« Back to Results

Race and Gender in Employment and Education

Paper Session

Saturday, Jan. 6, 2018 8:00 AM - 10:00 AM

Pennsylvania Convention Center, 203-B
Hosted By: Labor and Employment Relations Association
  • Chair: Marta Lachowska, W.E. Upjohn Institute for Employment Research

The Wage Penalty for Motherhood in Developing Countries

Jorge M. Aguero
University of Connecticut
Mindy Marks
Northeastern University
Neha Raykar
Public Health Foundation of India


Despite the growing size and importance of female employment worldwide, there have been limited efforts to explore the magnitude and causes of motherhood wage penalties in developing countries. Data from almost 130,000 women in 21 developing countries reveal a robust negative relationship between family size and female earnings, with a per child penalty that is three time larger in middle-income than in low-income countries. To address the endogeneity of family size, we instrument for the number of children using infecundity shocks and show that the negative relationship is causal. We find that for all women the negative impact of children diminishes as children age. For mothers in low-income countries we find a differential impact by child's gender, with adolescent daughters increasing their mother's earnings. Working from home, occupations and seasonal work explain little of the family gap for mothers in low-income countries; these variables account for one-third of the gap for mothers in middle-income countries.

The Black-White Gap in Non-Cognitive Skills Among Elementary School Children

Todd Elder
Michigan State University
Yuqing Zhou
University of California-Los Angeles


A vast literature has examined black-white gaps in cognitive skills, but racial differences in non-cognitive skills have attracted relatively little attention. Using data from two cohorts of the Early Childhood Longitudinal Study, we find large black-white gaps in teacher-reported measures of non-cognitive skills, even after controlling for detailed student and family characteristics. We show that these measures likely understate true racial disparities in non-cognitive skills because of systematic differences across schools in what teacher reports actually represent. Correcting for the resulting bias nearly doubles the size of the estimated gaps, to roughly the same magnitude as analogous gaps in test scores. Our estimates are remarkably stable across cohorts, suggesting that black children have neither made nor lost ground in recent decades. Finally, supplemental analyses based on the British Cohort Study of 1970 suggest that non-cognitive skills may account for sizeable portions of black-white disparities in adult outcomes.

Let the Girls Learn! It Is Not Only About Math ... It's About Gender Social Norms

Natalia Nollenberger
IE Business School
Nuria Rodriguez-Planas
City University of New York-Queens College


Using PISA test scores from 11,527 second-generation immigrants coming from 35 different countries of ancestry and living in 9 host countries, we find that the positive effects of country-of-ancestry gender social norms on girls’ math test scores relative to those of boys expand to other subjects (namely reading and science). We further find that gender norms shaped by beliefs on women’s political empowerment and economic opportunity affect the gender gaps in test scores in general. Interestingly, gender norms do not seem to particularly influence math-related stereotypes, but instead, preferences for math. Finally, the evidence indicates that these findings are driven by cognitive skills, suggesting that social gender norms affect parent’s expectations on girls’ academic knowledge relative to that of boys, but not on other attributes for success--such as non-cognitive skills. Taken together, our results highlight the relevance of general (as opposed to math-specific) gender stereotypes on the math gender gap.

Average Gaps and Oaxaca-Blinder Decompositions

Tymon Sloczynski
Brandeis University


In this paper I develop a new version of the Oaxaca-Blinder decomposition whose unexplained component recovers a parameter which I refer to as the average wage gap. Under a particular conditional independence assumption, this estimand is equivalent to the average treatment effect (ATE). I also provide treatment-effects reinterpretations of the Reimers, Cotton, and Fortin decompositions as well as estimate average wage gaps, average wage gains for men, and average wage losses for women in the United Kingdom. Conditional wage gaps increase across the wage distribution and therefore, on average, male gains are larger than female losses.
Brantly Callaway
Temple University
Ina Ganguli
University of Massachusetts-Amherst
JEL Classifications
  • J1 - Demographic Economics