The 30 Years' War and Violent Crime in the Late 19th Century

We introduce a unique data set of estimated population losses in Germany as a result of the 30 Years’ War (30 YW), fought between 1618 and 1648, which we match to German counties as defined in 1900. We find that counties with larger population losses experience significantly more youth assault crimes in 1900. To support a more causal interpretation of our results, we identify a set of counterfactual counties via propensity score matching. On this matched sample, we find that a county’s 30 YW population losses are largely unrelated to a wide range of pre-30 YW social, economic, geographic and religious characteristics, indicating that our result is driven by differential exposure to the violence of the war and not by pre-existing county differences. Although, owing to the historical nature of our data, it is difficult to uncover precise mechanisms, we present additional evidence that suggests 30 YW population losses lead to a general deterioration of county social and economic conditions over time: Differential exposure to the violence of the war led otherwise similar counties to experience more conflict following the 30 YW as well as higher rates of child mortality, less welfare expenditure, and increases in suicides some 250 years later. (JEL: N33, N43, D74, Z10)