Some Republican policymakers continue to propose basing eligibility for assistance programs on participants’ ability to meet strict work requirements — most recently with House Agriculture Committee Chairman Michael Conaway’s proposal to reauthorize SNAP (formerly food stamps)[1] — despite a lack of credible evidence that the requirements would work as intended.[2] To build support for work requirements that take away assistance from adults who cannot work a set number of hours per month, conservative policymakers are pointing to three methodologically flawed studies touting the policy’s alleged success in Kansas’s and Maine’s cash and food assistance programs. The three studies misrepresent or omit key findings, and in many instances, make inappropriate claims about the impact of work requirements on work and earnings that the facts do not support.
Our analysis of the same data sharply contradicts the studies’ findings. We found that many adults in these programs already worked or would likely work soon anyway, but many of them found it difficult to find steady work and had earnings far below the poverty line or would have otherwise still qualified for assistance after their exits from the program.
Renewed attention to these flawed studies began last week when Chairman Conaway unveiled his 2018 farm bill calling for sweeping, aggressive SNAP work requirements that would likely prove unworkable,[3] along with a White House executive order[4] that cited these studies while encouraging similar work penalties in other assistance programs.
Background on the Studies and Their Key Flaws
The Foundation for Government Accountability (FGA) — an organization that actively promotes onerous work requirements in SNAP and Medicaid nationwide — authored two of the three flawed studies. One study presented employment and earnings data for a group of non-disabled, childless adults who lost SNAP assistance after Kansas re-imposed a harsh three-month time limit for people not working at least 20 hours a week. The second study assessed earnings data for a group of adults who exited Temporary Assistance for Needy Families (TANF) due to a work sanction after Kansas made it more difficult for families to qualify for, maintain, and regain cash assistance. State officials in Maine authored the third study, which presented employment and earnings data for a group of non-disabled, childless adult SNAP recipients subject to the return of the three-month time limit. [5]
The studies have at least four key flaws.
Claiming or suggesting that work requirements caused employment and earnings to increase requires a counterfactual — that is, a comparison of families that were subject to the new work sanction policy to an otherwise similar comparison group that is subject to the old work sanction policy. Differences between the two groups then could be attributed only to the policy change. Because all families were immediately subject to the harsher work sanctions, FGA and Maine state officials can’t make such a comparison. Given that families often turn to SNAP and TANF at times of crisis or other serious need — that is, when earnings are at a low point — it is not surprising that after leaving the programs some are able to get back on their feet to some extent and secure jobs or increase their earnings. Further, the states’ economies were improving during the times these studies covered, so we would expect earnings gains simply due to the strengthening economy, especially for those with low skills and education.[6]
Many program participants already work, or would likely work soon even without the work requirement. The two SNAP studies misleadingly emphasize only what happened after the time limit took effect, omitting the finding that work rates were similar before and after sanctioned adults exited the programs. The employment data in the Kansas SNAP study are particularly distorted, presenting the share of exiting recipients who ever worked in a quarter over the post-exit year, thereby showing a cumulative increase in each successive quarter. Rather than showing an isolated impact of the policy change, this captures typical movement in and out of the labor force. This group tends to work in high-turnover jobs: in any quarter some people lose jobs and some get new jobs, so the share that ever worked increases. The trends in the share who ever worked likely followed a very similar pattern in earlier years when the time limit was not in effect (though the authors do not present such data).[7] In the TANF study, the authors didn’t present exiting families’ employment outcomes, but based on our analysis, work rates were essentially unchanged in the year before and after exiting TANF due to a sanction.
As such, the data contradict the premise underlying work sanctions — that it is necessary to take away people’s food and cash assistance for an extended period to encourage them to work. Rather, the data suggest that the barriers to work that many work-sanctioned adults face, such as lower education and volatile work schedules,[8] remain after they leave SNAP and TANF.
The SNAP studies claim that, among adults who had their food assistance taken away due to the time limit, total earnings increased by 127 percent and 114 percent in Kansas and Maine, respectively, in the year after exit. Again, the studies are misleading because they claim that the work requirement caused the earnings increase when the observed data cannot tell us why the increase occurred. Given that people often turn to SNAP at times of need, it is not surprising that after leaving the program some later get back on their feet to some extent and secure jobs or increase their earnings. These earnings figures are also highly misleading because they do not include the loss of SNAP benefits, and in the case of the Kansas findings, are based on a low comparison point or artificial baseline. When lost benefits are accounted for, total income did not increase substantially in Maine and actually decreased slightly in Kansas.
The TANF report’s claims that sanctioned parents’ incomes more than doubled within a year after exit and more than tripled within four years after exit are particularly egregious. These claims are based on an inappropriate extrapolation and grossly overstate the earnings increase observed in the data. Further, by only presenting aggregate earnings, the report masks the fact that parents exiting TANF because of work sanctions had earnings far below the poverty line. For the parents exiting TANF due to a work sanction for whom four years of post-exit data are available, nearly 7 in 10 had either no earnings or deep-poverty earnings in the fourth year after exit; only 17 percent had incomes above the poverty line that year.
The studies fail to discuss the circumstances of the individuals who are subject to these work requirements and the consequences for increased hardship when they lose the support that SNAP and TANF provide. Without addressing this side of the equation, the studies misrepresent the effect that these punitive policies have on the well-being of those cut off from SNAP and TANF. They do not discuss or attempt to assess what happens to individuals who lose their assistance and are unable to find employment, meaning they are left without either food or cash assistance even though they have no income. Further, considerable research suggests that extremely low family income is likely to have long-term negative consequences for the health and development of the children living in families exiting due to a sanction in TANF.[9]
Despite work requirement proponents’ claims, the reports about the experiences in Kansas and Maine do not provide persuasive evidence that such requirements for adults in TANF and childless adults in SNAP increase work effort or well-being. Policymakers should not draw conclusions about work requirements or time limits based on these faulty assessments of the data.
