We use data on 44,000 twin pairs observed as adolescents in the 2000 Census, linked to their earnings 20 y later, to study the heritability of labor market outcomes. We extend the Classical Twins Design (CTD), which identifies heritability from contrasts between monozygotic and dizygotic twins, to settings where an analyst does not observe zygosity but can measure outcomes for same-sex and opposite-sex twins. We also allow for the presence of a family- and sex-specific component affecting same-sex siblings but only partly shared by opposite-sex siblings. Our extended model identifies heritability from the difference-in-differences of same-sex vs. opposite-sex twins vs. siblings. The estimates indicate strong heritability (h-sq = 0.36) of log earnings. Using an AKM [J. M. Abowd et al., Econometrica 67, 251-333 (1999)] decomposition that separates the person-specific component of earnings from employer-specific pay premiums, we find that both components are heritable. The heritability of employer pay premiums, however, is mainly driven by same-sex twins who work at the same firm, suggesting that social interactions may lead to an overestimate of the genetic component in earnings. Excluding siblings who work together leads to estimates of the heritability of log earnings that are 15% lower. Similar biases may be present in CTD-based estimates of heritability for other social outcomes like education.

American colleges and universities are highly stratified by pre-college academic achievement, family background, and institutional resources. We study the meritocratic consensus in American higher education: colleges that high-testing students (who are generally also from high-income families) attend spend dramatically more on instruction than do those that enroll lower-testing students. Stratification by test scores has been largely stable since the 1960s, but the stratification of instructional resources has risen sharply since 1970 at both private and public institutions. Non-academic admissions criteria like athletics, legacy, and affirmative action are second-order in determining the allocation of students to universities. Potential economic justifications for the positive association of instructional expenditures with student prior achievement—q-complementarity between achievement and resources, convex social returns to high human capital, and incentives to invest in learning prior to college—have little empirical support. Resource stratification across universities has not increased in the past decade, largely due to increased public funding of universities that enroll lower-testing students through financial aid programs like California's CalGrant, but stratification within institutions is now rising swiftly. Institutional subscribers to the NBER working paper series, and residents of developing countries may download this paper without additional charge at www.nber.org.

We use linked employer-employee data to study the causal effects of location on earnings in the United States. We estimate a model with employer and employee effects, then aggregate to the commuting zone ( CZ ) level. Sorting across firms biases traditional "movers" designs. Our model accurately predicts earnings changes for CZ movers after accounting for firm sorting. Worker skills explain half of observed earnings differences across CZs; observable characteristics understate this. Industry composition explains little of average place effects. Costs at least offset CZ earnings premia on average; workers who move to higher-wage CZs have equal or lower real consumption. ( JEL J24, J31, R23, R32)

Behaviorally informed "nudges" are widely used in government outreach but are often seen as too modest to address poverty at scale. In four field experiments over 2 y (n = 542,804 low-income households), we test whether more proactive communication, varying message framing, and more precise targeting can boost take-up of tax-based benefits in California above and beyond traditional light-touch approaches. Our interventions focused on extremely vulnerable households, most with no prior-year earnings, who were at risk of missing out on two crucial benefits: the 2021 expanded Child Tax Credit and pandemic-relief Economic Impact Payments. Light-touch outreach consistently increased take-up of these benefits by 0.14 to 2 percentage points-a 150% to over 500% relative increase-regardless of message, sample, timing, or modality. These light-touch approaches resulted in over $4 million disbursed, with a highly cost-effective return of $50 to over $8,000 per $1 spent. However, higher-touch proactive outreach, varying messaging, and more precise targeting yielded minimal additional benefits, with proactive outreach even showing negative returns. These findings demonstrate that light-touch outreach can effectively shift behavior among very vulnerable households in contexts with reduced compliance burdens, but also underscore an urgent need to rethink the role of higher-touch strategies in closing take-up gaps in social safety net programs.

Rising interest rates can create “mortgage rate lock” for homeowners with fixed rate mortgages, who can hold onto their low rates as long as they stay in their homes but would have to take on new mortgages with higher rates if they moved. We show mobility rates fell in 2022 and 2023 for homeowners with mortgages, as market rates rose. We observe both absolute declines and declines relative to homeowners without mortgages, who are unaffected by mortgage rate lock. Mobility declines are not explained by changes in home values. Overall, our estimates imply that rising interest rates reduced mobility in 2022 and 2023 for households with mortgages by 16% and caused $20bn of deadweight loss.

We use linked employer–employee data to study the causal effects of location on earnings in the United States. We estimate a model with employer and employee effects, then aggregate to the commuting zone (CZ) level. Sorting across firms biases traditional “movers” designs. Our model accurately predicts earnings changes for CZ movers after accounting for firm sorting. Worker skills explain half of observed earnings differences across CZs; observable characteristics understate this. Industry composition explains little of average place effects. Costs at least offset CZ earnings premia on average; workers who move to higher-wage CZs have equal or lower real consumption. (JEL J24, J31, R23, R32)

Since their inception in 1989, the U.S. News & World Report law school rankings have influenced how schools, students, and the legal profession itself think about legal education. In the Fall of 2022, however, several of the most selective law schools formally withdrew from the annual rankings. In so doing, these schools laid bare longstanding criticisms of the rankings' questionable criteria and opaque methodology. While the long-term effect of this boycott remains to be seen, school rankings are likely here to stay. In this Article we design a more informative approach to rankings, based on actual decisions students make. Using individual-level data provided by the Law School Admissions Council (LSAC), we analyze the universe of applicants to U.S. law schools for the period 1988 through 2017. In so doing, we are the first to create a revealed preference ranking based solely on where applicants matriculate given offers of admission. Our approach relies neither on potentially faulty data collection from schools nor arbitrary decisions about which factors to emphasize in rankings, thereby minimizing the scope for manipulation. It also allows us to quantify the magnitude of differences in preferences among schools and to test their statistical significance. Matriculants reveal a strong preference for a handful of the most selective schools; outside of the top tier, however, matriculants do not appear to draw meaningful distinctions between schools ranked adjacently or even near to each other. While existing school rankings sow more confusion than clarity, our analysis provides a rigorous and transparent alternative, and a blueprint for redesigning school rankings.

Using Longitudinal Employer-Household Dynamics data, we demonstrate several facts that are not consistent with the “spatial mismatch” hypothesis that residential segregation and uneven distribution of jobs limit Black workers' opportunities. We show that (a) there is no Black-White gap in the firm premium component of wages in an Abowd-Kramarz-Margolis wage decomposition; (b) there are both more jobs and more good jobs within commuting distance of Black than White workers; and (c) Black workers' commutes are shorter. We conclude that geographic proximity to good jobs is not a major source of racial earnings gaps in major US cities today.

As US college costs continue to rise, governments and institutions have quadrupled financial aid. Yet, the administrative process of receiving financial aid remains complex, raising costs for families and deterring students from enrolling. In two large-scale field experiments (N = 265,570), we test the impact of nudging high-school seniors in California to register for state scholarships. We find that simplifying communication and affirming belonging each significantly increase registrations, by 9% and 11%, respectively. Yet, these nudges do not impact the final step of the financial aid process – receiving the scholarship. In contrast, a simplified letter that affirms belonging while also making comparable cost calculations more salient significantly impacts college choice, increasing enrollment in the lowest net cost option by 10.4%. Our findings suggest that different nudges are likely to address different types of administrative burdens, and their combination may be the most effective way to shift educational outcomes.

We revisit the estimation of industry wage differentials using linked employer-employee data. Cross-sectional industry differences overstate pay premiums due to unmeasured heterogeneity. Estimates based on models with person and industry effects understate true premiums: workers who switch to a higher-premium industry typically move from higher-paying firms in their origin industry to lower-paying firms in their destination (and vice versa). The corrected standard deviation of log wage effects is 0.122 across narrowly defined industries and is similar at higher levels of aggregation. Higher-skilled workers sort to higher-pay industries. Premiums and worker sorting are more variable in cities with higher-wage firms and higher-skilled workers.

We use data from the Longitudinal Employer-Household Dynamics program to study the causal effects of location on earnings.Starting from a model with employer and employee fixed effects, we estimate the average earnings premiums associated with jobs in different commuting zones (CZs) and different CZ-industry pairs.About half of the variation in mean wages across CZs is attributable to differences in worker ability (as measured by their fixed effects); the other half is attributable to place effects.We show that the place effects from a richly specified cross sectional wage model overstate the causal effects of place (due to unobserved worker ability), while those from a model that simply adds person fixed effects understate the causal effects (due to unobserved heterogeneity in the premiums paid by different firms in the same CZ).Local industry agglomerations are associated with higher wages, but overall differences in industry composition and in CZ-specific returns to industries explain only a small fraction of average place effects.Estimating separate place effects for college and non-college workers, we find that the college wage gap is bigger in larger and higher-wage places, but that two-thirds of this variation is attributable to differences in the relative skills of the two groups in different places.Most of the remaining variation reflects the enhanced sorting of more educated workers to higher-paying industries in larger and higher-wage CZs.Finally, we find that local housing costs at least fully offset local pay premiums, implying that workers who move to larger CZs have no higher net-of-housing consumption.

We use data from the Longitudinal Employer-Household Dynamics program to study the causal effects of location on earnings. Starting from a model with employer and employee fixed effects, we estimate the average earnings premiums associated with jobs in different commuting zones (CZs) and different CZ-industry pairs. About half of the variation in mean wages across CZs is attributable to differences in worker ability (as measured by their fixed effects); the other half is attributable to place effects. We show that the place effects from a richly specified cross sectional wage model overstate the causal effects of place (due to unobserved worker ability), while those from a model that simply adds person fixed effects understate the causal effects (due to unobserved heterogeneity in the premiums paid by different firms in the same CZ). Local industry agglomerations are associated with higher wages, but overall differences in industry composition and in CZ-specific returns to industries explain only a small fraction of average place effects. Estimating separate place effects for college and non-college workers, we find that the college wage gap is bigger in larger and higher-wage places, but that two-thirds of this variation is attributable to differences in the relative skills of the two groups in different places. Most of the remaining variation reflects the enhanced sorting of more educated workers to higher-paying industries in larger and higher-wage CZs. Finally, we find that local housing costs at least fully offset local pay premiums, implying that workers who move to larger CZs have no higher net-of-housing consumption.Institutional subscribers to the NBER working paper series, and residents of developing countries may download this paper without additional charge at www.nber.org.

We use the introduction of the Texas Top Ten Percent rule to estimate the effect of access to a selective college on graduation and earnings outcomes for two groups of students. For highly ranked students at more disadvantaged high schools, who gained access under the policy, college enrollment and graduation increased. Less highly ranked students at more advantaged schools, who tended to lose access, shifted toward less-selective colleges under the policy, but did not see declines in overall college enrollment, graduation, or earnings. The policy thus benefited students targeted for admission without evidence of adverse effects on displaced students. (JEL I21, I23, I24, I26)

In a pilot program during the 2016-17 admissions cycle, the University of California, Berkeley invited many applicants for freshman admission to submit letters of recommendation. This proved controversial within the university, with concerns that this change would further disadvantage applicants from disadvantaged groups. To inform this debate, we use this pilot as the basis for an observational study of the impact of submitting letters of recommendation on subsequent admission, with the goal of estimating how impacts vary across predefined subgroups. Understanding this variation is challenging in an observational setting because estimated impacts reflect both actual treatment effect variation and differences in covariate balance across groups. To address this, we develop balancing weights that directly optimize for "local balance" within subgroups while maintaining global covariate balance between treated and control units. Applying this approach to the UC Berkeley pilot study yields excellent local and global balance, unlike more traditional weighting methods, which fail to balance covariates within subgroups. We find that the impact of letters of recommendation increases with applicant strength. However, we find little average difference for applicants from disadvantaged groups, although this result is more mixed. In the end we conclude that soliciting letters of recommendation from a broader pool of applicants would not meaningfully change the composition of admitted undergraduates.

The Earned Income Tax Credit distributes more than $60 billion to over 20 million low-income families annually. Nevertheless, an estimated one-fifth of eligible households do not claim it. We ran six preregistered, large-scale field experiments with 1 million obser-vations to test whether "nudges" could increase EITC take-up. Despite varying the content, design, messenger, and mode of our messages, we find no evidence that they affected households' likeli-hood of filing a tax return or claiming the credit. We conclude that even the most behaviorally informed low-touch outreach efforts cannot overcome the barriers faced by low-income households who do not file returns. (JEL C93, D91, H24, I38)

Education is a credence good. While the virtues of education are widely embraced, its qualities are difficult to discern, even among its consumers. The sizeable and increasing cost of tuition – as in the case of U.S. law schools – only add to the stakes. In response, law school rankings have emerged, with the purported goal to help students make more informed choices. While these rankings have generated both interest and debate, an important question has remained unanswered: how do prospective law students perceive these schools? Drawing upon data provided by the Law School Admissions Council (LSAC), we analyze the universe of law school applications for the period 1989 through 2017, creating a revealed preference ranking of law schools based solely on where applicants choose to matriculate given their offers of admission. We find that applicants strongly prefer Yale, Stanford, and Harvard, and to a lesser extent other schools in the top 20, but do not draw such sharp distinctions outside of these schools. For all but the very top schools, we cannot rule out that schools adjacent in the rankings are equally preferred by admitted students. We also separately analyze the application, admission, and matriculation stages of the law school matching process. Applicants apply broadly, we find, but that admissions and matriculation decisions hew closely to academic indicators. Our revealed preference rankings are similar those of the U.S. News law rankings at the top but bear little resemblance for the remaining schools. Our rankings offer a compelling alternative to commercial rankings, which are opaque and highly manipulatable. Our analyses also highlight the limitations of ordinal rankings, which by themselves can suggest meaningful differences amongst alternatives where they do not exist.

A subset of undergraduate applicants to the University of California, Berkeley were invited to submit letters of recommendation as part of their applications. I use scraped text of the submitted letters, natural language processing tools, and a within-subject experimental design wherein applications were read in parallel with and without their letters to understand the role that this qualitative information plays in admissions. I show that letters written on behalf of underrepresented applicants were modestly distinctive. I also construct an index of letter strength, measuring the predicted impact of the letter on the student's application score. I show that underrepresented applicants tend to get weaker letters, but that readers pay less attention to letter strength for underrepresented students. Overall, the inclusion of letters modestly improved application outcomes for the average underrepresented student.