Akbulut-Yuksel and Kugler 2016 — Intergenerational Persistence of Health in the U.S. Do Immigrants Get Healthier as They Assimilate

Summary

Using matched mother-child data from the National Longitudinal Survey of Youth 1979 (NLSY79) and Children of NLSY79 Young Adult File ($3{,}921$ native and $914$ immigrant pairs), this paper estimates intergenerational health persistence for immigrants and natives in the United States. Health — measured by weight, body mass index (BMI), asthma, and depression — transmits more strongly from immigrant mothers to their children ($\rho \approx 0.58$–$0.71$) than from native mothers ($\rho \approx 0.45$–$0.50$), but persistence declines sharply across generations as immigrants assimilate. Crucially, assimilation for weight and BMI runs toward American obesity norms rather than toward better health.

Key Claims

• Intergenerational weight persistence $\rho = 0.50$ for natives, $0.71$ for immigrants (ordinary least squares (OLS)); controls for maternal socioeconomic status (SES) narrow the gap but it remains positive. BMI persistence: $\rho = 0.45$ (natives) vs. $0.58$ (immigrants). Asthma persistence $\approx 17\%$ (natives) vs. $22\%$ (immigrants). Depression persistence $\approx 8\%$ (natives) vs. ${\sim}0\%$ (immigrants).
• Grandparents' fixed effects (FE) (a "cousin design" that controls for shared genetic endowment) do not materially change the persistence estimates, indicating that cultural, behavioral, and SES transmission — not genetics — drives most of the measured correlation.
• Assimilation by generation: BMI persistence declines from $\rho = 0.51$ (1st-generation immigrant mothers) to $0.49$ (2nd) to $0.32$ (3rd), converging toward the native level of ${\sim}0.45$.
• The direction of assimilation is toward American norms: because recent immigrant mothers have lower BMI than native-born mothers on average, convergence means immigrant children gain weight across generations toward the U.S. mean.
• Mother's education reduces child weight by $\approx 2.4$ lbs per year of schooling (for native mothers). Single motherhood adds $\approx 15$ lbs to immigrant children's weight, suggesting family structure is an important transmission channel for immigrants.
• Identification relies on OLS with rich controls; the grandparent FE design provides a robustness check against genetic confounding.

Concepts Introduced or Extended

• Intergenerational Health Transmission
• Cultural Assimilation

Entities Mentioned

• Mevlude Akbulut-Yuksel
• Adriana Kugler

Quotes

"We find that health persistence is higher for immigrants than for natives, but that it declines with assimilation, converging to native levels by the third generation."

"The direction of assimilation is toward American norms: since immigrants start with lower BMI than natives, their children's BMI rises across generations toward the U.S. average."

My Take

A carefully executed paper with a provocative implication: immigrant health assimilates, but toward American obesity norms, so "assimilation" is not straightforwardly good news on this health dimension. The grandparent FE design is clever but assumes cousin pairs are comparable on unobservables — a strong assumption. The small immigrant sample ($914$ pairs) limits precision for the generational breakdown, and the heterogeneous patterns across health domains (weight/BMI vs. asthma vs. depression) hint at distinct assimilation pathways that deserve separate treatment. The paper does not decompose the mechanism (diet convergence? neighborhood effects? SES transmission?), leaving the policy lever unclear.