Richard Florida is a co-founder and editor at large of CityLab and a senior editor at The Atlantic. He is a University Professor and Director of Cities at the University of Toronto’s Martin Prosperity Institute, and a Distinguished Fellow at New York University’s Schack Institute of Real Estate.
Where the rich live with the rich, and the poor live with the poor.
This is the first post in a five-part series examining economic segregation in U.S. metros.
Debates in the U.S. over income inequality have taken center stage in recent years, but its existence in our cities is of long standing. Major metro areas have been magnets for both the rich and the poor since ancient times; in fact they owe a great deal of their dynamism to their economic and social diversity. But growing economic segregation—the increasing tendency of affluent people to live in neighborhoods where almost everyone else is affluent, and poor people to live in neighborhoods where almost everyone else is poor—may be a more insidious problem. The emergence of a new urban geography of concentrated wealth and advantage juxtaposed to endemic poverty and concentrated disadvantage poses troubling implications for the economic mobility of people and the economic health of cities.
Just as lower-skill, higher-pay manufacturing jobs have dropped out of the labor market, and work in America has bifurcated into high-skill, high-paying professional and knowledge occupations and much lower-paying, low-skill service jobs in fields like food service and retail trade, America’s once middle-class neighborhoods have also begun to disappear.
In 1970, roughly two-thirds (65 percent) of Americans lived in neighborhoods that could be described as middle income; today that number is just slightly more than four in ten (42 percent), according to a study by Cornell University’s Kendra Bischoff and Sean Reardon of Stanford University. Over the same time span, the proportion of families living in affluent neighborhoods rose from 7 to 15 percent, and the share living in poor neighborhoods increased from 8 to 18 percent. The share of Americans living at both extremes grew from 15 percent in 1970 to 33 percent in 2009. Income segregation grew in nearly nine in ten of all U.S. metros with populations over 500,000 people, according to the study. A 2012 report by the Pew Research Center found economic segregation of upper- and lower-income households to have risen in in 27 of America’s 30 largest metros.
Working closely with Charlotta Mellander and my Martin Prosperity Institute team, I have charted the level and extent of segregation by income, of the rich and the poor, of the highly educated, and by socioeconomic class across tracts in all of America’s more than 350 metros (To track income segregation, I will today use data modeled after the Pew Research Center’s methodology. Future posts will use a method based on the landmark work of Douglas Massey and Nancy Denton.) I will be sharing additional findings over the next several weeks.
We measured income segregation by calculating the share of low-income households that are located in neighborhoods with a majority of low-income households in comparison to the share of upper-income families who live in neighborhoods with a majority of upper-income households, following the methodology used by the Pew Research Center to calculate their Residential Income Segregation Index (here, neighborhood is defined as a Census tract). We defined upper income households as those with annual incomes of $100,000 or more and lower-income households as those with annual incomes of $34,000 or less. The analysis covers all 70,000-plus Census tracts in the U.S. and matches them to more than 350 metros. My MPI colleague Zara Matheson mapped the results.
• • • • •
The map below shows how U.S. metros stack up on income segregation. Dark blue reflects high levels of income segregation; light blue significant levels; green moderate levels, and yellow low levels.
As the map clearly shows, there are high levels of income segregation in metros along the Boston-New York Corridor, in Greater Miami, in Northern and Southern California, across the Sunbelt to Texas, and in the major Midwestern cities.
America's Most and Least Segregated Large Metros
The tables below show the ten large metros (those with one million or more people) that rank highest and lowest in terms of income segregation.
|America's Most Income-Segregated Large Metros|
|Rank||Metro||Index||Rank of All Metros|
|1||San Antonio-New Braunfels, TX||0.700||7|
|3||New York-Northern New Jersey-Long Island, NY-NJ-PA||0.653||12|
|4||Houston-Sugar Land-Baytown, TX||0.647||13|
|6||San Francisco-Oakland-Fremont, CA||0.619||20|
|8||Dallas-Fort Worth-Arlington, TX||0.595||28|
|10||Austin-Round Rock-San Marcos, TX||0.576||32|
San Antonio has the highest level of income segregation, followed by Memphis, New York, Houston, Washington, San Francisco, Philadelphia, Dallas, Denver, and Austin. (There is substantial overlap with the findings of the Pew Report, even though their list is limited to just 30 metros. Five metros — Houston, Dallas, New York, Philadelphia, and Washington, D.C. — appear on both lists.
The most highly segregated metros are actually smaller and medium sized, many of them in Texas. El Paso tops this list, followed by second and third-ranked Laredo and McAllen. College Station comes in sixth place. San Antonio, which is first out of large metros, is eighth overall, and Brownsville is ninth. Outside of Texas, Bridgeport, Connecticut is fourth; Trenton, New Jersey fifth; Memphis eighth; and Jackson, Tennessee tenth.
|America's Least Income-Segregated Large Metros|
|Rank||Metro||Index||Rank of All Metros|
|49||Virginia Beach-Norfolk-Newport News, VA-NC||0.283||256|
|46||Las Vegas-Paradise, NV||0.323||221|
|44||Salt Lake City, UT||0.366||182|
|43||Minneapolis-St. Paul-Bloomington, MN-WI||0.373||173|
On the flip side of the equation, Orlando has the lowest level of income segregation of large metros. Portland, Virginia Beach, Seattle, Jacksonville, Las Vegas, Sacramento, Salt Lake City, Minneapolis, and Rochester make up the ten least segregated large metros.
The least segregated metros in the country are also smaller metros. In fact, there are about 80 small- and medium-sized metros that have lower levels of income segregation than the least segregated large metro, Orlando, Florida. Holland-Grand Haven, Michigan has the lowest level of income segregation in the nation. St. George, Utah; Palm Coast, Florida; Glen Falls, New York; Sheboygan, Wisconsin; and Harrisonburg, Virginia round out the five least segregated metros in America.
• • • • •
We know which metros have the greatest and least levels of income segregation. But what do they have in common? What are the underlying factors associated with higher or lower levels of income segregation?
To get at this, Mellander ran a basic correlation analysis between income segregation and a number of key economic, social and demographic characteristics of metros. As usual, I note that correlation does not equal causation and points only to associations between variables.
First off, income segregation is higher in larger metros (with a correlation of. 36). It is positively, but more modestly related to density (.27). This is not surprising, as large metros have more households and a greater range of incomes and types of housing; there is more gentrification and “superstar” neighborhoods in such places, which price out less affluent people, enabling the better off to ensconce themselves in exclusive enclaves.
America has long been divided and segregated along racial lines. And race is significantly related to income segregation, according to Mellander’s analysis. Income segregation is higher in metros where Blacks and Latino make up greater shares of the population (with positive correlations of .33 and.25 respectively); and it is lower in metros with a greater share of the population is White (-.27). There is no statistical association, however, between income segregation and the share of the population that is Asian.
You would think that income segregation would be higher in more affluent metros. But income segregation is only weakly associated with average wages (.14) and not significantly related to either per capita income or economic output per person. This result is curious and a bit counterintuitive. In part, this could be because average wages and income say little about the distribution of money across a region’s residents.
It is therefore important to note that income segregation is much more closely associated with income inequality as measured by the Gini coefficient (with a correlation of .54). Income inequality explains roughly 28 percent of the variation in income segregation, according to the simple regression analysis Mellander conducted. In other words, income segregation is more closely related to the gap between the rich and the poor (captured by income inequality) than the overall level of development or affluence (measured by income, wages or economic output per capita).
One might also suspect that income segregation would be greater in metros that have more advanced high-tech, knowledge-based economies. Income segregation is only modestly associated with the concentration of high tech industry (.26) and also with the share of creative class workers (.27); and, it is not statistically associated at all with the share of adults that are college grads, a common measure of human capital. Again, this may be an artifact of our measure, which identifies income segregation overall, but not the isolation of the rich or poor individually.
While income inequality and residential segregation do go together to a degree, it is important to remember that they are not the same thing. A city or metro might be quite unequal but not particularly segregated if lower and upper income groups are distributed evenly across neighborhoods. Likewise, a city or metro could be highly segregated but relatively equal, if its income groups reside in different neighborhoods across the same metro area.
Though the geographic segregation of income is not a direct consequence of income inequality, there are several ways that it compounds these economic disadvantages. Schools, parks, policing and all manner of other public services, goods and amenities are more abundant and better-quality in higher income neighborhoods. This improves the life chances and opportunities for children born into those communities while limiting the mobility of children born into lower income communities, who are more dependent upon public services and investments to compensate for limited family resources. Numerous studies (including several I have discussed here on this site) document the lower levels of socioeconomic mobility and lasting neighborhood effects that go along with concentrated disadvantage.
As middle class neighborhoods have declined, America’s economic landscape is increasingly polarized. My next two posts examine the growing segregation of the poor and the increased isolation of the affluent across the United States.
Top Image: The Queensbridge Houses, the largest public housing development in America (left), and an affluent street in Manhattan's Upper East Side (right) (Wikimedia Commons users Wikiwiki718 and Gryffindor).