Many families use child care subsidies only for short periods of time (median subsidy spells range from four to nine months across states). Policymakers have raised concerns that short subsidy spells may lead to disruptions in children’s care arrangements, and arrangement instability has been associated with negative effects on children’s development. In this study we have a unique opportunity to assess stability in both subsidized and non-subsidized arrangements. We use both survey data on low-income families with young children and administrative data from the child care subsidy program to investigate the patterns of changes in child care arrangements. In the survey data, we first developed profiles of the arrangements for the focal child, and then identified changes in arrangements, based on five waves of survey data (covering about two years). We complement the survey data with analysis of matched administrative data on those families using child care subsidies to examine the stability of their subsidized care arrangements. Each of these data sources has limitations, so by combining them we can examine the stability of arrangements from different angles. We find a high rate of turnover in child care arrangements for the children in these low-income families. Over a period of six months, half of the children changed primary provider. Children receiving child care subsidies were, if anything, less likely to change arrangements. Families tended to have multiple types of arrangements and multiple providers, and children receiving subsidies also often had additional non-subsidized arrangements.

The multi-method design of the study combines two data sources. The survey data were collected as part of a multi-year longitudinal study, conducted by telephone with low-income parents who had at least one child age six or younger. Detailed information was collected about the child care arrangements for the focal child at each wave and parents were asked about reasons for choosing or changing arrangements. The five waves were conducted between 2010 and 2012. The baseline sample included 323 families. The Minnesota Department of Human Services provided administrative records for all children participating in the Child Care Assistance Program (CCAP) between 2009 and 2012. Survey responses and administrative records were linked based on respondent name, date of birth, gender and the focal child’s name. The administrative data include detailed information on the child care arrangement(s) paid for by the subsidy program, including the provider name, type, hours of care, and amount of payment. 

The main objective of the research is to analyze the stability of subsidized and nonsubsidized child care arrangements over the time period of the study. Measures of instability include number of arrangements, length of arrangements, and frequency of changes. The two data sources provide different but complementary information about the stability of child care arrangements. This paper provides a unique exploration of the stability of child care arrangements to assess whether subsidized arrangements contribute to overall instability or frequent disruptions of care arrangements.