Search for a command to run...
Identifying causal effects of prenatal psychological stress on birth outcomes is challenging because stressful events typically bundle psychological stress with material disruptions. The 2011 Fukushima nuclear accident provides a unique setting to overcome this challenge: while physical radiation exposure was geographically limited and well-documented, fear of radiation spread nationwide. We exploit this geographic separation to examine how maternal anxiety independently affects fetal development. Using universal Japanese birth records linked to census data, combined with a novel Google Trends-based measure of radiation-specific anxiety, we employ three complementary identification strategies: population-level comparisons of in-utero exposed versus unexposed cohorts, within-family sibling analysis controlling for time-invariant family characteristics, and dose-response estimation exploiting geographic variation in anxiety intensity. Experiencing the accident during pregnancy increased preterm births by 17% and reduced birth weights by 22-26 grams. Birth outcomes exhibit a clear dose-response relationship with anxiety intensity, with radiation-specific anxiety accounting for 72-79% of the overall preterm birth effects and 28-37% of the overall birth weight effects. Effects are concentrated among socioeconomically disadvantaged mothers and during first-trimester exposure, with the most severe impacts on already-vulnerable infants in the very low and extremely low birth weight categories. Our findings indicate that invisible threats generate measurable intergenerational health impacts through psychological stress pathways, with implications for disaster preparedness and risk communication during contemporary crises from pandemics to climate change.
Published in: Journal of Health Economics
Volume 107, pp. 103125-103125