The fall-off is particularly steep among children born in the middle class, and for those born in several states in the industrial Midwest.
If policymakers care about giving more Americans a chance to out-earn their parents, Chetty said in an interview, "these results show, you have to care about inequality.”
Previously, Chetty's team studied a different measure of mobility: the ability of children to move up or down America's income ladder as they grow up, when compared to other Americans. The new research attempts, for the first time, to quantify so-called "absolute mobility," which people often associate with the American Dream: the odds of a child earning more as an adult than his or her parents earned at the same age.
The researchers say rising concentration of income among the richest Americans explains 70 percent of what has been a steady decline in absolute mobility from the baby boom generation to millennials, while a slowdown in economic growth explains just 30 percent.
That sort of mobility "is something that was a feature of the American economy for kids born around 1940, that baby boomer generation," said Nathaniel Hendren, a Harvard economist who is one of the authors of the study. "As we look forward, there’s just been a dramatic decline in that measure" - which, he said, runs parallel to a decades-long stagnation in earnings for typical American workers, even as earnings for the very rich have soared.
"If you don’t have that kind of widespread economic growth across the income distribution, it’s tough to grow up and earn more than your parents,” Hendren said. “This is a distinct reason to focus on inequality."
That finding suggests Trump's tax-cut plans would do little to improve economic mobility for struggling blue-collar families, even if they help accelerate growth, unless they are somehow able to deliver growth that is much more broadly shared by workers of all income levels. Independent analysts, though, predict the tax plans pushed by Trump in the campaign, and by Republicans in Congress, would benefit high earners disproportionately.
Trump did not make inequality a focus of his presidential campaign, but he appealed directly to the anxieties of workers who feel the economy is not delivering for them in the same way it did for their parents. The paper highlights several sources of those anxieties in the Trump coalition.
It finds barely 2 in 5 men born in the mid-80s grew up to earn as much, at age 30, as their father's did at the same age. It shows average rates of mobility falling particularly fast in Rust Belt states, most notably Michigan and Indiana. And it finds a much steeper drop in absolute mobility for the middle class than for the poor. (Other research suggests what young people earn early in their careers is a good indication of how much their earnings might grow over their lifetimes.)
The calculations come from a novel approach by the authors to marry data from anonymous tax records and detailed statistics on the distribution of income in the United States from the Census. Using those sources, the authors estimate the chances that a particular child will grow up to earn more than her or his parents.
“We’ve never had a good way of measuring" those changing odds for children to out-earn their parents, said Sean Reardon, an education professor at Stanford, who was not part of the Chetty team but reviewed the study. "This really quantifies it in a very elegant and thoughtful way, and an important way.”
The large decline in the past half-century remains steady even if they change how they calculate inflation over time, or take into account demographic shifts such as the changing size of U.S. households over time, or compare incomes across generations at age 40 instead of age 30. If added together, those tweaks could raise the odds of a child born in the 80s out-earning his or her parents to at least 2 in 3 - though that would still be substantially less than the odds for a child born in 1940.
"The paper provides really strong evidence that absolute mobility has fallen over time. It’s too pessimistic, though, to say that the American Dream is 'fading'," said Scott Winship, a visiting fellow who studies mobility at the Foundation for Research on Economic Opportunity, a think tank that studies market-based solutions for the poor and middle class. "Intergenerational advancement continues apace, but at a diminished rate."
The surge in inequality over the past half-century is well documented. New research this week, from economists Thomas Piketty, Emmanuel Saez and Gabriel Zucman, shows the bottom 50 percent of U.S. income earners only gained 1 percent in earnings from 1962 to 2014, after adjusting for inflation. From 1980 to 2014, they found, nearly 70 percent of income gains went to the top 10 percent.
Even knowing the data on rising inequality, Chetty said, the mobility researchers were surprised by their new findings.
“We thought mobility would be lower," he said. "We didn’t realize how much lower it would be. And we didn’t realize that the distribution of growth would matter more than the level of GDP growth.”
Other research supports Chetty, Hendren and their colleagues, who also include Stanford sociologists David Grusky and Maximilian Hell, Harvard sociologist Robert Manduca and Berkeley economist Jimmy Narang. Brown University economist Nathaniel Hilger, for example, has found a similar decline over time in absolute educational mobility - which is to say, the odds that an American child will obtain more education than his or her parents did.
Based on his work and the Chetty team's work, Hilger said, "I do believe that kids born after the 1930s are doing worse relative to their parents than kids born 1900-1930, and kids born in 1970 are not typically doing better than their parents in terms of educational attainment or income. It's closer to a coin flip than it used to be earlier in the century--now some kids do better than parents, and some kids do worse."