Housing First is Not Housing Only
Much of BHHI’s work is rooted in a simple premise: The solution to homelessness is housing. While lack of a permanent home is not the only challenge facing most unhoused people, it is the one challenge they all have in common. And whatever other challenges they may have—whether they include persistent unemployment, physical disability, mental illness, or none of the above—are significantly harder to manage for people who are simultaneously dealing with the stress of homelessness. Giving people homes creates the conditions necessary for them to heal and flourish. Or, as BHHI Director Margot Kushel, MD, likes to say: “There is no medicine as powerful as housing.”
This is the central insight of the Housing First approach to ending homelessness. Prior to the widespread adoption of Housing First-style social programs, public anti-homelessness measures typically adhered to a discipline-and-punish strategy. Individual behaviors like substance use were assumed to be the main cause of homelessness. As a result, the solution was to incentivize lifestyle adjustments on the part of the unhoused. Housing was a reward for good behavior.
Now we know better. People who are homeless don’t suffer from a lack of motivation to lift themselves up; they suffer from homelessness. That is why the presidential administration of George W. Bush first adopted Housing First as federal policy, and why, more recently, California has invested billions in Housing First-style anti-homelessness interventions such as Project Homekey.
But while experts and policymakers generally favor the Housing First approach, it does not by any means enjoy universal popularity. Opponents such as Pacific Research Institute for Public Policy fellow Wayne Winegarden tend to unfavorably contrast it with what they call the “treatment first” model. Without making access to housing conditional on first receiving treatment, says Winegarden, “All we’re doing is taking the problem from the street to the hotel room.”
Most critiques of Housing First reiterate three key points:
- Housing First is “one-size-fits-all
- Housing First interventions don’t provide adequate treatment to clients. Or, as one critic put it, “Housing-first policy is indistinguishable from housing-and-nothing-else.”
- Housing First interventions are ineffective. (The right-leaning Manhattan Institute has observed that homelessness continued to rise in California following the adoption of Housing First as a statewide strategy.)
Arguments 1 and 2 are based on a conceptual misunderstanding. Argument 3 is simply inaccurate.
What Housing First Really Means
Let’s start with Argument 1, which is the most mysterious of the three. What does it mean for Housing First to be “one-size-fits-all”? Critics tend to invoke the phrase without explanation, as if having a coherent statewide plan for solving homelessness was a priori objectionable. But who are the unhoused people who would not benefit from receiving housing? Calling Housing First “one-size-fits-all” is a bit like complaining that blood transfusions are the Emergency Department’s “one-size-fits-all” response to major blood loss. The patient’s overall treatment plan may vary based on the nature of the blood loss—or the blood type of the patient—but everyone needs blood to live.
Similarly, everyone needs a home. But beyond access to permanent housing, the needs of people experiencing homelessness vary. While some individuals may need additional support in order to achieve lasting housing security, there is also a large share of the homeless population that only needs housing assistance.
This was one of the key insights from the federal Department of Housing and Urban Development’s Family Options Study, which evaluated various interventions for helping homeless families. Households that received a permanent housing subsidy—but no other form of specialized assistance—saw a sharp drop in homeless shelter utilization and cycling between unstable living situations. For many of the families who received this subsidy, “housing-and-nothing-else” (or, rather, rental assistance and nothing else) really was exactly what they needed.
But the Housing First approach is not only for those families. As the title of this post indicates, housing first is not housing only. And for people with severe physical, behavioral, or substance use challenges—challenges that could force them back into homelessness if not addressed—we should be offering wraparound services on a Housing First basis.
For people wrestling with chronic homelessness and of the aforementioned challenges, the evidence supports permanent supportive housing (PSH), an intervention that prioritizes moving unhoused people with particularly high needs into permanent housing while also offering them a variety of optional services. These services may include physical rehabilitation, mental health care, treatment for substance use disorders, and even home care. The specific treatment program is voluntary and customized to meet the needs of the client—precisely the opposite of “one-size-fits-all.”
Housing First Works
So far, I’ve only addressed critics’ (often willful) mischaracterization of the theory behind Housing First. But despite what Argument 3 claims, we also have a significant body of evidence demonstrating its effectiveness in real-world conditions. This data visualization from the National Alliance to End Homelessness makes the case better than I ever could: Housing First interventions have been tested again and again, with unambiguous results.
I already mentioned HUD’s finding on the effectiveness of simple housing subsidies for many unhoused families. But what about interventions targeted at the highest-need individuals? My own favorite study of the PSH model—the study I find myself returning to again and again—comes from here at BHHI. Margot, fellow BHHI faculty member Maria Raven, MD, and Matthew Niedzwiecki, PhD, conducted a randomized control trial of a PSH intervention offered in Santa Clara County on a Housing First basis. (For those who don’t consume much social science research: Randomized control trials are just about the closest you can get to replicating ideal laboratory conditions when studying a policy intervention out in the field.) The target population for this intervention was people with extremely high needs; as the researchers noted in their writeup of the study for Health Services Review, “Participants averaged five hospitalizations, 20 visits to the emergency department, five to psychiatric emergency services, and three to jail in the two years prior to being enrolled.”
In other words, this Housing First-aligned treatment was specifically for the hardest-to-treat members of Santa Clara’s unhoused community. The results of the intervention were extraordinary: 86% of those who received the treatment were successfully housed and remained housed for the vast majority of the follow-up period (which averaged around three years). Similarly, there was a sharp drop in utilization of emergency psychiatric services among the treatment group, corresponding to a rise in scheduled mental health visits.
Not only does Housing First work, but the evidence shows that it can work for even the highest need population of people experiencing homelessness. The 86% success rate cited in the Health Services Review article, while impressive, actually understates the intervention’s effectiveness. When BHHI researchers revisited Santa Clara for additional data, they found that more than 90% of participants had been housed and remained housed over the long term.
So why is California still experiencing a homelessness crisis? It’s not because we don’t know how to house people. Rather, it’s because the scale of public investment in Housing First infrastructure doesn’t match up to the enormity of the problem. While the state took a step in the right direction this year when it allocated $12 billion out of its most recent budget to fighting homelessness, it will take years of work to adequately build up California’s existing Housing First infrastructure.
Furthermore, while Housing First programs are the best way to help most people who are already homeless, they do not prevent people from becoming homeless in the first place. If Housing First programs were to successfully re-house 10,000 people over one year, but rising housing costs forced 20,000 people into homelessness over the same period, then the homeless population will have grown on net. That is why Housing First interventions need to be paired with policies that bring down housing costs, offer protection to struggling renters, and prevent people from falling into homelessness in the first place.
A full write-up of the revised and updated Santa Clara study is forthcoming. In future posts, I’ll dive deeper into the new findings and what BHHI has learned from watching this Housing First program up close. Stay tuned.