Marriage, migration, and work: Three essays on mobility in the United States, 1850-1930

This dissertation studies three forms of mobility in the United States during the late nineteenth and early twentieth centuries. The first chapter uses newly collected data from Union Army widows' pension files to isolate the causal effect of women's income on their decisions about marriage. Making use of exogenous variation in the processing time of pension applications, I show that receiving a pension caused widows to remarry at a significantly slower rate. This suggests that women's income directly influenced marital outcomes, largely by making women more selective in the marriage market. The second chapter explores the extent to which nineteenth century internal migrants in the United States were motivated by the possibility of upward occupational mobility. Drawing on the literature on contemporary migrant selection and sorting, I argue that workers with greater potential for occupational upgrading should have selected themselves out of counties with low skill premiums and sorted themselves into counties with high skill premiums. Using linked data from the U.S. Census and county-level wage data, I present results consistent with this argument. The third chapter of the dissertation (co-authored with Claudia Olivetti and Daniele Paserman) examines intergenerational income mobility across three generations between 1850 and 1930. Making use of the socioeconomic content of names, pseudo-panels of three generations are created by grouping samples of individuals by first name. Using G1, G2, and G3 to index generations one two and three, respectively, we find a significant correlation between G1 and G3, controlling for G2. We also find differences in this correlation by gender, suggesting that the process by which income was transferred from fathers to daughters was not the same as the process by which it was transferred from fathers to sons.