Explaining variations in state foster care maintenance rates and the implications for implementing new evidence-based programs

BackgroundU.S. Child Welfare systems are involved in the lives of millions of children, and total spending exceeds $26 billion annually. Out-of-home foster care is a critical and expensive Child Welfare service, a major component of which is the maintenance rate paid to support housing and caring for a foster child. Maintenance rates vary widely across states and over time, but reasons for this variation are not well understood. As evidence-based programs are disseminated to state Child Welfare systems, it is important to understand what may be the important drivers in the uptake of these practices including state spending on core system areas.Data and methodsWe assembled a unique, longitudinal, state-level panel dataset (1990–2008) for all 50 states with annual data on foster care maintenance rates and measures of child population in need, poverty, employment, urbanicity, proportion minority, political party control of the state legislature and governorship, federal funding, and lawsuits involving state foster care systems. All monetary values were expressed in per-capita terms and inflation adjusted to 2008 dollars. We used longitudinal panel regressions with robust standard errors and state and year fixed effects to estimate the relationship between state foster care maintenance rates and the other factors in our dataset, lagging all factors by one year to mitigate the possibility that maintenance rates influenced their predictors. Exploratory analyses related maintenance rates to Child Welfare outcomes.FindingsState foster care maintenance rates have increased in nominal terms, but in many states, have not kept pace with inflation, leading to lower real rates in 2008 compared to those in 1991 for 54% of states for 2 year-olds, 58% for 9 year-olds, and 65% for 16 year-olds. In multivariate analyses including socioeconomic, demographic, and political factors, monthly foster care maintenance rates declined $15 for each 1% increase in state unemployment and declined $40 if a state's governorship and legislature became Republican, though significance was marginal. In analyses also examining state revenue, federal funding, and legal challenges, maintenance rates increased as the federal share of maximum TANF payments increased. However, > 50% of variation in foster care maintenance rates was explained by unobserved state-level factors as measured by state fixed effects. These factors did not appear to be strongly related to 2008 Child Welfare outcomes like foster care placement stability and maltreatment which were also not correlated with foster care maintenance rates.ConclusionsDespite being part of a social safety net, foster care maintenance rates have declined in real terms since 1991 in many states, and there is no strong evidence that they increase in response to harsher economic climates or to federal programs or legal reviews. State variation in maintenance rates was not related to Child Welfare outcomes, though further analysis of this important relationship is needed. Variability in state foster care maintenance rates appears highly idiosyncratic, an important contextual factor to consider when designing and disseminating evidence-based services.