A new study from Ben Sommers and other researchers at Harvard University finds that Medicaid work requirements fail to promote employment but do result in more people losing their health coverage and may promote other negative health outcomes. The study, published in Health Affairs, found that negative economic consequences ensued as well – with medical debt rising.

Proponents of work requirements, led by CMS Administrator Seema Verma, have insisted they will promote better health and higher rates of employment. Both of these unfounded claims are refuted by the study of experience in Arkansas, the only state in the nation to have implemented Medicaid work requirements. Federal courts have intervened to stop the implementation of Medicaid waivers promoting work requirements and imposing other limits on coverage.

What did the study find?

More people in Arkansas became uninsured. For adults in Arkansas between ages 30-49, work requirements were associated with a significant increase in the uninsured rate relative to other age groups and other comparable states (TX, LA, and KY). However, after the 2019 court decision, which halted the requirements, these coverage losses were largely reversed. These findings confirm what many, including us, feared – that the requirements would result in the loss of coverage for beneficiaries’ subjected to the ill-conceived policy.

Individuals who lost Medicaid or Marketplace coverage were faced with large medical bills that they couldn’t pay and experienced adverse outcomes. Almost half of Arkansans ages 30-49 who lost coverage reported serious problems paying off medical debt. Over 55 percent of these individuals delayed needed care because of cost while almost two-thirds delayed taking medications because of cost — all three of these indicators were significantly higher than those in Arkansas who maintained Medicaid or Marketplace coverage. The median medical debt for those who disenrolled was $1,001 compared to $0 for those who did not disenroll.

The study found no evidence that the work requirements had any effect on employment or community engagement. The authors did not find any significant changes in number of hours worked nor that employment or community engagement increased among low-income adults when the work requirements were in effect or after the policy was blocked. In fact, more than 95 percent of individuals in the policy’s “target population” qualified for an exemption or were already meeting the work requirements through work or community engagement.

The majority of Arkansas residents were unsure about the current status of work requirements, or incorrectly believed the policy was still in effect. This is a key finding, since the notion that you can change people’s behavior by threatening them with the loss of their health coverage is the foundation upon which work requirement policy is based. That is a flawed theory for many obvious reasons, but taking it at face value, the policy was a failure as 70 percent of all Arkansas residents were not sure whether work requirements were in effect. Of those that had heard of work requirements, 40 percent incorrectly said the policy was in effect while almost 53 percent were unsure if the policy was in effect. Only 65 percent of individuals ages 30-49 in AR (the targeted population) with Medicaid or Marketplace coverage had heard of work requirements.

This study adds to the growing body of evidence (examples from AR, KY, and NH)  showing that work requirements do not promote work, but rather cause substantial harm to beneficiaries and create confusion.