Instead of Work Requirements, Why Not a Jobs Guarantee?

Paying people to work would be a more direct—and less expensive—way to lift them out of poverty.

Centers for Medicare and Medicaid Services Administrator Seema Verma at a meeting with President Trump
Centers for Medicare and Medicaid Services Administrator Seema Verma (Evan Vucci / AP)

The Trump administration has been signaling for months that it plans to  implement conservative reforms to core federal welfare programs, including by allowing states to have work requirements for Medicaid. So it was no surprise on Thursday when the Centers for Medicare and Medicaid Services issued guidance for “state efforts to test incentives that make participation in work or other community engagement a requirement for continued Medicaid eligibility.”

In a series of tweets announcing the policy shift, CMS Administrator Seema Verma explained the agency’s rationale that requiring eligible able-bodied adults to have jobs to qualify for Medicaid will make them healthier and less reliant on welfare in the future. “Our fundamental goal is to make a lasting and positive difference in the health and wellness of Medicaid beneficiaries,” she tweeted. She also cited a 2014 meta-analysis that concluded that “employment is beneficial for health, particularly for depression and general mental health.”

There’s one gaping hole in that reasoning, however: If the federal government really wants to harness the benefits of employment to improve beneficiaries’ health and wealth, wouldn’t providing them with jobs be a simpler and more effective answer?


So far, it’s unclear how widely adopted work requirements will be and how exactly states will implement them under CMS’s new guidance. On Friday, Kentucky was the first state to have its 1115 waiver creating work requirements approved by CMS. On Thursday, Verma noted that nine other states had already submitted waivers asking the federal government to approve incentives or requirements for some Medicaid beneficiaries. In addition to allowing strict job mandates, CMS will also allow requirements for “other community-engagement activities,” including volunteering, job training, and caregiving. (These rules only apply to specific adults; CMS carves out people with disabilities, the elderly, children, and pregnant women.)

1115 waiver applications from Kentucky, Maine, and Wisconsin provide examples of what constraints on welfare programs might look like at their strictest. Kentucky’s—which Verma helped create when she was a health-care consultant—applies to all able-bodied non-elderly adults, and requires them to complete 80 hours per month of work, jobs training, education, or community service. It also requires recipients to pay an income-adjusted premium or risk losing coverage for up to six months. Wisconsin wants to require all adults without dependent children who live in poverty to work or complete community-engagement activities in order to qualify for Medicaid. Maine would require even some parents in poverty to work for their benefits.

Yet if states want work requirements to increase the health and self-sufficiency of Medicaid beneficiaries—their stated goal—most available data suggest they’ll fall short. As the Kaiser Family Foundation reported in 2017, most people on Medicaid who can work do work. Around 60 percent of adult enrollees have a job, and for the most part those who don’t report impediments in their ability to work. Even those who are not officially disabled often attest to having debilitating conditions—like severe back problems—that make full-time jobs difficult or impossible. Others may be in school, work as primary caretakers for loved ones, or may have retired.

Instead of spurring employment among the sliver of people left able and wanting for jobs, it’s more likely that the requirements will simply push those people who can’t work out of the program—leading to fewer people on Medicaid overall. Evidence suggests that a work requirement in the 1996 welfare-reform bill, within the Temporary Assistance for Needy Families cash-transfer program, led to a reduction of millions of qualified TANF recipients over time, even as exogenous factors increased employment generally among people in poverty. Indeed, the reform likely pushed those with impediments to work into even deeper rungs of poverty.

If those effects were repeated in Medicaid, it could prove disastrous for the health of the program’s beneficiaries. Especially in states that expanded Medicaid under Obamacare, work requirements could create a new underclass of people ineligible for any health insurance. That includes a large contingent of people with disabilities who don’t qualify for Supplemental Security Income and vulnerable populations like young men with felonies. Caught in a vicious cycle, those people would then be less healthy and less financially secure, and thus less likely to be able to work and make it out of poverty.

To be sure, Medicaid is an incomplete on-ramp to prosperity, and joblessness and financial instability do indeed plague people on welfare. But that’s why they’re on welfare in the first place. And while the research Verma cites is fundamentally correct—stronger employment does lead to stronger mental-health outcomes, which are, in turn, key factors in improving general well-being and financial stability—CMS’s synthesis of that research is utterly paradoxical. Why would a proposal to better the health and wealth of millions of people include a policy for which there’s a strong possibility people will lose access to health care?

The body of evidence suggests that another policy would do a much better job at making Medicaid a more robust all-purpose poverty program, a goal that CMS says it wants to accomplish: a jobs guarantee, perhaps accompanied by a basic income for enrollees.

Scholars such as William Darity of Duke University and Darrick Hamilton of the New School have written at length about the prospect of a federal jobs guarantee, and the proposal recently gained steam on the political left thanks to a 2017 report from the Center for American Progress. According to my colleague Annie Lowrey, a federal jobs guarantee—one that creates full employment and pays a minimum of $15 an hour, with benefits, to prime-age people without college degrees—would cost somewhere north of $158 billion a year for 4 million jobs. While that’s a large amount of money, it’s only a fraction of the budgets of several large federal programs, including Medicaid.

Such a program would have its most drastic effects on wages, productivity, and reducing racial and class-based wealth inequality if it were implemented as a universal program. But it could probably achieve CMS’s goals of long-term health benefits and poverty reduction if it were instituted solely for current Medicaid beneficiaries. If the 4.4 million non-elderly adults who aren’t working; aren’t caregivers, retired, or students; and don’t qualify for disability insurance are used as a floor, providing jobs for them would cost a little more than Lowrey’s total of $158 billion, around 30 percent of Medicaid’s annual budget of over $550 billion. If people who self-report as ill or disabled are excluded from that number, Medicaid would need to pay for a maximum of 880,000 jobs, or $35 billion a year, 6 percent of the annual Medicaid budget.

A jobs guarantee—provided perhaps through reimbursements for state programs or under a federal umbrella—would be useful for the states currently requesting the ability to implement work requirements. For example, Pennsylvania’s unemployment rate of 4.6 percent still hovers above the national average, and features hot spots like the city of Erie, where 47 percent of the black population lives at or under the poverty line and 25 percent of that population is unemployed. And even in states where the unemployment rate is lower than average, stubborn and wide disparities in employment between white people and people of color—who are usually disproportionately found on Medicaid rolls—persist.

More broadly, Medicaid already is a jobs program. Obamacare expansion notwithstanding, Medicaid enrollment and spending already track with unemployment, and spending spikes when unemployment does. So it stands to reason that the reverse might hold true: If Medicaid itself can provide jobs, it might decrease the unemployment rate, and thus the number of people who need Medicaid in the first place. Although one of the more common refrains against expanding Medicaid to all low-income adults—and indeed a common refrain against poverty programs generally—is that its rich benefits will disincentivize work, four years into the second-biggest Medicaid spike in history, unemployment rates are low and the unemployment rate for black people is at an all-time low.

In fact, for many people Medicaid seems to make it easier to work. A recent study by researchers Jean P. Hall, Adele Shartzer, Noelle K. Kurth, and Kathleen C. Thomas in the American Journal of Public Health indicates that access to Medicaid’s supports and services actually makes employment possible for many of its beneficiaries with disabilities. The expansion itself—by way of dramatically enlarging the health-care services sector—has also increased economic activity and increased the number of jobs available. In short, it’s reasonable to view Medicaid as both a personal and a macroeconomic stimulus as much as it is a safety net.

A Medicaid jobs guarantee could serve to amplify both of those roles. It could essentially set a wage floor for Medicaid enrollees, who often work near the bottom of the wage scale and often barely crack the poverty line even while working full-time hours (or more). Integrating Medicaid into bespoke job structures for people with disabilities could provide transportation and rehabilitation, and further increase the accessibility of those positions, thus creating more synergy between health and employment.

Such a program could still provide the training and community-engagement opportunities that have proved successful in getting people employed, serving as a launching pad for vulnerable parents with children in deep poverty, men with felonies who face recidivism, and people who struggle with mental-health and substance-abuse issues. But it could do so by replacing stigmatization with empowerment, and without encumbering people with exhaustive and even discriminatory job searches.

Similar to how employer-sponsored insurance has become a backbone to the economic growth of the middle class, a jobs guarantee for Medicaid would take the largest health-insurance program in America and transform it into a nexus of anti-poverty policy and health equity. Put more simply: The easiest way to make sure people receive the health benefits of employment could be to employ them.