Job Market Paper

Who Benefits Most From a Same-Race Mentor? Evidence From a Nationwide Youth Mentoring Program


We identify the impacts of assigning a mentor of the same race or ethnicity on the social, emotional and academic development of youth relative to assigning a mentor of a different race or ethnicity. Using the universe of matches from a nationwide youth mentoring program, we document that a rich set of pre-match observables are balanced across same-race/ethnicity match status. We find that Black and Hispanic youth assigned a same-race/ethnicity mentor had slightly faster growth in self-perceived school ability and attitudes toward risky behaviors after twelve months of mentoring, relative to cross-race matches. On the other hand, cross-race matched Hispanic youth had improvements in course grades and cross-race matched Black youth were more likely to report having a "special adult" in their life. In contrast to previous work on race-matching, we do not find improvements in grades or expectations for future educational attainment. Given that racial/ethnic minority mentors are often in short supply, these results imply matching on race or ethnicity at the expense of other desirable traits is not necessarily optimal.

Presented at IBS Speaker Series, IBS Poster Symposium, PAA 2021 Annual Meeting, WEAI 2021 Annual Conference
Fig 2. Within-Agency Supply of Mentors/Youth by Race & Ethnicity

White mentors are overrepresented relative to racial/ethnic minority youth


Thirty-Five Years Later: Long-Term Effects of the Matlab Maternal and Child Health / Family Planning Program on Women's Well-Being
Proceedings of the National Academy of Sciences, vol. 118, no. 28, July 2021, p. e2101160118, doi:10.1073/pnas.2101160118.


The success of large-scale family planning programs depends on potential long-term benefits for women’s health and economic empowerment. They are presumed to reduce total pregnancies and family size, which may free up women’s time and resources. However, few studies have established long-term effects on health. We investigate the highly influential Matlab Maternal Child Health / Family Planning quasi-experiment effects on lifetime fertility and multiple dimensions of health 35 years after introduction of services. For cohorts of women defined by age at program initiation, using baseline and follow-up survey data, we find the program led to fewer children but few significant effects on health or economic production with one exception: women born 1950-1961, who experienced the largest MCH/FP effects on contraception and childbearing, have significantly poorer metabolic and functional health. Despite strong arguments in favor of long-term benefits, we observe no positive effects of this family planning program on long-term health.

Media coverage: Time Magazine
Fig 2. MCH/FP Program Effects on Women's Health Domains

Women born 1950-1961 who were in their prime fertility years during the program had worse overall and metabolic health

Working Papers

Early Childhood Health and Family Planning: Long-Term and Intergenerational Effects on Human Capital


Improving the health and nutrition of young children is important not only for immediate well-being, but also because it is believed to reduce poverty in the long-run through improved human capital. In addition, there may be intergenerational transfers of endowments and investments from improved health and nutrition that augment the human capital of the next generation. Little, however, is known about the long-term and intergenerational effects of programs targeted to improve health and nutrition in early childhood on human capital. This paper examines the effects of the quasi randomly placed maternal and child health and family planning in Matlab, Bangladesh on the human capital of the generation that directly received benefits and their children. Results demonstrate sustained impacts on height into adulthood for the generation that was directly impacted, and improved height and cognition for daughters in the next generation.

Fig 2. Timeline of Interventions, Birth Cohorts and Data Collection

MCH/FP program interventions relative to affected birth cohorts and data collection events

Works in Progress

Long Run Impacts of Famine Exposure: A Study of the 1974-1975 Bangladesh Famine
(with Gisella Kagy)
Mental Health Parity and Depression
Dynamic Social Network Effects of Refugee Resettlement