Working Papers

Aireti, D. (2026). “Impressionable-Year Conflict Exposure and Locus of Control: Evidence from the Conflict-Affected Regions in Sri Lanka.” RIEB Discussion Paper Series, No. 2026-16, May 2026. Under review at Review of Development Economics. (Full paper is available upon request)

Work in Progress

Aireti, D. and Yamazaki, K. “Prenatal War Exposure and Birthweight: Evidence from War-Affected Regions in Sri Lanka.”

This paper analyzes the effects of in utero war exposure on birthweight using original survey data collected in war-affected regions of Sri Lanka. While earlier studies find that prenatal exposure to violence negatively affects birth outcomes, relatively little attention has been paid to gender-specific in utero selection and its implications for estimated effects among live births. Using within-sibling comparisons, we account for shared family-level characteristics and further address potential gender-specific selection into live birth by distinguishing comparisons by sibling sex composition. We find that exposure to violence during pregnancy reduces birthweight, with effects varying by the child’s gender and type of violence. In particular, the results show that prenatal exposure to property damage reduces the birthweight of girls by approximately 293 grams, whereas the estimated effects are smaller and statistically insignificant for boys and for exposure to human victimization or forced displacement. Our findings suggest a culling effect that prenatal exposure disproportionately prevents less healthy male fetuses from surviving to live birth, leaving the surviving boys with stronger health endowments at birth than the girls in our sample. The results also indicate that negative effects of prenatal war exposure on birthweight are driven primarily by maternal undernutrition during pregnancy rather than by maternal stress.

Aireti, D. “Start Off on a Bad Note: The Long-Term Effect of the Sri Lankan Civil War on Mental Health and Occupational Choice.”

This paper examines the long-term effects of early-life exposure to war on adult mental health and labor-market choice. Using individual-level measures of war exposure, we find that property damage experienced by age five is associated with poorer mental health in adulthood, as measured by the Hopkins Symptom Checklist (HSCL), whereas the loss of family members or friends has no statistically significant effect. In particular, the results indicate that each additional incident of property damage increases the adult HSCL score by approximately 1.5 points, equivalent to approximately 7.6 percent of the sample mean. We also find that early-life war exposure does not significantly affect labor-force participation. However, individuals who experienced one additional incident of property damage by age 5 are nearly 7 percentage points more likely to work part-time, a form of employment often characterized by lower and less stable earnings. These labor-market consequences may slow the economic recovery of war-affected households.