Definition
The degree to which parental health status predicts child health status, measured by the ordinary least squares (OLS) regression coefficient ρ from a child-health-on-parent-health regression. Analogous to the intergenerational income elasticity but applied to continuous (weight, body mass index (BMI)) or binary (asthma, depression) health outcomes. Values of ρ above zero indicate positive transmission; values around 0.5 mean roughly half of the parental deviation from the population mean is passed on to children.
Key Ideas
- Persistence coefficient ρ: The OLS slope of child health on parent health, controlling for child and parent socioeconomic status (SES) characteristics. Raw ρ captures total transmission including SES pathways; conditional ρ isolates transmission beyond education and income.
- Immigrant–native gradient: Immigrant mothers exhibit higher health persistence to their children (ρ≈0.58–0.71 for weight/BMI) than native mothers (ρ≈0.45–0.50). This reflects greater overlap in cultural environment, diet, and health behaviors between immigrant mother and child (Akbulut-Yuksel and Kugler 2016).
- Generational convergence: BMI persistence declines from ρ=0.51 (1st-generation immigrant mothers) to 0.49 (2nd) to 0.32 (3rd), converging toward the native level of ≈0.45. Consistent with Cultural Assimilation gradually dissolving the immigrant-specific transmission channel.
- Assimilation direction paradox: For weight, immigrant health assimilates toward American norms, which are worse (higher BMI) than recent immigrant norms. The "healthy immigrant effect" (initial health advantage of recent arrivals — see Epidemiological Paradox) erodes across generations partly because assimilation runs upward in BMI.
- Genetics vs. environment: Grandparent fixed effects (exploiting variation across cousins who share maternal grandparents) do not materially alter persistence estimates, suggesting genetics accounts for a small share of the measured correlation — cultural, behavioral, and SES channels dominate.
How It Works
Intergenerational health persistence is estimated as ρ in:
Child health=α+ρ×Parent health+β×Controls+ε
with controls for child and parent SES, mother's education, family structure, birth order, and age. The grandparent fixed effects (FE) ("cousin") design uses sibling mothers with the same grandparents to difference out shared genetic endowment, providing a cleaner estimate of the behavioral/cultural component. The National Longitudinal Survey of Youth 1979 (NLSY79) + Children of NLSY79 data enable this design because they follow both generations longitudinally.
Why It Matters
- Immigration policy: Generational convergence in health persistence implies immigrant health outcomes are not permanent — they fade across generations. But when assimilation runs toward unhealthy American norms (obesity), it worsens the immigrant health trajectory rather than improving it.
- Mechanism identification: The small role of genetics (grandparent FE robustness) implies that health disparities are addressable through SES policies (income, education, family structure) and behavioral/cultural interventions, not exclusively through genetic or pre-birth channels.
- Connection to economic mobility: Health persistence operates in parallel with Intergenerational Income Mobility — both measure how strongly parental circumstances constrain children's outcomes, but through distinct pathways that are largely uncorrelated across sending-country groups.
Open Questions
- Does the generational decline in immigrant health persistence hold uniformly across health domains? Akbulut-Yuksel/Kugler find heterogeneous patterns: weight/BMI converge, but asthma and depression show different trajectories.
- Is the "assimilation toward American obesity norms" finding robust to destination-country selection? It may not generalize to countries with healthier baseline norms.
- What fraction of the immigrant–native persistence gap is driven by neighborhood vs. household factors? NLSY79 data cannot fully separate residential environment from family environment.
- How does health persistence interact with SES Health Behavior Gradient? Education (−2.4 lbs/yr of schooling) and single motherhood (+15 lbs for immigrants) are significant channels but account for only part of the gradient.
Related
Sources