The 92 Percent Who Stayed

Homeward Bound has worked on homelessness in Western North Carolina since 1987, and it adopted the Housing First approach long before the current national argument over it. Its premise is the one the research validates: move people into permanent housing as quickly as possible, without requiring sobriety or treatment first, then surround that housing with ongoing support.

The cumulative number is the part that is easy to miss in any single year. Since 2006, Homeward Bound has placed more than 2,760 people into permanent homes. In 2025 alone, 102 people exited homelessness through its programs. These are not shelter beds or nights off the street. They are leases.

Permanent supportive housing, or PSH, is the high-intensity end of Housing First. Homeward Bound's program is built for people who meet the federal test for chronic homelessness. That test already requires a disabling condition, so the disability is part of the definition rather than something added to it. Homeward Bound says the people in this program typically need a higher level of care because of how long they have been homeless and because of long-term health conditions, among them physical disabilities, mental illnesses and substance use disorders. It is the population the staircase model failed most completely. A quick rental subsidy alone cannot stabilize this group.

PSH pairs a permanent home with services that do not end on a clock: case management, connections to health care and treatment, and help holding onto the housing despite setbacks that used to end it. Sobriety is not a condition of getting a home, and treatment is encouraged rather than required. Meeting with a case manager is different. Homeward Bound told Asheville Watchdog in November 2025 that regular meetings with caseworkers are required to stay in the program. The staff weigh each resident's situation and match the services to what that person actually needs. This is exactly the fully funded version of the model that the national evidence rewards, and the version that produces the retention numbers below.

In 2025, 92 percent of the people in Homeward Bound's permanent supportive housing stayed housed, by the organization's own impact reporting. Put the other way: fewer than one in ten returned to homelessness in a year, among a group that had cycled through the street, the hospital, and the jail for years before.

The 92 percent is a one-year figure from the organization's own reporting, so it does not line up exactly with the Housing First trials, which followed people for two years. But it points the same way. In the founding New York trial, the Housing First program kept roughly four in five of its participants housed. Homeward Bound's number is not a miracle. It came from funded services, done consistently, for people the old model wrote off.

For most of Homeward Bound's history the federal money behind permanent supportive housing was bipartisan and steady. That changed in 2025. A federal turn against Housing First began redirecting the Continuum of Care program, the main homelessness grant, away from permanent housing and toward shorter-term, treatment-conditioned options.

Locally the stakes are concrete. Of the area's roughly $2 million in Continuum of Care money, about $1.8 million funds permanent housing. Homeward Bound's permanent supportive housing draws the biggest single share, about $1.4 million. The grant pays roughly $14,000 a household for a year of rent, case management, and counseling. In November 2025 it housed 102 households, at Homeward Bound's small Key Commons building and in private rentals across the county, and the people in them are counted in the 92 percent. Compass Point, the largest building, is paid for from a different source. One early rule in that shift would have capped funding and cut the permanent-housing programs to at most $568,000. A federal court threw out that rule on June 29, 2026. The near-term risk is the 2026 grant competition. Its first tier protects only 60 percent of the money each community needs for renewals. It also sets aside about $1.3 billion for new projects, with priority for transitional housing and services-only projects, and it awards points for providing treatment and for requiring people to take part in services. A federal court set that notice aside on August 7, 2026. On September 16, 2026 an appeals court paused that order while HUD appealed, and HUD reopened the competition on September 18, 2026. On September 21 the challengers asked the appeals court to vacate the August ruling and send the cases back to the trial court, and on September 24 it did. HUD's notice stayed in effect, and applications closed on September 30, 2026. The cases are back before the trial judge, with new briefing due through October 14, 2026. The 60 percent is the minimum Congress set (HUD must select projects worth at least 60 percent of each community's renewal needs), so it holds whatever the courts decide next, and the $1.4 million is not out of danger. We tell that policy story in full in Strings Attached; the point here is narrower. The 92 percent did not come from a program that withholds housing until people earn it. It came from the opposite. Defunding that does not make the model work better, it breaks it.

Not all of the organization's funding is exposed, and there is no cause to panic. Its veterans' work, funded through the Department of Veterans Affairs, sits outside this fight. One option is to fold in county and opioid-settlement dollars, which in North Carolina can pay for recovery-oriented housing and the services beside it. The one condition is that the home itself is not made conditional. That is the same home-and-help model, funded from more than one direction so the 92 percent does not fall when the federal share does.

The most visible piece of this work is Compass Point Village, the 17.8 million dollar conversion of a former Tunnel Road motel into 85 permanent supportive apartments. It opened in September 2023, and by June 2026 very few of its rooms were empty. It is also where the local version of the national disorder debate played out. After it opened, and after a separate group, Sunrise Community for Recovery and Wellness, opened a recovery drop-in center next door the following year, reported crime along that stretch of Tunnel Road rose. In May 2026, Asheville police compared 2022 and 2023 with the two years after and counted reports of assaults up 63 percent and trespassing up 118 percent.

The honest reading concedes those conditions and still holds two things at once. Homeward Bound and Sunrise say the conditions on Tunnel Road also reflect a countywide rise in unsheltered homelessness, driven mainly by housing costs. And inside the building, the case management worked. People who had once been chronically homeless stayed housed: of the residents admitted since January 2026, 94 percent were still there, Homeward Bound told Asheville Watchdog for a June 2026 story. Six residents have moved on to more independent living since the building opened. Two hard years for a neighborhood are not the same as a failed model.

The numbers turned only recently. In 2026 the police put more officers on the corridor, and a department spokesperson told Asheville Watchdog, for the same June 2026 story, that the overall rate of reported crimes there had fallen 12 percent. Property crimes rose even so, and the police credit their own enforcement for the drop. Early in 2026, neighbors began forming the Tunnel Road Corridor Coalition, a group of homeowners and business people, to fight the rise in reported crime since Compass Point and the drop-in center opened.

That local fight has a national echo. The conditions neighbors organized against, the encampments, the open drug use, the strain on the police, are the same ones, scaled up, that the 2025 executive order was named for: "Ending Crime and Disorder on America's Streets." Asheville lived the small version of the argument now driving federal policy, and Compass Point's retention number is the part that argument leaves out.

Permanent supportive housing is the deep end, but it is only part of the story. The research is clear that the model is built for the high-need minority, and that most homelessness needs lighter, faster help. Homeward Bound covers that range too, which is what a working local system looks like.