Labor Studies

Members of the NBER's Labor Studies Program met November 7-8 in Chicago. Program Directors David Autor of MIT and Alexandre Mas of Princeton University organized the meeting. These researchers' papers were presented and discussed:

Emily Breza, Harvard University and NBER; Supreet Kaur, University of California, Berkeley and NBER; and Yogita Shamdasani, University of Pittsburgh

Labor Rationing: A Revealed Preference Approach From Hiring Shocks

Gizem Kosar, Federal Reserve Bank of New York; Ayşegül Şahin, University of Texas at Austin and NBER; and Basit Zafar, Arizona State University and NBER

The Work-Leisure Tradeoff: Identifying the Heterogeneity

Labor force participation is a key determinant of important aggregates in the economy, such as employment and hours worked. Understanding and predicting the behavior of participation in the macroeconomy requires identifying individuals' trade-off for work and leisure correctly, and the underlying heterogeneity. However, since observed work-leisure choices are influenced by various other determinants such as year, age, and cohort effects, it is difficult to identify the heterogeneity in these preference parameters using standard data without imposing some additional assumptions. In addition, observational data suffers from issues such as labor market frictions and unobservables, which may lead to biased inference. Kosar, Şahin, and Zafar address this challenge by designing a novel survey in which respondents are presented with multiple scenarios, in each of which they are asked to choose between two different job offers. The scenarios vary work hours, wage offers, and the outside option of non-work. These scenarios are individual-specific, and take into account the individual's household income, consumption, and current and past labor market history. Using the variation in this rich data, the researchers estimate a labor supply model to recover the unique preference parameters separately for each demographic group (such as gender, education, and income) without imposing any parametric assumptions on the underlying distribution of the heterogeneity in preferences. The elasticities implied by the estimated parameters vary systematically across these demographic groups. For example, the wage elasticity with respect to leisure is negative for only part-time hours for younger females (that is, younger females are willing to increase labor supply in response to an increase in wages, but only in the range of part-time hours). On the other hand, the estimated elasticity is negative for younger males for all (except very long) hours. There is also substantial heterogeneity in estimated elasticities within each of the demographic groups. In order to highlight the importance of this heterogeneity, assess the impact of policy changes in the tax policy and childcare subsidies on labor supply in a labor supply model with and without this heterogeneity.

Paul Mohnen, University of Michigan

The Impact of the Retirement Slowdown on the U.S. Youth Labor Market

Employment among older Americans has sharply risen since the mid-1990s, particularly in high-skill jobs. How has this labor-supply increase affected other participants in the labor market, and new entrants in particular? Exploiting cross-commuting zone differences in age composition among the old driven in large part by historical birth patterns, Mohnen explores the impact of retirement trends on youth employment outcomes between 1980 and 2017. Mohnen finds that in commuting zones where fewer older workers retire due to the initial age structure, the share of younger workers in high-skill jobs declines while the share of younger workers in low-skill jobs rises. Fewer retirements also lead to a rise in the share of younger workers who have higher educational attainment than their job typically requires, declining youth wages, and lower job mobility. Together, the results suggest that the retirement slowdown has contributed to stagnant early career outcomes in recent decades, explaining 30 percent of the rise in the share of younger workers in low-skill jobs between 1980 and 2017.

Henrik Kleven, Princeton University and NBER

The EITC and the Extensive Margin: A Reappraisal (NBER Working Paper 26405)

Kleven reconsiders the impact of the Earned Income Tax Credit (EITC) on labor supply at the extensive margin. He investigates every EITC reform at the state and federal level since the inception of the policy in 1975. Based on event studies comparing single women with and without children, or comparing single mothers with different numbers of children, the Kleven shows that the only EITC reform associated with clear employment increases is the expansion enacted in 1993. The employment increases in the mid-late nineties are very large, but they are influenced by the confounding effects of welfare reform and a booming macroeconomy. Based on different approaches that exploit variation in these confounders across household type, space and time, Kleven shows that the employment effects align closely with exposure to welfare reform and the business cycle. Single mothers who were unaffected by welfare reform (but eligible for the EITC) did not respond. Overall and contrary to consensus, the case for sizable extensive margin effects of the EITC is fragile. Kleven highlights the presence of informational frictions, widely documented in the literature, as a natural explanation for the absence of extensive margin responses.

Peter Q. Blair, Harvard University and NBER, and Benjamin Posmanick, Clemson University

When Does Labor Market Flexibility Reduce Gender Wage Gaps?

Ellora Derenoncourt, Princeton University, and Claire Montialoux, University of California, Berkeley

Minimum Wages and Racial Inequality

The earnings difference between white and black workers fell dramatically in the United States in the late 1960s and early 1970s. Derenoncourt and Montialoux show that the extension of the minimum wage played a critical role in this decline. The 1966 Fair Labor Standards Act extended federal minimum wage coverage to agriculture, restaurants, nursing homes, and other services which were previously uncovered and where nearly a third of black workers were employed. The researchers digitize over 1,000 hourly wage distributions from Bureau of Labor Statistics industry wage reports and use CPS micro-data to investigate the effects of this reform on wages, employment, and racial inequality. Using a cross-industry difference-in-differences design, Derenoncourt and Montialoux show that wages rose sharply for workers in the newly covered industries. The impact was nearly twice as large for black workers as for white. Within treated industries, the racial gap adjusted for observables fell from 25 log points prereform to zero afterwards. Using a bunching design, the researchers find no effect of the reform on employment. The researchers can rule out significant dis-employment effects for black workers. The 1967 extension of the minimum wage can explain more than 20% of the reduction in the racial earnings and income gap during the Civil Rights Era. The findings shed new light on the dynamics of labor market inequality in the United States and suggest that minimum wage policy can play a critical role in reducing racial economic disparities.

Stefano DellaVigna and Ulrike Malmendier, University of California, Berkeley and NBER; John A. List, University of Chicago and NBER; and Gautam Rao, Harvard University and NBER

Estimating Social Preferences and Gift Exchange with a Piece-Rate Design

DellaVigna, List, Malmendier, and Rao design two field experiments to estimate the nature and magnitude of workers' social preferences towards their employers. Unlike previous gift-exchange field experiments, the researchers vary piece rates in addition to gift treatments. This piece-rate design allows them to estimate the elasticity of effort to motivation and in turn identify aspects of the workers' social preferences. The first experiment measures productivity -- units of output produced in a fixed amount of time. The second experiment measures a form of labor supply -- the willingness to work for extra time. Using the piece-rate treatments, the researchers document that productivity is rather unresponsive to motivation, while labor supply is very responsive. In terms of social preferences, DellaVigna, List, Malmendier, and Rao document, first, that workers provide effort for their employer, but are insensitive to the return to the employer. This result is consistent with models of 'warm glow' or social norms, rather than pure altruism towards the employer. Second, while the researchers do not detect any effect of the gifts in the productivity experiment, they find sizable positive impacts in the labor-supply experiment. The researchers show that, at least in part, this different response to gifts is explained by different elasticities of productivity and labor supply, highlighting the importance of the piece-rate design.

Brent R. Hickman, Washington University in Saint Louis, and Jack Mountjoy, University of Chicago

The Returns to College(s): Estimating Value-Added and Match Effects in Higher Education

Students who attend different colleges in the United States end up with vastly different educational and labor market outcomes. Hickman and Mountjoy estimate value-added of individual institutions to decompose these differences into causal contributions of colleges versus selection bias from student sorting. Linking administrative registries of high school records, college applications, admissions decisions, enrollment spells, degree completions, and quarterly earnings spanning the Texas population, they identify institutional value-added across the diverse distribution of Texas public universities by comparing the outcomes of students who apply to and are admitted by the same set of institutions, as this “matched applicant” approach strikingly balances student ability measures across college treatments and delivers value-added estimates impervious to additional controls. The researchers find that differences in colleges’ causal value-added on earnings are nearly an order of magnitude smaller than, and uncorrelated with, their raw mean earnings differences. Earnings value-added is also uncorrelated with traditional measures of selectivity and peer quality, but some non-peer inputs like instructional expenditures and the share of faculty who are full-time are moderately predictive of value-added. Examining intermediate outcomes, Hickman and Mountjoy find colleges that modestly boost persistence, BA completion, and STEM degrees also tend to modestly boost earnings. Finally, they probe the potential for (mis)match effects by allowing value-added to vary flexibly by student characteristics, including race, gender, family income, and pre-college measures of cognitive and non-cognitive skills. At first glance, black students appear to face small negative returns to attending more selective colleges, but this pattern of modest mismatch is entirely driven by the presence of two large historically black universities in Texas that have low average incoming SAT scores but yield value-added for black students; across the non-HBCUs, black students face very similar (null) returns to selectivity as their peers from other backgrounds.

NBER Videos

National Bureau of Economic Research, 1050 Massachusetts Ave., Cambridge, MA 02138; 617-868-3900; email:

Contact Us