Settlement Money Bought $340 Million of Housing Last Year. Most Agreements Never Recommended It.

Jason Fazekas slept in building stairwells on rainy nights and woke on winter mornings with fingers too numb to bend. He had been addicted to heroin for decades and had spent most of his adult life on the street or behind bars. In September he moved into a one-bedroom apartment in East Haven, Connecticut. In October he marked nine months of recovery. He makes grilled cheese sandwiches, takes occasional jobs cleaning out trucks, attends a daily treatment program and virtual meetings, and is usually in bed by 8:30.

The apartment came through a state program called Housing Empowering Recovery from Opioids, funded with $58.6 million in opioid settlement cash. Participants need a verified substance use disorder to get in. They receive a rental voucher, a case manager, and money for a security deposit or basic furniture.

Multiply that story across the country and you get a number that landed this week with almost no ceremony: state and local governments reported spending or committing more than $340 million of settlement money on housing-related initiatives in 2025. That is roughly 9% of the $3.7 billion publicly reported as spent or committed last year, according to the annual review by KFF Health News, the Johns Hopkins Bloomberg School of Public Health and the nonprofit Shatterproof, published Nov. 12 — a catalog of more than 17,500 individual uses of settlement cash.

Housing drew about as much as the prevention bucket and more than the core strategy covering overdose-reversal drugs like naloxone. It is also not one of the core strategies most settlement agreements recommend. It is merely allowed.

What Nine Percent of a Settlement Actually Buys

The national settlements will pay out something near $60 billion over almost two decades, which is why every category share matters. At least 85% of the money going directly to states and localities must be used to address the epidemic. Inside that requirement, local officials decide how tightly each project has to target addiction, and that is where the review found daylight.

Housing and homelessness overlap with addiction in the public imagination far more than they do in the data. A study published last year using California data, a state that accounts for about a quarter of the nation's homeless population, found that 37% of homeless people had used illicit substances regularly in the previous six months and 25% had never used drugs at all. Twenty-three percent began regular use only after they lost housing — addiction did not put them on the street; the street pushed them toward self-medication, sometimes stimulants to stay awake and avoid assault.

"Most people who are homeless do not have an addiction. And most people who have addictions are not homeless," Margot Kushel, who directs the UCSF Benioff Homelessness and Housing Initiative and co-authored that study, told KFF Health News.

That single sentence is the test. Money that houses people with substance use disorders is opioid remediation. Money that reduces the visibility of encampments is homeless policy wearing a settlement invoice.

In Connecticut, the Eligibility Bar Is the Program

The Connecticut commitment is the largest highlighted in the review, and its structure explains why it reads as remediation rather than general housing. The $58.6 million breaks into $14.25 million a year for four years, enough for up to 500 people with substance use disorders, plus $400,000 a year for an evaluation by the University of Connecticut. As of mid-September, about 170 people had moved in and roughly 180 were still looking for a place.

Two things make it work as a model and one makes it fragile. The eligibility bar — a verified diagnosis — keeps the money aimed. The case management is what turns a voucher into a lease in a rental market that does not want complicated tenants. What it does not have is speed: roughly half the enrolled group was still searching months in, which is a housing-supply problem no settlement check solves.

In Louisville, a Million Dollars for Picking Up Litter

Louisville, Kentucky put more than $1 million of settlement money into a Goodwill program called Another Way, which gathers people living on the street, takes them around the city to work — often picking up litter for a few hours — and pays them $50 plus lunch. About 1,800 people took part in 2025. Roughly 270 landed full-time jobs and about 240 were connected to treatment. Participants are not screened for substance use disorder.

Mayor Craig Greenberg's spokesperson, Matt Mudd, defended the spending on outcome grounds and offered a single sentence of connection to the epidemic: "Overdose is the number one cause of death for unhoused people." Brittany Morris, who works at a Louisville nonprofit advocating on overdose and poverty policy, was not persuaded. Another Way "helps people that are on the streets," she said. "But do I feel like it's related to the opioid crisis? That's basically implying every homeless person is a drug addict hooked on heroin."

The Louisville contrast is the sharpest in the review because the same city contained the program advocates wanted funded. Arthur Street Hotel, a transitional housing site for people with mental illness and substance use disorders, housed more than 400 people over its tenure on an annual budget of about $3.5 million, with intensive case management that sometimes took weeks to earn a person's trust. It closed in July for lack of funding. Dominic Sitz, housed there in 2022 and now four years into recovery, spent his first 48 hours in his own apartment with the door locked, sleeping on the tile floor, no food, no furniture, just the sense of safety.

An Allowable Use Is Not a Recommended One

New Brunswick, New Jersey listed $520,000 for "homeless services." Asked what was funded and how it related to addiction, city spokesperson J.T. Miller wrote that the money supported staff and services engaging the homeless population on "assistance, prevention and drug addiction programs," and pointed to a city document describing the use as "budgetary offsets and allocations" in its Department of Human and Community Services.

That phrase is the tell for the failure mode recovery advocates have warned about since the first checks arrived: settlement dollars substituting for money a city was already spending, rather than adding to it. The difference does not show up in a spending total. It shows up two budget cycles later, when the general fund line item that used to cover a program quietly shrinks.

Housing also sits inside a broader category the review labeled recovery support — transportation, education, job training — that totaled more than $786 million, about a fifth of 2025 spending. Dan Kelly, who advises local governments on settlement spending for the Michigan Association of Counties, told KFF Health News that the important thing is "maintaining that connection to opioid remediation and the populations impacted by opioid use disorder." He is describing a discipline, not a rule, because the settlements do not supply one.

The Rules Were Written to Allow This

Nothing in the agreements prevents a county from spending opioid money on a litter-pickup program. The national settlements require that 85% of direct payments address the epidemic, then hand the definition of "address" to the same officials who are under pressure about encampments and visible homelessness. That is a governance design, not a scandal, and it means the accountability layer is local: county commissions, city councils, opioid advisory committees, and the handful of residents who show up to public comment.

The data has its own ceiling. The database reflects government actions through Dec. 31, 2025, collected as of July 2026, and many jurisdictions publish nothing at all, so the $340 million is a floor rather than a total. A state that reports nothing is not necessarily spending badly; it is simply invisible.

How to Read Your County's Housing Line Item

Three questions separate remediation from relabeling, and none of them require a public records request. Is the program reserved for people with substance use disorders, or open to anyone who is unhoused? Does it connect residents to treatment, or only to a bed? Does it track outcomes that would matter to a family — housing stability at twelve months, overdoses among participants, time retained in care?

What to ask before choosing a recovery residence

For anyone choosing a recovery residence right now, the questions are more immediate: whether the home is state-certified or follows national standards, who staffs it overnight, what it costs, and whether residents may continue prescribed medications for opioid use disorder. Some homes still bar those medications, which forces a person to choose between housing and the treatment that keeps them alive. Neither a settlement check nor a voucher fixes that, though the difference between opioid addiction treatment that continues and treatment that stops is often the difference between a lease renewal and a return to the stairwell.

Housing is the intervention with the least glamorous evidence base and the most plausible mechanism: a person with a stable address can hold a dose schedule, keep appointments, and stop sleeping with one eye open. Whether the $340 million buys that, or buys quieter streets, depends entirely on eligibility criteria that are set in rooms most families never enter. The levels of care a person can actually reach — including the recovery residence at the end of the sequence — are decided as much by a county's line-item choices as by anything a clinician recommends.