Women and Post-WWII Wages

11/01/2002
Summary of working paper 9013
Featured in print Digest

Increases in female labor supply decreased both female and male wages, but had a stronger effect on women.

In Women, War, and Wages: The Effect of Female Labor Supply on the Wage Structure at Mid-Century (NBER Working Paper No. 9013), authors Daron Acemoglu, David Autor, and David Lyle study the effect of women's work on wages, looking at the period before and after WWII. The authors focus on the growth of female employment from 1940 to shortly after the war, in 1950. In 1940, only 28 percent of women were working; by 1945, this figure exceeded 34 percent. In fact, the 1940s saw the largest proportional rise in female labor during the entire twentieth century. Although more than half of the women drawn into the workforce by the war left at the end of the decade, a significant number remained. The focus of the Acemoglu, Autor and Lyle study is the increase in female labor supply caused by the WWII mobilization.

To isolate the mobilization-induced labor supply shift, the authors exploit the fact that the fraction of males serving in the war was not uniform across states. For example, in Massachusetts, Oregon, and Utah, almost 55 percent of males between the ages of 18 and 44 left civilian work to serve in the war. In Georgia, the Dakotas, and the Carolinas, this number ranged between 40 and 45 percent. The state differences in war mobilization actually reflect a variety of factors. The Selective Service's guidelines for deferments were based on marital status, fatherhood, essential skills for civilian war production, and temporary medical disabilities, but left considerable discretion to the local boards. Because of the importance of maintaining a strong food supply to support the war, an important consideration for deferment was farm employment. States with a high percentage of farmers had substantially lower mobilization rates, and this explains a considerable share of the state variation in mobilization rates.

The authors show that in states with greater war mobilization of men, women worked more after the war and in 1950, but not in 1940. This differential does not appear to be explained by other cross-state differences or possible demand factors, and is not present in the 1940 data nor does a similar trend recur in the decade of the 1950s. The authors interpret these differentials as labor supply shifts induced by the War. Acemoglu, Autor, and Lyle believe these cross-state changes in female employment were caused by greater participation of women during the war years, with some of those women staying on. War changed women's preferences, opportunities, and information about available work.

Using the cross-state changes in women's employment caused by the mobilization, the authors show that increases in female labor supply decreased both female and male wages, but had a stronger effect on women's wages. The impact of women working on male earnings was not uniform across male education levels, however: greater female labor participation led to increased earnings inequality between male college and high school graduates (that is, it lowered what a male high school graduate was paid relative to a college graduate) but reduced earnings inequality between male graduates of high school versus eighth-grade graduates. The authors' results demonstrate a closer degree of substitutability between males and females in the labor force than has been suggested by previous economic analyses. Contrary to common belief, women do not substitute most closely for the least educated males, but rather for male high school graduates -- or at least that was the case at mid-century.

Because rising female labor force participation raised inequality substantially in the top half of the male wage distribution and reduced it only slightly in the bottom half, the authors' estimates suggest that its net effect on wage inequality was positive (that is, more inequality). However, the authors explain that education levels and characteristics of women who increased their labor supply during the 1940s differ substantially from those women increasing their labor supply today. The experience of the WWII era therefore provides an intriguing but imperfect guide to the effect of female labor supply on male earnings inequality in recent decades.

-- Marie Bussing-Burks