The Limits of Housing First

Los Angeles makes a $1.2 billion bet on a dubious solution to homelessness.

In 2016 influential political leaders, activists, and media outlets in Los Angeles said they had a simple solution to homelessness: build more housing. Echoing an argument heard across the country, they claimed that rising rents have thrown people onto the streets and that by directly providing free “permanent supportive housing,” cities can reduce the number of people on the streets and save costs on emergency services.

In response, 77% of Los Angeles voters approved a $1.2 billion bond for the construction of 10,000 units for the city’s homeless. That commitment made Los Angeles the most significant testing ground for the “Housing First” approach that has become the dominant policy idea on homelessness for West Coast cities. Even before the passage of the bond, the concept’s creator, Sam Tsemberis, was lavished with praise by the national media. In 2015, the Washington Post wrote that Tsemberis had “all but solved chronic homelessness” and that his research “commands the support of most scholars.”

In the years since, “Housing First” has taken even greater hold in California and the across the West. Los Angeles Mayor Eric Garcetti recently declared that “we need to have an entitlement to housing.” California Gov. Gavin Newsom went a step further, arguing that “doctors should be able to write prescriptions for housing the same way they do for insulin or antibiotics.”

Five years in, the project has been plagued by construction delays, massive cost overruns, and accusations of corruption. The Los Angeles city controller issued a scathing report, “The High Cost of Homeless Housing,” which shows that some studio and one-bedroom apartments were costing taxpayers more than $700,000 each, with 40% of total costs devoted to consultants, lawyers, fees, and permitting. The project is a boon for real estate developers and a constellation of nonprofits and service providers, but a boondoggle for taxpayers. The physical apartment units are bare-bones — small square footage, cheap flooring, vinyl surfaces — but have construction costs similar to luxury condos in the fashionable parts of Los Angeles. Meanwhile, unsheltered homelessness has increased 41%, vastly outpacing the construction of new supportive housing units. Los Angeles magazine, which initially supported the measure, now wonders whether it has become “a historic public housing debacle.”

Before completing a single housing unit, the city reduced its projected construction from 10,000 units to 5,873 units over 10 years, with the potential for further reductions in the future. But the long-term problem runs much deeper: Even if one accepts that permanent supportive housing is the solution, there are currently more than 66,000 homeless people in Los Angeles County. Under the best-case scenario, Proposition HHH will solve less than 10% of the problem over the course of a decade.

Despite Housing First’s uncertainties, other West Coast cities desperate to solve homelessness, including Seattle and San Francisco, have been captured by its seductive messaging and promise of respite. As Los Angeles grapples with the unforeseen consequences of its big bet on “Housing First,” the federal, state, and local governments, especially in major metropolitan areas, are preparing to commit billions of dollars to the program, whose track record remains woefully underexamined.

Ever since clinical psychologist Tsemberis pioneered the model in New York City in the 1990s, political leaders, activists, and academics have insisted that Housing First is an “evidence-based” intervention that reduces homelessness, saves taxpayer money, and improves lives. Supporters frequently argue that the program reduced costs in a study of chronic alcoholics in Seattle, consistently demonstrates high retention rates in multiple academic surveys, and eliminated chronic homelessness in Utah. “We're going to stem this crisis by building supportive housing in every neighborhood throughout Los Angeles,” City Council member Herb Wesson recently claimed.

These studies, however, are not as persuasive as activists suggest. Although the study of chronic alcoholics in Seattle does show a net reduction in monthly social service costs of $2,449 per person, this figure does not include $11 million in capital and construction costs for the housing units themselves; in other words, Housing First saves money if the cost of housing is not included. Even on its own favorable terms, the study’s purported savings aren’t as dramatic as they appear: While the Housing First participants showed a 63% reduction in service costs over six months, a wait-listed control group that was not provided housing showed a 42% reduction in service costs over the same time period, raising questions about the specific effectiveness of the intervention.

Claims that studies show one-year retention rates of roughly 80% for Housing First participants are open to question. In a meta-study of three best-in-class Housing First sites, researchers found that 43% remained in housing for the first 12 months, 41% were “intermittent stayers” who left and returned, and 16% abandoned the program or died within the first year. These findings challenge the argument that Housing First is a long-term solution to homelessness.

Finally, advocates and the media have long touted Utah as the gold standard of Housing First. “The Daily Show” called the state’s program “mind-blowing,” the Los Angeles Times reported in 2015 that Utah “is winning the war on chronic homelessness,” and dozens of media outlets announced that the state “reduced chronic homelessness by 91%.” These miraculous results, however, were not the result of Housing First policies, but apparently clerical manipulation by state officials. According to the Deseret News and economist Kevin Corinth, “As much as 85% of Utah's touted reductions in chronic homelessness … may have been due to changes in how the homeless were counted.” It’s not that all of the chronically homeless were housed; they were simply transposed onto a new spreadsheet. Moreover, between 2016 and 2018, the number of unsheltered homeless in Utah nearly doubled – hardly the victory that Housing First activists had declared.

The recent debate surrounding Housing First has predominantly been focused on the physical and budgetary metrics of housing retention and cost reductions. But these surface-level concerns obscure a deeper question: What happens to the human beings in these programs? The results, according to the vast majority of studies, point to a grim conclusion: Housing First does not meaningfully improve human lives.

Although housing programs are often an effective solution for families experiencing a temporary loss of shelter, Housing First programs do not have a strong track record improving the lives of the unsheltered homeless — the people in tents, cars, and on the streets — who often suffer from more severe challenges. According to research by the California Policy Lab, 75% of the unsheltered homeless have substance abuse condition, 78% have mental health conditions, and 84% have physical health conditions. In theory, Housing First would address these problems. In every program, residents are offered a wide range of services. At the Pathways to Housing program in New York City, a flagship program founded by Sam Tsemberis himself, residents are served by an “interdisciplinary team of professionals that includes social workers, nurses, psychiatrists, and vocational and substance abuse counselors who are available to assist consumers 7 days a week 24 hours a day.” However, despite this massive intervention, the Pathways program shows no reduction in substance abuse or psychiatric symptoms over time – in fact, those conditions often worsened.

This basic finding is confirmed by a range of studies showing that residents of Housing First programs show no improvement regarding addiction and mental illness. They are housed but broken, wracked by the cruelest psychoses, compulsions, and torments – all under the guise of medical care.

A Housing First experiment in Ottawa, Canada, illustrates this paradoxical outcome in stark terms. Researchers divided the study into two populations: an “intervention” group that was provided Housing First and access to primary care, medically assisted treatment, social workers, and on-demand services; and a non-intervention “control” group that was not provided housing or services – they were simply left on the streets. To the shock of the researchers, after 24 months the non-intervention control group reported better results regarding substance abuse, mental health, quality of life, family relations, and mortality than the Housing First group. In other words, doing nothing resulted in superior human outcomes than providing Housing First with wraparound services.

One explanation may be that Housing First programs are deliberately not oriented toward recovery, rehabilitation, and renewal. They operate on the “harm reduction” model, which allows residents to continue using drugs such as alcohol, heroin, and methamphetamine, and does not require mental health treatment as a condition of residency. In theory, this permissive policy would help “reduce harm” to the individual; in practice, however, it may create a community-level effect that makes it hard for any individual to find recovery. Here is the basic chain of events: Homeless individuals with substance abuse and psychiatric disorders are placed together in a residential facility where they are allowed to continue the way of life they had on the streets. Despite the availability of services, there is no incentive to use those services and no disincentive to the problematic behavior associated with street homelessness. Consequently, widespread addiction often becomes the norm within Housing First programs. 

Preferring Homelessness

This chain of events is not just a thought experiment. In Birmingham, Ala., researchers inadvertently created this exact problem when they put participants of two different programs – one “recovery” program and one “harm reduction” program – in the same apartment complex. Immediately after beginning the experiment, the recovery group “began abandoning the provided housing, complaining that their proximity to persons not required to remain abstinent (i.e., the other trial group) was detrimental to their recovery. They claimed that they preferred to return to homelessness rather than live near drug users.” The researchers quickly stopped and reorganized the trial, writing that “this unexpected reaction shows one possible risk to housing persons with active addiction."

Still, Housing First advocates insist that their policy is working. When reached for comment, Tsemberis insisted that the Washington Post headline declaring that he had “solved homelessness” is true. “The most effective way to end homelessness for people with mental health and addiction is to provide housing and wraparound support,” Tsemberis said. He points towards rates of “housing stability” as the key metric, while conceding that Housing First does not provide “a cure for mental illness and addiction.” This is a suggestion that policymakers have “solved homelessness” simply by bringing people indoors, no matter their addictions, mental illnesses, and human torments.

Advocates portray Housing First as a science that transcends politics. The policy was first adopted by the George W. Bush administration and has gained support from Republicans and Democrats alike. As the Washington Post observed, it is “a model so simple children could grasp it, so cost-effective fiscal hawks loved it, so socially progressive liberals praised it.

However, the real-world evidence from cities such as Los Angeles challenges this narrative. If Housing First has demonstrated anything, it is this: It provides a stable residential environment for the homeless to live out their pathologies, subsidized by the public and administered by the social-scientific sector. It does, not however, address addiction, mental illness and other factors that limit human potential and lead to homelessness.

In Los Angeles, despite the insistence that Housing First is the answer, some uncertainty is creeping in. Mayor Garcetti is now on the defensive, as homelessness in Los Angeles continues to increase despite billions in spending. After the federal government released a study questioning the premises of Housing First, Garcetti backed away from the unidimensional approach, telling reporters with irritation in his voice: “Sometimes people parody Housing First as ‘only housing.’ Nobody embraces only housing. It’s got to be housing with services together.”

In more bad news for public officials and supporters of Housing First, there is an emerging body of evidence that calls into question the “cost savings” of the program. A recent study in Massachusetts shows that Housing First does not reduce rehospitalization and service utilization, while another study in Chicago suggests that Housing First might increase overall costs. Furthermore, researchers have concluded that the purported cost savings in earlier Housing First studies would not apply to the 82% of the homeless population that is not chronically homeless.

In Los Angeles, this could spell disaster. In the most optimistic scenario laid out by the controller’s office, the city will build 5,873 supportive housing units at an initial cost of $1.2 billion, plus an estimated $88 million in annual service costs associated with the Housing First model. The recipients of this housing will not meaningfully improve their lives in terms of addiction, mental illness, and spiritual well-being — and there will still be 60,000 people on the streets across Los Angeles County. In other words, even under its own theoretical assumptions, Proposition HHH is doomed to fail.

The City of Los Angeles did not return a request for comment.

The potential silver living might be that a reconsideration of the Housing First approach could lead to a wider reckoning for policymakers and political leaders. At the end of the Housing First experiment in Los Angeles, the city will be responsible for thousands of wards of the state with little hope for recovery, as well as tens of thousands of campers in its public spaces. A few curious citizens will read through the academic literature and find a vast discrepancy between the ideological promises of Housing First and its real-world outcomes. They might then conclude that proponents should have known better.

This article was adapted from research for the new book "No Way Home: The Crisis of Homelessness and How to Fix It with Intelligence and Humanity."

Originally published at RealClearInvestigations.