HIGHLIGHTS
  • Intervention: A program to quickly move welfare recipients into the workforce.
  • Key Findings: Randomized controlled trial shows sizable increase in employment rates and job earnings, reduction in welfare dependency, and savings to the government, especially for single parents.

Description of the Intervention

Riverside County California’s Greater Avenues for Independence Program (GAIN) program was a welfare-to-work program for (i) single parents receiving or applying for welfare whose children were over a certain age (generally five); and (ii) the principal wage earner in all two-parent families receiving or applying for welfare. It was a mandatory program, in that participation was required as a condition for receiving full welfare benefits under the federal Aid for Families with Dependent Children (AFDC) program. Riverside’s GAIN program was part of a larger California GAIN initiative, operating statewide from the late 1980s through the mid-1990s, that was designed to increase the employment and self-sufficiency of welfare recipients. The GAIN model was implemented differently in various counties — for example, some counties focused more on providing basic education services, and others focused more on providing job search assistance.

Riverside’s program was administered by county welfare offices in partnership with contractor organizations that provided program services (primarily community colleges and adult schools). The welfare office first tested participants’ basic math and reading skills and assigned each program participant a case manager who placed him or her in one of the following two service streams:

  • Participants deemed not to need basic education. These included participants who scored above a cutoff score on the reading and math tests, had a high school diploma or GED, and were proficient in English. These participants were immediately assigned to one of two job search activities: (1) a job club, in which participants learned basic job-seeking and interviewing skills in a small-group setting; or (2) a supervised job search, using resources available at the welfare office and contractor sites that included employment counseling, direct links with local employers, personalized assistance in locating a job, and telephones. Both job search activities typically lasted for three weeks.
  • Participants deemed to be in need of basic education. These included participants who scored low on the reading and math tests, did not have a high school diploma or GED, or were not proficient in English. These participants could choose, as their initial activity, either to attend a basic education program (i.e. GED preparation, Adult Basic Education, or English language classes) or participate in one of the job search activities described above.

Participants in either service stream who failed to find a job after completing their initial activities were reassessed by their case manager, in order to determine their next assigned activity (e.g., skills training, vocational education, on-the-job training, unpaid work experience, or further job search assistance). Those in the basic education stream were normally expected to complete a job search assignment before other options were tried.

Overall, of program participants who started an activity during their first 11 months in the program, 57% participated in job search, and 60% participated in education or training.

Riverside GAIN was unique among the programs studied in the larger California GAIN initiative in that it placed much more emphasis on moving participants quickly into the workforce — and much less emphasis on providing basic education — even for those deemed to be in need of basic education. For instance: (1) the program conveyed a strong message to all participants about the importance of employment, and encouraged participants to accept job offers even if they were low paying; (2) the county used successful job placements as one of its criteria for assessing staff performance; and (3) participants assigned to basic education were also strongly encouraged to participate in job search activities, and about 30% did so.

Also, Riverside GAIN was strongly mandatory compared to most other programs studied in the California GAIN initiative, with about one-third of program participants placed in an enforcement process for noncompliance during their first year, and 6% ultimately losing some welfare benefits.

The program cost approximately $2,963 per participant in 1994 dollars ($3,988 in 2006 dollars). These costs were more than offset by savings to the government (as discussed below).

Click here for a more detailed description of the Riverside GAIN program. See especially pp. xxiii to xxvi and pp. 24 to 63.

EVIDENCE OF EFFECTIVENESS

This program was evaluated in one randomized controlled trial of 8,033 welfare applicants or recipients who attended a mandatory orientation to the Riverside GAIN program. These individuals were randomly assigned between August 1988 and March 1990 to either a group required to participate in the Riverside GAIN program or a control group that was not eligible for the program, but could participate in other services in the community. 33% of the sample had been receiving welfare for at least two years; 49% were White; 29% were Hispanic; and 13% were African American.

Five other California county GAIN programs participated in the statewide evaluation, but their effects were considerably smaller than Riverside’s and are not reported here.

Effects on single-parent program participants 5 years after entering the program (versus the control group):

  • 42% higher annual job earnings per person over the five-year period ($3,395 in annual earnings for the Riverside GAIN group vs. $2,387 for the control group).
  • 38% higher average quarterly employment rate over the five years (32.6% for the Riverside GAIN group vs. 23.6% for the controls).
  • 15% lower welfare payments per person over the five years ($3,135 in annual welfare payments for the Riverside GAIN group vs. $3,676 for the controls).
  • The sizeable effects above applied to both the less and the more disadvantaged subgroups in the full sample (e.g., both those with higher and lower levels of prior education).
  • Program participants experienced no net financial gain or loss, as their increased earnings were offset by reduced welfare and other benefits.
  • The program produced a large net savings to the government through increased tax revenues and reduced welfare and food stamps payments (as an estimate, $2.84 saved for every $1.00 invested over five years).

Effects on the principal wage earner in two-parent families 5 years after entering the program (versus the control group):

  • 14% higher annual job earnings per person over the five-year period ($3,845 in annual earnings vs. $3,362). This effect was statistically significant at the .10 level but not the .05 level.
  • 15% higher average quarterly employment rate over the five years (32.8% vs. 28.4%).
  • 11% lower welfare payments per person over the five years ($3,717 in annual payments vs. $4,173).
  • Program participants experienced no net financial gain or loss, as their increased earnings were offset by reduced welfare and other benefits.
  • The program produced a sizeable net savings to the government through increased tax revenues and reduced welfare and food stamps payments (as an estimate, $1.61 saved for every $1.00 invested over five years).

Effects on program participants 7-9 years after entering the program (versus the control group):

  • For single-parent families, the program’s effects on their average annual earnings, quarterly employment rate, and annual rate of welfare receipt during years 7-9 diminished to between 5 and 15%, and in most cases were not statistically significant.
  • For the principal wage earner in two-parent families, the program’s effects on these outcomes diminished to near zero during years 7-9.
  • One possible explanation for these diminished effects is that as of mid-1993 (i.e. approximately 3-5 years after random assignment), control group members were no longer excluded from the program if they were still on welfare, and may have received program services and become subject to the program’s participation requirements. Another possible explanation is that the program’s large effects on participants in the first five years simply wore off with time, and control group members “caught up” in annual earnings and other outcomes (although they still lagged in cumulative earnings over the 7-9-year follow-up period).

Description of Study Quality (click here for a glossary of terms)

  • The study had very low attrition and a long-term follow-up: Outcome data on employment, earnings, and welfare payments were obtained for 97% of the sample at the 5-year follow up, and 96% of the sample at the 7-9 year follow up.
  • The study measured outcomes using an intention to treat analysis.
  • To measure employment, earnings, and welfare outcomes, the study used official state and county administrative records, rather than relying on sample members’ self reports.
  • This was a large study which evaluated Riverside GAIN as implemented across the entire county, thus providing evidence of the program’s effectiveness in real-world conditions.

REFERENCES

(Click on linked authors’ names for their contact information)