About 90 percent of American children born in 1940 grew up to earn more than their parents did at the same age. For children born in the 1980s, that share dropped to roughly 50 percent. That is the finding at the centre of The Fading American Dream, published in December 2016 by a team led by the economist Raj Chetty.
The number makes a slippery idea concrete. If the American dream means each generation doing better than the last, the odds of that happening have gone from near-certain to a coin flip. As Chetty put it when the study came out, “It’s basically a coin flip as to whether you’ll do better than your parents.”
What the number actually measures
The measure is simple: the share of children who out-earn their parents at the same age. Both incomes are adjusted for inflation, so the comparison is real and not just a side effect of rising prices. A summary by the Washington Center for Equitable Growth notes that only 50 percent of children born in 1984 cleared that bar. For those born in 1940, it was close to nine in ten.
The number is narrower than it can sound. It does not say that half of today’s young adults are poor, or that the economy stopped growing. It says something specific: how you compare with your own parents at the same stage of life. The biggest declines showed up in the middle class, which is where the phrase “American dream” tends to live in the first place.
The standard explanation, and the study’s more precise verdict
The intuitive story is that opportunity has become more unequal. The economy grew, but the gains piled up at the top. So the typical child had less chance of clearing a bar set by a parent who lived through a more evenly shared boom. The Chetty team leans this way. They ran what-if simulations and concluded that the uneven sharing of growth mattered more than the slowdown in growth itself.
The simulations are specific. Keep today’s slower growth but share it as evenly as the 1940s did, and mobility climbs back to about 80 percent, reversing more than two-thirds of the fall. Keep today’s uneven sharing but bring back the fast growth of the 1940s and 50s, and mobility rises only to 62 percent. On those numbers, the sharing problem dwarfs the growth problem. Chetty argued that reviving the dream would likely require “policies that foster more broadly shared growth.”
The mirror-image reanalysis
That conclusion about cause is where the agreement ends. In a 2022 working paper, Scott Winship of the American Enterprise Institute reproduced the headline drop, from about 91.5 percent for the 1940 group to 50 percent for the 1980 group. He accepts the number. But when he rebuilds the assumptions behind the simulations, he gets the opposite answer.
In Winship’s version, faster growth would raise mobility from 50 percent to 81 percent, while lower inequality would lift it only to 57 percent. He describes this as “almost the mirror image of the Chetty findings.” His conclusion is blunt: the results, he writes, “deeply undermine the conclusion that falling absolute mobility is primarily due to rising income inequality.”
This is one scholar’s contested reanalysis, not a settled overturning. Brookings, writing when the Chetty paper first appeared, noted that earlier estimates of absolute mobility had ranged from about 65 to 85 percent, largely because of data limitations and sensitivity to statistical assumptions. The exact figure depends on choices about which inflation measure to use, how to define income, and whether you account for household size. The 50 percent number is real. It also sits among a set of dials, and turning them changes the story about why.
Why the coin toss matters more than the argument over it
The data leaves a tension on the table. Everyone in the debate agrees the odds have collapsed. Chetty and Winship disagree, sharply, about whether that collapse is mainly a growth problem or mainly a sharing problem. That disagreement is not merely academic, because the two diagnoses point to almost opposite fixes.
If the fall is about sharing, the lever is how the gains from growth get divided up. If it is about growth, the lever is the size of the pie itself. The Chetty team’s own modelling shows how heavy that lever is: under today’s distribution, the economy would need to grow above 6 percent a year to get mobility back to 1940s levels, a rate the postwar United States rarely hit even at its peak.
The near-guarantee of out-earning one’s parents no longer holds; the evidence on that point is not in serious dispute. Why it no longer holds, and what could shift it back, is where the evidence runs out and the disagreement begins.
