Race and Work Relief in the New Deal Program

José Aveldanes, University of California, Berkeley

This paper investigates the role of discrimination in explaining racial disparities within the Works Progress Administration (WPA), focusing on the under-representation of Chinese and Japanese American male heads of households. The analysis aims to discern to what extent discrimination can explain racial disparities in WPA employment. I hypothesize that discriminatory practices may underlie this observed under-representation of Chinese and Japanese Americans. I exploit 1940 census full-count microdata, employing Oaxaca-Blinder decompositions to estimate employment probabilities among male American households. Pair-wise comparisons between Whites and various racial groups (i.e., Chinese, Japanese, Black) reveal differences in the likelihood of WPA employment. Comparisons with Black Americans help us understand how Chinese and Japanese Americans were treated relatively, with Japanese Americans experiencing the greatest disparity in employment. The analysis dissects factors contributing to the under-representation of Chinese and Japanese Americans in WPA employment, considering covariates such as family size, household wages, state-level relief ratios, age, and the presence of non-whites at county levels. These decompositions reveal that Chinese and Japanese Americans could have benefitted more from the WPA relative to whites based on employment criteria. However, these advantages are counterbalanced by the intercept in the Oaxaca-Blinder decompositions, reasonably suggesting the presence of unmeasured discrimination.

See paper

 Presented in Session 166. Labor, Patents, and Regulation: Insights from IPUMS Full Count Census Data