Newsweek is promoting Suketu Mehta’s short piece on New York as a celebration of the role of immigrants in the city’s revival. But the piece is really a celebration of inequality:

People often explain the problems in European cities by citing unemployment or inequality. But New York today is one of the most unequal cities in America. In 2007, accord-ing to a study by the Fiscal Policy Institute, 1 percent of New Yorkers earned 45 percent of its income (in 1980 this single percent earned 12 percent of income). That works out to an average of $3.7 million a year for the city’s top 34,500 households. The average daily income of this group is greater than the average annual income of the city’s bottom 10 percent.

As Edward Glaeser has made clear, however, city are magnets for poor people because they offer poor people greater economic opportunity. Density and the availability of mass transit options reduces the effective cost of mobility across economic space, and the concentration of affluent consumers creates demand for service workers. And this is exactly the dynamic Mehta goes on to describe:

They come because any newcomer stepping off the plane at JFK can find a place in the hierarchy of New York. If you look at a New York City restaurant, for example, the chef might be French, the people washing dishes might be Mexican, the hostess might be Russian, the taxi driver bringing the customers might be Pakistani, the owner might be British. They are not all equal. They earn different rates. But they work together, to get food to hungry people. It’s like the Hindu caste system: it’s not equitable, but everybody has a place.

There is a small problem with this analogy, which is that New York’s caste system is not a caste system at all. Rather, New York is an imperfect machine for upward mobility, in which immigrants arrive in enclave communities, build assets, and then fan out to other cities and towns and regions. For many new arrivals, New York is a launch pad into middle class prosperity elsewhere.

It is often said that the United States is not as socially mobile as Americans believe, and that it is near the bottom of the pack among other rich democracies. This is an artifact, I suspect, of the real American caste system, or what we might call African American, and to a lesser degree Latino, exceptionalism. Consider the following from the Center for American Progress:

African American children who are born in the bottom quartile are nearly twice as likely to remain there as adults than are white children whose parents had identical incomes, and are four times less likely to attain the top quartile.

The difference in mobility for blacks and whites persists even after controlling for a host of parental background factors, children’s education and health, as well as whether the household was female-headed or receiving public assistance.

While downward mobility is relatively rare for white Americans who make it into the middle class, it is woefully common for African Americans. This lack of class stickiness for middle-class blacks is, in my view, a real problem. But it is a distinctive problem that calls for narrowly-tailored solutions, and certainly not a blanket condemnation of, for example, flexible labor markets.