Fatherless: The Long-Term Effects of Losing a Father in the U.S. Civil War

We use the U.S. Civil War, in which more than 650,000 soldiers perished, as a natural experiment to estimate the causal effect of losing a father on children’s long-run socioeconomic outcomes. We link military records from the 2.2 million Union Army soldiers with the 1860 U.S. population Census and then track their sons into adulthood by linking them to the 1880 Census. Compared to the sons of soldiers who returned, sons of soldiers who died had a lower occupational score in 1880 and were less likely to have a highor a semi-skilled occupation. Our results are robust to instrumenting paternal death by participation in one of the top 10 bloodiest battles of the war. We show how record linkage errors can attenuate OLS and inflate IV estimates. We also provide evidence that income is an important channel and that wealth is a mitigating factor. The negative effects are persistent and even affect the generation of the grandchildren observed in 1900. JEL codes: N11, J13, J62