Growth, and Increase Inequality (Oxford University Press).


If you want to understand the connection between rent-seeking and America’s economic ills of slow growth and high inequality, nowhere is that connection more consequential—and less obvious—than in the case of land-use regulation. Since rules governing the development of real property are generally administered at the local level, it’s natural to suppose that their effects are local as well. But in recent years, it’s become increasingly clear that these rules add up to a massive drag on national economic growth—and they do so by locking in economic inequality between different parts of the country.


Zoning ordinances and the like have been endemic in the United States for the better part of a century. These laws have always influenced the location of housing within a given metropolitan area—that was their whole point. But until relatively recently, they didn’t have much of an impact on the total amount of housing built. Since the 1970s, though, increasing restrictiveness in America’s big coastal cities has caused new housing supply to lag behind rising demand, resulting in a big, artificial boost to housing prices. According to Harvard economist Edward Glaeser’s calculations, this “regulatory tax” equals roughly 20 percent in Baltimore, Boston, and Washington, DC. In Los Angeles and Oakland, it surpasses 30 percent. And in Manhattan, San Francisco, and San Jose, the regulatory tax has reached roughly 50 percent.


But why do these local rules matter nationwide? Even if zoning restrictions are growing progressively worse in big urban areas, the United States remains a largely empty country. If people are priced out of building or living in one location, alternatives always exist. Yes, zoning alters where people choose to live, but how does the location of America’s population affect prospects for expanded output and higher incomes? To connect artificially high housing prices on the coasts with big macroeconomic effects on output and inequality, it’s necessary to delve a bit into economic geography.


The Geography of Inequality


It turns out that most of our country is empty for a very good reason: people derive great value from concentrating together in urban areas. First, proximity reduces transportation costs, so producers benefit from being close to their suppliers and customers. Second, more people living in one place means deeper and more diverse markets for both products and labor. With a large enough urban population, niche markets that appeal to only a small fraction of consumers become profitable to serve. Employers have a better pool of potential workers to draw from, while workers have greater choice in prospective employers. And third, people living and working close to one another can take advantage of “information spillovers”: cities expand opportunities for exchanging ideas and information, thereby facilitating both innovation and the accumulation of human capital.


Economists call these benefits of urban concentration, which combine economies of scale and network effects, “economies of agglomeration.” For most of history, these centralizing forces were held in check by the requirements of traditional, low-productivity agriculture: growing food required not only lots of open land but also relatively large numbers of people to work that land. But as the mechanization of agriculture freed people from work on the farm, the gravitational attraction of cities asserted itself.


But all cities are not created equal. Whether due to natural factors (e.g., climate, proximity to bodies of water) or accidents of history (e.g., William Shockley moved to northern California to take care of his ailing mother and thereby set in motion a train of events that would result in Silicon Valley), some urban areas end up with much higher populations, output, and output per worker, than others. Thus, as of 2005 the top 50 most productive metropolitan areas in the United States (out of a total of 363) combined to produce 60 percent of the nation’s GDP. Meanwhile, in the top ten cities the unweighted average of output per worker exceeded $91,000 – more than double the average of just under $40,000 in the country’s ten least productive metro areas.


As of 2005 the top 50 most productive metropolitan areas in the United States (out of a total of 363) combined to produce 60 percent of the nation’s GDP. Meanwhile, in the top ten cities the unweighted average of output per worker exceeded $91,000 – more than double the average of just under $40,000 in the country’s ten least productive metro areas.

These big differences in productivity translate into big differences in incomes. We normally think of income inequality as a function of differences in class or socioeconomic status: workers in high-skill occupations with high levels of educational attainment make more than workers with less education and lower skill levels. But much more than generally realized, geographic differences are a major source of inequality as well. Indeed, geographic inequality can sometimes outweigh the more familiar socioeconomic inequality. For example, the average income of high school graduates in Boston is now over 40 percent higher than the average income of college grads in Flint, Michigan.


Geographic inequality, though, tends to be self-liquidating when people can move from poorer areas to richer areas. Wage differences across cities and regions, by creating incentives for moving, spur an arbitrage process whereby those wage differences are reduced and overall output and wages rise.


Through much of the twentieth century, geographic mobility was a significant contributor to both income convergence and economic growth. Between 1880 and 1980, per capita incomes in US states converged at an average rate of 2 percent a year. In other words, per capita incomes in poorer states rose faster than in richer ones. Many factors worked together to produce this result, but one was labor mobility: over this period, people moved on net from poorer to richer areas. These dynamics had strongly egalitarian consequences: approximately one third of the decline in hourly wage inequality between 1940 and 1980 was due to cross-state convergence.


Mobility was also an engine of growth. The lure of higher wages in the innovative, high-productivity cities of the Northeast and industrial heartland led to explosive population gains in those cities. Between 1870 and 1950, New York City’s population grew over 700 percent, Chicago’s climbed over 1,100 percent, and Detroit’s population skyrocketed by over 2,200 percent. In analyzing the productivity gains from industrialization, economists regularly stress the important role played by reallocating labor from low-productivity agriculture to high-productivity factory and office jobs. Much of this move from one economic sector to another was accomplished by a physical move from one place to another. The sectoral and spatial reallocation of labor from less productive to more productive uses went hand in hand.


Rising Barriers to Mobility


In recent decades, however, this reallocation process has sputtered and broken down. According to economists Peter Ganong of the University of Chicago and Daniel Shoag of Harvard, between 1990 and 2010 the rate at which income gaps across states narrowed was less than half the long-term historical rate. With the shift from an industrial to an information economy, the innovative, high productivity cities of today are no longer manufacturing hubs. Rather, the cities that now feature big wage premiums, and which therefore should be attracting big influxes of new workers, are human capital hubs: urban areas with large numbers of college graduates. Unfortunately, these are the very same cities that have led the way in constraining housing growth with ever more restrictive land-use regulation.


Education levels in American cities have been diverging over time: the smart cities get smarter while other cities fall farther and farther behind.

Since the 1970s, college graduates have increasingly tended to congregate in particular urban areas—namely, those that started out with initially high shares of college grads. In other words, education levels in American cities have been diverging over time: the smart cities get smarter while other cities fall farther and farther behind. The contrast between the most educated and least educated cities in the United States is now truly remarkable. In the five top human capital hubs, the average share of workers with a college degree or better was 49 percent as of 2006-08; among the bottom five metro areas in the rankings, the average rate of college completion among workers was only 12 percent. To put the disparity in perspective, this four-to-one ratio is equal to the difference between the college completion rates in the United States overall and Ghana.


The increasing geographic concentration of highly skilled workers is a response to those economies of agglomeration we discussed earlier: in particular, the information spillovers that accelerate skill acquisition and innovation when smart people are working together in close proximity.


And those payoffs don’t just benefit the college grads. Less skilled workers in the city are even bigger winners—in part because they are providing services to lots of affluent managers and professionals who can afford to pay them high wages. Let’s look again at the top five and bottom five metro areas in terms of share of college-educated workers. In the top five, college grads earn an average annual salary of $87,689 as of 2006-08—61 percent more than the average salary of $54,518 earned by college grads in the bottom five. By comparison, high school grads in the top five make $71,483 a year on average—a whopping 137 percent more than their counterparts in the bottom five. According to calculations by Enrico Moretti of the University of California, Berkeley, a percentage point increase in a city’s share of college-educated workers boosts the earning of college grads in that city by 0.4 percent—and lifts the earnings of high school grads by 1.6 percent, or four times as much.


America’s innovative, high-productivity human capital hubs should be experiencing rapid population growth as both highly skilled and less skilled workers flock there in search of higher pay and wider opportunities. But they’re not. In recent decades, migration flows in the United States have been going in the opposite direction: away from the richer coastal cities and toward the poorer exurbs of the Sunbelt. To take an especially striking example, consider the experience of San Jose, California—the heart of Silicon Valley. Between 1995 and 2000, at the height of the dizzying Internet boom, 100,000 more Americans moved out of the San Jose metro area than moved in. In his 2011 book The Gated City, Ryan Avent of The Economist refers to this inversion of traditional migration patterns as “moving to stagnation.”


Why move away from higher and faster growing wages? The answer is clear: housing costs. The rise of the Sunbelt has been a major demographic story of the post–World War II era, but the plot of the story has taken a big twist in recent decades. As Edward Glaeser has documented, between 1950 and 1980, population gains in the Sunbelt were propelled primarily by above-average productivity growth. Since 1980, however, the rapid growth in the Sunbelt’s housing supply—and thus its growing advantage in offering affordable housing— is the main factor behind the region’s continuing attractiveness.


Meanwhile, in the high-productivity human capital hubs that should be growing robustly, artificially high housing costs act as a regressive filter. Since housing costs bulk larger in the budgets of less skilled, lower-income workers than for the highly skilled and well paid, those costs have a differential impact in deterring in-migration. Even though the gross wage premium for college grads is smaller than it is for high school grads, the situation changes when you look at wages net of housing costs. High-income workers still enjoy a net wage premium and thus still have an incentive to move, but for less skilled workers the regulatory tax on housing wipes out that nominal wage premium. As a result, college grads continue to move to human capital hubs, while less educated workers—who would stand to gain the most by moving—are kept away by artificial housing scarcity.


Enrico Moretti, working with Chiang-Tai Hsieh of the University of Chicago, has attempted to quantify the cumulative impact these local distortions have on the national economy. Their analysis reveals that the scale of these distortions, and the combined toll they have taken, are staggeringly large. When you look at wages across metropolitan areas (as opposed to across states or regions), there has been no convergence at all over the past half century. On the contrary, there has been sizeable divergence: geographic inequality has gotten dramatically worse. Specifically, Moretti and Hsieh find that the standard deviation of wages across U.S. cities in 2009 was twice as large as it was in 1964. This worsening spatial misallocation of labor has exacted a stiff price. The increasing dispersion of wages has reduced total U.S. economic output by an average of 0.3 percentage points a year.


As to the cause of growing geographic inequality, Moretti and Hsieh conclude that the main culprit is land-use restrictions. And what is truly striking, the lion’s share of the harm is being caused by the highly restrictive policies of just three cities: New York, San Francisco, and San Jose. If regulatory barriers to new housing construction in those three cities had been pared back to just the median level of restrictions nationwide, Moretti and Hsieh estimate that the resulting influx of workers would have raised overall U.S. output by 9.7 percent over the period in question.


With land-use regulation, slumping growth and rising inequality are thus inextricably connected. It is precisely through perpetuating and entrenching geographic inequality that potential economic output is squandered. Accordingly, liberalizing the rules for new housing construction offers an attractive two-for-one deal: ensuring both a bigger economic pie overall and more equally sized slices.


Brink Lindsey (@lindsey_brink) is vice president for policy at the Niskanen Center.


Steven M. Teles is associate professor of political science at Johns Hopkins University and a senior fellow at the Niskanen Center.


Disclaimer: The ProMarket blog is dedicated to discussing how competition tends to be subverted by special interests. The posts represent the opinions of their writers, not necessarily those of the University of Chicago, the Booth School of Business, or its faculty. For more information, please visit ProMarket Blog Policy.