Recovery Housing as a Solution to Homelessness
Photo: Mel Melcon / Los Angeles Times via Getty Images
Executive Summary
The Trump administration has changed federal policy toward homelessness and addiction. Its efforts reflect broad dissatisfaction with the prevailing policies of the past decade. First, in homelessness services, the administration wants to expand programs that pursue goals beyond keeping the formerly homeless housed, such as better behavioral health. Second, within addiction services, the Trump administration has promoted nonmedical approaches to addiction.[1]
Recovery housing advances both those aims.[2] A residential program for people in recovery from drug or alcohol addiction, recovery housing uses social supports to help residents maintain sobriety. Thousands of recovery housing residents are formerly homeless. Recovery housing provides a means of sustaining sobriety in a cost-effective way that minimizes overmedicalization.
This report will discuss recovery housing’s potential as a solution to homelessness. It will explain the connection between addiction and homelessness, the theory behind the Trump administration’s push to reform homelessness policy, and what recovery housing is and how it differs from other residential programs serving the homeless. The report concludes with the following recommendations:
- State government is, and should remain, the main theater of recovery housing policymaking.
- States should expand certification for recovery houses.
- States should create exceptions for recovery housing in eviction law.
In general, government should exercise caution when integrating recovery housing into homelessness services systems.
Recovery Housing: Definition and Data
Recovery housing is an intermediate-length residential program serving people with substance use disorders that prohibits drug and alcohol use among residents.[3] It is not “treatment.” Residents enter recovery housing in a non-acute phase of their condition, often, after having completed treatment in detox or an inpatient program. The average stay in a long-term residential treatment program is 66 days; in a short-term program, 25 days.[4] Relapse remains highly likely after a few months of sobriety. The longer someone’s period of sobriety lasts, the likelier he is to remain permanently sober.[5] Staying in recovery housing helps residents extend their sobriety, increasing the likelihood of permanent recovery.
Recovery housing is shared housing, offering less privacy than a private apartment. Sharing enables operators to keep costs down and promotes recovery, by encouraging the formation of social supports among residents. Addiction scholar David Best terms sobriety-promoting social factors “recovery capital.”[6] Recovery capital consists of the resources that someone, often alone and facing an uncertain future immediately out of treatment, accumulates that he can then “spend”—for example, on resisting relapse when faced with a traumatic life experience six months into his sobriety. Sustainable sobriety cannot be secured in a mere couple of weeks because it takes longer to develop recovery capital.[7]
Recovery housing’s model has changed over the years, but community (social supports) and time (lengthening sobriety) remain its two foundational elements, making it more of a social than medical approach to addiction.[8] The setting is the service. Recovery housing’s emphasis on social supports makes it a more cost-effective approach to addiction than more professionalized treatment programs.[9]
The oldest and best-known model of recovery housing is Oxford House, whose programs are nonprofit, peer-run (no paid staff), and self-governing (instead of a landlord, fellow tenants interview and invite new tenants). The other major model is houses certified by the National Alliance for Recovery Residences (NARR), founded in 2011. Both Oxford House and NARR operate through a system of state-based chapters responsible for helping open and for certifying recovery housing programs. NARR certifies four levels of recovery housing (Level 1 corresponds to the Oxford House model) (Table 1).[10]
TABLE 1
NARR Levels of Support
| Level 1 (“Peer-Run”) | Level 2 (“Monitored”) | Level 3 (“Supervised”) | Level 4 (“Service Provider”) | |
| Administration | Democratically run | Assigned house manager or senior resident | Formal organizational hierarchy Administrative oversight for service provider | Formal organizational hierarchy Clinical and administrative supervision |
| Services | Drug screening; house meetings Encouraged attendance in self-help meetings | House rules provide structure; peer-run groups; drug screening; house meetings; involvement in self-help or treatment services | Emphasis on life-skill development Clinical services utilized in outside community; other services provided in house | Clinical services and programming provided in house Emphasis on life-skill development |
| Form of Residence | Overwhelmingly single-family residences | Primarily single-family residences; possibly multiunit | Various types | Various types; more likely to be formal step-down phase within care continuum of a treatment center |
| Staff | No paid positions; at most, an overseeing officer | At least one paid position | Facility manager; certified staff or case managers | Credentialed staff |
Calculating the number of recovery housing programs (Oxford House, NARR, and uncertified) is a long-standing problem.[11] The federal government has estimated that about 8,000 certified programs exist nationwide.[12] One analysis estimated about 18,000 total programs nationwide.[13] A 2026 peer-reviewed study estimated a range of 10,000–18,000.[14] Equally difficult is calculating the rate of growth in recent years. The number of Oxford Houses has almost tripled since 2010 (from 1,450 to 4,029).[15] NARR-affiliated programs are also believed to have expanded. An estimated 250,000–300,000 people use recovery housing every year.[16]
Nationwide Fiscal Landscape
One reason for the likely increase in these programs is the expansion of public funding sources for recovery housing of late.[17]
The traditional revenue source for recovery housing programs—and still the largest source—is resident fees.[18] Reliance on resident fees limits access to recovery housing but also promotes self-sufficiency.[19] A 2025 survey of hundreds of recovery housing programs, mostly in Appalachia, found strong employment outcomes among those programs. Because recovery housing is largely self-pay, “individuals are strongly encouraged and supported in obtaining gainful employment.”[20]
The second most significant revenue source is federal and state resources distributed by state governments. The U.S. Department of Housing and Urban Development (HUD) operates the Recovery Housing Program, the only dedicated federal source of recovery housing support. Congress authorized it in the 2018 Substance Use Disorder Prevention that Promotes Opioid Recovery and Treatment for Patients and Communities (SUPPORT) Act. HUD’s Recovery Housing Program has provided $25–$30 million annually since 2020.[21]
Within the U.S. Department of Health and Human Services (HHS), the Substance Abuse and Mental Health Services Administration’s two main addiction-specific funding sources are the Substance Use Prevention, Treatment, and Recovery Services Block Grant and State Opioid Response grant programs. Combined, these provide states with $3.6 billion.[22] Recovery housing is an officially authorized use of both programs, though it must compete with other addiction-related programs such as treatment and prevention.
State governments play a major role in crafting recovery housing policy. Some states operate their own unique non-federally funded programs supporting recovery housing. States have the power to direct other agencies, such as corrections, to use recovery housing programs. In addition, states play a major role in the distribution of opioid settlement funds, which can be used to support recovery housing. There have been several “opioid settlements” brokered during the 2010s and 2020s between state governments and corporations that held the latter liable for the recent epidemic. Nationwide, these funds have generated significant resources that are technically private, though administered by states and localities. Michigan, Colorado, and Vermont have used opioid settlement funds on recovery housing.[23]
Several states, including Virginia, Iowa, and Oregon,[24] used 2021’s American Rescue Plan Act (ARPA) funds, while they lasted, for recovery housing. Some states, such as Vermont, will be using Rural Health Transformation funds (from that $50 billion program, established by the 2025 One Big Beautiful Bill Act) to fund recovery housing. Recovery Kentucky, a joint effort between that state’s corrections, housing, and local government support agencies, supports recovery housing via resources from Low-Income Tax Credits, the Community Development Block Grant, and correctional services.[25]
Philanthropy is the third largest revenue source for the many recovery housing programs that are nonprofits.[26] There has also been private investment, for better or worse, motivated by online influencers who promote recovery housing as a business venture.[27] Small for-profit (“mom and pop”) operators play a large role in the provision of recovery housing. One 2025 survey of 420 programs (about 80% NARR-certified) found that one-third were for-profit.[28]
Insurance (private or public) does not pay for recovery housing—or, at least, not the core housing benefit. Medicaid has been used by Kentucky, for example, to pay for peer services in some programs.[29] And clinically enhanced programs (Level 4 in Table 1) rely on Medicaid to pay for whatever health services residents are using. Some surveys have found that about three-fourths of recovery housing residents are enrolled in Medicaid.[30] The Affordable Care Act’s expansion of Medicaid, some have suggested, may have “crowded in” other resources to spend on recovery housing by increasing funding for treatment proper.[31] But it remains unclear as to how robust that substitution effect is, as California and New York have expensive Medicaid programs but are not considered outstanding supporters of recovery housing.[32]
Recovery housing needs funding for three purposes: capital or startup operations; resident fees; and support for the state NARR and Oxford House affiliates in their technical assistance and accreditation efforts. In some cases, states provide direct financial support to their accreditors, including Washington and Vermont. States may also provide temporary rental assistance (sometimes termed “indigent care”) to residents of recovery housing. States that provide tenant subsidies for recovery housing include Virginia, Ohio, Maine, Connecticut, Arkansas, and South Carolina.[33] State indigent-care programs expand recovery housing’s reach. One recent survey involving more than 300 homes found an average rent of $692,[34] a sum not necessarily onerous but difficult to support for someone coming directly out of treatment with no job.
Funding for recovery housing stands at a historical high. Twenty years ago, there was no dedicated HUD funding, the State Opioid Response grant program did not exist, and there was no opioid settlement funding. Funding for recovery housing has expanded along with funding for proper treatment, through the Affordable Care Act’s expansion of Medicaid, the largest source of addiction services funding in the U.S., and imposition of insurance regulations requiring carriers to pay for care for substance use disorders.[35]
Recovery housing has expanded in recent years in terms of model design. The traditional Oxford House model is rigorous in terms of expectations, eligibility standards, tenants’ financial commitment, and commitment to the social model of recovery.[36] But in the past two decades, recovery housing has embraced a degree of medicalization through the affirmation of medication-assisted treatment and the expansion of NARR Level 4 houses. Much as with indigent-care programs, service-enriched recovery housing allows the model to serve tenants with greater needs, creating more of a continuum of care for recovery housing than existed two decades ago.
Recovery housing has expanded along with all varieties of addiction services programs. Specialized addiction treatment centers, use of medication-assisted treatment, and the number of trained clinical professionals have all increased since 2010.[37] And sobriety-oriented recovery housing’s expansion also coincided with the increased normalization of “harm-reduction” approaches to addiction.
The harm-reduction movement began with the deployment of methadone and other medication-assisted treatments, expanded to needle exchanges justified by the crisis of the AIDS epidemic, and, by the 2010s, had embraced the position that substance abuse was a health condition to be treated with the same respect as any other.
The growing embrace of harm reduction has coincided not only with recovery housing’s expansion but also that of other sobriety-oriented programs such as drug courts.[38] About 65,000 Alcoholics Anonymous groups exist nationwide, with more than 1 million members.[39] The Biden administration, which directed over 80% of homelessness grant funding for permanent housing, embraced more of a range of solutions (harm reduction but also drug courts and recovery housing) in its addiction services budget request.[40] Today, the status quo on addiction services embraces a spectrum of solutions.[41]
Recovery Housing and Homelessness Policy
The Housing First Debate
Unlike addiction services, homelessness services systems have less of a spectrum-like character than they did 20 years ago. Proponents of Housing First, the now-dominant philosophy inspired by harm reduction, specifically questioned the need for a continuum of care in homelessness services. In their view, intermediate-length programs were unnecessary because homeless individuals are capable of being moved rapidly from the streets into permanent housing. Because it was never integrated into mainstream homelessness services systems, recovery housing was not affected by the increasing influence that Housing First exerted over homelessness funding sources. Transitional housing, by contrast, was severely affected, and funding was cut. Transitional housing is an intermediate-length (24-month) residential program that requires sobriety and participation in services (social, employment, health).[42]
Under Housing First’s influence, homelessness funds have been heavily shifted to permanent supportive housing (PSH). That model pairs subsidized rents with supportive services, but there is no requirement for using those services or for behavioral change. Over the last two decades, homelessness services systems have become more barbell-shaped, comprising short-term crisis care (shelter and outreach), on the one hand, and permanent housing, on the other. In the late 2000s, transitional housing received about one-third of the total funds awarded through the Continuum of Care (CoC) program, the main federal homelessness grant. In 2024, it received only 1%.[43] During that same period, transitional housing units have fallen from 210,000 to 84,000.[44] Over the last 15 years, funding for permanent supportive housing units has more than doubled, and the number of beds stands at about 410,000.[45] Currently, over 80% of federal homelessness funding goes to permanent housing, mostly PSH.[46]
A portion of the homeless population will never be self-sufficient.[47] But while over 80% of federal funding flows to PSH and other permanent housing programs, only 30% of the homeless population is considered “chronic.”[48] The shift in funding toward Housing First coincided with homelessness reaching historical highs. In the late 2000s, about 640,000 Americans were homeless, compared with 750,000 in 2026.[49] The Trump administration has called for a restoration to the status quo prior to Housing First by shifting funds away and back to more sobriety-oriented programs, including transitional and recovery housing.[50]
Homelessness and Addiction
Addiction causes homelessness. According to HUD’s official point-in-time estimate (based on self-reported data), about 25% of the unsheltered population have a substance abuse disorder.[51] More targeted surveys have found rates in the 70%–80% range.[52] Even on a self-reported basis, some studies have found that half the street homeless cite substance use as having contributed to their loss of housing.[53] Homeless people constitute 20% of admissions to substance use treatment programs.[54]
Addiction makes people homeless because it consumes their income, makes them unable to maintain employment, and alienates them from social support networks. Addiction also keeps people homeless because, especially in the case of West Coast encampment culture, their lives become organized exclusively around substance abuse. Encampments are known, among the broader non-homeless population in their host communities, as drug markets.[55] Accommodating encampments reduces users’“search costs”—in effect, making drugs cheaper. That would not be the case if addiction and homelessness were only loosely connected.
Formerly homeless people constitute a large share of recovery housing residents. A 2025 survey of 420 programs found that “most organizations indicated that most of their population were individuals with a history of homelessness (73%).”[56] Oxford House reports that nearly 75% of its residents are formerly homeless and had been homeless three times on average, and that their longest period of homelessness lasted about a year, on average; nearly 20% were homeless immediately prior to move-in.[57] Recovery Kentucky, a recovery housing program directed by that state’s government, reports that about 33% of residents are “currently homeless” at intake.[58] A 2025 survey of about 10,000 recovery housing residents in Ohio found that about 20% had been homeless during the month prior to move-in.[59] Similarly, 19% of participants in Maine’s Housing Subsidy Services Program entered recovery housing from homelessness.[60]
It stands to reason that expanding recovery housing could reduce homelessness. Making more homeless grant funding available for recovery housing would be one way to accomplish that aim, which will require shifting funds from other programs, just as expanding Housing First required shifting funds from transitional housing. In November 2025, the Trump administration released a new Notice of Funding Opportunity (NOFO) for the CoC program that capped funding for permanent housing at 30%,[61] down from the current 80%-plus.[62] That 2025 NOFO was successfully challenged in court, thus leaving in place the two-year (2024 and 2025) NOFO that the Biden administration announced in July 2024.[63]
In February 2026, Congress passed the “T-HUD” Appropriations bill (Transportation, Housing and Urban Development, and Related Agencies Appropriations Act). This act set a baseline expectation that at least 60% of the CoC program’s budget go toward the renewal of previously funded programs, which will mean mostly, in practice, permanent housing programs.[64] In June, HUD released the $4 billion 2026 NOFO, which set renewal funding at 60% but also provided a $1.3 billion set-aside for new transitional housing and other non-permanent housing programs.[65] That has been challenged in court, too; but after a mid-September victory at the First Circuit Court of Appeals level, HUD looks likely to be able to proceed with the FY26 NOFO.[66]
Comparing Housing Models
Evaluated as a solution to homelessness, recovery housing has different virtues from permanent supportive housing and transitional housing. These differences are summarized in Table 2 below.
Transitional Housing vs. Recovery Housing
In the current debate over federal homelessness policy reform, transitional housing has received more emphasis than recovery housing because, as noted earlier, the former used to receive significant homelessness funding whereas the latter was never a major recipient.[67]
Recovery housing and transitional housing both address the “root causes” of homelessness and serve similar cohorts. Everyone in recovery housing is in recovery—but not necessarily everyone in transitional housing is in recovery (though many are).
Effective recovery housing and transitional housing programs promise externalities beyond helping their residents. Both prove that sobriety and upward mobility are possible even for formerly homeless drug addicts. Recovery is common. At least 25 million American adults are in remission from a serious substance abuse disorder.[68] Effective recovery and transitional housing programs build confidence that we can and should aim higher than accommodating addiction.
Recovery and transitional housing provide social validation for people fighting addiction. People in recovery or coming out of homelessness are often alienated from friends, family, and society generally. Programs for addicts and the homeless thus aim at social reintegration. That goal may be impossible for some; but for others, it may be possible if friends, family, and potential employers can rely on objective evidence of personal transformation. Both recovery and transitional housing can provide objective evidence of personal transformation. Certain employment programs for ex-offenders help them get past the point of an initial performance review, thereby providing an objective way to vouch for their employability.[69] Recovery and transitional housing work in a similar fashion for their clients.
Recovery housing differs from transitional housing in four ways:
Homelessness Status
First, HUD considers transitional housing clients still homeless but not residents of recovery housing. Recovery housing program participants enjoy stronger eviction protections than transitional housing program clients.
Reliance on Peers
Second, while peer pressure contributes to the success of transitional housing, it is considered foundational to recovery housing. Indeed, some recovery housing programs rely almost exclusively on the benefits of living with peers who are working on their sobriety to achieve behavioral change. In contrast to such pure social model approaches, transitional housing is much more structured. Transitional housing programs are always service-enhanced and utilize paid staff.
Housing Services Nexus
Third, transitional housing provides services and housing as part of the same benefit package. Recovery housing places behavioral expectations on program participants and may sometimes require service participation. But by law, and as an ethical matter,[70] programs must maintain a careful distinction between the provision of health services and housing benefits.
The legal requirement was the result of overly entangled relationships between housing and health services providers in the 2010s.[71] Some operators used the core housing benefit as a loss leader to secure patients whose treatment services those operators could bill Medicaid for. In response, Congress passed the 2018 SUPPORT Act, which contained a provision prohibiting “Medicaid Inducement.” Any resident of recovery housing who is receiving medical treatment for his addiction must have a meaningful choice in his providers. Thus, what the Trump administration and other Housing First critics find appealing about transitional housing—the tight interweaving of treatment and housing provision—is prohibited in recovery housing.
Reputation and Evidence Base
Fourth, recovery housing has a more mixed reputation than transitional housing. The reputation of transitional housing is less freighted by scandal, partly because of the model’s decline but also because it is overwhelmingly operated by nonprofits funded by faith-based charities or the government. For-profit operators funded by self-paying clients are major players in recovery housing. That means that, occasionally, the face of recovery housing is an internet user promoting a get-rich-quick scheme, which is never the case for transitional housing.
At the same time, recovery housing has a stronger evidence base than transitional housing.[72] One disadvantage that the Trump administration has faced in trying to promote transitional housing is the dearth of studies rigorously evaluating that model. To a degree, this reflects the homeless research community’s lack of interest in non–Housing First–aligned programs and, again, transitional housing’s decline, which has left few programs to be studied. The addiction research community has taken a greater interest in recovery housing.
In terms of experimental approaches utilizing a comparison group, a 2025 survey of the peer-reviewed literature on recovery housing identified five studies, all from this century, that generated 11 publications.[73] They looked mainly at Oxford House programs with residents formerly involved in the criminal-justice system. All found that, over the 18–24-month observation period, the recovery housing program under evaluation generally promoted recovery from addiction to a “moderately strong” degree, in contrast to the other program (or no program) with which it was compared. Male participants especially seemed to benefit.
Other studies have measured people’s outcomes before they entered the housing and again afterward (in more technical terms, they utilized a “pre-post” naturalistic design). This is more representative of how recovery housing truly operates, because people are not assigned randomly to a program. Instead, programs and potential residents vet each other by a “mutual selection” process.[74]
Two related 2010 studies of sober living (non–Oxford House) programs in California followed 300 residents over 18 months.[75] One program saw its residents’ abstinence rate rise from 11% to 68% at six months, and then 46% at 18 months. The other program rose from 20% to 40% at six months, and then stood at 42% at 18 months.[76] One large (~900 n) 2007 study interviewed Oxford House residents in 16 states. At the end of the year, researchers found that only 13.5% of participants reported using substances, 80% reported being employed, and only 5% were incarcerated.[77]
Even when not evaluating outcomes, recovery housing researchers have helped define the model—laying out how its function and goals differ from those of other residential programs. They study the qualities that distinguish good programs and how the model might support other systems (such as criminal justice and outpatient addiction services), as well as the model’s debt to other models (such as 12-Step) that have their own evidence base.[78] This literature has been invaluable in the development of policies to support and regulate recovery housing, and its equivalent does not exist for transitional housing, whose virtues can be difficult to debate because even fairly informed homelessness policy experts sometimes do not have a clear conception of how transitional housing differs from other residential programs such as halfway houses and shelters. No NARR or Oxford House–type association promotes transitional housing.
Permanent Supportive Housing vs. Recovery Housing
Different Cohorts Served
PSH and recovery housing serve different cohorts: disabled formerly long-term homeless people, in the case of PSH; and people in recovery with a chance at self-sufficiency, in the case of recovery housing. PSH serves the seriously mentally ill, whereas recovery housing mostly does not. Recovery housing is more suitable than PSH for nondisabled individuals.[79] The best fit for PSH is someone who does not need long-term institutionalization but needs more services and supervision than mainstream affordable housing provides.
Different Measures of Success
Related to these differences in clientele, PSH and recovery housing measure success differently. PSH has an extensive evidence base but one that mainly focuses on the model’s ability to promote “residential stability.” It achieves that goal partly because PSH staff tend to be highly progressive individuals who are profoundly reluctant to evict tenants for rent nonpayment or other lease violations that would ordinarily justify eviction. Furthermore, PSH programs located in jurisdictions with powerfully pro-tenant housing laws will boast a high program success rate—again, defined as “residential stability”—that is just as much due to the onerousness of eviction protocols as program excellence.
Recovery capital forms no part of PSH’s definition of success. A PSH program that builds no recovery capital among residents would be judged just as successful as one that did, if both ensure that all their residents remain “stably housed.” A recovery housing program that did not build recovery capital would be considered underperforming.
In terms of externalities, PSH can reduce the number of homeless individuals on the streets. But PSH, even when effective by its own metrics, cannot promise the same behavioral externalities as recovery housing because it does not improve health outcomes.
Homelessness Status
Both PSH and recovery housing residents are considered “housed” according to HUD and enjoy normal tenant protections. For-profit landlords play a large role in PSH, particularly in the voucher-style model, which is more common nationwide than project-based PSH.[80] But PSH, even when provided by private ownership, is always subsidized. A recovery housing program may be provided by a for-profit property owner receiving no government subsidy and yet still qualify as “recovery housing.”
TABLE 2
Major Differences Between Recovery Housing, Permanent Supportive Housing, and Transitional Housing
| Recovery Housing | Permanent Supportive Housing | Transitional Housing | |
| Length of residents’ stay | Intermediate | Permanent | Intermediate |
| Service provision | Services range from non-existent to light touch to robust; service participation can be required; residents must have choice in health services provider | Services are provided; participation is voluntary | Services are provided; participation is required; choice in provider is immaterial |
| Operator status | Mix of nonprofit and for-profit operators | Mix of nonprofit and for-profit operators | Nonprofit-operated |
| Residents’ most relevant behavioral health condition served | Substance use disorder | Serious mental illness and substance use disorder | Substance use disorder |
| What level of government is the most important policymaker? | State government | Federal government | Federal government |
| Residents’ homeless status | Residents are not considered homeless | Residents are not considered homeless | Residents are considered homeless |
| Housing accommodation: shared or private? | Shared | Private | Shared |
| Size of stock/ population served | 250–300,000 served every year (estimated) | 410,000 beds | 85,500 beds |
Recommendations: An Agenda for Recovery Housing
Over the last two decades, recovery housing has expanded in terms of number of units and model design. Many who now have access to recovery housing would not have had access back in 2000. The urgency to serve a larger population was driven partly by the opioid crisis. That crisis has since eased slightly: fatal overdoses have declined in each of the last three years.[81] But the basic rate of addiction remains high.
The federal government estimates that 44.6 million Americans had a substance abuse disorder in 2025, down only slightly from 46.8 million in 2021.[82] Substance use disorders destroy lives even when not resulting in fatal overdoses. More than 100,000 homeless Americans have a substance abuse disorder.[83]
According to one estimate, only 1%–2% of Americans with substance abuse disorders use recovery housing annually.[84] Many recovery housing programs report wait lists.[85] “Financial resilience” is a major concern of operators who believe that more revenue sources must be identified to expand or stabilize their program.[86] Meanwhile, the homelessness services system is imbalanced toward PSH and needs more investment in sobriety-oriented programs. Integrating recovery housing and homelessness services therefore seems like a win-win.
Successful integration will require safeguards to ensure recovery housing’s integrity. The quality of American recovery housing stock is variable. Federal and state governments should not fund low-quality programs, and funding can be used as leverage to strengthen quality. Whatever happens with federal homelessness policy and recovery housing, policymakers should focus as much on increasing the quality of the existing stock as increasing the raw numbers of units. The following recommendations explain how to do that.
State government is, and should remain, the main theater of recovery housing policymaking.
State governments make the most important decisions about recovery housing quality. That will remain the case for the foreseeable future.
Recovery housing is a highly decentralized policy sector where thousands of private entities, including many for-profit, form a response to the pressing social problem of addiction. As a question of simple math, government oversight of a policy sector becomes more difficult the more providers there are. It is hard enough for states, given that even the total number of recovery housing programs is unknown, to ensure quality standards. It is nearly impossible for the federal government to do so.
States should expand certification for recovery houses.
In the broader discussion about affordability, concerns over housing overregulation are acute. In recovery housing, under-regulation is a larger problem. In the long term, overregulation could harm the recovery housing sector; but in the intermediate term, states need to expand their regulation efforts—in the form of certification—to strengthen quality.
Certification validates that a given recovery housing program meets certain basic quality standards.[87] Certification is done at the state level, by state governments, state Oxford House or NARR chapters, or some other state-designated entity. Certification efforts have been ongoing for the past 10–15 years, and the number of certified houses is now at a historical peak. At the same time, an unknown number of recovery programs—potentially reaching into the thousands nationwide—remain uncertified. Some states directly support their certifying NARR and Oxford House chapters, whereas other states do not even have such chapters.[88] About one-third of states have no official definition of recovery housing, and most do not maintain a database of homes.[89] Fraud and abuse remain major reputational liabilities for the sector as a whole.[90] Certification is the only certain route to improving data collection about the number of programs and their characteristics.
As much as possible, certification should be expanded via carrots (the benefits of certification) rather than sticks (defining noncertified houses as illegal, shutting them down, and perhaps even imposing civil or criminal penalties on their owners). American social history has a long tradition of government shutting down—or failing to prevent the closure of—housing programs that are seen as substandard but that could not easily be replaced by better programs serving the same population. Two highly relevant examples to the homelessness situation: single-room occupancy hotels; and board and care residences for the mentally disabled in California. While it is tempting to call for shutting down objectively bad recovery housing programs, doing that en masse would leave potentially thousands of residents homeless. Mandating certification could also run afoul of disability law, as it would apply requirements to housing serving people in recovery that don’t apply to the broader population.
Carrots include the marketing benefit of being a certified residence, referrals from government agencies, and access to exemption from state eviction law (discussed below). In some states, these benefits will be more meaningful than in others.
Another carrot is funding. States that directly fund recovery housing only if such housing is certified will have a more robust certification regimen than states that do not. As noted, the fiscal landscape of recovery housing is complicated and some programs are now dependent on funding sources that will not last forever, such as state opioid settlement funds. Allocating homeless grant funding to certified recovery housing operators could serve as another carrot, thus increasing quality as well as quantity.
Carrots alone may not suffice to eliminate all abuse, fraud, and low-quality programs. But no state should begin resorting to sticks until it has succeeded in developing a competent certification process.
States should create exceptions for recovery housing in eviction law.
The nature of recovery shapes the design of recovery housing. Recovery takes time—a person who is two years into sobriety is more likely to stay sober permanently than a person who is only days or weeks into sobriety—and recovery capital formed by social supports.
The entire recovery housing model is undermined when a resident returns to substance abuse and does not leave. Such a person could easily be removed from a homeless shelter, residential treatment center, or transitional housing program, but recovery housing participants have tenant rights defined by state law. Some recovery housing operators try to avoid state law by persuading residents at move-in to sign a highly pro-landlord contract that effectively relinquishes their tenant rights. But that is legally questionable.
More promisingly, some states have passed legislation giving certified recovery housing operators special discharge powers that most landlords don’t have. An exemplary case is Maine’s “Act to Improve Conditions for People Living in Recovery Residences by Requiring Approved Discharge and Transfer Policies,” passed in 2023.[91]
One study found the average stay in recovery housing to be about seven months.[92] In states with highly pro-tenant laws, it could take longer to evict a resident who has returned to use. Therefore, a resident of a “recovery residence” could theoretically live with an actively using roommate through his entire stay. Recovery housing programs are often small: 10–15 residents is fairly standard.[93] One single actively using resident, in a house of 10 individuals trying to avoid the “people, places, and things” that trigger relapse, can be devastating. To the extent that a state’s housing policy allows recovery housing to be undermined in that manner, a federal homeless authority, eager to expand high-quality sober options, should not fund programs in that state.
Excessively pro-tenant laws undermine recovery housing’s ability to socially validate residents to community and family members and potential employers. A recovery housing program that 100% of residents are evicted from would not be considered a good program. But a program where no residents are evicted is probably also not a good program. Functional social programs acknowledge failure and manage it responsibly.
Not every resident will be equally assertive in contesting an eviction, and landlord–tenant laws vary by state. But many blue states have recently passed pro-tenant legislation. Architects of these laws have generally not appreciated the harmful unintended consequences that the laws hold for recovery housing.
Exemption from eviction law should be granted only to certified recovery housing programs. As previously noted, such authority can serve as one of the “carrots” that entice programs to pursue certification. Exemption legislation can also mandate that programs, as part of the eviction process, refer tenants to another program better suited to their condition.
Conclusion: Exercise Caution When Integrating Recovery Housing
A more balanced homelessness services system would provide more sobriety-oriented options than the Housing First status quo. Investing in transitional housing could provide balance. Recovery housing may be even better positioned to do so, since it was not defunded during the recent Housing First era but rather expanded.
Some beneficial relations between homelessness services agencies and recovery housing already exist, such as when the former acts as a referral source for the latter. Providing homelessness grants to recovery housing, such as the Trump administration has called for, would lead to deeper system integration. Recovery housing would thereby function as part of the continuum of care of homelessness services.
Integration should proceed with care, in a manner that ensures the integrity of recovery housing programs. The culture of addiction services exhibits a tolerance for a spectrum of approaches, both sobriety-oriented and harm reduction–style. In homelessness services, harm reduction–style Housing First approaches remain dominant, making it not necessarily a natural fit for recovery housing.
The federal government’s pursuit of integration makes it even more incumbent on state governments to improve program quality. Through expanding certification and reforming eviction law, states can strengthen the integrity of their current stock of recovery housing programs at this time of potentially sweeping change for the sector as a whole.
About the Author
Stephen Eide is a senior fellow at the Manhattan Institute and contributing editor of City Journal. He researches social policy questions such as homelessness and mental illness. He is the author of Homelessness in America: The History and Tragedy of an Intractable Social Problem, published by Rowman & Littlefield in 2022. He was previously a senior research associate at the Worcester Regional Research Bureau and holds a B.A. from St. John’s College in Santa Fe, New Mexico, and a Ph.D. in political philosophy from Boston College.
Endnotes
Are you interested in supporting the Manhattan Institute’s public-interest research and journalism? As a 501(c)(3) nonprofit, donations in support of MI and its scholars’ work are fully tax-deductible as provided by law (EIN #13-2912529).