Harvard economist Raj Chetty explains how insights from big data can be used to restore opportunity and upward mobility.
As a father of two young children, I like to think that my wife and I will be the most central factor in shaping our children’s future.
But my research tells a more complex story. Our team’s data shows that there are a broader set of individuals and institutions in our neighborhood who shape our children’s life outcomes. ZIP codes may not be destiny, but where you start out and whom you grow up with matters tremendously.
Consider a child growing up in a low-income family in Roxbury, a historically Black community in Boston, Massachusetts. On average, children who grow up in Roxbury earn little more than their parents did as adults. They are also less likely to be employed and more likely to be incarcerated than a Black child growing up in a similarly low-income family in Cambridge, only a few miles from Roxbury, near my office at Harvard University.
Here’s where things get really interesting: our research shows that if a low-income family moves from Roxbury to Cambridge their economic prospects shift — dramatically. Their children’s trajectory starts to look more like that of the children around them in their new neighborhood. And the younger the children are when they move, the more completely their life chances transform to match those of their new surroundings.
As researchers, policymakers, and social investors continue their efforts to address recent trends toward greater inequality and a reduction in economic mobility in the U.S., these findings about the local roots of economic opportunity provide new insights into how we can make smarter, more impactful investments to improve outcomes at scale.
Just over half a century ago, climbing up the economic ladder in the U.S. was a near certainty, with virtually all kids going on to earn more than their parents did. Today, it is a coin toss. Children starting their careers today have only a 50-50 shot of doing better than their parents. But restoring the American dream doesn’t require traveling back to the halcyon days of the 1940s or ’50s. Instead, helping the U.S. live up to its “land of opportunity” moniker is about understanding what promotes economic mobility in some communities and limits it in others.
For the past decade, my colleagues and I have harnessed big data to do exactly that — studying, with precision, the science of opportunity.
What we have learned about fostering equal opportunity and upward mobility is meaningful not only for the U.S., but for communities all over the world. Our findings point to powerful, data-based levers for strategic social investors.
One tool we’ve built is the Opportunity Atlas. Developed in partnership with the U.S. Census Bureau using anonymized data from tax records following 20 million children over their lives, this interactive mapping tool shows children’s chances of climbing the income ladder for every one of the nearly 80,000 neighborhoods across America.
The Opportunity Atlas provides sharp insights into the roots of economic opportunity, demonstrating that the country is a patchwork of low-opportunity and high-opportunity neighborhoods side by side. This patchwork exists in every city and in every region. The Opportunity Atlas is also inspiring similar work across the world. Researchers in India, Spain, and other countries are now using similar methodology and approaches to map economic mobility and understand what underpins opportunity in other countries.
Now that we’ve recognized this patchwork, the question becomes: what can we do about it?
One approach is to help willing low-income families move to high-opportunity neighborhoods.
The U.S. federal government spends US$ 21 billion each year on housing vouchers for low-income families. Using our Opportunity Atlas, we determined that most of the two million low-income families who receive this rental assistance use it to pay for housing in neighborhoods that currently do not offer great prospects for upward mobility.
For example, families use the rental assistance to stay in or move to neighborhoods such as Roxbury, rather than move to nearby, affordable areas such as East Cambridge where their children would have better chances of rising up. Often, families stay in neighborhoods like Roxbury simply because that is where they currently live or where they know of available housing. But remaining in or moving to places like Roxbury is not the best use of government funding, nor is it in the long-term interests of these children.
We leveraged data from our Opportunity Atlas to improve the impact of the housing voucher program. We designed and tested a pilot program that helps families in Seattle with a housing voucher find housing in neighborhoods where their children have a better shot at upward mobility.
Through a randomized trial, we found that this pilot program dramatically increased the number of families who chose to move to high-opportunity neighborhoods. We estimate that children in families who move to these higher-opportunity areas will earn US$ 200,000 more in their lifetimes. Early success from this pilot in Seattle led to bipartisan support and the passage of a US$ 60m bill to conduct similar pilots in other cities across the U.S., with discussions now underway for an additional US$ 5 billion of investment in vouchers per year.
Of course, helping families move is not a scalable or
sustainable solution to improve upward mobility for everyone. Families ultimately need to have access to opportunity in every community. Our research highlights
four major factors shared by high-opportunity neighborhoods:
Some of these factors, such as social capital and family structure in a community, are difficult to change directly through policy. But we can design programs that could serve their likely function like the place-based program, Year Up, which provides highly effective career mentorship and Harlem Children’s Zone, which provides high-quality early education and mentorship programs. These and similar organizations launch children on a trajectory above and beyond their parents’ position on the economic ladder in neighborhoods where such an ascent has traditionally been less likely.
Other factors, such as education and poverty rates, can be directly influenced by policy, and again, big data can shed light on which investments can be most effective.
Schools are a key agent for upward mobility.
Following millions of students from kindergarten to adulthood, we find that better teachers and smaller class sizes, starting in kindergarten, yield large gains decades later. Such investments improved college attendance rates, reduced teen pregnancy rates, and produced higher earnings.
Good teachers matter. Students taught for a single year by a great teacher in elementary school, instead of an average teacher, earn US$ 50,000 more over the course of their careers. This adds up to an increase in future earnings of US$ 1.4m for an entire class taught by a high-quality teacher.
Supporting access to high-quality higher education can also be a powerful pathway for upward mobility. A few schools, including the City University of New York system and schools in the California State University system, stand out as high-mobility institutions where many students rise from the bottom to the top of the income distribution. Understanding how to create more such engines of mobility — by increasing the number of low-income students at elite colleges that deliver great outcomes and improving outcomes at community colleges that serve the most low-income students — should be a priority for philanthropy in the higher education space.
Our data shows that educational investments, while very valuable, must be complemented by other interventions to deliver equal opportunity to all.
Racial disparities cast a long shadow on the future of children. Black and white boys who grow up on the same city block and attend the same school still have very different outcomes in adulthood. Why? Our data provides some hints. The Black-white gaps are smallest in areas with low levels of racial bias among whites. Our Opportunity Atlas is often a deep crimson (signaling a lack of opportunity) in Southern states with a history of deep racial bias.
But we don’t have to rewrite history to improve opportunity for Black Americans.
Our data also shows that childhood exposure to neighborhoods with less economic and racial segregation and a larger share of Black fathers improves outcomes for Black children and produces smaller gaps in the next generation. These findings suggest that mentorship programs such as My Brother’s Keeper Alliance may be able to help close the racial gap.
Equality of opportunity is often framed as an issue of justice.
But it is also an economic issue.
Consider the role of innovation, widely viewed as a major driver of economic growth. Looking at data on a million American patent holders linked to tax data over a 20-year period, we found that it takes a village to create an Albert Einstein or a Steve Jobs.
Innovators aren’t lone wolves working out of their garage. Rather, like my children, they start off as sponges. They are inspired by a network of role models, supported by the community around them, and financed by families or institutions with means.
To take one example of how this influence works, we find that girls are more likely to become inventors if they are exposed to female inventors (but not male inventors) while growing up. And the type of inventor a person grows up to become is shaped by the types of innovation they happen to be exposed to while growing up. Silicon Valley produces a self-reinforcing cycle of innovation in technology. Unfortunately, neighborhoods like Roxbury also produce a self-reinforcing cycle, where its compounding disadvantages make it harder for its residents to rise up.
The data shows we can do something about upward mobility. Every extra year of childhood spent in a better neighborhood seems to matter.
By our calculations, the U.S. is missing 3.6 million innovators because women, people of color, and low-income people are not inventing at the same rate as white men who grow up in high-income families. Closing this gap, with programs that identify, mentor, and support these “lost Einsteins” is not only about fairness, it is vital for economic growth. If women, minorities, and children from low-income families were to invent at the same rate as white men from high-income families, the number of inventors in America would quadruple.
As parents, it can sometimes be difficult to accept the tremendous role others will have in shaping our children’s future. But for social investors, these findings about the role of community present us with a gift — powerful levers that can be used to democratize opportunity in every neighborhood.