In this analysis, we did not have comparative data to measure the relative effectiveness of imposing sanctions or imposing different kinds of sanctions. However, the data from Illinois and New Jersey strongly suggest that the imposition of a gradual full-family sanction does promote compliance with work requirements. Over an 18-month period, Illinois and New Jersey imposed initial partial sanctions on 26 and 38 percent of TANF recipients, respectively. As Table III.7 shows, in both states the majority of recipients who experience an initial partial sanction eventually come into compliance with work requirements (67 percent in Illinois and 60 percent in New Jersey). Eighty percent of recipients who come into compliance in Illinois and 60 percent in New Jersey do so before a full-family sanction is ever imposed. These results suggest that the imposition of an initial partial sanction is sufficient to encourage a substantial number of families to participate in program activities. What we cannot tell from these data is whether families would have responded differently if the initial grant reduction was not followed by a full-family sanction. It is also important to note that about one-quarter of recipients who received an initial partial sanction left TANF for reasons other than the imposition of a full-family. Some of these families may have left because they had access to other resources, including unreported earned income or because they found employment on their own.
|% of Families with an Initial Partial Sanction Imposed||% of Families with an Initial Partial Sanction Imposed|
|Initial partial sanction imposed||100||100|
|Evidence of compliance1 after a sanction is imposed||67||60|
|Full tanf grant restored before full-family sanction imposed||55||36|
|Return to tanf after full-family sanction imposed||12||24|
|Some employment, no return to tanf||n.a.||8|
|Record of compliance or employment after a sanction is imposed||n.a.||68|
|Exited tanf for reasons other than the imposition of a full family sanction||23||26|
|1 We define compliance to include all cases where the full TANF grant was restored after an initial partial sanction was imposed and all cases that returned to TANF after a full family sanction was imposed. Using this definition, families who started participating in assigned activities and those who received an exemption or modification of their work participation requirements are considered to be compliant with work requirements.|
(3) The predicted probabilities presented here and in Table III.2 are based on the results from estimating logistic regression models for sanction rates within 10 months in South Carolina and 12 months in Illinois and New Jersey. They represent the likelihood of the outcome in question for a client who has the particular characteristic in the table but who otherwise has the average characteristics of all clients. In addition to the client characteristics in the table, the models included and controlled for clients' gender and whether they had earnings either in the baseline month (New Jersey) or baseline quarter (Illinois and South Carolina).
(4) Similar data are not available in New Jersey. For a complete definition of employment assets and liabilities see Kirby, Fraker, Pavetti, and Kovac (June 2003), "Families on TANF in Illinois: Employment Assets and Liabilities." Washington, DC: Mathematica Policy Research, Inc.
(5) From a recipient's perspective, a full-family sanction could also end when she finds employment. However, this information is not necessarily known to the welfare office.
(6) In Illinois, cases officially remain open for three months while under a full grant sanction. For purposes of comparison to New Jersey and South Carolina, we considered these cases closed immediately upon being fully sanctioned. Similarly, cases that were in a "zero grant" status for reasons other than a sanction were considered closed.
(7) In contrast, among those who were not employed when they received a full-family sanction, 58 percent reported leaving TANF because of a sanction while 21 percent reported leaving for work.