Joel Kotkin examines for Forbes readers some interesting data about income inequality in the United States.

Scholars of the geography of American inequality have different theses but on certain issues there seems to be broad agreement. An extensive examination by University of Washington geographer Richard Morrill found that the worst economic inequality is largely in the country’s biggest cities, as well as in isolated rural stretches in places like Appalachia, the Rio Grande Valley and parts of the desert Southwest.

Morrill’s findings puncture the mythology espoused by some urban boosters that packing people together makes for a more productive and “creative” economy, as well as a better environment for upward mobility. A much-discussed report on social mobility in 2013 by Harvard researchers was cited by the New York Times, among others, as evidence of the superiority of the densest metropolitan areas, but it actually found the highest rates of upward mobility in more sprawling, transit-oriented metropolitan areas like Salt Lake City, small cities of the Great Plains such as Bismarck, N.D.; Yankton, S.D.; Pecos, Texas; and even Bakersfield, Calif., a place Columbia University urban planning professor David King wryly labeled “a poster child for sprawl.”

Demographer Wendell Cox pointed out that the Harvard research found that commuting zones (similar to metropolitan areas) with less than 100,000 population average have the highest average upward income mobility. …

… The most profound level of inequality and bifurcated class structure can be found in the densest and most influential urban environment in North America — Manhattan. In 1980 Manhattan ranked 17th among the nation’s counties in income inequality; it now ranks the worst among the country’s largest counties, something that some urbanists such as Ed Glaeser suggests Gothamites should actually celebrate.

Maybe not. The most commonly used measure of inequality is the Gini index, which ranges between 0, which would be complete equality (everyone in a community has the same income), and 1, which is complete inequality (one person has all the income, all others none). Manhattan’s Gini index stood at 0.596 in 2012, higher than that of South Africa before the Apartheid-ending 1994 election. (The U.S. average is 0.471.) If Manhattan were a country, it would rank sixth highest in income inequality in the world out of more than 130 for which the World Bank reports data. In 2009 New York’s wealthiest one percent earned a third of the entire municipality’s personal income — almost twice the proportion for the rest of the country.

The same patterns can be seen, albeit to a lesser extent, in other major cities. A 2006 analysis by the Brookings Institution showed the percentage of middle income families declined precipitously in the 100 largest metro areas from 1970 to 2000.

The role of costs is critical here. A 2014 Brookings study showed that the big cities with the most pronounced levels of inequality also have the highest costs: San Francisco, Miami, Boston, Washington, D.C., New York, Oakland, Chicago and Los Angeles. The one notable exception to this correlation is Atlanta. The lowest degree of inequality was found generally in smaller, less expensive cities like Ft. Worth, Texas; Oklahoma City; Raleigh, N.C.; and Mesa, Ariz. Income inequality has risen most rapidly in the bastion of luxury progressivism, San Francisco, where the wages of the 20th percentile of all households declined by $4,300 a year to $21,300 from 2007-12. Indeed when average urban incomes are adjusted for the higher rent and costs, the middle classes in metropolitan areas such as New York, Los Angeles, Portland, Miami and San Francisco have among the lowest real earnings of any metropolitan area.