Citation
Hsiao, C., Shen, Y., Wang, B., & Weeks, G. (2007). Evaluating the impacts of Washington State repeated job search services on the earnings of prime‐age female TANF recipients. Journal of Applied Econometrics, 22(2), 453-475.
Highlights
- The study’s objective was to examine the impacts of Job Search Services (JSS) on employment and earnings outcomes of women ages 25 to 35 who participated in the WorkFirst program in Washington State.
- The authors used a nonexperimental study design to examine the earnings and employment outcomes of women who took part in JSS. Women were categorized as being in the treatment group if they participated in JSS at least once from the second quarter of 1998 to the fourth quarter of 2000. Women who did not participate in JSS during this time made up the comparison group.
- The study found that the first instance of participation in JSS increased the likelihood of employment by 4 percent for women seeking employment. The study also found a positive and statistically significant relationship between quarterly earnings and participation in JSS one or more times for employed women.
- The quality of causal evidence presented in this report is low because the authors did not ensure that the groups compared exhibited similar employment or earnings outcomes before the intervention. This means we are not confident that the estimated effects are attributable to JSS; other factors are likely to have contributed.
Intervention Examined
Job Search Services (JSS)
Features of the Intervention
JSS is a job search assistance service of WorkFirst, a welfare reform program, in Washington State. The JSS program assists people for up to 12 weeks with finding employment. Participants could receive one or more of the following program components: classroom instruction; job search support services with a specialist or group of unemployed peers; or pre-employment, high-wage, or high-demand trainings. The authors did not describe the pre-employment, high-wage, and high-demand trainings. Though JSS is mandatory on first entry into WorkFirst, people are not required to participate in JSS if they find more than 20 hours of unsubsidized employment, have a child younger than three months old, provide an accepted excuse, need additional skills or experience as determined by a specialist, or require other support services because of substance abuse or domestic violence. If participants remain unemployed after completing JSS for the first time, they may choose to participate in the program again.
Features of the Study
The study included 13,626 women who entered and exited WorkFirst from the second quarter of 1998 to the fourth quarter of 2000. The study participants were ages 25 to 35 and were receiving welfare. Among the unemployed women, 21 percent were married, 70 percent were white, 10 percent were Hispanic, and 14 percent had completed high school. The unemployed women in the sample had, on average, 2.4 children and had been receiving Aid to Families with Dependent Children or Temporary Assistance for Needy Families for an average of 23 quarters. Among the employed participants, 14 percent were married, 63 percent were white, 14 percent were Hispanic, and 15 percent had completed high school. The employed women had on average 2.4 children and had been receiving Aid to Families with Dependent Children or Temporary Assistance for Needy Families for an average of 24 quarters.
The authors compared employment and earnings outcomes of women who participated in JSS zero times, one time, or more than one time. Of the 13,626 women in the study, 4,791 women did not participate in JSS, 5,100 women participated once, and 3,735 women participated more than once. The authors analyzed the effect of participating in JSS one or more times relative to what an individual would have experienced without participating in JSS using an econometric technique that can examine employment and earnings measures collected over time (a Type II Tobit model). The authors did not specify the source of the data used in the study.
Findings
- The study found that the first instance of participation in JSS significantly increased the likelihood of employment by 4 percent for women seeking employment. Subsequent JSS participation had no additional significant impacts on employment.
- The study also found a positive and statistically significant relationship between quarterly earnings and participation in one or more JSS for employed women. The impact of JSS increased quarterly earnings by, on average, 40 percent for participation in the first JSS and by 31 percent, on average, in quarterly earnings for the second and third JSS.
Considerations for Interpreting the Findings
The main empirical approach employed by the authors compared the outcomes of participants who participated in JSS one or more times, thereby examining differences for individuals over-time as well as differences across individuals at any given time. The study identified the effect of an individual’s participation in JSS (1, 2, or 3 or more times) relative to the outcomes an individual would have experienced without participating in JSS. However, the statistical analysis did not account for employment or earnings more than one year prior to joining the Workfirst program. Consequently, even accounting for study members’ characteristics, participation in JSS, and current employment and earnings, may still be strongly influenced by past employment and earnings.
Causal Evidence Rating
The quality of causal evidence presented in this report is low because the authors did not ensure that the groups compared exhibited similar employment or earnings outcomes before the intervention. This means we are not confident that the estimated effects are attributable to JSS; other factors are likely to have contributed.