Friday, May 11, 2007


A friend of mine recently gave me (by tearing a paper copy out of the magazine and leaving it in my mail box; he’s charmingly old-fashioned that way) an article from The Atlantic (only available online as a summary, unless you have a subscription) on a growing body of research indicating that intergenerational income mobility in the United States is declining. When you compare the extent to which a person’s income depends on his parents’ income in the US versus, for example, Sweden or Denmark, you find that the correlation is much higher in the US – as much as 40 to 60%. And so Europe, not the US, is increasingly the land of opportunity. If you are born poor in the US, you’re much more likely to stay poor.

But I think that this misconceives what "opportunity" is really about. Opportunity is not about whether everyone can achieve a particular income level, or about how much generations drift between portions of the income distribution. Opportunity is about the ability to achieve as much as other people are willing to grant you in voluntary exchange. And by that standard, the US, while far from perfect, may still be a considerably more mobile society, properly defined.

Suppose that what you earn depends on things transmitted to you directly from your parents – intelligence, attitudes toward work, norms that lead to better and better-paying jobs, etc. – and on things that you get from the state – public education, public health care, etc. Suppose further that there are two societies. One - call it the America - is a complete meritocracy, in the sense that you are entirely paid on the basis of what you are worth in the market, but there is no attempt to adjust anyone's endowment of the second input. In the other - call it Europe - the state resdistributes the second input so that everyone enters the market with equal amounts of that input.

So which society has more opportunity? That depends on the nature of the production function. If the second input matters the most, Europe does. No matter which inupt matters, in Europe anyone can move up in the income distribution through state redistribution , while in America the distribution is cemented in time for eternity. But if the first input matters, America is the opportunity society. But if the second input is the most important, Europe equalizes the initial unequal distribution, and thus provides more opportunity. Since it doesn't matter in the market, those with a lot of the first input (those who earn the most in the market) find that most of the proceeds from having it are taken by the state. Poor European children still end up moving up in the distribution, but high-achieving European parents find that their children move down. So there is lots of mobility, but little growth, and little rewarding of the most productive, arguably a better definition of "opportunity." In America everyone can earn his full potential, but there is little mobility. If in Europe the high-achieving emigrate, European mediocrity, slow growth and American achievement at the high end all coincide. This interpretation seems to better fit the facts.

Numerous arguments have been put forth over the years to suggest that the US is in fact becoming a highly stratified society in terms of things that might affect market productivity - in other woerds, a society where earnings depend more and more on parental endowments (i.e. the first input). James Q. Wilson argued that the primary divide in contemporary America is between couples who marry and couples who don’t. Those who marry raise their children to get married, are able to invest more (by virtue of having two parents instead of one always on the scene) in their childrens' skills and attitudes, and their children go on to live a lifestyle substantially different from those who don’t, or those who get married but divorce quickly. Some years prior, Richard Herrnstein and Charles Murray notoriously contendedthat how intelligent you are is a major predictor of how much money you go on to earn, and that America is increasingly undergoing what they called “cognitive stratification,” the tendency of people to create children largely through union (married or not) with people of similar intelligence. The bottom line, then, is that "opportunity" does not necessarily equal "lots of shuffling among generations in the income distribution," nor does extensive income redistribution for the purposes of public investments necessarily improve "opportunity." That depends on the extent to which people are awarded purely by their talents, and also on the extent to which your talent depends on your parents’ talent. And these are very open questions.

It is a simple model, and leaves out a lot. Thomas Sowell has noted that changes in the overall age distribution of the population matters a lot too, and one could also argue that the huge influx of low-skilled immigrants, especially from Latin America, into the US, combined with increasing returns to the skill set possessed by people at the top of the income distribution almost unavoidably make intergenerational mobility slower. But the notion that whether a society is one of open opportunity or not should be judged by how likely it is that the children of someone in the low end of the income distribution ends up at a much higher percentile (or, equivalently, how likely it is that the children of someone at the top of the distribution end up at the low end) clearly leaves much to be desired.