The Great Gatsby Curve: Not So Great After All

by J. D. Vance

In my recent essay in the September 2 issue of National Review, I argue that the connection between economic growth and economic opportunity is less direct than many conservatives presume. Many on the left make a similar mistake with regard to inequality — there’s little evidence that inequality causes economic immobility. Yet despite that lack of evidence, this myth refuses to die.

The best evidence linking inequality per se with stagnating upward mobility is the “Great Gatsby” curve, popularized by Professor Alan Krueger in a speech last year. The chart shows a very direct correlation between the level of inequality in a country and inelasticity of incomes:

Intergenerational earnings elasticity is an excellent measure of economic mobility. At the low end, a value of .2 (about where Finland is) indicates that every $100 increase in a father’s income predicts a $20 increase in the son’s. An earnings elasticity of 1 would show that a son’s earnings are completely predicted by the father’s – this would be the polar opposite of a classless society. The fact that the United States scores so high is cause for concern by itself. I’d rather not live in a society in which so much of a child’s life is tied to his parent’s identity. I suspect that this view is widely shared across the political spectrum.

That said, I’ve always doubted the utility of Krueger’s Gatsby curve. First, as Scott Winship noted in a piece in this space, the relationship Krueger documents depends on studies with widely varying methodologies. Moreover, it’s always struck me as a bit of an apples-to-oranges comparison: Norway is a different country with different demographics and a different culture. The Gatsby chart, with one fell swoop, reduces those differences to a single liberal talking point. That’s a reasonable tactic if you want to score political points; it’s virtually useless if you want to understand what drives economic immobility (which, again, is too high).

Nevertheless, the chart has received a lot of airplay. The White House highlighted Krueger’s work and what it allegedly proves on the official presidential website. Paul Krugman wrote on his blog that the curve showed “a clear negative relationship between inequality at a point in time and intergenerational social mobility.” It even has its own Wikipedia page.

I always wondered what an apples-to-apples comparison might look like – whether, for example, inequality in U.S. cities is related to economic mobility in those cities. But until recently, we didn’t have the data to make that comparison. Thanks to the Equality of Opportunity study by Professors Raj Chetty and Emmanuel Saez, we now do.

The study’s authors measured economic mobility in over 750 “commuting zones” across the country. These CZs map closely onto Census-defined metropolitan areas, with slight variations. So I decided to make a chart of my own comparing the absolute upward mobility in a given CZ with the inequality in the metropolitan area. I’ll admit that after the chattering classes erupted over the Krueger chart, I expected at least some relationship between inequality and upward mobility. So what I found was striking:

In short, in our 48 largest metro areas, there is no meaningful relationship between inequality and upward mobility.

To explain the chart: There are 48 dots, one for each of the largest metropolitan areas in the United States. The numbers on the bottom, .38 to .52, measure inequality — the higher the number, the higher the inequality in a given area. The numbers on the left measure absolute mobility – the expected income percentile of a child born poor in those cities. Higher numbers mean greater upward mobility. As you can see, as inequality increases, nothing really happens. There are cities with high inequality and high mobility, low inequality and low mobility, and everything in between.

If Krueger’s chart is the Great Gatsby curve, then this is the small Gatsby curve. But the small Gatsby curve has the advantage of comparing things that are actually somewhat alike – specifically, large, densely populated urban areas. And when you do that, the “strong, negative relationship” between upward mobility and inequality disappears.

Let me emphasize that this doesn’t mean we should ignore upward mobility or the fact that America could use a lot more of it. It does, however, show that the Left’s obsession with inequality is a distraction from the real issue. Inequality is an inevitable consequence of living in a society with people of different aptitudes and ambitions. Upward mobility, on the other hand, is about the American Dream. And the connection between the two is tenuous at best.

This gives conservatives an incredible opportunity in both policy and politics. Voters on the bottom half of the income ladder voted overwhelmingly for the Democrats in 2012. To do better, we need to offer something that speaks to those voters’ concerns. If we don’t, I suspect they’ll continue voting for bad ideas over no ideas. 

The Agenda

NRO’s domestic-policy blog, by Reihan Salam.