Participation mandates in welfare-to-work programs are intended to change welfare recipients' perceptions and behavior in several ways. First, they send the message that receipt of welfare is not an unconditional entitlement. Second, owing to the time investment required by program activities, they reduce the perceived value of welfare grants relative to that of employment. Third, they compel some people who would otherwise not do so to participate in activities that program operators believe will enhance their employability.
As discussed earlier, FSA established participation mandates and means by which to enforce them. Most single parents had to participate in employment preparation activities for as long as they remained on welfare. Case managers could use a variety of informal and formal responses, including financial sanctions, when people did not cooperate. In the NEWWS programs, sanctions typically reduced the family's monthly welfare grant by approximately 20 percent until the sanctioned individual complied with the mandate; there was no minimum sanction length for the first "offense," a minimum of three months for the second, and a minimum of six months for the third. As a result, some sanctioned individuals experienced a penalty for a short time, while others were under sanction much longer. In Grand Rapids, for example, almost half of those sanctioned remained in this status for at least half of the first two years of the study period.
Like FSA, PRWORA put in place mandates and enforcement mechanisms, but it also made the compliance requirements for welfare-to-work programs stricter and gave states the option to increase the severity of penalties for noncompliance. As a result, many states now impose full-family sanctions (revocation of the family's whole grant) as opposed to the partial-family sanctions (revocation of only the adult's portion of the grant) imposed under FSA.
Participation mandates and their enforcement could contribute to a program's impacts on earnings and welfare payments by, for example, getting more people to participate or getting nonparticipants to take a job to avoid participating in the program. It is very hard to determine the size of these contributions, however, because participation in program activities and the requirement to participate are usually bundled together. An innovative substudy was conducted as part of NEWWS in an effort to assess the effects of a mandate per se. Coupled with the extensive cross-program comparisons possible in the evaluation, this investigation points to several lessons and cautions.
- Simply communicating to welfare recipients that they are required to participate in a welfare-to-work program before they actually enroll in any program can affect their employment behavior.
The substudy, which was undertaken in Grand Rapids and Riverside, involved the random assignment of recipients to one of two groups before the random assignment process described in the first section of this document began. In the substudy, some welfare recipients were randomly assigned to a group that was informed that it was subject to a welfare-to-work program participation mandate; the others were assigned to a group that was not. By comparing the employment and welfare outcomes for the two groups, it is possible to isolate the effects of simply communicating to people that they are under a mandate to enter a welfare-to-work program before they learn how the program works or receive an activity assignment.
The findings from the substudy indicate that, in the site with the healthier labor market, the mandate had a large initial effect on average earnings but left rates of employment and welfare receipt unchanged. This suggests that some recipients reacted to the mandate by finding a job more quickly, finding a job with longer hours, or working more hours at a job they already had than they would have without a mandate. The earnings impact was concentrated among the more job-ready recipients, that is, among those who had been employed at some point during the year before study entry. In the site with the weaker labor market, the mandate had no effect on employment, earnings, or welfare receipt, even among the more job-ready. These results suggest that employment opportunities play a role in how welfare recipients react to a welfare-to-work mandate: The mandate alone can have its intended effects, but it is more likely to work in situations where jobs are available and people have the recent work experience needed to obtain a job.
- High-enforcement programs lead to higher participation rates than do low-enforcement programs.
As discussed in the first section, the nine high-enforcement NEWWS programs aimed to enroll most targeted people in the program, monitored participation in program activities moderately or very closely, and stressed the mandatory nature of the program through sanctions and means such as positive encouragement. About half of these programs had high rates of sanctioning; the other half had moderate sanctioning rates. The two low-enforcement NEWWS programs tended to give preference in enrollment to those who volunteered for the program (as opposed to vigorously trying to enroll all targeted individuals), did not closely monitor participation in program activities, and rarely imposed sanctions for nonparticipation. As shown in Table 6, the low-enforcement programs produced lower participation rates (calculated using method 3 in Table 5) and smaller impacts on participation than did the high-enforcement programs. The low-enforcement programs' particularly small participation impacts (of about 10 percentage points) are partly due to the fact that some control group members engaged in activities on their own. A program that lacks a significant "push" or "pull," like the low-enforcement programs in NEWWS, will not engage many more people than would participate in activities anyway.
- Among high-enforcement programs, higher sanctioning rates do not necessarily increase participation levels.
Some of the high-enforcement programs in Table 6 sanctioned nonparticipants at very high rates, imposing partial-family sanctions on at least one-third of welfare recipients who enrolled in the programs. As shown in the table, however, these programs were no more successful in engaging people in activities than were programs in the high-enforcement category that had more moderate sanctioning rates. The programs with the highest sanctioning rates had neither the highest participation rates nor the largest participation impacts.
- More disadvantaged recipients are more likely to be sanctioned in welfare-to-work programs.
In the NEWWS programs, sanctions were more likely to be imposed on and lasted longer for the more disadvantaged welfare recipients than the less disadvantaged ones. One possible reason for this is that more disadvantaged people remain on welfare longer, lengthening the period of time during which they could fail to comply with the participation mandate and be sanctioned.