This paper is part of the Next Generation?s working paper series. The Next Generation is a project that examines the effects of welfare, antipoverty, and employment policies on children and families. Drawing on rich data from recent welfare reform evaluations, the project aims to inform the work of policymakers, practitioners, and researchers by identifying policy-relevant lessons that cut across evaluation. Although there is a growing research on the types of child care used among welfare recipients and low-income women in general, there is less information on child care stability. The research shows, for example, that many women use informal care by relatives, but there is less information about how often they change types of care from month to month and about how often they have interruptions in care. There is also limited evidence on how child care instability affects employment, in part because of a lack of appropriate data but also because it is difficult to prove that one factor causes the other. Although the same women who have unstable child care typically have unstable work patterns, this association may arise simply because both of these outcomes are caused by some other factor, such as low education levels or family instability. Using a unique data set consisting of over 3,500 women targeted for welfare-to-work programs in three states, this paper examines child care stability and its association with employment stability among current and former welfare recipients. Each of the three programs was evaluated using a random assignment design, in which some women were assigned to the new program being tested while others were assigned to the then-existing welfare system in the state or to Aid to Families with Dependent Children (AFDC).