Effects on Employment and Public Assistance Receipt

This section presents the effects of the Tier 2 program, compared with Tier 1, on work, TANF receipt, and food stamp receipt after one and a half years. Administrative records data are used to compare outcomes for the two groups, for the full sample of single parents, and for several key subgroups. The tables present effects on summary measures. Effects on the full set of outcomes, as well as outcomes from the 12-month survey, are shown in Appendix D.59

Outcomes for the Tier 1 Group

Before assessing the effects of the Tier 2 program, it is worthwhile to describe the benchmark outcomes set by the Tier 1 group. These outcomes serve as the counterfactual against which Tier 2 is measured, showing what would have happened to recipients in the absence of Tier 2.

Table 10 presents outcomes for both the Tier 1 and the Tier 2 group. For the purposes of this section, only data for the Tier 1 group are considered. These data show that many recipients left welfare over time — by Quarter 7, only 54 percent were receiving benefits. In contrast, quarterly employment rates increased fairly modestly, from about 30 percent in Quarter 2 (not shown) to 43 percent in Quarter 7. A fair amount of employment instability can be seen from the fact that 65 percent of the Tier 1 group worked at some point during follow-up, but only 43 percent were still working in Quarter 7. Average earnings among those who worked in that quarter were also low, at $3,130.60

Data from the 12-month survey indicate that people in the Tier 1 group who worked earned, on average, $9.60 per hour, which should yield about $5,000 per quarter for full-time consistent work (shown in Appendix Table D.13). The relatively low quarterly earnings among workers, indicated in Table 10, reflect unstable employment — many people do not work all weeks or months of a quarter, and many work only part time. At the time of the survey, a third of those employed worked part time. Few of those who worked were offered employer-provided benefits. For example, an employer offered health insurance to one out of seven respondents who were working at the time of the survey (Appendix Table D.13).

Effects of the Tier 2 Program

• The Tier 2 program, compared with Tier 1, had little effect on employment or earnings over the follow-up period of one and a half years. An early increase in employment did not persist.

The Tier 2 program led to a modest increase in employment, but only in the quarter just after program entry (not shown in Table 10). The impacts faded largely because many of those who went to work in that quarter did not stay employed through subsequent quarters. As a result, the program had little effect on employment or earnings during the first year and a half of follow-up (Table 10). Data from the 12-month survey (not shown) indicate that Tier 2 had little effect on other employment outcomes, such as wages, hours worked, or benefits.

• The Tier 2 program, compared with Tier 1, led to a modest increase in employment early on for those who had recent work experience, although this effect did not last beyond Year 1. A large fraction of the employment effect appears to be due to the placement of the Tier 2 group in supported employment positions.

Employment programs tend to work differently for people who have recent work experience, compared with their counterparts who have been out of work for long periods. Although the evaluation sample was intended to be restricted to Tier 1 participants who had not worked in the previous three months, about a third of the sample did have some, if very limited, recent work experience.61 As mentioned earlier, this outcome may reflect a lag in TANF program data on employment. For this group, the program led to notable increases in employment during the first year (Figure 3). Employment rates for those in the Tier 2 group, for example, were 9 to 10 percentage points higher than for the Tier 1 group in Quarters 2 through 4.

The two groups also differed with respect to program participation (see Appendix Tables D.10 and D.11). For the group who had not worked recently, the Tier 2 program led to large increases in basic job preparation activities, such as help with supplies, job preparation, and job search. For the recently employed group, in contrast, the program had no effects on these outcomes but led to a 12 percentage point increase in participation in supported employment. Additional analyses suggest that, for this group, a large part of the employment increase in the early quarters is associated with these supported work positions.62 It is not clear why the recently employed, and thus the most job-ready, would be more likely to be placed in supported employment positions, although they may be less risky placements from a caseworker’s point of view. What is evident, however, is that increased placement in these supported work positions did not lead to longer-term effects on employment.

[D]

• The Tier 2 program had no effect on public assistance receipt over the follow-up period as a whole but led to a small increase in the last three months.

For the entire follow-up period of a year and a half, the Tier 2 and Tier 1 groups had similar rates of TANF and food stamp receipt (Table 10).63 Receipt gradually fell over the follow-up period at roughly the same rate for both groups, with the exception of the last one or two quarters. As a result, the Tier 2 group was somewhat more likely than the Tier 1 group to receive TANF and food stamps in Quarter 7, with impacts of 4.3 percentage points and 5.6 percentage points, respectively. Data for an early cohort suggest that these effects are likely to persist beyond Quarter 7. It is not clear why the program led to a late increase in public assistance receipt in the absence of effects on employment and earnings.

• In general, the effects of the Tier 2 program did not vary across subgroups defined by barriers to employment or period of study entry.

Effects were estimated for several subgroups defined by potential barriers to work, such as depression, poor health, poor child health, low education level, and learning disabilities (not shown). There were few consistent and statistically significant differences in impacts between those with a given barrier and their counterparts without that barrier. One exception is for the small group of recipients who reported limitations in the ability to perform moderate physical activities, such as cleaning or climbing stairs. For this group, the Tier 2 program led to a reduction in employment.64

Effects were also examined for an early versus a later cohort. A program might have different effects for early versus later entrants, for example, if it takes time to become fully or strongly implemented. In the case of Minnesota, one reason to expect a cohort difference in effects is that the process of random assignment changed in October 2002, when recipients began to be randomly assigned by letter rather than in person. A cohort difference might emerge if the earlier group became more engaged in the program because of the initial in-person contact. The results (shown in Appendix Table E.3) suggest no systematic differences in effects between cohorts.