Reversing a two-decade consensus

For roughly twenty years, federal homelessness policy rested on a single organizing idea: get people into stable housing first, then address the problems — addiction, mental illness, unemployment — that often accompany homelessness. That idea, called “Housing First,” was not a partisan project. It was adopted under the George W. Bush administration, expanded under Obama, and retained under the first Trump administration. It was the framework HUD used to score and award most of its homelessness grants.

In late 2025 and through 2026, HUD has moved to dismantle that framework. The Department now publicly describes Housing First as a failure, and is rewriting the rules of its largest homelessness grant program to steer money toward transitional and treatment-conditioned housing instead. About twenty states and a coalition of cities and nonprofits have sued, and federal courts have so far blocked the change.

This brief explains what Housing First is, what the evidence shows, what HUD is changing, and where the legal fight stands.

What “Housing First” means

Housing First is a model, not a single program. Its core claim is sequencing: housing is provided without first requiring sobriety, treatment compliance, or employment. Services — addiction treatment, mental-health care, case management — are offered alongside the housing and are voluntary rather than a condition of keeping the unit.

It is usually contrasted with the older “treatment first” or “linear” model, in which a person must move through stages — sobriety, then transitional housing, then permanent housing — and can be removed for failing a stage.

In HUD’s grant system, Housing First is delivered primarily through permanent supportive housing: long-term subsidized housing paired with services, with no time limit.

What the evidence shows

Housing First is among the more heavily studied interventions in US social policy, including randomized controlled trials.

• Housing retention. Permanent supportive housing programs report long-term housing retention rates as high as 98 percent. Reviews comparing Housing First to treatment-first models have found substantially better housing stability (National Low Income Housing Coalition).
• Substance use. Studies have generally found no worse substance-use outcomes under Housing First than under treatment-first models — undercutting the claim that providing housing without a sobriety requirement enables addiction (systematic review and meta-analysis of randomized controlled trials).
• Public cost. A systematic economic review found that for permanent supportive housing with Housing First in the US, benefits exceeded costs, with reductions in hospital visits, emergency care, and criminal-legal-system contact (Community Guide systematic economic review).

This does not mean Housing First solves every problem. It is most effective at keeping people housed; it is not, by itself, a treatment program, and outcomes on employment and self-sufficiency are more modest. Critics argue that the model under-invests in recovery. But the central empirical finding — that housing people directly keeps them housed better than requiring them to earn housing — is well supported.

What HUD is changing

The mechanism is HUD’s Continuum of Care (CoC) program, the roughly $4 billion in annual grants that fund local homelessness systems (National Association of Counties). HUD does not run shelters directly; it sets the terms of a competition, and local “continuums” apply for renewals and new projects. By controlling how applications are scored, HUD effectively decides what kind of housing gets built.

In November 2025, HUD issued new CoC funding rules that would (HUD):

• Cap permanent supportive housing. Limit the share of CoC funds going to permanent housing — in practice often around 90 percent in local systems — to as little as 30 percent, redirecting the rest toward transitional housing, street outreach, job training, and addiction treatment.
• Condition assistance. Prioritize programs that tie housing to work or treatment participation.
• Restrict certain uses. Bar CoC funds from supporting “harm reduction” overdose-prevention strategies, services specifically for transgender clients, and diversity-and-inclusion efforts.

HUD Secretary Scott Turner has framed this as ending “failed ‘Housing First’ policies” in favor of programs that “promote treatment, recovery, and self-sufficiency” (HUD). HUD says it will publish a revised CoC funding notice and continues to defend the shift.

The structural concern is that transitional housing is generally more expensive per person and time-limited, and that research has found it tends to screen out the people with the most serious needs — the population permanent supportive housing was designed to reach. A sharp cap on permanent housing would, on its face, put current permanent-housing residents at risk if their projects lost renewal funding.

The legal challenge

The change has been blocked, so far, in court.

In November 2025, a coalition of about twenty states — including Arizona, California, New Jersey, New York, and Vermont — plus the District of Columbia sued HUD (Washington State Attorney General). A second suit followed on December 1, 2025, from organizations including the National Alliance to End Homelessness and the National Low Income Housing Coalition, joined by the cities of Boston, Tucson, and San Francisco and Santa Clara County (Rhode Island Current).

The core legal argument is about authority: that Congress created and funds the CoC program, and that HUD cannot unilaterally restructure it — capping permanent housing, adding conditions — without congressional authorization. The plaintiffs also argued the last-minute timing would itself cause harm by freezing grant renewals.

• A federal judge in Rhode Island temporarily blocked the rules in December 2025 and ordered HUD to process applications under the original criteria (Rhode Island Current).
• In March 2026, the same court found HUD had violated the law, describing the rollout as a “slapdash imposition of political whims” (Atlanta Civic Circle).
• In April 2026, after a First Circuit panel declined to lift the lower-court order, the administration dropped its appeal — leaving the block in place while the broader case continues (Minnesota Lawyer).

HUD has signaled it intends to try again through a future funding notice. Congress, in the meantime, directed HUD to renew expiring CoC projects so that local systems would not lose funding during the litigation.

Why it matters

This is a contest over both policy and process. On policy: whether two decades of evidence favoring Housing First should be set aside for a model — transitional, treatment-conditioned housing — that the research record treats more skeptically. On process: whether an agency can rewrite a congressionally funded program by changing its scoring rules, or whether that requires Congress.

The outcome will shape what kind of housing hundreds of thousands of formerly homeless people can rely on, and whether the local nonprofits that run these systems can plan past the next grant cycle.

What to ask your representatives

• Should HUD be able to cap permanent supportive housing and redirect Continuum of Care funds without new authorizing legislation from Congress?
• Do they accept the evidence base for Housing First — and if they support shifting toward transitional and treatment-conditioned housing, what evidence supports that model?
• Will they support legislation that protects existing permanent-supportive-housing residents from losing assistance during the dispute over CoC funding rules?
• How are they monitoring HUD’s next Continuum of Care funding notice, given the courts’ findings against the earlier attempt?