Medicaid Waivers to Address Homelessness: Political Development and Policy Trajectories Free

Abstract

This article reviews the role of Medicaid waivers in homeless policy and their emerging role as a mechanism to address homelessness. The authors evaluate the political development of waivers in housing and homeless policy over the past thirty years, and they investigate the status of current and approved waivers targeting homelessness. They then consider how waivers may shape homeless policy governance going forward (including the success of existing systems), and they address implementation and efficacy questions related to the role of health care payers in solutions to homelessness. The authors find that the scope of Medicaid waivers for addressing homelessness has always been present, but it significantly expanded after enactment of the Affordable Care Act and more notably following the COVID-19 pandemic. These expansions brought new opportunities for states to fund responses to homelessness through Medicaid social determinants of health provisions providing wraparound medical services for populations at risk of or experiencing homelessness and through time-limited coverage of direct housing costs paired with payment for essential medical services. More than one third of states have an 1115 waiver specifically targeting homelessness, with nearly one in five states including provisions that cover direct housing costs (e.g., rent). Going forward, Medicaid's involvement in homeless policy has the potential to reshape state and local responses to homelessness.

Homelessness has reached crisis proportions in the United States, with more than 1.2 million individuals and 1.3 million school-aged children experiencing sheltered homelessness (NCHE 2023; Henry et al. 2023).¹ The majority of the population of persons experiencing homelessness consists of people sleeping in sheltered locations. Yet, following the onset of the COVID-19 pandemic in 2020, rates of unsheltered homelessness skyrocketed from 20% of the total population of people experiencing homelessness to nearly 40% (de Sousa et al. 2023). Successfully ending homelessness requires barrier-free access to housing paired with essential wraparound social and medical services to maintain stable housing (NASEM 2018; Kushel and Moore 2023).² Crucially, barrier-free access to housing includes affordable and low-income housing, the lack of which is the primary driver of the homelessness crisis (Colburn and Clayton 2022). Wraparound sociomedical services are necessary, but needs vary importantly across groups, with greater support required for high-risk populations facing multiple complex medical conditions and longer durations of homelessness (Kushel and Moore 2023).

The severity of this crisis has garnered attention from the US Department of Health and Human Services (HHS) in unprecedented ways. The current policy mechanism for addressing homelessness uses locally organized groups of primarily nongovernmental actors, known as Continuums of Care (CoCs), to identify and address homelessness. Yet although CoCs wield enormous expertise, they have very limited capacity, including staffing and funding, and little to no governing authority to design or implement policy solutions (Willison 2021). Given these constraints and implementation failures, HHS partnered with the Department of Housing and Urban Development (HUD) to use Medicaid waivers as a mechanism to bolster the existing system, or alternatively to create new structures to address and end homelessness. Most acutely, HHS has allowed Medicaid waivers to fund some housing costs directly for the first time in the history of the 1115 program. This consequential decision builds on more than a decade of investments by Medicaid to pay for social and medical services for persons experiencing homelessness. Pairing housing with sociomedical services will allow Medicaid to begin to bring together the necessary components to help address the homelessness crisis.

This article reviews the role of Medicaid waivers in homeless policy and their evolving role as a mechanism to address homelessness in the United States. We evaluate the political development of waivers in housing and homeless policy over the past 60 years, illustrating the shift in the program from informally supporting housing programs to playing a direct role in homelessness policy. To this end, we investigate the status of current and approved waivers targeting homelessness, including those that use Medicaid's novel ability to pay for direct housing subsidies. We then consider how waivers may shape homeless policy governance going forward (including the success of existing systems), and we address implementation and efficacy questions related to the role of health care payers in solutions to homelessness.

Overall, we find that the scope of Medicaid waivers to address homelessness has long been present through various housing programs, but it significantly expanded and explicitly targeted homelessness following enactment of the Affordable Care Act (ACA) and more notably following the onset of the COVID-19 pandemic. These expansions brought new opportunities for states to first fund responses to homelessness through Medicaid social determinants of health (SDoH) provisions providing wraparound medical services for populations at risk of or experiencing homelessness, and now through new coverage of time-limited direct housing costs paired with essential medical services through health-related social needs (HRSN). More than one third of states have an 1115 waiver specifically targeting homelessness, with one in five states including provisions that cover direct housing costs (e.g., rent). Going forward, Medicaid's involvement in homeless policy has the potential to reshape state and local responses to homelessness.

Research Methods

To investigate the political development of Medicaid waivers to address homelessness, we traced Medicaid policy changes related to homelessness over the duration of the waiver programs. Policy changes included services available, structure, and policy goals. We gathered qualitative data from primary sources including federal policy documents from the Centers for Medicare and Medicaid Services (CMS), HUD, and HHS. We also compared federal agency documents with executive branch policy announcements as appropriate, for a more comprehensive understanding of policy changes.

We collected primary and secondary data on current 1115 waivers targeting homelessness to measure the current status of Medicaid waivers to address homelessness. We selected these waivers because they are the most common waiver mechanism that explicitly includes people experiencing homelessness as a target population. Because 1915 waivers are often indirect mechanisms to address homelessness, given the overlap between disability and homelessness eligibility, we coded them as a submechanism within 1115 waivers.

We used the Kaiser Family Foundation dataset (KFF 2024a) to collect information on 1115 waivers. This database disaggregates waivers based on SDoH categories, including homelessness. We collected all currently approved 1115 waivers as of April 2024 that include at least one provision targeting homelessness.³ For many of the waivers, homelessness is the primary focus. In some cases, homelessness is one SDoH category concurrent with other provisions. Each current 1115 waiver was subsequently collected from the state Medicaid agency website. We also collected data on state levels of homelessness and Medicaid expansion as relevant indicators of waiver presence.⁴

We qualitatively coded each Medicaid waiver for the type of services provided, specifically evaluating whether or not direct housing services were included (e.g., rent), as compared to tenancy support services (services to help clients identify and maintain housing). We also coded for the target population for homeless services as well as the structure of the waiver, regarding which entity is responsible for implementing the waiver. The extent to which waivers include tenancy support services compared to direct housing costs is not only relevant for understanding development of Medicaid policy goals over time relative to existing homeless policies but also is important for evaluating the potential efficacy of Medicaid waivers to address homelessness if they begin to more concretely incorporate housing and sociomedical services simultaneously. Measuring the implementation structure of waivers is critical for comparing the relationship between waivers and the existing governance system, which consists of CoCs—locally organized groups of nonprofits that receive funding from HUD to mitigate and end homelessness (HUD 2017). A baseline understanding of these relationships is necessary for understanding and predicting future governance changes.

Policy Development of Medicaid Waivers in Housing Policy and Homelessness Pre-ACA

Medicaid as a program has, in some ways, always had a role in housing groups facing complex medical needs and socioeconomic risks. At the inception of the program, Medicaid began funding housing for beneficiaries needing “postconvalescence services” after hospitalization, with Medicaid paying for long-term care and skilled nursing services in a long-term care facility rather than at home (IOM 1986; KFF 2015). In 1971, Medicaid began to pay for residence in intermediate care facilities as an alternative to longer residential stays (IOM 1986). Medicaid waivers to address homelessness emerged in the late 20th century, giving states alternative ways to spend Medicaid dollars. Medicaid waivers also always embedded housing as at least a tertiary part of the model, but in contrast to the federal Medicaid program, the development of waivers focused on housing services outside of institutional environments.

Beginning in 1981, 1915 waivers for the Home and Community-Based Services program were developed to help stabilize and support living arrangements in the community (as opposed to institutionalization), allowing states to fund in-home medical services (without room and board) (Duckett and Guy 2000). Housing came into the picture in 1986 when Congress created an option for states to develop “targeted case management” services under their plan (Smith et al. 2000). The purpose of this amendment was for states to facilitate Medicaid beneficiaries’ “access to needed medical, social, educational, and other services,” including housing (O'Keeffe et al. 2010: 26, 97).⁵ While homelessness is not specifically referenced in this statute, it authorizes services to be provided for medically high-risk or high-needs individuals who require additional services to maintain stable housing in the community. In other words, the populations receiving services through this program overlap with groups facing high rates of homelessness or housing insecurity (Fazel, Geddes, and Kushel 2014).

In contrast, 1115 waivers were authorized in 1962 under the Social Security Act, and they were then integrated into the 1965 Medicare and Medicaid Act (Albanese 2019). The use of 1115 waivers grew over time with coverage expansion opportunities through Medicaid managed care in the 1990s, followed by alternative funding mechanisms and increased eligibility for formerly excluded groups during the Bush administration (Albanese 2019). In the existing literature, it is not clear to what extent states were using 1115 waivers to provide housing-related services during this period. In 2010, before implementation of the ACA, four states (Arizona, Hawaii, Rhode Island, and Vermont) offered residential care services for high-risk groups under an 1115 waiver (O'Keeffe et al. 2010: 133) to provide supportive medical services in noninstitutional home settings.

Changes Following Enactment of the ACA

Following the ACA's passage in 2010 under the Obama administration, CMS provided states with increased support to address housing, and explicitly homelessness, through Medicaid waivers. Over the next decade, these post-ACA changes broadly rationalized the use of Medicaid waivers to address the complex health needs of individuals experiencing homelessness, while also targeting high costs associated with this group's unaddressed needs.

In 2013, CMS issued guidance for states to address disproportionate share spending on high utilizers, including people experiencing homelessness (Mann 2013). CMS encouraged states to use integrated care models (for care coordination including housing and social services) through either 1915 or 1115 waivers, and targeted case management through 1915 waivers (Mann 2013: 17–18).

In the same year HUD released guidance on how states can use Medicaid waivers to develop supportive housing programs through both 1915 and 1115 waivers (HUD 2013). Supportive housing, or programs that provide simultaneous access to housing and any necessary supportive medical services,⁶ is the only evidence-based mechanism to successfully end homelessness (NASEM 2018). In this HUD guidance, 1115 waivers were noted for their flexibility in covering “specific populations for services not typically reimbursed by Medicaid” (HUD 2013). HHS followed HUD with a primer in 2014 reemphasizing waiver mechanisms to address chronic homelessness and develop supportive housing programs. The 2014 guidance emphasized that coverage of people experiencing chronic homelessness through Medicaid expansion made these new provisions possible, acting as a mechanism to address the needs of a highly vulnerable population (OASPE 2014).

In 2015, CMS released additional guidance for coverage of housing-related activities and services for persons with disabilities, including people experiencing chronic homelessness, noting a partnership with the Substance Abuse and Mental Health Services Administration to provide additional guidance to states for coverage for this population specifically (Wachino 2015: 1). The 2015 guidance highlighted the role of 1915 waivers in providing long-term housing-related services (transitional services) for persons experiencing chronic homelessness as well as emphasizing the new role of greater flexibility in innovative 1115 waivers to provide housing-related services as outlined throughout the bulletin (Wachino 2015: 7). From the ACA's enactment in 2010 through 2016, states began to initiate 1115 waivers to address homelessness, including New York and Washington, while other states leveraged 1915 waivers, including Michigan, Louisiana, California, and Connecticut (Wachino 2015; HUD 2013).

Notably, throughout the Obama administration, policy documents discussing Medicaid's increased role in providing housing-related services reiterated that Medicaid could not pay for direct housing costs. For example, one document stated that “consistent with statute, CMS does not provide Federal Financial Participation (FFP) for room and board in home and community-based services, but can assist states with coverage of certain housing-related activities and services” (Wachino 2015: 1).

Under the Trump administration during 2017–2020, officials were supportive of waivers to address homelessness, despite broad Republican opposition to the ACA. Secretary of Health and Human Services Alex Azar (Azar 2018) publicly stated support for such programs and approved new waiver applications. During the Trump administration, more states (including Maryland, California, Illinois, and North Carolina) initiated 1115 waivers with tenancy support services as detailed in the 2015 bulletin as part of the CMS effort to promote Medicaid flexibility to address homelessness and housing for high-needs groups (Hinton et al. 2019; Thompson et al. 2021). In 2018, focusing on behavioral health and substance use disorders in the context of the opioid epidemic, CMS issued guidance to state Medicaid directors about the ability to use waivers to improve housing transitions for persons at risk of homelessness who were exiting psychiatric institutions (HHS 2018). Importantly, while administration officials were publicly supportive of the program, the Trump White House did not make any public statements in favor of the waiver program.⁷

The Biden Administration Targets Housing through SDoH Policy

Where the post-ACA period focused on the complex needs of individuals, the Biden administration pivoted to directly target the upstream causes of these complex needs, that is, the SDoH. Immediately after Joe Biden became president in January 2021, CMS released a state health official letter outlining ways in which Medicaid and the Children's Health Insurance Program could be leveraged to address SDoH (CMS 2021). CMS's goal of applying waivers to address SDoH, including homelessness, was intended to “improve population health, reduce disability, and lower overall health costs” (CMS 2021: 1).

In this letter, CMS (2021) encouraged the use of several Medicaid provisions—section 1915, section 1115, and managed care in lieu of services and settings (ILOSs) (CMS 2023a)—to cover “housing-related services and supports” including home accessibility modifications, one-time community transition costs, and housing and tenancy supports. Housing and tenancy supports provide aid in the search for housing as well as assistance during the tenancy with landlord dispute resolution, eviction prevention, and tenant rights education (CMS 2023b: 5; Thompson et al. 2021). Section 1115 demonstrations were explicitly offered up to address homelessness and housing insecurity, with examples provided by CMS from Washington state and North Carolina that enumerate best practices for how existing 1115 programs provided housing services through their respective waivers (CMS 2021: 21). In Washington, 1115 waivers were used for the Foundational Community Supports program, which provides community transition services, including housing and employment support, for those exiting behavioral health institutions (Lindeblad 2020). North Carolina's Enhanced Case Management and Other Services Pilot Program used the 1115 waiver to provide case management services, including housing support, to “high-need Medicaid beneficiaries,” a population that explicitly includes those with housing instability (Richard 2019).

Direct Housing Costs through Health-Related Social Needs

In the winter of 2022, CMS announced pivotal guidance allowing Medicaid to pay for direct housing costs through a variety of mechanisms for the first time in the history of the agency. Citing its previous statement on SDoH and Medicaid, CMS released a statement that introduced the term health-related social needs (HRSNs). CMS defined HRSNs as “an individual's unmet, adverse social conditions that contribute to poor health” including access to housing, food, transportation, and employment (CMS 2022). The new HRSN framework expanded CMS's focus on SDoH to target individual adverse outcomes, or HRSNs, arising as “a result of their community's underlying SDOH” (CMS 2022: 8). The framework further emphasized housing and nutrition as key supplements to health care services, rationalizing Medicaid's funding of these services as a mechanism for CMS to mitigate health disparities (CMS 2022: 8; Hinton and Diana 2024). Furthermore, CMS enumerated eight areas of housing support for eligible populations, including persons experiencing homelessness,⁸ that would be considered under 1115, including rent or temporary housing for special populations up to six months, pretenancy and tenancy supports, housing transition case management, and assistance with one-time relocation costs (CMS 2022). CMS defines direct housing costs, also called “room,” as “hotel or shelter-type expenses, including all property-related costs (e.g., rental or purchase of real estate and furnishings, maintenance, utilities, and related administrative services)” (CMS 2023b).

In November 2023, the Biden administration (DPC/OSTP 2023) and CMS in a new bulletin (CMS 2023b) further clarified the scope and ability of 1115 and 1915 waivers to pay for direct housing costs. The flurry of activity emphasized interagency partnerships between HUD and HHS as a mechanism for addressing HRSN for persons experiencing homelessness. The guidance explicitly stated that waivers will require collaboration with housing providers to help enrollees obtain long-term non-Medicaid housing. Importantly, it emphasized that waiver coverage of direct housing costs (room) are time limited and may only occur for high-risk groups in medically designated circumstances. For example, 1115 waivers may cover posttransition housing with room and board for up to six months after transition, once during the demonstration period (CMS 2023b: 5). Preprocedure and posthospitalization housing in 1115 waivers may cover up to six months of room and board per client on an annual basis (CMS 2023: 4). By comparison, ILOS through 1915 waivers (Tsai 2024: 6) does not cover rent but can cover housing transition and stability services, and first month's rent (CMS 2023b: 4).

In the same month, the HUD and HHS announced the Housing Services Partnership Accelerator, a program intended to support and promote states with Medicaid 1115 waiver programs targeting homelessness in high-risk groups (HHS 2023). High-risk groups are defined as those exiting the carceral system, emergency shelters, the foster care system, and care facilities for older adults and adults with disabilities (CMS 2023b). The accelerator provides “technical assistance” with program development, including a state needs assessment, meetings with federal agencies like HUD and CMS, and one-on-one support from subject matter experts (ACL 2023). The program does not provide additional funding to the 1115 waiver programs but is intended to bolster the existing program by providing these tailored services for waiver design and implementation. In February 2024, eight states and the District of Columbia were selected for the 12-month accelerator program (HHS 2024).

Current Status of Medicaid Waivers and Homelessness

As of April 2024, 16 states had Medicaid 1115 waivers (approved or pending) with provisions specifically targeting homelessness. One additional state, Connecticut, does not explicitly target homelessness in the waiver documents, but the governor (Office of the Governor 2021) has made public statements about the intention of the waiver's housing provisions for persons transitioning from carceral institutions as a mechanism to address homelessness. We therefore include Connecticut in the results (see table 1 of the online appendix).

Most waivers are located in states that have expanded Medicaid. The only nonexpansion state with a waiver is Florida (only 10 states currently have not expanded Medicaid eligibility under the ACA) (KFF 2024b). The majority of waivers were located in states with the greatest levels of homelessness. Twelve waivers were located in states falling in the top two quartiles of homelessness (highest and high; see table 2 of the online appendix for state levels of homelessness). Five were located in states with low rates of homelessness, and no waivers were located in states in the lowest quartile (fig. 1).

Regarding the distribution of housing services among Medicaid waivers targeting homelessness, specifically measuring whether or not waivers include direct housing provisions (property-related costs such as rent and utilities) compared to tenancy support services (CMS 2023b), as of April 2024, all Medicaid waivers targeting homelessness included tenancy support services. As shown in figure 2, nearly 60% of waivers (10 out of 17) include direct housing provisions. Of waivers that do provide direct housing provisions, most are located in states with the highest rates of homelessness, as shown in figure 3. By comparison, states with waivers not providing direct housing provisions, or only providing tenancy support services, are primarily located in states with low rates of homelessness (fig. 3).

There is also important variation in the types of direct housing provisions states use in their waivers. Most waivers (seven out of 10) providing direct housing provisions include up to six months of rent or temporary housing (see table 1 of the online appendix for complete list of provisions by state). A majority of states also provide pre- or posthospitalization housing (also known as medical respite) for up to six months. Some states only provide one of these services independent of the other, while four states provide both rent/temporary housing and medical respite housing (Hawaii, Massachusetts, New York, and Washington; see table 1). Just five states provide utilities.

In terms of the implementation mechanisms states use for delivering provisions to address homelessness, Medicaid waivers are almost equally split in their use of health care entities compared to community-based providers (i.e., CoC structures) to provide homeless services. Twenty-one percent (five out of 17) use health care entities (i.e., managed care organizations, primary care providers, institutes of mental disease, hospitals) as implementing entities in predominant capacity. Comparably, 35% (six out of 17) of waivers use community-based organizations in full capacity as implementing entities to carry out services to address homelessness. Notably, a separate 29% of waivers explicitly use partnerships between health care entities and community-based providers to deliver services.

Discussion and Conclusion: Future Directions for Medicaid Waivers and Homelessness

The use of Medicaid waivers as a policy alternative to address homelessness has expanded considerably since the program's inception in 1965. Although Medicaid waivers were never intended specifically to address homelessness, Medicaid has always served populations at risk of homelessness, and in many cases has provided services for persons who were formerly homeless through wraparound services for permanent supportive housing. After Medicaid eligibility expansions under the ACA, Medicaid waivers were newly touted as a mechanism to explicitly address the complex health needs of persons experiencing homelessness. This emphasis on individual health needs evolved to provide flexibility in 1115 waivers to pay for SDoH needs of this population, including tenancy support services to improve access to housing. Despite this focus, waivers were always constrained in their ability to successfully mitigate homelessness because of their inability to pay for housing. However, following the onset of the COVID-19 pandemic, waivers are now able to pay for time-limited direct housing costs. This relatively incremental yet newly rapidly increasing expansion of waiver resources available to address homelessness is significant. Currently, 34% of states have waivers targeting homelessness. One in five states have waivers covering direct housing costs beyond tenancy support services, including California and New York, the states with the largest Medicaid enrollment nationally (CMS 2024).

Ultimately, there are three key policy implications that policy makers, practitioners, and persons at risk of or experiencing homelessness should be aware of pertaining to the trajectories of homeless policy governance and the United States's ability to successfully mitigate or end homelessness: (1) efficacy, (2) implementation, and (3) governance trajectories.

Efficacy of Policy Solutions

The only effective solution for mitigating and ending homelessness is the provision of barrier-free housing paired with essential social and medical services necessary to maintain stable housing (NASEM 2018). The evidence for these solutions, often known as “housing first” or “permanent supportive housing,” is substantial, with decades of controlled trials from around the world. However, such approaches are often not fully implemented. For example, many housing first programs fail because they either do not provide access to key sociomedical services or do not provide sufficient levels of these services. By comparison, medical programs alone, absent housing, are not sufficient to end homelessness (NASEM 2018). Since 2015 the CoC system has been required to use a housing first approach (Goodloe 2015). Despite this, high levels of constraints on CoCs’ ability to carry out policy tasks—such as being gatekept out of local government policy decisions pertaining to homelessness, and having little to no governing authority—make it nearly impossible in most cases for these structures to establish true housing first or barrier-free supportive housing programs (Willison 2021).

Yet Medicaid waivers are not currently doing so either. The ability to pay for direct housing costs is a dramatic change, but it is not a panacea. Waivers’ ability to pay for direct housing costs come with time-limited durations. Furthermore, unforeseen implementation challenges will surely arise as Medicaid attempts to pay for and coordinate access to non-Medicaid-funded community-based housing services. In the face of an affordable housing crisis that is the backbone of the homelessness crisis (Colburn and Clayton 2022), Medicaid offers an important policy lever for affordability but does not provide solutions to the supply-side problem. Despite these limitations, the widespread availability of federal dollars to directly fund housing and sociomedical services in tandem, even though time-limited, may make Medicaid a key apparatus through which people experiencing homelessness can become housed, and will hopefully facilitate more opportunities for long-term access to housing and social services.

Policy Implementation: Medicaid and CoCs

Beyond policy efficacy, it is important to consider the extent to which Medicaid may improve the efficacy of policy implementation. Medicaid waivers, through their partnerships with CoC structures as implementing entities, may serve as a way to address the gaps in the existing system. HUD funding has continued to languish since being cut by 80% in the 1980s (Biles 2011), with additional cuts of 15% under the Trump administration (Parrot et al. 2018). While the Biden administration nearly doubled the HUD budget from the Trump administration (CRS 2024; HUD 2020), these long-term deficits and dramatic shifts across administrations cause deep instability for homelessness services programming (NASEM 2018). Medicaid federal funding, by comparison to HUD, has remained more robust and stable across administrations, increasing substantially since the 1980s and withstanding attempts to cut the program (Campbell 2024; Brown and Sparer 2003). Federal Medicaid match funding also mitigates some state budget fluctuations, retaining state support for the program (Sommers and Gruber 2017). Barring threats to Medicaid funding following the 2024 elections, waiver funding to CoC systems, directly and indirectly, may offer a substantial and stable investment in housing programming and in CoC policy capacity that the program has never seen (Willison 2021).

In an ideal world, HUD funding would supplant these gaps to strengthen the CoC system. However, HUD's long-standing retrenchment and limited political power make this very unlikely (Willison, Unwala, and Klasa 2024). Thus, while some scholars rightly question the scope of Medicaid and health care actors’ engagement in responses to homelessness (Glied and D'Aunno 2023; Gondi, Beckman, and McWilliams 2020), reliable investment from more durable funding structures seems preferable in the face of the long-standing and growing homelessness crisis.

Despite this opportunity, it is also important to consider that while Medicaid may boost gaps in the existing CoC system, there are questions about which model is most effective for delivering supportive housing services. To date, we do not have any data evaluating the efficacy of strengthened CoC models as compared to housing delivered through health care systems. We also have relatively limited evaluations of Medicaid waiver interventions to address homelessness as a result of data constraints for measuring outcomes among persons experiencing homelessness (Willison et al. 2023) and coordination challenges between Medicaid and CoCs that constrain policy implementation (DeGrazia et al. 2023; Thompson et al. 2019, 2020; Willison et al. 2021). Crucially, research on the effects of Medicaid waivers to reduce homelessness is also nonexistent to date, likely because all existing evaluations of Medicaid interventions are on program waivers without direct housing provisions because of the recency of the program. As the California Whole Person Care Waiver evaluation found, “a major issue in addressing housing challenges for enrollees experiencing homelessness was lack of funding to directly provide housing and insufficient housing supply” (Pourat et al. 2022: 43).

However, research has demonstrated the protective nature of Medicaid against housing insecurity (Allen et al. 2019; Zewde et al. 2019). Some initial evidence from waivers in California, North Carolina, and Maryland finds that investments in CoC partnerships with health care entities and Medicaid may improve access to sociomedical services for persons experiencing homelessness (DeGrazia et al. 2023; Pourat et al. 2022; Silberberg, Biederman, and Carmody 2022). More research demonstrates the needs of this population, focusing on high-service utilization by persons experiencing homelessness who are also Medicaid enrollees (Cantor et al. 2020; Lin et al. 2015; Moulin et al. 2018). Some new work finds that enrollment in permanent supportive housing for homeless Medicaid beneficiaries reduces costs, inpatient hospitalization, and emergency room visits (Hollander et al. 2021). These findings align with widespread results of permanent supportive housing evaluations in reducing emergency room visits and hospital stays along with homelessness (NASEM 2018).

A counterfactual to the CoC system that may provide insight into waivers’ responses to homelessness is the Veterans Administration (VA). In a manner that mirrors Medicaid waivers using health systems as implementing entities, the VA provides flexible funding directly to VA medical centers for the provision of housing and supportive services (VHA 2023). Access to federal property and zoning for health care services similarly expedite the permitting process through the VA system compared to regular processes (Einstein, Glick, and Palmer 2019). The VA, as both a payer and a health care provider, is the only institution in the United States that has made substantial progress toward functionally ending homelessness (Petrovich 2019; Veterans Administration 2024; White House 2023). Ultimately, it will be crucial to evaluate the effectiveness of these models in ongoing and future research.

Shifting Governance Trajectories

Despite the promise of increased capacity and policy efficacy, it is possible that trajectories of homeless policy governance could shift. CMS claims that “Medicaid-covered services and supports to address HRSN will not supplant the work or funding of another federal or state non-Medicaid agency, and must be complementary to existing social services such as those provided by the US Department of Housing and Urban Development and the US Department of Agriculture Supplemental Nutrition Assistance Program” (Tsai 2023). However, the split in waiver implementation across CoC structures and health care systems raises cause for concern about the implications of waiver investments for existing homeless policy governance structures. For example, depending on the success of waivers using CoC structures for homeless services, waivers may lean more toward health care entities, siloing essential expertise and risking retrenchment of HUD funding (Glied and D'Aunno 2023; Willison, Unwala, and Klasa 2024).

Going Forward

Ultimately, Medicaid waiver investments in homelessness programming are increasing over time, with important implications for the ability to mitigate and end homelessness in the United States. Waivers may present new opportunities to formalize supportive housing programming and may generate more stable policy capacity for homeless governance systems. Bipartisan political support for waivers across the past three presidential administrations—Obama, Trump, and Biden—and participation from conservative states grappling with homelessness and housing crises, such as Utah, Florida, and Arkansas (as shown in tables 1 and 2 of the online appendix), suggest continued support for waivers in the future. As states will likely continue to seek waivers as new alternatives to address homelessness, policy makers and communities should remain vigilant to ensure that waivers bolster and bridge existing systems without compromising them, through ongoing evaluation.

Acknowledgments

The authors would like to thank Heather Howard, Jonathan Oberlander, and Frank Thompson for their helpful feedback in conceptualization of the article. We also would like to thank Phil Singer for his comments and expertise, and Naquia Unwala for her support in data collection.

Notes

References