How Will Medicaid Work Requirements Affect Hospitals’ Finances?

Abstract

• Issue: The recent debate regarding Medicaid Section 1115 demonstration waivers that include work requirements has focused on potential loss of coverage for Medicaid beneficiaries, but there has been little discussion of the potential impact on providers that serve Medicaid patients.
• Goal: To assess the potential financial impact on hospitals in states that have approved or pending Section 1115 demonstration waiver applications or bills that have been passed by state legislatures for implementing work requirements in their Medicaid programs.
• Methods: This brief updates our prior analysis (published March 2019) on the financial impact that Medicaid work requirements may have on hospitals by incorporating results from a recent study of potential Medicaid coverage loss because of work requirements performed by Leighton Ku and Erin Brantley for the Commonwealth Fund, updated hospital financial data, and expanding our analysis to additional states that are considering implementing work requirements in Medicaid.
• Key Findings and Conclusion: The results show that Medicaid work requirements could weaken hospitals’ financial positions, especially rural hospitals, in states that implement these requirements as a condition of coverage. However, the design of states’ Medicaid work requirement programs will play a key role in how many beneficiaries lose coverage and the resulting financial impact on hospitals.

Background

Much of the recent debate regarding Medicaid Section 1115 demonstration waivers that impose work requirements as a condition for eligibility has focused on potential loss of coverage for beneficiaries, but there has been little discussion about the impact on providers. In states that impose work requirements, Medicaid beneficiaries will lose health insurance coverage if they cannot find work, are unable to document the required number of hours of work activity, or cannot document an exemption. Their loss of coverage will impact hospitals by reducing revenue and increasing uncompensated care costs. These adverse outcomes will not only affect the hospitals and Medicaid patients, but the entire community if hospitals must reduce staff or eliminate important services because of lower revenues and increased uncompensated care.¹

In this brief, we examine the potential impact on hospitals in states that have approved or pending Section 1115 waiver applications or bills that have been passed by state legislatures for implementing work requirements in their Medicaid programs. This brief updates our prior analysis of the financial impact that Medicaid work requirements may have on hospitals. In this analysis, we use a study of potential Medicaid coverage loss because of work requirements, which estimated that 600,000 to 800,000 adults in nine states could lose Medicaid coverage because of implementing work requirements.² For the other states included in this brief that were not covered in the abovementioned study, we used a similar methodology to estimate coverage losses. In addition, this updated brief expands our analysis to additional states that are considering implementing work requirements in Medicaid.

At the time of publication, nine states had received approval, six states had submitted applications, and three states had bills approved by their legislatures that would require nondisabled adults to work a certain number of hours to receive Medicaid coverage.³ States with bills that have been approved by the legislature have not yet formally applied for a waiver nor have been approved. Exhibit 1 shows the status of these efforts.

Reductions in Medicaid coverage will have an impact on hospitals by reducing Medicaid payments and increasing uncompensated care costs, which will result in lower hospital operating margins. How the work requirements are designed will play a key role in how many beneficiaries lose coverage and the resulting financial impact on hospitals.

The following analysis estimates the impact of Medicaid coverage loss on revenues, uncompensated care costs, and operating margins for hospitals in the affected states. For modeling purposes, we assume that work requirements in all states are fully implemented in 2019. We present impact estimates under two scenarios: a low coverage loss assumption and a high coverage loss assumption. See “How We Conducted This Study” for details.

Impact on Medicaid Revenues

The loss of Medicaid coverage because of implementing work requirements will have a significant impact on Medicaid revenues for hospitals in nearly all of the study states. However, the impact will vary across states because of the design of the various work requirement programs. Nine states (Arizona, Arkansas, Idaho, Michigan, Montana, Nebraska, New Hampshire, Ohio, and Utah) target work requirements only to adult enrollees who obtain eligibility through the ACA expansion. Indiana, Kentucky, and Virginia will apply work requirements to both the traditional Medicaid and expansion populations. Six states that did not expand Medicaid (Alabama, Mississippi, Oklahoma, South Dakota, Tennessee, and Wisconsin) will apply work requirements to adults in the traditional Medicaid program. All states have a maximum age limit that ranges from 49 to 64. Exemptions from the work requirements vary significantly by state, but typically focus on enrollees who are medically frail, full-time students, or caregivers.

Nebraska is proposing to implement a modified Medicaid expansion beginning in 2020 for adults with incomes below 100 percent of the federal poverty level. In the second year of expansion, enrollees would need to meet certain work requirements to receive an enhanced benefits package. Those that do not will receive only a basic benefits package. However, enrollees will not lose coverage for noncompliance.

Wisconsin’s project permits 48 months of work-requirement noncompliance prior to losing coverage. As a result, coverage losses because of work requirements may not occur in the first year.

Exhibit 2 shows the estimated reductions in Medicaid revenues for acute care hospitals.⁴ We estimate that Medicaid revenues will decline by 11 percent to 18 percent on average for hospitals in Kentucky and by 15 percent to 23 percent for hospitals in Virginia (Exhibits 2 and 3). These two states apply work requirements to both traditional and expansion eligible beneficiaries up to age 64. In contrast, Arizona, Arkansas, Idaho, and Ohio will apply work requirements only to the expansion population up to age 49. We estimate that Medicaid revenues will decline by a lesser degree (6% to 8%) for hospitals in these states.

Across all 18 states, we estimate that Medicaid revenues will decline by 8 percent to 12 percent on average. This would result in a decline in Medicaid payments to hospitals of $2.2 billion to $3.1 billion in 2019 assuming that work requirements are fully implemented in that year.

Impact on Uncompensated Care Costs

Most of the individuals losing Medicaid coverage will be ineligible for premium subsidies in the health insurance marketplaces because their incomes will be below the federal poverty level (or below 138% of poverty for those in expansion states).⁵ Many will be unemployed or have jobs that do not offer employer-sponsored insurance. Therefore, many beneficiaries losing Medicaid coverage will become uninsured and will contribute to rising hospital uncompensated care costs.

A recent study on insurance coverage “churning” among Medicaid beneficiaries nationally showed that nearly one-third of nonelderly Medicaid beneficiaries churned off Medicaid over a two-year period for various reasons.⁶ Of those that left, about 74 percent became permanently or temporarily uninsured. Many individuals that experienced a temporary uninsured period later reenrolled in Medicaid. However, Medicaid beneficiaries who lose coverage because of work requirements may be “locked out” of reenrolling for a certain time period. For example, under Arkansas’ waiver, enrollees lose coverage for the remainder of the calendar year after not meeting the work requirements for any three months and cannot reapply until the following January. Even after the lock-out period, these individuals will need to prove they are working the required number of hours to regain coverage. As a result, a large portion will be permanently uninsured and others will have extended gaps in coverage. This increases hospital uncompensated care costs.⁷

Exhibits 4 and 5 show the estimated increase in uncompensated care costs per hospital from implementing Medicaid work requirements. Hospitals in states that expanded Medicaid will experience the largest increases in uncompensated care in both dollar amounts per hospital and in terms of percentage increases. This is because there will be a larger proportion of Medicaid beneficiaries losing coverage in expansion states. In addition, hospitals in expansion states have benefited from reduced uncompensated care costs, a benefit which will now be undone. Hospitals in Kentucky could see the largest uncompensated care increases from implementing work requirements, as the condition will apply to both traditional and expansion populations up to age 64.

Across all 18 states, we estimate that uncompensated care costs will increase by 15 percent to 29 percent, on average. This would result in an increase in uncompensated care costs for hospitals of $1.5 billion to $2.8 billion in 2019 assuming that work requirements are fully implemented in that year.

Impact on Hospital Operating Margins

The reduction in Medicaid revenues and increases in uncompensated care costs will lead to reduced operating margins for hospitals in states that implement Medicaid work requirements.⁸ Exhibits 6 and 7 show the estimated changes in hospital operating margins by state. For example, we estimate that hospital operating margins in Alabama hospitals will be –3.7 percent in 2019 without Medicaid work requirements. Implementing work requirements in the state would reduce margins by an additional 0.2 to 0.4 percentage points, resulting in margins of –3.9 percent to –4.1 percent.

To put operating margin impacts into dollar terms, we estimate that hospital operating income — defined as net patient revenues less operating expenses — for hospitals across the 18 states could decline by $0.8 billion to $2.0 billion in 2019 assuming that work requirements are fully implemented in that year.

Impact on Rural Hospital Operating Margins

Rural hospitals may be adversely impacted by work requirements relative to hospitals located in urban areas.⁹ Rural hospitals are projected to have negative operating margins, on average, and lower operating margins than urban hospitals in most of these states prior to implementation of work requirements, meaning they are already operating at a loss on patient care (Exhibit 8).

Implementing work requirements will further reduce operating margins for these already struggling hospitals. Hospitals in rural communities have recently been closing at an alarming rate. Since 2010, 113 rural hospitals have closed in the U.S.; a reduction in operating margins may intensify the issue.¹⁰ Nearly a quarter of rural hospitals nationwide are near insolvency. An analysis of 2,045 rural hospitals nationwide showed that 21 percent across 43 states are at a high risk of closing unless their financial situations improve.¹¹

Most states that are planning to implement work requirements expanded Medicaid under the ACA. The Medicaid expansion was associated with improved hospital financial performance and substantially lower likelihoods of closure, especially in rural markets and counties with large numbers of uninsured adults before Medicaid expansion.¹² Implementing work requirements may undo many of the benefits of Medicaid expansion realized by rural hospitals.

We estimate that operating income — defined as net patient revenues less operating expenses — for rural hospitals across the 18 states could decline by $215 million to $545 million in 2019 assuming that work requirements are fully implemented in that year.

Discussion

This analysis was based on a recent study of Medicaid coverage losses because of implementing work requirements in nine states with approved Medicaid waivers.¹³ It shows that Medicaid work requirements may weaken hospitals’ financial positions but will impact hospitals’ operating margins differently across states. Several factors help to explain these differences:

• Hospital payer mix. Hospitals in states that have a high Medicaid payer mix are more dependent on Medicaid revenues and will be adversely affected more than hospitals in states with a lower Medicaid payer mix.
• Medicaid enrollees subject to work requirements and those that lose coverage. States that subject a large portion of enrollees to work requirements, by setting higher age limits and applying work requirements to both traditional and expansion groups, will experience a greater negative impact than other states. Hospitals in Kentucky, for instance, will be adversely impacted because of the design of the program, which applies work requirements to both traditional and expansion eligible beneficiaries up to age 64.
• Portion of Medicaid enrollees that become uninsured. If a large portion of enrollees that lose Medicaid coverage are unable to obtain private coverage, hospital uncompensated care costs will increase and operating margins will decline.

While the estimates provided by Ku and Brantley are extremely helpful for understanding the program’s impact on Medicaid coverage, more research is needed to understand the risk profile of Medicaid beneficiaries who lose coverage. Enrollees with disabilities or with health conditions that keep them from working have substantially higher costs than the average Medicaid beneficiary. If even some of these individuals fall through the cracks, it could have a significant impact on hospitals’ uncompensated care. While most states plan to exempt people deemed “medically frail,” it’s likely that many people with disabilities won’t qualify for an exemption or will be unable to prove that they do.

Additional research is also needed to explore whether Medicaid enrollees that lose coverage will be able to obtain other insurance coverage or will become uninsured. Much of the current research regarding churning in Medicaid indicates that most people who lose coverage experience permanent coverage loss or significant gaps in coverage. If a high percentage of Medicaid enrollees that lose coverage because of work requirements are unable to obtain private insurance coverage, this also will increase the uncompensated care burden for hospitals.

The improved financial stability experienced by many hospitals following the ACA coverage expansion has allowed them to hire new staff and maintain or offer new services to their communities. The improvements in hospital finances may be jeopardized if the Medicaid coverage losses experienced by Arkansas are seen in other states.

This adverse financial impact will not only affect hospitals and Medicaid patients but the entire surrounding communities. This may be especially critical for rural communities. Since many rural hospitals already experience negative operating margins, the increased pressure of further reducing Medicaid revenue and increasing uncompensated care could exacerbate closures. Closing a hospital has rippling effects through the community, particularly if it is the only one in the community. It has an effect on emergency and physician care. When a hospital closes, many physicians relocate to another hospital or leave the area. In addition, economic effects are felt immediately, with per capita income falling and the unemployment rate rising.¹⁴

How We Conducted This Study

This brief updates our prior analysis (published March 2019) on the financial impact that Medicaid work requirements may have on hospitals using a recent study by Ku and Brantley of potential Medicaid coverage loss because of work requirements,¹⁵ updated hospital financial data, and an expanded analysis to include additional states that are considering implementing work requirements in Medicaid. The Ku and Brantley study estimated Medicaid coverage losses to be lower than those used in our prior analysis, which in effect reduces the impact on hospitals as compared to our prior brief.

This analysis uses the Dobson | DaVanzo Hospital Finance Simulation Model (HFSM) to produce estimates of the financial impact of Medicaid work requirements on hospitals. The model is built using 2017 Medicare Hospital Cost Reports (MCRs) as the primary data source. This data source allows us to determine revenues and expenses by payer (i.e., Medicare, Medicaid, other government payers, and all other payers) for each U.S. hospital. Hospital revenues and costs for each payer category were projected from 2017 through 2027 based on trends in population growth, utilization, service intensity, and medical inflation.

HFSM uses these data and applies assumptions about the impact of Medicaid work requirements on coverage loss within each state. The model then incorporates dynamics of how the assumptions impact hospital utilization, costs, and revenues. Coverage loss assumptions were developed using the following steps:

1. We first simulated the impact of Medicaid expansion on hospitals’ Medicaid revenues and uncompensated care costs for hospitals in Idaho, Nebraska, Utah, and Virginia. Since these states did not expand Medicaid until 2019 (Utah and Virginia) and 2020 (Idaho and Nebraska), the effect of Medicaid expansion is not reflected in the 2017 hospital financial data used for this analysis. For modeling simplification purposes, we assumed that Medicaid expansion is fully implemented in these four states in 2019 in order to estimate and present results of work requirement impacts for all states in 2019.
2. We used range estimates of Medicaid coverage loss because of implementing work requirements in nine states with approved Medicaid waivers from Ku and Brantley.¹⁶ For the other nine states in our study that were not included in the Ku and Brantley study, we incorporated a similar methodology to estimate coverage losses in these states. This was accomplished by obtaining data from state websites on the number of Medicaid enrollees in specified eligibility categories or state’s estimates of expansion enrollment for states that have recently implemented Medicaid expansion or are planning to implement expansion in 2020. We then estimated the number of enrollees that would meet the age criteria for each state’s work requirement program based on national age distribution of nondisabled Medicaid adults. We obtained high- and low-range Medicaid coverage loss percentage estimates from Ku and Brantley for states with similar populations subjected to work requirements. The coverage loss percentages were multiplied by the estimated number of enrollees subjected to work requirements to obtain Medicaid coverage loss estimates for these additional states.
3. An important factor for providers will be the health care utilization or risk profile of Medicaid beneficiaries that lose coverage. Our analysis of the national Medical Expenditure Panel Survey (MEPS) data for 2015 found that hospital spending for working Medicaid nonelderly adults is about 16 percent less costly than the average Medicaid enrollee, and nonworking adults that would not meet the criteria for a potential exemption are 52 percent less costly. However, enrollees that could potentially meet one of the exemptions are substantially more costly than the average Medicaid enrollee. While most states plan to exempt people deemed “medically frail,” it is likely that many people with disabilities won’t qualify for an exemption or will be unable to prove that they do.
4. Finally, we estimated the number of individuals losing Medicaid coverage who will become uninsured. A recent study on insurance coverage “churning” among Medicaid beneficiaries nationally found that about 63 percent of people losing Medicaid coverage would become permanently uninsured and the remaining 37 percent would experience a gap in insurance coverage of about four months over the 24-month study period.¹⁷ This would result in about 69 percent (63% + 37% x (4/24)) of people who lose Medicaid coverage because of work requirements would be uninsured at any given point in time. We applied this assumption to the low-range Medicaid coverage loss estimate in order to present a low-range health insurance coverage loss estimate.
5. Another recent study of the impact on enrollees of the suspension of the Tennessee adult Medicaid expansion found no evidence that adults who lost Medicaid coverage gained private insurance. Therefore, as a high-range estimate, we assume that nearly all people who lose their Medicaid coverage because of work requirements would become uninsured. We applied this assumption to the high-range Medicaid coverage loss estimate in order to present a high-range health insurance coverage loss estimate.

Acknowledgments

The authors thank Leighton Ku and Erin Brantley of the Milken Institute School of Public Health at the George Washington University and Melinda Abrams and Sara Collins of the Commonwealth Fund for their support of this project.