In 2002, the average number of beneficiaries of federal adoption assistance surpassed for the first time the average number of recipients of federal foster care payments. By 2008, it is projected that the number in federally assisted permanent homes will exceed the number in federally funded foster homes by an order of 2 to 1. The shifting balance from foster care to family permanence, however, does not mean that the work of supporting and strengthening these new families necessarily ends. Even though regular casework and judicial oversight are no longer required, these homes still need occasional support to ensure child well-being and sometimes more intensive interventions to preserve family stability. Unless federal and state governments adapt existing funding mechanisms to address the challenges of family support in a post-permanency world, many of the gains achieved over the past decade in bringing safety and permanence to the lives of thousands of former foster children could potentially be lost.

The design of a responsive post-permanency service system can be sketched in only preliminary detail since the long-term challenges posed by the late 1990s surge in adoption and guardianship are only now coming into view. This paper offers some preliminary insights into the scale and nature of family vulnerability and resilience after adoption and guardianship by drawing on follow-up data collected in conjunction with the nation's largest federal subsidized guardianship experiment—the Illinois Subsidized Guardianship Waiver Demonstration.

Westat, Inc., an independent survey firm, conducted two waves of in-person interviews in 1998 and 2000 with 1,084 adult caregivers in the experimental group and 1,076 in the control group. Because only 20 percent of experimental children were actually discharged to guardianship, a supplementary random sample of 91 legal guardians was also interviewed. Administrative data on subsidy outlays and placement changes permitted the tracking of the stability of most of these living arrangements through December 30, 2004.

In prior analyses, I employed an intent-to-treat (ITT) approach to compare the outcomes of children assigned to the experimental and control groups regardless of whether or not the families selected a permanency option. I found that the stability of living arrangements, children's feelings of belongingness, and caregiver's intentions of raising them to adulthood were no greater in the experimental group than in the control group. This current study supplements my prior analyses with a treatment-on-the-treated (TOT) analysis. This TOT approach narrows the comparison to only foster children actually discharged to subsidized guardianship in both the experimental and supplementary samples with a matched control sample of adopted and foster children with similar propensities but denied the option of subsidized guardianship. Prior findings of no differences are replicated. Quantitative and qualitative data on post-permanency rates of displacement and family preservation are used to draw inferences about the potential scope and organization of a 21st century child welfare system that can fulfill traditional foster care obligations as well as support and strengthen these newer forms of state-created kinship.