On January 11, 2018, the Centers for Medicare and Medicaid (CMS) sparked immediate controversy by announcing new policy [PDF] for states seeking §1115 Medicaid waivers. For the first time, CMS invited states to impose work or related requirements (job training, job searching, education, volunteering, or caregiving) as a condition of eligibility for Medicaid expansion benefits. The next day, CMS approved Kentucky's 2016 waiver application, which included work requirements for some beneficiaries. The new guidance raises both legal and practical questions about the purpose of Medicaid, the relationship between work and health, and the scope of administrative discretion.
CMS appears to have two goals: reducing federal and state spending on Medicaid; and reframing Medicaid as a "welfare" program like Temporary Assistance for Needy Families (TANF) and Supplemental Nutrition Assistance Program (SNAP). TANF and SNAP are income-support programs with statutory authorization for work requirements. Medicaid, in contrast, is designed to pay providers for medical care for those who cannot pay for it and does not mention work. Work requirements can reduce government spending, because they tend to force many enrollees out of assistance programs for failing to meet complicated eligibility requirements. Yet, work requirements are popular with those who believe that the poor are idle freeloaders and should "earn" public assistance. If all such programs are framed as part of unearned charity from taxpayers, the stage may be set to consolidate funding for all into a single block grant to the states. Such funding makes a tempting target for budget cuts, leaving fewer benefits for fewer beneficiaries.
The new policy signals that CMS is open to restricting eligibility for Medicaid - reversing its historical preference for waivers that expand eligibility and cover additional benefits or more efficient models of service delivery. Congress has not changed the definitions of the populations eligible for Medicaid nor the statutory provisions for waivers. A threshold legal question, therefore, is whether CMS has the statutory authority to approve new limits on eligibility.
Section 1115 of the Social Security Act [42 U.S.C. §1315] authorizes CMS to waive some statutory requirements (in 42 U.S.C. §1396a) to allow states to conduct an "experimental, pilot or demonstration project which, in the judgment of the Secretary, is likely to assist in promoting the objectives" of Medicaid. These objectives are stated in the statute's first section authorizing the appropriation of federal funds "[f]or the purpose of enabling each State . . . to furnish (1) medical assistance on behalf of families with dependent children and of aged, blind, or disabled individuals, whose income and resources are insufficient to meet the costs of necessary medical services, and (2) rehabilitation and other services to help such families and individuals attain or retain capability for independence or self-care." It also requires that medical "assistance shall be furnished with reasonable promptness to all eligible individuals.
The statute gives CMS considerable discretion (as delegated from the Secretary of Health and Human Services). Regulations add no substantive limits. [42 C.F.R. §§ 431.400-431.416] But the demonstration project must both (a) serve the purpose of Medicaid and (b) be conducted as an experiment whose results can be evaluated.
Work and Health
The CMS policy asserts that work requirements can improve health, because people who are employed generally have better health than those who are unemployed. It draws on the social determinants of health literature, which does support a correlation between employment and health status. But, as statisticians insist, correlation is not causation. Causation is more likely to go in the opposite direction: good health enables people to work. A study [PDF] of programs in Medicaid expansion states that allow persons with disabilities to remain enrolled while earning more than the Medicaid income limit suggests that Medicaid coverage helps them to become employed. By itself, however, employment does not necessarily improve health. Indeed, some occupations pose substantial health and safety risks. Lack of employment, as well as very low wage jobs, typically lead to poverty, which is a much stronger risk factor for poor health.
CMS implies that work requirements will also enable financial independence. But Medicaid's purpose is to "furnish medical assistance" and "rehabilitation and other services," not economic independence. The statute specifies "rehabilitation and other services" to help people - primarily those with disabilities - attain independent physical or mental functioning so that they are not institutionalized. A few court decisions have upheld work requirements for TANF, but those decisions depend on a wholly different statute and have no precedential value for Medicaid. Moreover, historical evidence suggests that work requirements do not significantly increase employment. Income earned by TANF enrollees required to work is rarely enough to rise above the poverty level. Many can find only low-wage or part-time jobs without health insurance or other benefits.
A practical question is whether proposed new eligibility requirements are likely to affect enough Medicaid enrollees to significantly reduce state expenditures. Kentucky seeks to reduce its Medicaid payment by $2.6 million over five years by moving almost 95,000 people from Medicaid into private commercial health insurance plans. Kentucky has a much greater chance of achieving disenrollment than private coverage or substantial cost savings, because most low-wage jobs do not offer any (or any affordable) health insurance benefits.
The proportion of the Medicaid population that would be subject to work or community engagement requirements is small and does not generate significant costs. The new policy would exempt non-elderly, non-pregnant adults, as well as students and those caring for dependents. According to a Kaiser Family Foundation analysis, 63% of Medicaid dollars pay for the care of elderly (21%) and disabled (42%) individuals, even though these groups together comprise less than 25% of the Medicaid population. (Long term care alone consumes about 21% of national Medicaid expenditures.) In contrast, children are almost half (48%) the Medicaid population, but account for 21% of Medicaid costs. Adults, including pregnant women, make up the remaining 27% of the population, accounting for 15% of total costs.
Thus, only a small fraction of a state's Medicaid population is likely to qualify for work requirements. Among the 24.5 million non-elderly, adult Medicaid beneficiaries not on SSI in 2016, 42% were working full-time and 18% part-time, while 14% were ill or disabled, 12% were family caregivers, and 6% were in school. That left 7% of a less costly population who were not working for other reasons. It is likely that many of those want a paid job, but were unable to find or hold one, for lack of education, relevant skills, stable housing, transportation, or childcare.
Most successful welfare-to-work programs followed a "human capital development" approach, offering education and training. Even these have had limited success, increasing employment by only a few percentage points. It is unclear how the states would help this group. Federal Medicaid dollars do not pay for non-medical services, so the states would have to find state or private financial and human resources to provide job training and search assistance.
The new CMS policy also encourages state waiver plans to include additional eligibility requirements beyond work or community engagement. Kentucky's waiver, which is being challenged in the federal district court for the District of Columbia, requires enrollees to "earn" medical benefits that are no longer covered by engaging in healthy behaviors, avoiding the emergency department for non-emergencies, and complying with work requirements. Kentucky enrollees must also report any changes in their work/service hours and income within ten days, which can be challenging, especially for people with variable schedules and limited internet access. Paperwork requirements are understood to make compliance difficult and result in disenrollment of eligible, as well as ineligible, individuals. As Gail Wilensky writes, there is no dignity in periodically having to prove you are still poor enough for help.
Kentucky also now requires that enrollees pay premiums for Medicaid, which can be up to 4% of household income. Since the target population has income just below or above the federal poverty level (FPL: $12,140 - individual; $25,100 - family of four), premiums can eat up the rent and food budget. The national median income eligibility level for adult Medicaid beneficiaries is 44% FPL. Yet failure to pay renders a Kentucky enrollee over 100% FPL no longer eligible (locked-out) for 6 months.
A practical and legal issue is whether states applying for waivers will conduct real experiments that can be evaluated, as required by the statute, regulations, and the new policy. [42 U.S.C. §431.424] Most important, an experiment must test whether it furthers Medicaid's purpose: access to "medical assistance." Employment is a means to that end, not the goal. Well-designed evaluations are necessary to confirm or reject assumptions about the value of the new eligibility criteria, as well as how many people are affected. It may be difficult, however, to find people who were removed from Medicaid for noncompliance to compare their results.
We don't need to wait to predict the results of such "experiments." Tennessee's experience with reducing Medicaid enrollment left the former beneficiaries in worse financial condition and health. In contrast, the Oregon Health Insurance Experiment found that eligible persons selected at random to enroll in Medicaid improved their financial condition by avoiding catastrophic medical bills, compared with eligible persons not selected for enrollment. Rates of depression also decreased, which may enable people to perform better in all facets of life, including employment. A recent review of studies concludes that health insurance coverage saves enrollees money and improves access to care, as well as adherence to medications. Improvements in health outcomes, especially for chronic diseases, have been harder to measure, perhaps because of the time required to alter the trajectory of such conditions.
Section 1115 waivers give CMS and the states a great deal of flexibility to adapt their programs to more modern and efficient structures. That is the value of administrative discretion. Yet administrative discretion can be abused. When exercised for budgetary or ideological reasons untethered to evidence, it undermines the agency's credibility. It seems clear that the purpose of these new waiver criteria is not to "furnish medical assistance," but to reduce Medicaid enrollment and federal and state spending. The results are likely to cause unnecessary suffering. If this is found to be within CMS's authority, it may be time to rethink the scope of its statutory authority.
Professor Wendy Mariner's research focuses on laws governing health risks, including social and personal responsibility for risk creation in conceptions of insurance, as well as national health systems, including the Affordable Care Act and ERISA, health information privacy, and population health policy. She has published more than 100 articles in the legal, medical and health policy literature on patients and consumers' rights, health care reform, insurance benefits, insurance regulation, public health, AIDS policy, research with human beings, and reproductive rights, and co-authored the law school textbook, Public Health Law (with Ken Wing, George Annas, and Dan Strouse). Professor Mariner has served on state, national, and international boards and commissions, including the Massachusetts Health Facilities Appeals Board, the Massachusetts Health Care Quality and Cost Council Advisory Committee, the National Institutes of Health's AIDS Policy Advisory Committee, Institute of Medicine Study Committees, the CIOMS/WHO Steering Committee for the International Ethical Guidelines for Biomedical Research Involving Human Subjects, and the Executive Board of the American Public Health Association.
Suggested citation: Wendy K. Mariner, Medicaid Waivers, Work Requirements, and Administrative Discretion, JURIST - Academic Commentary, Feb. 3, 2018, http://jurist.org/forum/2018/02/Wendy-Mariner-medicaid-waivers.php
This article was prepared for publication by Henna Bagga, an Assistant Editor for JURIST Commentary. Please direct any questions or comments to her at