America’s Lost Einsteins

Millions of children from poor families who excel in math and science rarely live up to their potential—and that hurts everyone.

Saul Loeb / AFP / Getty

Consider two American children, one rich and one poor, both brilliant. The rich one is much more likely to become an inventor, creating products that help improve America’s quality of life. The poor child probably will not.

That’s the conclusion of a new study by the Equality of Opportunity project, a team of researchers led by the Stanford economist Raj Chetty. Chetty and his team look at who becomes inventors in the United States, a career path that can contribute to vast improvements in Americans’ standard of living. They find that children from families in the the top 1 percent of income distribution are 10 times as likely to have filed for a patent as those from below-median-income families, and that white children are three times as likely to have filed a patent as black children. This means, they say, that there could be millions of “lost Einsteins”—individuals who might have become inventors and changed the course of American life, had they grown up in different neighborhoods. “There are very large gaps in innovation by income, race, and gender,” Chetty told me. “These gaps don’t seem to be about differences in ability to innovate—they seem directly related to environment.”

The discrepancy in who gets patents is not the result of innate abilities, Chetty and his team, Alex Bell of Harvard, Xavier Jaravel of the London School of Economics, Neviana Petkova of the U.S. Treasury Department, and John Van Reenen of MIT, conclude. Children from many different backgrounds excel in math and science tests in third grade, for instance. But it’s the wealthy children who do well in math and science that end up getting patents. Why? Because they have more exposure to innovation in their childhood, the researchers say. This exposure comes mostly from interacting with people who are themselves inventors. If young kids know people who are inventors, or hear conversations at the dinner table about research and innovation, they’re more likely to become interested in pursuing careers in that field, Chetty told me. “Opportunity broadly, and exposure to innovation in particular, are really the keys to increasing innovation,” he said. Chetty, for instance, grew up in a family of academics, and overheard conversations about science and making discoveries, which, he says, influenced his decision to pursue a career in academia.


Inventors Are More Likely to Come From High-Income Families
Just one out of 1,000 children from families in the 30th percentile of income are likely to become inventors. By contrast, eight of 1,000 children from families in the 100th percentile of income are. (Steven Johnson / The Atlantic)

Aaron Hertzmann, who is now a principal scientist at Adobe Systems, has eight patents. He grew up in Palo Alto, where he had a computer before he was 10 years old. He had a lot of exposure to inventors as a child—his father was a “tinkerer,” Hertzmann told me, and his stepfather was an academic in the field of computer science. “He exposed me to approaching math as something to explore, rather than it just being a homework assignment,” Hertzmann said, about his stepfather. Hertzmann and his mother would tag along to his stepfather’s conferences in places like Greece, meeting other academics and hearing them talk about their work. From that, the idea bloomed in Hertzmann’s mind that academia, and specifically math and science, were areas where he could have an impact. When he graduated from college, his stepfather guided him along the process of applying to study for a Ph.D. Hertzmann, now 43, has a Ph.D. in computer science, and develops new ideas and algorithms at the intersection of art and computer science.

Indeed, exposure to certain specific fields makes children more likely to pursue a career, and a patent, in those fields, the researchers found. This is how they know that exposure, in addition to neighborhoods, is important to innovation: It would be unlikely that growing up in a good neighborhood would inspire many children to patent in the same small field. People who grew up in Minneapolis, where there are many medical-device manufacturers, were especially likely to get patents in medical devices, for instance. Among people living in Boston as adults, those who grew up in Silicon Valley were especially likely to patent in computers. Children whose parents have patents in a specific field—say, antennas—are also likely to patent in exactly the same field as their parents did.

Women who grew up in an area where women held a higher share of patents in a certain field were more likely to themselves get patents in that area when they grew up. Strikingly, it was especially important for children to see people who looked like them as innovators for them to pursue the same career path—girls in an area with a lot of male innovators wouldn’t necessarily envision themselves in the same career, while boys would. If girls were as exposed to female inventors as boys are to male inventors, the gender gap between male and female inventors would fall by half, the researchers estimate. (They find that 82 percent of 40-year-old inventors today are male.)

These findings have big implications for the state of the U.S. economy, which has seen innovation decline in recent decades. Innovation is often measured by what’s known as “total factor productivity,” which essentially tracks advances that have been made in using existing resources to increase output. Prior to 1973, total factor productivity increased at an annual rate of 1.9 percent—but since then, that growth rate has fallen to 0.7 percent, according to the Brookings Institution. Innovation is central to economic growth, Chetty says. About half of U.S. annual GDP growth is attributed to innovation. Innovation, says Chetty, is what allows people to live richer, healthier, and more productive lives.

Much of the past work out of the Equality of Opportunity project has been motivated by ideas about justice, and the idea that everyone, regardless of where they are raised, should have a fair shot at the American Dream, Chetty told me. But these results indicate that equality of opportunity is important for another reason too: It makes the economy stronger. “Opportunity might be vital for economic growth even if you don’t care about inequality or fairness concerns,” Chetty said. “If you give kids from lower-income families better training and better opportunities, maybe they would end up contributing more to the economy and that would help everyone essentially.” Chetty and his team estimate that if women, minorities, and children from low- and middle-income families invented at the same rate as white men from high-income families, there would be four times as many inventors in America as there are today.

Chetty and his team came up with these results by linking patent applications in the U.S. between 1996 and 2014 to federal income tax returns to create a dataset of 1.2 million inventors. (The study uses patents as a way to measure an individual’s contribution to innovation.) They tracked inventors’ lives from birth to adulthood to determine who becomes an inventor. They then linked data on math test scores from third to eighth grade from children who attended New York City public schools to see if the differences in who got a patent could be related to innate ability (they aren’t). They then showed that children who grew up in commuting zones with higher patent rates are “significantly” more likely to become inventors than children who did not. They also showed that children from both low-income and high-income families who attend universities like MIT go on to patent at relatively similar rates, suggesting that it’s factors in a child’s earlier life that determine whether they go on to patent.

This follows on earlier research from the Equality of Opportunity project that shows that growing up in an impoverished area can hurt a child’s chances of achieving many of the pieces of the American Dream. Living in certain neighborhoods makes it less likely that a child will attend college, that they’ll earn more than their parents did, and that they’ll postpone having children until they marry.

This paper’s results suggest that policies that increase exposure to innovation childhood could go a long way in stimulating economic growth. Internships or mentorship programs could link children interested in math and science with innovators, for example, which might make them more likely to pursue careers in that field. Integration could also help—if children have exposure to more types of people, the people they think of as their peer group changes, and they might be more likely to pursue a career that is dominated by people who don’t look like them. That will help them succeed individually, and it could have a positive effect on the economy as a whole.

Alana Semuels is a former staff writer at The Atlantic.