Americans pride themselves on their intergenerational mobility, the American dream: No matter where you come from and no matter who your parents are, you can rise to the top of the economic ladder, so long as you are willing to commit yourself and work hard.

In recent years, however, a great deal of comparative research has raised legitimate questions about whether the United States stands out as a land of opportunity. In 2006, a widely reported study found that in terms of intergenerational mobility, the U.S. lagged behind Nordic nations (Denmark, Finland, Sweden and Norway) as well as the United Kingdom.

For example, Danish men born to households in that nation's bottom "quintile" (the lowest 20 percent of incomes) are far more likely than their U.S. counterparts to make it into the higher quintiles.

A comprehensive study published in July found that the United States shows less intergenerational mobility than a number of other countries, including Germany, New Zealand, Canada, Australia, France and Japan. The U.S. is marked by a degree of "stickiness" in the top and bottom 10 percent. If an American is born to poor parents, he has a decent chance of staying poor; if his parents are wealthy, it is a pretty good bet that he will end up in the economic elite.

However illuminating, the data raise many questions. The U.S. is a big country, and the aggregate numbers don't tell us about variations across states.

Does the reality of the American dream depend on whether you are born in Mississippi, Colorado, Kentucky or New Hampshire? The answer would help us to establish what, exactly, is the source of the problem.

Harvard economist Raj Chetty and his coauthors have started to provide an answer. The big news is that intergenerational mobility is indeed variable across regions, states and cities.

If you are born in Pittsburgh, Boston, San Francisco, Minneapolis or New York, you have a fair chance of getting to the top fifth of the income distribution, even if you start out in the bottom fifth. But in other cities — Atlanta, Charlotte, Nashville, Raleigh — children who are born into the bottom fifth are significantly more likely to get stuck.

State-by-state data suggest an even more vivid picture. In almost all of the West, there is a high degree of mobility. The Northeast and the Southwest also look pretty good. In the Midwest, the picture is more mixed, with generally high levels of mobility in Minnesota, Nebraska and North Dakota but disturbingly low levels in some cities, including Detroit, Indianapolis, and Columbus, Ohio.

Mobility is markedly low in one region above all: the Southeast. That region's cities dominate the list of those showing the worst levels of mobility, with heavy representation from Georgia, Mississippi, South Carolina, North Carolina and Tennessee.

It is tempting to think that economic growth is the crucial factor — we will observe mobility where we also observe growth. Surprisingly, however, regional differences in mobility have little to do with differences in growth. It is also tempting to think that racial differences are involved. But essentially the same regional differences appear after controlling for race.

The good news is that the United States as a whole is not suffering from especially low levels of intergenerational mobility. The problem is fairly localized. But two fundamental questions remain: Why is intergenerational mobility high in some regions and low in others? And where it is low, what can be done to increase it?

We don't yet have answers, but Chetty and his coauthors offer some preliminary clues. In general, "tax expenditures aimed at low-income taxpayers have significant impacts on economic opportunity." More specifically, intergenerational mobility is higher in states that have a generous earned income tax credit as well as progressive income taxes and tax expenditures.

To be sure, correlation is not causation. Progressive state taxes could be correlated with other policies (such as a strong educational system) that are responsible for intergenerational mobility. Researchers aren't yet in a position to explain the high regional variations in mobility. But we know enough to appreciate the immense importance of getting to the bottom of that particular puzzle.

The American dream depends on it.

-------------

Cass R. Sunstein is a professor at Harvard Law School. He is the former administrator of the White House Office of Information and Regulatory Affairs.