At nearly 326,000, the number of new U.S. patents has more than doubled from 2005 to 2015. But in every year since 2008, the patents granted to foreign inventors have outpaced those of U.S. inventors, and a new study suggests the nation could be overlooking thousands of potential young inventors.
As part of the ongoing Equality in Opportunity Project, researchers from Harvard and Stanford Universities, the Massachusetts Institute of Technology, the London School of Economics, and the U.S. Treasury Department analyzed data from 1.2 million inventors, using patents issued from 1996 to 2014, linked to tax records, to trace the role of early grades and environment to their later innovation. The working paper was circulated by the National Bureau of Economic Research.
Nationwide, the researchers found income and academic talent played a role in how likely a student was to become a patent-holding inventor by 2014. For example, they found a little less than two of every 1,000 children born to the wealthiest 20 percent of families in New York City between 1979-85 had earned a patent by 2014, compared to 0.52 out of 1,000 children from families below the median income. And children of the wealthiest 1 percent of families were 10 times as likely to become inventors than children from lower-income families.
Moreover, among New York City public school students, 3.3 of every 1,000 Asian students and 1 in 1,000 white students grew up to be a patent-holding inventor by 2014. That rate was 0.5 for black students and only 0.2 for Hispanic students. (See chart.)
“If women, minorities, and children from low-income families were to invent at the same rate as white men from high-income families, the rate of innovation in the economy would quadruple,” the researchers noted.
Higher-income students on average had significantly higher math test scores than lower-income students, but differences in math performance in 3rd grade accounted for less than one-third of the gap in their inventing.
Communities Nurturing ‘Innovation’
Alexander Bell, a Harvard postdoctoral economist and the lead author on the study, said they found income and academic prowess alone didn’t fully account for the gaps in inventing among young people. Rather, communities with higher concentrations of adult inventors were disproportionately more likely to produce young people inventing in those fields.
“If you grow up in the Bay area, you are more likely to invent in computers; if you grow up near the Mayo Clinic, you are more likely to invent in drugs and medicine,” Bell said. “That speaks to me that it’s not a story of schools just being better in some areas—or it’s not just that. There is likely to be something about the social network, the internships, the mentorships, the role models of that area.”
For example, if a child with a promising math ability in 3rd grade moved from New Orleans—a city in the 25th percentile of innovation—to Austin, Texas, a city in the 75th percentile—it would increase her chance of becoming an inventor by 17 percent.
Moreover, the researchers found the concentration of female inventors significantly boosted the likelihood young girls would grow up to be inventors in the adults’ fields. For example, girls in Modesto, Calif. grew up in a community where more than 40 percent of inventors were women, while in Santa Rosa, Calif., women made up fewer than 10 percent of inventors.
“There’s a phrase thrown around a lot: ‘You can’t be what you can’t see,’” Bell said. “It’s hard for women and other minorities to be innovators if they don’t see people like them as role models.”
“I think the biggest lesson to draw from this is that the pipeline is leaky,” Bell said. “There are a lot of talented kids that don’t go on to fulfill their talents.”
The findings suggest schools could help boost communities’ capacity for innovation and economic growth not just by closing achievement gaps, but by identifying academically talented students from low-income and minority groups—what they dub the “lost Einsteins"—with mentors in the field and giving them more opportunities to form their own social networks of students interested in inventing.
Chart: A student’s likelihood of becoming an inventor varies based on family income and the concentration of other inventors in the area, according to a new study. Source: National Bureau of Economic Research
Related: