Last week, I wrote about state diversion programs. These provide an alternative to traditional welfare, helping to solve the short-term and immediate needs of welfare applicants, while keeping them out of the long-term welfare system. This week, I want to look at the evidence showing how successful these alternatives can be.

Unfortunately, there is little recent scholarship. Between 1998 and 2012, there were some 16 studies evaluating the effectiveness of diversion programs. In the decade since, however, there has been only one—and that a very narrow study looking only at diversion into a single alternative government program, Parents and Children Excel. We are left with a limited number of studies, all of which are significantly dated. Those early studies largely looked at individual state programs and are difficult to compare because of variations in population size and program characteristics, as well as the follow-up time among other things. Researchers have also noted that states did not devote substantial resources to tracking diversion participants.

The decade following passage of the Personal Responsibility and Work Opportunity Act was a period of significant experimentation with welfare reform, when other changes were being implemented that had an impact on participation rates. It was also a period of economic growth and low unemployment, factors that historically have reduced welfare use. From 1992 to 2005, for instance, there was an overall 56 percentage point reduction in the number of people receiving Temporary Assistance for Needy Families (TANF). Teasing out which factors resulted in welfare caseload and/​or poverty reductions has been the subject of much debate. However, the consensus appears to suggest that policy changes are responsible for about a third of caseload changes, with the rest attributable to economic conditions or other factors.

How one views the interaction between welfare reform and the overall economy may be particularly relevant to evaluating the effectiveness of diversion programs. A 2008 study published in the Journal of Policy Analysis and Management concluded that diversion programs, particularly lump sum programs, are extremely sensitive to macroeconomic factors. Other factors that were found to impact the success of diversion programs were the availability of supportive services such as childcare and earnings disregards. The composition of the applicant pool also played an important role, with job-ready applicants who had not previously been on welfare, most likely to benefit from diversion programs. Therefore, we should be careful about overstating the results. Still, it is worth noting that most of those earlier studies showed positive outcomes, with recipients generally less likely to rely on welfare and more likely to be part of the labor force, without increasing poverty.

For example, a 1998 study of programs in Maryland, Utah, and Virginia found that 96 percent of lump sum recipients in Maryland, 75 percent in Utah, and 85 percent in Virginia had not applied for cash assistance within a year. Another Maryland study which was conducted in 2002 for the University of Maryland’s School of Social Work compared outcomes for recipients participating in both lump sum payments and employment search requirements. Researchers found that both groups of participants were more likely to be in the labor force in the year following than welfare recipients; however, those who participated in the job search program were slightly more likely to be employed a year out than those who received lump sum benefits. Overall, diversion was found to reduce welfare spending while producing generally positive outcomes for participants and “no obvious warning flags.”

The Department of Health and Human Services conducted a review of diversion programs in Denver and Illinois. Only ten percent of participants in Denver’s program were found to have gone on welfare subsequent to diversion, compared to 19 percent of TANF leavers overall. In Illinois, work participation rates increased substantially, from 51.6 to 57.5 percent after the adoption of the state’s Work Pays Earned Income Disregard program, which included a diversion component. A 2005–2006 report on Tennessee’s program from the University of Tennessee’s Center on Business and Economic Research found that 80 percent of participants remained off TANF within 18 months of receiving diversion assistance.

And, while not as academically rigorous as other studies, an investigation of Minnesota’s job search diversion program, which also contained lump sum elements since the state paid rent and utility bills during the job search, by the Minneapolis Star-Tribune found that 49 percent of participants found employment before they would have been enrolled in the traditional TANF system.

Not every study yielded uniformly positive results. A study of Oregon’s lump sum payment program found that two-thirds of participants did not return to the welfare rolls within 19 months. However, that was a slightly higher recidivism rate than TANF recipients who left the rolls for other reasons. And a 2002 study by the Urban Institute found that 17–34 percent of diversion participants would have been financially better off by remaining in traditional TANF. However, that still meant that the overwhelming majority of participants fared as well or better under diversion. Perhaps the worst outcomes were found by Rebecca London in 2003. She concluded that diversion was associated with lower rates of employment and higher rates of subsequent food stamp and Medicaid utilization. Her study does appear to be an outlier, however.

Clearly, additional research would be extremely valuable. Opportunities and data may be limited because states have neglected these programs and utilization is extremely low. Still, scholars and state welfare agencies should take up the challenge and determine whether the earlier evidence of success still holds.

As noted above, even in states that have diversion programs on their books, utilization is very limited. Next week, we will examine why that is the case, and how both state and federal governments can remedy this.