Housing Markets First: Housing Supply and Affordability Are Key to Reducing Homelessness

The administration’s plan rests on an approach known as Housing First, which emphasizes the need for permanent housing before tackling other issues, such as mental health problems or substance use disorder. Proponents of this strategy advocate providing homeless people with housing regardless of whether they have a job or overcome other obstacles to stability. Once the need for housing is addressed, advocates argue, formerly homeless people will be able to tackle underlying problems, find a job, and stabilize their lives.

The Biden administration’s Housing First strategy is not new. Utah began implementing Housing First policies in 2005, and California, Houston, Seattle, Denver, Canada, and Finland followed in subsequent years.

This brief examines policy outcomes in three major Housing First locations: Utah, California, and Houston. In Utah and California, Housing First policies failed to reduce homeless numbers; chronic homelessness grew 95 and 93 percent, respectively, during their implementation. However, chronic homelessness in Houston declined by 68 percent following policy implementation.

Unlike California and Utah, Houston’s low-cost housing and lack of a traditional zoning code make it easier to produce and retain housing. National data indicate that a state’s relative land-use freedom explains 38 percent of the variation in homelessness between states and that housing affordability explains 34 percent of the variation.

Utah implemented Housing First policies in 2005 with the goal of ending chronic homelessness by 2015. The state claimed that it reduced chronic homelessness by 91 percent from 2005 to 2015, which initially prompted commentators and policy analysts to promote Utah’s Housing First policies as the gold standard.⁴ Unfortunately, the largest decline in homelessness during this period occurred between 2009 and 2010, when Utah changed its definition of chronic homelessness and stopped counting people in transitional housing as chronically homeless. As a result, methodological changes exaggerated the purported decline.⁵

Perhaps more significantly, since 2017, chronic homelessness grew rapidly both in the state generally and in Salt Lake County specifically (Figure 1 and Figure 2). In 2010, the state counted 406 chronically homeless individuals, and by 2017, this number had fallen to 185 (a 54 percent decrease). However, from 2017 to 2022, the number of chronically homeless people reached 792, a 95 percent increase from 2010 and a 328 percent increase from 2017.⁶

Most of the state’s homeless population resides in Salt Lake County, which follows a similar pattern (Figure 2). Chronic homelessness there fell around 29 percent between 2010 and 2017 and rose sharply (297 percent) between 2017 and 2022.

Utah’s Housing First initiatives specifically target the chronically homeless population, a subset of the total homeless population. However, it is worth considering changes to the total homeless population as well. Appendix A and Appendix B show that the number of people who are homeless on a given night did not change much between 2005 and 2022; although the number of homeless individuals trended down after 2012, counts later returned to former levels both statewide and in Salt Lake County.

California is another state that put Housing First at the core of its homeless policy response, and the state provides another case study. California began its foray into Housing First in 2004 when San Francisco implemented a policy based on Housing First principles called “Care Not Cash.”⁷ Gavin Newsom pushed the policy, which diverted funding from cash benefits for homeless people toward building permanent housing and shelters.⁸

In 2016, California adopted Housing First statewide.⁹ As a result, California now requires any state‐​funded homeless program to abide by the principles of Housing First, including allowing tenants to stay housed regardless of substance use.¹⁰ Governor Newsom revamped these efforts in 2020 with the introduction of Project Homekey, a program that aims to convert existing buildings into permanent supportive housing.¹¹

These efforts have proven expensive, and California has spent $3.7 billion on the Homekey program since announcing it in 2020.¹² Cities also are spending millions of dollars: San Diego spent over $62 million on permanent supportive housing between 2010 and 2018, significantly more than the $30 million it spent in that same period on homeless shelters.¹³

However, California’s chronically homeless and homeless populations grew despite the significant funding. Chronic homelessness in California fell by 51 percent between 2005 and 2016, but the trend reversed after 2016, the year Housing First was implemented statewide (Figure 3). Between 2016 and 2022, chronic homelessness increased by 93 percent, reaching levels not seen since 2005.

A similar trend is observed among the overall homeless population (Appendix C). Overall homeless numbers rose following policy adoption, and according to a 2022 report from the Department of Housing and Urban Development, 30 percent of the nation’s homeless population now live in California, which also has the highest homelessness rate in the country (44 people per 10,000).¹⁴

Results in San Francisco—where Housing First–based policies were in place more than a decade before they were adopted statewide—are also disappointing. While the number of chronically homeless individuals varies after 2005, it remains elevated: San Francisco was only able to reduce numbers below 2008 levels once, in 2015 (Figure 4). Overall, chronic homelessness increased 53 percent between 2005 and 2022.

California and San Francisco’s Housing First initiatives prioritize the chronically homeless, but it is also worth considering changes in the overall homeless population. Appendix D shows that the total number of homeless individuals in San Francisco rose steadily after implementing Care Not Cash. Specifically, between 2005 and 2022, overall homelessness increased 43 percent.

It is possible that California and Utah’s policies still reduced homeless counts over a counterfactual without the policies, but they are not delivering on lofty promises to end homelessness or even reduce homeless numbers from the baseline. In both states, the number of homeless people is increasing more quickly than they are being housed.

In contrast to Utah and California, Houston appears to have substantially reduced homelessness since adopting Housing First policies in 2011. Houston is part of a Continuum of Care, a regional entity that receives federal funds and administers a homelessness response, known as The Way Home.¹⁵ The Coalition for the Homeless, a nonprofit organization that helps coordinate the efforts of public and private stakeholders, leads the region’s efforts.¹⁶ Houston adopted a Housing First approach to homelessness in 2011 with a successful pilot initiative focused on housing homeless veterans.¹⁷ It subsequently expanded the program to focus on the broader chronically homeless population.

In 2020, regional partners invested additional money to fund Housing First projects with the Community-wide COVID-19 Housing Program. Phase One included a two-year, $65 million budget with the objective of housing 5,000 individuals. Phase Two, announced in January 2022, put $100 million toward the goal of housing 7,000 more individuals.¹⁸

Houston data show chronic and total homelessness declined following policy implementation. Chronic homelessness fell substantially in the years before Housing First was implemented, but the downward trajectory continued after the introduction of Housing First. Chronic homelessness declined by 68 percent since policy implementation in 2011 (Figure 5), and although numbers began to rise after 2017, Houston reversed the trend by 2022.

Houston shows a similar trend in total homelessness, where counts fell before policy adoption but increased sharply between 2010 and 2011 (Figure 6). However, total homeless numbers fell again after the implementation of Housing First: between 2011 and 2022, overall homelessness in Houston fell by 63 percent. Remarkably, this is despite the Houston area adding additional counties to its Continuum of Care in 2014 and 2017.¹⁹

Houston’s comparative success in reducing homelessness is partly attributable to the low cost of housing and elastic housing supply in that city, among other factors. Abundant, low-cost housing is an asset for governments working to reduce homelessness. Under the Housing First approach, low-cost housing makes it easier to buy or build permanent supportive housing units for homeless people. Average home values in Houston ($261,071) are a fraction of those in San Francisco ($1,277,409), the state of California ($728,134), and Utah ($506,072).²⁰ As a result, Houston was able to move 1,080 people into permanent supportive housing, provide short-term rental assistance for 3,180 people, and stabilize 2,780 people on the verge of homelessness with the $65 million spent on homelessness during Phase One.²¹

In contrast, Los Angeles spent almost $18.4 million buying and renovating motels with just 59 units to house homeless individuals, with an average cost of over $311,000 per unit.²² An audit in Los Angeles found that 14 percent of housing units for the homeless population cost more than $700,000 to build.²³ Last year, the Corporation for Supportive Housing estimated that California needs to build 112,527 units at a total cost of $67.9 billion, or $603,410 per unit, to end homelessness.²⁴ Subsidizing operating costs and rents for these units would cost an additional $22 billion over 12 years.

Ned Resnikoff, the policy director at California YIMBY stated that “the Housing First model can only work when homeless services agencies actually have enough units to meet their clients’ needs.”²⁵ Due to the availability of low-cost housing, Houston spends between $17,000 and $19,000 to house an individual for a year, while San Francisco must spend between $40,000 and $47,000 to do the same.²⁶

Similarly, Utah’s Office of the Legislative Auditor General identified housing availability and affordability as challenges for the state during an audit of Utah’s homelessness response in 2021.²⁷ Permanent housing in Utah is expensive, with an estimated cost of $250,000 to $275,000 per unit built. Utah’s Office of the Legislative Auditor estimates that Utah is short 1,200 units to meet current demand. Thus, it would cost the state between $300 million and $330 million to catch up with current demand.

To put this number in perspective, Governor Spencer Cox’s proposed fiscal year 2024 budget, which has been called “Utah’s highest budget ever,” recommends $20 million in state funding for “deeply affordable housing,” at least some of which would benefit the homeless population. This is in line with recent years’ spending, in which the state allocated between $3 million and $23 million annually for homelessness.²⁸

Utah officials agree that the high cost of permanent housing is making Housing First difficult to sustain. Joseph Jensen, who oversees data collection and management at the Utah Office of Homeless Services, said, “We do feel like [the increase in homelessness is] being driven, at least in part, by the tightening housing market.” Unfortunately, Housing First initiatives become more expensive when housing is expensive, and housing in Utah costs more than twice as much on average today than it did in 2015.²⁹

This is particularly problematic since Housing First’s focus on permanent placements means that there are rarely vacancies in the Housing First stock.³⁰ According to Utah’s 2021 audit, between 92 and 95 percent of homeless residents placed in permanent housing between fiscal years 2017 and 2020 remained during the reporting period. This creates added pressure on the program’s finances because the state must continually build or acquire additional permanent housing units. Producing this housing is difficult or impossible to do successfully when housing is high-cost and supply is limited.

In addition to housing availability and affordability making it easier to sustain a Housing First approach, abundant, low-cost housing also averts homelessness by ensuring residents stay housed to begin with. In line with this, studies have long found a relationship between housing prices and rates of homelessness. For example, the book How to House the Homeless found that rent costs explained about 40 percent of the variation in homelessness rates.³¹ In a recent book, authors Gregg Colburn and Clayton Page Aldern found that the supply of affordable housing was a more accurate predictor of homelessness rates than mental illness, substance abuse, or poverty measures.³²

Besides low-cost housing and abundant housing supply, there are other factors that increase the effectiveness of Houston’s homeless response efforts. One factor is that Houston has better coordinated efforts over a considerably smaller geography. Houston’s homeless approach is implemented over a single Continuum of Care, whereas the state of California has 44 distinct Continuums of Care, all with local governments, counties, law enforcement agencies, and state departments of their own.³³

Another possibility is that the coordinating entity in Houston, Coalition for the Homeless, is more effective because it is a nonprofit rather than a government agency. Thus, it is better positioned to innovate, take smart risks, and respond in flexible ways to challenges compared to a government department tasked with overseeing homeless response efforts. As one example, Houston’s coalition actively reaches out to local landlords to negotiate additional housing units for the homeless population.

Houston’s “compassionate enforcement” policies that prohibit public camping, and politicians that discourage panhandling, have also potentially helped reduce homelessness in Houston.³⁴ Of course, clearing encampments only reduces the homeless population if individuals have somewhere else to stay. In 2019, Houston spent about $3.4 million clearing encampments, but the city ensured that affected residents had access to either permanent supportive housing or a housing voucher.³⁵

Although a variety of factors influence homeless policy outcomes, policies that increase prices or limit housing supply appear to be among the most significant. There are a variety of reasons for high home prices, but zoning and land-use regulations that limit housing supply, density, and housing innovation are a major factor. A large body of research finds that increased land-use regulation leads to a decrease in the supply of housing, which in turn raises prices.³⁶ In Manhattan, San Francisco, and San Jose, researchers found that zoning regulations pushed up the cost of apartments by about 50 percent.³⁷

Given the relationship between regulation and costs, it should not be a surprise that research finds an association between housing regulation and homelessness. A 2019 Council of Economic Advisors paper titled The State of Homelessness in America found that “if the 11 metropolitan areas with significantly supply-constrained housing markets were deregulated, overall homelessness would fall … by 31 percent in these 11 metropolitan areas, which currently make up 42 percent of the United States homeless population.”³⁸

The Freedom in the 50 States’ land-use freedom measure combines the Wharton Residential Land Use Regulatory Index with rent control measures and other proxies for land-use restrictiveness.³⁹ Comparing land-use regulation with homelessness, the authors of this brief find a statistically significant, direct relationship between greater land-use regulation and homelessness (Figure 7). A state’s relative land-use freedom explains 38 percent of the variation in homelessness between states.⁴⁰ Moving from the lowest level of regulation to a median level of regulation increases homelessness by more than 3.5 times.

Similarly, there is a statistically significant, inverse relationship between state housing affordability and homelessness (Figure 8), and state housing affordability explains 34 percent of the variation in homelessness.⁴¹

Because Houston is the only major city that lacks a traditional zoning code, low regulation is a significant factor driving both its housing affordability and successful homelessness response.

Reforms that eliminate zoning, improve permitting speed, legalize greater housing density, and remove barriers to housing innovations—including co-living units, tiny homes, and manufactured homes—are all part of a successful strategy to reduce homelessness.

Given Utah and California’s experience with Housing First, adopting a Housing First approach nationally is unwise. Houston is unique among major American cities in its lack of a traditional zoning code, pro-development culture, and responsive housing supply. American cities adopting the Housing First approach absent these characteristics are unlikely to experience similar success.

Moreover, the average home price in the United States is almost twice as high as it was a decade ago, and a nationwide Housing First approach would come with an enormous price tag.⁴² Rather than pushing Housing First nationwide, national policymakers should allow local governments to continue experimenting with strategies to reduce homelessness.

At the local level, there are many constructive measures that policymakers can take to reduce homelessness. Although homelessness is a complex issue, and housing affordability is not the only contributor, access to low-cost housing plays a critical role in prevention and resolution. Reforms to improve access to low-cost housing are independently useful, and a low regulatory environment also seems to improve the outcomes of Housing First policy initiatives.

There are many ways to improve housing affordability, including dramatically reducing regulatory barriers to housing development and increasing housing supply.⁴³ Local policymakers should consider Houston’s example and act swiftly to address homelessness with reforms to increase housing supply and reduce costs. Deregulating housing markets is key to reducing homelessness and must be prioritized first.