Commentary: Administration Can’t Justify Re-Approving Waiver Taking Coverage Away from Kentuckians

After a federal court vacated the Trump Administration’s approval of Kentucky’s Medicaid demonstration project (or “waiver”) to impose onerous new requirements on most beneficiaries because it failed to consider the proposal’s impact on coverage, the Administration opened a new 30-day comment period on the proposal. Yet even before that comment period ended on August 18, Administration officials signaled their intent to again approve Kentucky’s proposal,^[1] despite comments mostly in opposition to the plan filed by over 11,000 individuals and organizations. The Administration should reconsider, given the overwhelming evidence in favor of rejecting the proposal.^[2]

The Administration should reconsider, given the overwhelming evidence in favor of rejecting the proposal.Section 1115 of the Social Security Act gives the Secretary of the Department of Health and Human Services (HHS) authority to approve demonstration projects that promote the objectives of Medicaid. HHS’ approval of Kentucky’s waiver proposal was “arbitrary and capricious,” the court found, because providing affordable coverage to people who would otherwise be uninsured is a “central objective” of the Medicaid program, and HHS failed to consider the proposal’s impact on coverage. In signaling their intent to re-approve the waiver, Administration officials reportedly said “they could sidestep the ruling by providing a better explanation of the rationale for work requirements,” and “by compiling a fuller record and showing that they have thoroughly reviewed the evidence.”^[3]

It’s not that simple. In vacating HHS’ approval of the waiver, the district court judge noted that an agency’s action is arbitrary and capricious if it “entirely failed to consider an important aspect of the problem, offered an explanation for its decision that runs counter to the evidence before the agency, or is so implausible that it could not be ascribed to a difference in view or the product of agency expertise.”^[4] That’s not a standard the Administration can sidestep. HHS can’t get around the substantial evidence that large numbers of Kentuckians — including many who should remain eligible under the proposal — would lose coverage and become uninsured under the state’s plan.

Kentucky’s own estimate is that the waiver would lead to 1.14 million lost coverage months, equivalent to a 15 percent drop in adult Medicaid enrollment, by the fifth year. These lost coverage months are equivalent to 95,000 people losing coverage in a typical month, meaning that well over this number would likely lose coverage for some part of the year due to lockouts for failing to meet work requirements, pay premiums, or report changes or renew coverage in a timely manner. Early evidence from Arkansas, which began implementing its work requirement in June, suggests the coverage loss in Kentucky could be even greater.^[5]

The substantial loss of coverage that would result from Kentucky’s plan to take Medicaid coverage away from those not meeting work or other program requirements can’t be minimized, and it can’t be excused or justified by claims that those affected will get other coverage or benefit in some other way from approval of Kentucky’s waiver. Whatever rationale the Trump Administration tries to advance, it can’t overcome or legitimize the substantial loss of coverage and resulting harm Kentucky’s proposal would cause.

Substantial Numbers of Medicaid Beneficiaries Would Lose Coverage, Become Uninsured

The percentage of uninsured Kentuckians with incomes under 138 percent of the poverty line dropped from 40 percent in 2013 to 7.4 percent at the end of 2016. Much of that reduction was driven by large coverage increases under the Affordable Care Act (ACA) in high-poverty areas, which eliminated pre-ACA disparities in uninsured rates across these areas.^[6] And coverage mattered to those who obtained it. A multi-year study comparing Kentucky and Arkansas, which expanded Medicaid under the ACA, to Texas, which hasn’t, found that by the end of 2016, low-income adults in Kentucky and Arkansas received more primary and preventive care, visited emergency departments less often, and reported better health than low-income adults in Texas. They also saw big savings in out-of-pocket medical costs.^[7]

Implementation of Kentucky’s waiver proposal would lead to a sharp reversal of the state’s progress in reducing uninsurance and improving the health of poor and low-income Kentuckians. Our conservative estimate, relying on research by the Kaiser Family Foundation and the Urban Institute, is that the work requirement would cause between 45,000 and 103,000 enrollees to lose coverage well before the fifth year of implementation, with most of them likely becoming uninsured.^[8] This estimate does not take into account coverage loss from the lockouts that the proposal would impose on people who don’t pay premiums, renew their coverage on time, or report changes that affect their eligibility.

Work Requirements Would Cause Substantial Numbers to Lose Coverage

Increased red tape would cause many working people and those qualified for an exemption to lose coverage. People would have to report their hours by phone or through an online portal, but 19 percent of adult enrollees in Kentucky lack any internet access, and 41 percent don’t have broadband access, according to an Urban Institute analysis.^[9] Many people who should qualify for an exemption likely won’t get one. For example, families sanctioned due to noncompliance with requirements under the Temporary Assistance for Needy Families (TANF) program were more likely than other families receiving TANF to have barriers that kept them from working and that should have made them exempt from the requirements, including having a child with a chronic illness or disability.^[10]

Claiming an exemption from the work requirement would be particularly difficult for people with mental illness. As Harvard Medical School Professor Richard Frank explains, “The burden of proving medical frailty in the Kentucky waiver program will generally fall on the recipient. In this case, that means it falls on a person with an illness that interferes with cognition, executive function and mood.”^[11] People with substance use disorders, many of whom should be exempt, would also have compliance problems, because they often have significant privacy concerns and may not trust Medicaid eligibility staff with information about their current or past substance use. People may also fear criminal ramifications if they are using illegal substances.^[12]

Moreover, many working people won’t be able to meet the 80-hour-a-month requirement every month. About a third of the 165,000 Kentucky adults receiving Medicaid who are working and not eligible for an exemption as a student or caregiver would be at risk of experiencing gaps in Medicaid coverage, because they don’t work enough hours or don’t work consistently enough throughout the year, according to the Urban Institute.^[13]

Premiums and Lockouts Would Also Cause Coverage Loss

Other provisions of the proposal have an even broader reach than work requirements, putting coverage at risk for parents and students who are exempt from the work requirement. Almost all adults would have to pay monthly premiums as high as 4 percent of monthly income. If people with incomes below the poverty line don’t pay their premiums within 60 days, they would be enrolled in Medicaid, but they would have to pay copayments for most health care services. People with incomes above the poverty line who don’t pay would not be enrolled and would remain uninsured. And people with incomes above the poverty line who do make a first payment will lose coverage for at least six months if they miss a premium payment and don’t pay it within 60 days.

Extensive research shows that premiums significantly reduce low-income people’s participation in health coverage programs.^[14] These studies show that the lower a person’s income, the less likely they are to enroll and the more likely they are to drop coverage due to premium obligations. People who lose coverage most often end up uninsured and unable to obtain needed health care services. Kentucky’s proposal is modeled on the Healthy Indiana Plan, where the state’s own evaluation shows premiums have kept large numbers of people from becoming eligible and caused others to lose coverage for non-payment.^[15]

Locking people out of coverage for six months if they don’t complete their annual eligibility renewal or don’t report changes in income or other factors affecting eligibility within the required timeframe would also cause Kentuckians to lose coverage. Renewal can be a confusing process, and beneficiaries often don’t receive notice that they are due for renewal and often have difficulty understanding what documents need to be turned in to complete the process.^[16] Similarly, Medicaid eligibility is complicated, and beneficiaries, particularly those with volatile work schedules, often do not understand when they must report a change in circumstances, or they may have difficulty reaching a caseworker or accessing an online portal.

Most Losing Coverage Will Become Uninsured

There’s little evidence that work requirements will meaningfully increase employment, and even less to support claims by Kentucky — or by the Centers for Medicare & Medicaid Services (CMS) in the initial approval letter — that work requirements would cause large numbers of enrollees to gain good jobs with health insurance. And CMS doesn’t even claim that those losing coverage when they don’t pay premiums or renew their coverage or report changes on time would obtain other coverage.

Studies of work requirements in federal cash assistance programs —TANF and its precursor, Aid to Families with Dependent Children — find that employment increases for those subject to work requirements are generally modest, fade over time, and don’t move many families out of poverty.^[17] Cash assistance programs generally provide at least some (albeit inadequate) resources for the supportive services that many low-income adults need to work, such as child care, job training, and transportation assistance. In contrast, the Administration’s Medicaid work requirements guidance says that states imposing these requirements need not offer any new work supports and may not use federal funding for such supports.^[18]

Moreover, even if some enrollees do find jobs because of work requirements, these will probably be mostly low-wage jobs. Such jobs are unlikely to boost enrollees’ incomes enough for them to shift from Medicaid into subsidized individual market coverage, and the large majority of such jobs do not offer affordable health insurance — meaning most enrollees would still need Medicaid. According to Labor Department data, among workers with earnings in the bottom quartile of the wage distribution, only 37 percent are offered health coverage, and less than a quarter actually obtain coverage, presumably in large part because required employee premium contributions are often higher than low-wage workers can afford.^[19] Similarly, only 37 percent of full-time workers with family incomes below the poverty line (and only 13 percent of such part-time workers) are even offered coverage.^[20] Consistent with these data, in Medicaid expansion states, 42 percent of workers with family incomes below 138 percent of the poverty line (the income limit for Medicaid in these states) obtain health insurance through Medicaid, more than twice the share that obtain insurance through an employer.^[21]

Taking Coverage Away, Leaving People Uninsured Is Contrary to Medicaid’s Objectives

As a rationale for approving policies that take health coverage away from people who don’t meet a work requirement, HHS has claimed that “work [will] promote health and well-being,” but this argument doesn’t support a claim that Kentucky’s waiver is consistent with the primary objective of Medicaid to provide coverage. While improving health and well-being are proper objectives of Medicaid, they are outcomes of having coverage rather than a result of taking coverage away from people who don’t work, pay premiums, or renew their coverage or report changes on time, which worsens health and well-being.

The evidence HHS cites on how work affects health is murky, since no studies in this area prove whether work improves health or if it’s simply that people in better health are more able to work. As noted in a recent comprehensive review of the literature conducted by the Kaiser Family Foundation, “There is strong evidence of an association between unemployment and poorer health outcomes, but authors caution against using these findings to infer that the opposite relationship (work causing improved health) exists.”^[22] The author of one study (which HHS nevertheless cited) clearly states: “Importantly, these findings do not necessarily imply that income has a causal effect on life expectancy…. [W]e caution that this correlational analysis does not uncover causal mechanisms.”^[23] And a brief in the Kentucky court case submitted by experts in health law, health policy, and health services research stated that “there is no such research” supporting the assertion that employment leads to improved health outcomes.^[24]

On the other hand, the evidence on how health coverage has a positive benefit for health is clear. A large and growing body of research on the ACA’s Medicaid expansion finds that coverage gains are generating large gains in access to care and financial security.^[25] And a new survey of Ohio enrollees who gained coverage under Medicaid expansion confirms earlier research in Ohio and Michigan finding that having Medicaid makes it easier for workers to work and for unemployed enrollees to look for a job.^[26] Taking coverage away and leaving people uninsured would move in the opposite direction.

Another way Kentucky tries to justify its waiver as a proper demonstration under section 1115 is by claiming that it would provide Medicaid beneficiaries the “tools” they need to successfully transition from Medicaid to commercial coverage. These “tools” include requiring premium payments, locking individuals out of coverage for nonpayment of premiums or for not complying with the work requirement, imposing deductibles, and using a health savings account to “purchase” additional benefits such as dental and vision. The objective of Medicaid is to provide medical assistance, or coverage — not to “train” low-income adults how to use commercial coverage. Moreover, neither Kentucky nor CMS provides any evidence that imposing premiums and other barriers to coverage in Medicaid would help Medicaid enrollees subsequently obtain, maintain, or use commercial coverage. Such evidence likely doesn’t exist, just as there is no evidence that taking coverage away from Kentuckians who don’t meet a work requirement would improve their health.