Untitled

Abstract

[Pre-Analysis Plan] Between 1911 and 1935, forty-six U.S. states implemented mothers' pension programs—the first government cash assistance targeting single mothers. A critical and understudied feature of these programs was the sharp eligibility cutoff: benefits terminated when the youngest child reached a specified age, typically 14. We propose to exploit this discontinuity using a regression discontinuity design with historical census data. Using simulated data calibrated to historical statistics, we demonstrate that our identification strategy can detect economically meaningful effects: the illustrative analysis suggests an 8.2 percentage point increase in maternal labor force participation at the eligibility threshold. The simulated validation shows that results are robust to bandwidth choice, pass placebo tests at non-cutoff ages, and that the effect appears only in states where the cutoff exists. These preliminary analyses will be replicated with actual IPUMS census data when available. If confirmed with real data, our findings would provide novel causal evidence on how welfare benefit loss affects maternal employment decisions. \\ Keywords: Mothers' pensions, labor supply, regression discontinuity, welfare policy, historical economics, pre-analysis plan \\ JEL Codes: H53, I38, J22, N32