The child care subsidy program is an important safety net program that is intended to provide low-income children with access to high-quality care and support parental employment. There is wide variation in state child care subsidy programs because states set their own rules regarding who is eligible for a subsidy, how much parents must contribute (copays), how often parents must recertify eligibility, and payment rates for providers. Previous studies have found that most families receive these subsidies only for short periods of time and many exit and return within a few months. These short spells and churning in the subsidy program raise concerns about the continuity of care arrangements for children on the subsidy. The four papers in this panel analyze how state child care subsidy policies affect continuity of child care subsidy receipt and stability in child care arrangements. Three of the papers use state-level child care subsidy administrative data and one uses a national survey data set. The first paper analyzes five years of subsidy administrative data to estimate Cox proportional hazard models of subsidy exit in order to examine the relationship between the timing of eligibility recertification and exits from the subsidy program, controlling for child, family, and provider characteristics. The authors exploit variation across counties and over time in local policies in implementation of eligibility rules. The second paper examines how state policy changes in child care subsidy affect child care subsidy receipt and child care stability among young children in Illinois. The study uses monthly child care subsidy administrative data from the state of Illinois from 1999 to 2013. Propensity score matching is used to compare how children who experienced 2006-2008 policy changes (increases in family income eligibility and provider reimbursement rates) differ from children who did not experience such changes (control group). In the third paper, using data from the Early Childhood Longitudinal Study-Birth Cohort, the author examines the associations between state subsidy policies in FY2000 and the use of multiple arrangements when children are 9 months and 2 years in age. Using difference-in-difference methods, analyses test whether the generosity and restrictiveness of state subsidy programs predict the use of multiple arrangements among low-educated, single mothers when compared to low-educated, married mothers. The last paper examines the relationship of policy generosity, child care usage, and continuity in the subsidy program as well as how these outcomes are associated with employment characteristics of participating parents. The presenter will discuss how unstable employment, as evidenced by frequent employment losses, job changes, and periods of unemployment, challenge the ability of a parent to remain in a program tightly linked to being employed. The panel represents diverse disciplines from Public Health, Human Sciences, Social Work, Economics and Sociology and a number of statistical methods, including propensity score matching and difference-in-difference estimators, to explore the impact of variation and changes in child care subsidy policies.