Research

World War II (WWII) was the largest mobilization of troops in US history. WWII veterans accounted for 37% of the US prime age labor force in 1950. This paper outlines a methodology that can be used to measure the effect of US military service during WWII on a variety of economic outcomes. The estimation uses randomization from the first WWII Draft Lottery as an instrumental variable to eliminate bias from voluntary enlistment and draft classification. This instrument requires the construction of a novel dataset of linked military records and US censuses. The instrument also requires adjustments based on an understanding of the complex institutional details of the WWII draft process to ensure exogeneity. As an illustration of this methodology, I estimate the effects of WWII military service on post-war self-employment and educational attainment and find suggestive evidence that military service led to a decrease in post-war self-employment.

We analyze a random sample of 15,049 white veterans and 5,329 black veterans of the US Civil War examined by physicians between 1890 and 1906. We calculate a period prevalence of STI of 1.2–1.7% among whites and 4.2–8.0% among blacks, even though blacks and whites had almost identical prevalence of STIs in their wartime medical records. Furthermore, we find evidence that Board physicians were on the lookout for STIs among black veterans that could be used to justify denial of pension support. With or without STIs, blacks were rejected at roughly twice the rate of whites during this time period. Currently, racial disparities are even higher today than in this historical period, with blacks currently having a 5–15 times higher incidence than whites. We invite a critical reflection upon practices of screening and measurement systems to assess properly the degree to which racial prejudice may be part of these systems.

We demonstrate the long reach of early social ties in the location decision of individuals and in their older age mortality risk using data on Union Army veterans of the US Civil War (1861-5). We estimate discrete choice migration models to quantify the trade-offs across locations faced by veterans. Veterans were more likely to move to a neighborhood or county where men from their same war company lived and were more likely to move to such areas than to areas where other veterans were located. Veterans also were less likely to move far from their origin and avoided urban immigrant areas and high mortality risk areas. They also avoided areas that opposed the Civil War. This co-location evidence highlights the existence of persistent social networks. Such social networks had long-term consequences: veterans living close to war-time comrades had a 6% lower probability of dying. 

This article overviews the research opportunities made possible by a National Institute on Aging-funded program project, Early Indicators, Intergenerational Processes, and Aging. Data collection began almost three decades ago on 40,000 soldiers from the Union Army in the U.S. Civil War. The sample contains extensive demographic, economic, and medical data from childhood to death. In recent years, a large sample of African-American soldiers and an oversampling of soldiers from major U.S. cities have been added. Hundreds of historical maps containing public health data have been geocoded to place soldiers and their family members in a geospatial context. With newly granted funding, thousands of veterans will be linked to the demographic information available from the census and vital records of their children.