Flint changed its public water source in 2014, causing severe water contaminations. We estimate the effect of in utero exposure to contaminated water on health at birth using the recent Flint water crisis as a natural experiment. Matching vital statistics birth records with various sources of data, this paper employs a difference-in-differences method as well as a synthetic control method to identify its causal impact on key birth outcomes. Our results suggest that the contaminated water modestly increased probability of low birth weight by 1.1-1.8% but only insignificantly reduced birth weight by 6-19 grams, presumably due to more salient changes in birth weight around the low birth weight cut-off than across the whole distribution. These results survive a rich set of placebo and falsification tests. There are larger effects for wider windows of in utero exposure and especially for Black Americans. We find little evidence on sex ratios at birth, indicating that the scarring effect in utero may dominate the channel of mortality selection. To better understand the physiological and psychological mechanisms behind, future studies need to distinguish the effect due to a surge in lead. (Author abstract)