This project explores the financial impact of marriage for women. It does this using data from the Civil War pension files, a unique source of information about the income and marriage histories of Union Army widows during the late nineteenth and early twentieth centuries. According to the theoretical economics model of marriage, women with better alternatives to marriage, or greater "outside options," will spend more time searching for husbands and will make higher quality matches. Intuitively, if a woman can support herself while single, she will be willing to wait longer for a better mate. Widows' pensions can be plausibly considered an "outside option" because they ceased upon remarriage. The researcher will construct a sample of widows from the Union Army database created by the Center for Population Economics at the University of Chicago, and collected information about these widows' pensions and subsequent marriages from the Civil War pension files at the National Archives in Washington, DC. Preliminary results suggest that receiving a pension significantly lowered the rate of remarriage.
This project contributes to a growing empirical literature on the gains from marriage, with a focus on women. It identifies a novel way of characterizing these gains by isolating a source of income that is unrelated to women's individual characteristics and which does not contribute to output within a marriage, which means that it should not affect women's "attractiveness" to potential mates. Pension eligibility depended only on the circumstances of a widow's husband's death, and it ceased upon remarriage. More broadly, this project will improve our understanding of the way in which marriage decisions were made during the nineteenth century, and will provide new evidence that will help to understand and predict the influence of labor mobility and employment options on marriage today.
The most significant impact of this project is the creation of an original database that other researchers will be able to use. This is significant not only because the database is new, but because data that follow women through marriage during the nineteenth century are exceedingly rare. The project has potential policy implications, as it will contribute to our understanding of the welfare effects of social programs that alter incentives around marriage and childbearing for women.
The main objective of this project is to undertake a study of the effect of Civil War pension income on the rate of remarriage among Union Army widows. This analysis offers some of the first insight into the value of marriage for women during this period. The project involved collecting data on widows' pension and remarriage outcomes from the Civil War pension files at the National Archives in Washington, DC. The project has two major outcomes: (1) the creation and dissemination of an original academic study of pensions and remarriage; (2) the creation and dissemination of a novel database that will be of interest to other researchers. These outcomes will be discussed in turn. The project has generated an original academic paper, "Women’s Income and Marriage Markets in the United States: Evidence from the Civil War Pension." In addition to forming the primary component of the co-PI’s dissertation, this work has been disseminated as a working paper in two separate series (National Bureau of Economic Research Working Paper no. 20201; Boston University IED Discussion Paper no. 244), it has been presented at multiple academic seminars and conferences, it has been reported on in the popular press (Wall Street Journal economics blog, June 17 2014), and it has been submitted for publication. The paper attempts to measure the effect of receiving a pension on the hazard rate of remarriage among widows, taking into account the potential endogeneity of the preferred measure of pension income to marital outcomes. Because the pension terminated upon remarriage, it would have made women more selective in the marriage market, thus causing them to remarry at a slower rate. The magnitude of this effect depends on how the value of the pension compares with the value of a typical marriage – if the pension is more valuable than many marriage, it should generate a larger effect on the remarriage rate. The paper shows that receiving a pension caused the hazard rate of remarriage to drop by 25%, implying an increase in the median time to remarriage of 3.5 years; the effect is greater for older women with more children. Because the pension was relatively small, this suggests that women during this period placed little value on marriage. This implies that the legal and social improvements in the terms of marriage for women that occurred over the last 150 years have had large positive welfare effects. The project has also generated a novel database of widows’ pension records, both in machine-readable form and in image form. The machine-readable database contains information on widows’ pension, marriage, demographic, and locational outcomes. Both databases will be made publically available on the co-PI’s professional website, and are expected to be useful for other project. For instance, the co-PI is undertaking a new project using the database, which is titled "Union Army Widows and the Historical Take-up of Social Benefits." This study will examine the determinates of program take-up among widows and will be the first study of participation in social programs before the 20th century.
