Raj Chetty
About
Economist
Cast within
No topic-region cast yet — this appears once Raj Chetty's compiled claims are aligned into a topic region's argument tree.
Claims by Raj Chetty (20 of 57)
Absolute Mobility Fell From 90% to ~50%
Using a baseline of the standard CPI and total household income, the fraction of kids who out-earn their parents fell from about 90% for those born in the 1940s-50s to about 50% for the current generation—meaning roughly half now do worse than their parents—a decline summarized as the 'Fading American Dream.'
Low-Income Friendships More Geographically Bound
Higher-income people's friendships are spread out across the country, while lower-income people's friends tend to be physically close to where they live; this physical proximity is part of why very low-income communities tend to be economically disconnected—low-income people are mostly friends with those around them, who are also low-income.
Experiments Validate Observational Predictions
Progress in empirical economics is verifiable via replication and experiments: overlaying the randomized Moving to Opportunity experimental results on the predictions from the observational Opportunity Atlas work produces a scatter plot showing the two line up well, meaning the team could forecast how randomly-assigned movers would do on average, giving confidence the estimates are stable.
Sibling Age Comparison Identifies Causal Effects
By comparing siblings of different ages within the same family who move from one neighborhood to another (e.g. a 7-year-old vs an 11-year-old), under a 'dosage' model where each extra year of exposure to a better environment helps, researchers find younger movers to high-mobility areas do better; this design neutralizes the confound of ambitious parents, since parental ambition would not create differential effects within the same family across siblings.
Neighborhoods Are a Package of Institutions
Neighborhood effects reflect not a fixed physical characteristic of a place but the people living there and a package of social institutions—school quality, crime, culture, family structure—all of which shift together; this means neighborhood effects are not stable over time, as shown by Derenoncourt's AER work on how the Great Migration changed causal neighborhood effects for African Americans through disinvestment and segregation.
Rural Areas Often Have High Mobility
Contrary to the intuition that big cities offer the most opportunity, rural areas (e.g. rural Iowa, much of North Dakota, small towns) often show higher levels of upward mobility and economic connectedness—everyone goes to the same school and connects with everyone—though the kids who succeed often move to cities like Chicago or New York as adults; the patterns are fairly linear across the distribution rather than driven by outliers.
Theory Remains Essential in Big-Data Age
Contrary to the interpretation that his work means 'just look at the data,' theory remains crucial in three domains: hypothesis formulation (you cannot organize data without a theory—e.g. testing Loury's cross-class interaction theory versus tight-knit-community theories), extrapolation to new policies/environments (which requires a theory to translate past findings), and general-equilibrium analysis (since the partial-equilibrium treatment-control paradigm cannot handle price changes and re-sorting because the control group gets contaminated).
Neighborhood Explains Only 5-10% of Variation
'Zip code is destiny' is false: the R-squared or explanatory power of neighborhoods for individual outcomes is only about 5-10%, meaning there is large variation within neighborhoods and many other factors (family, education, individual inspiration) matter far more than where you grow up; neighborhood is just one significant level of variation, not a determinant of fate.
Connectedness Survives Horse-Race Controls
When economic connectedness (univariate correlation ~0.7 with mobility) is put in a horse race against poverty rates, inequality, segregation, and racial composition, its correlation with mobility remains very similar, and it actually explains away much of the previously found relationship between poverty rates and mobility—poorer neighborhoods have worse mobility largely insofar as they have less cross-class interaction.
Closing Connectedness Gap Raises Income 20%
If low-income kids grew up in neighborhoods with the level of economic connectedness that the average high-income kid experiences—closing the connectedness gap by moving them to comparable neighborhoods—their adult incomes would increase by about 20%; a sizable effect, comparable to two-to-three years of college at a 7-10% annual return to education, though not a silver bullet.
Economic Connectedness Predicts Mobility
Economic connectedness—the extent to which low-income and high-income people in an area are connected with each other (class-bridging social capital)—is a very strong correlational predictor of differences in economic mobility across zip codes; growing up where low- and high-income people interact more makes you more likely to rise up the income distribution.
Community-Level Not Individual-Level Finding
The connectedness finding is a community-level relationship, not an established individual-level causal claim; the study has not shown that if a given person makes more high-income friends their own prospects will change, because connected communities may differ in many other ways (institutions, behaviors) that also affect outcomes.
Facebook Data as Social Capital Proxy
Using privacy-protected Facebook data on 72 million people aged 25-44 (85% of that age population) with 21 billion friendships, the researchers can measure cross-class interaction, cohesiveness, and volunteering at the zip-code level; benchmarking against the Add Health Survey and General Social Survey shows nearly identical national/state-level patterns, validating the Facebook measures.
Broad Network Proxies Close Friends
Although intuition suggests only close friends should matter, restricting the analysis to people's 5 or 10 closest friends (measured by wall posts and messages) yields almost identical results to the full ~500-friend network, because a person's broad social net is representative of who their close friends are; the two are highly correlated.
College Attendance Mediates Connectedness Effect
Part of the 20% mobility gain from connectedness operates not through friendships per se but through behavior change: kids who grow up seeing lots of other kids attend college are more likely to attend college themselves, which changes their educational choices, the jobs they get, and downstream outcomes.
Replication Across Eight to Ten Studies
The dosage finding—that moving to better neighborhoods at younger ages yields better outcomes—has been replicated by eight to ten subsequent studies using different methods, including the randomized Moving to Opportunity housing-voucher experiment and Eric Chyn's AER quasi-experiment using exogenous public housing demolitions, all recovering the same age-exposure pattern.
Immigrants Outperform and Validate Neighborhood Effects
Immigrants on average have higher upward mobility than natives even controlling for other factors (showing zip code isn't destiny), and a large part of their mobility advantage is explained by the fact that immigrants move to neighborhoods that have high upward mobility even for natives; immigrant and native outcomes are highly correlated across areas, validating that the neighborhood package has common effects across groups.
Sixty Percent Causal, Forty Percent Selection
About 60% of the variation in outcomes across neighborhoods (at the Census tract level) is due to causal effects of place, and about 40% is due to selection—different types of people living in different places; so selection clearly matters, but there is an important causal component.
Couple Resources With Social Support
Policy traditionally focuses on incentives and resources (how big the voucher budget should be), but there is a missed opportunity to couple resources with social-capital/connection interventions—like a counselor who helps people navigate systems—which can have a much bigger impact than simply handing over a check, voucher, or job-training program, because 'figuring things out' is harder than just having the resources.
Seattle Counseling Quadruples Moves
In a randomized trial (Creating Moves to Opportunity) with Seattle/King County Housing Authorities, providing voucher families with a counselor to help navigate the housing market raised the share moving to high-upward-mobility neighborhoods from 15% (control) to 60% (treatment), at a cost of about $2,500 per family—suggesting segregation stems from removable barriers (information, navigation) rather than deep-rooted preferences.
My Notes
Loading notes...