In this paper, I examine the effects of child care subsidies on female employment, including potential spillover effects on those without children, which can occur in the presence of limited information and statistical discrimination. I provide evidence from the Survey of Income and Program Participation (SIPP) that working mothers quit their jobs less often and childless women in low-wage occupations experience layoffs less frequently where childcare subsidies are more generous. To rationalize these patterns, I develop a labor search and matching model that incorporates statistical discrimination. Employers know that on average women quit more often than men due to childcare needs. In response, they terminate their employment relationships with all women more readily. Childcare subsidies reduce the average gender difference in quit rates through their effect on mothers, obviating the need for statistically discriminatory behavior by firms, and increasing the employment persistence of all observationally similar women. Estimates from the search model allow me to examine counterfactual scenarios including full information and changes to subsidy policy. Results reveal that a mechanism for policy-effect spillover via changes in discriminatory behavior by firms can have an economically significant effect on employment stability outside the policy’s target population, and that there is potentially a tight relationship between quit propensities and occupational sorting. (author abstract)