We analyze the effect of temperature during the pregnancy on the outcomes of live births in Hungary. Our main research question is whether temperature shocks (defined as the occurrence of extremely hot/cold days) in utero influence health at birth. In addition, we try to answer the question: how do the estimated effects differ between high‐ and low‐status families? Birth registry data of more than 1 million newborns between 2000 and 2014 are matched with daily temperature
data. Matching is based on the place of residence of the mother at the time of the delivery. Birth registry data that covers the entire population come from the Hungarian Central Statistical Office, whereas city‐level weather data come from the European Climate Assessment & Dataset project. Our dependent variables are indicators of health at birth (e.g. low birth weight, pre‐term birth). These indicators are regressed on the weather conditions during the pregnancy. Although
variation in weather over time supposed to be exogenous, we are able to control for important socio‐demographic factors that might influence newborns’ health. The effects of temperature are identified from interannual variation in weather conditions after adjustment for time‐invariant seasonal and regional effects, common shocks and time trend (by seasonal, time and location fixed effects). Our first results suggest that extremely hot days during pregnancy increase the probability of low birth weight and the probability of pre‐term birth. In addition, exposure to extremely hot weather during the pregnancy seems to have a stronger effect on newborns of low educated women.