Housing Supply and the Drivers of Homelessness
Homelessness is an increasingly salient policy issue across all levels of government—as well as a contentious political one. While urban communities and their representatives often frame the issue in terms of public safety, substance use, and mental health, some policy researchers emphasize the relationship between homelessness and housing markets.
A new Biden administration initiative, “All In: The Federal Strategic Plan to Prevent and End Homelessness,” also highlights the connection between homelessness and housing supply. The plan sets out goals to reduce homelessness by 25% by January 1, 2025, increase the supply of supportive housing for unhoused populations, and recommit the federal government to a “Housing First” approach—a framework that prioritizes providing unhoused people with permanent stable housing as the foundation for all other social service goals and seeks to minimize the requirements for housing access.
However, in response to an increase in unsheltered homelessness, some state and local governments are implementing policies that break with the priorities listed in “All In.” This includes clearing homeless encampments without alternative housing options and strengthening laws against loitering and sleeping on public-owned land. As the debate between different approaches to solving homelessness takes on new dimensions, policymakers must understand the fundamental interplay between homelessness and housing supply to craft sensible and effective responses.
This report reviews recent demographic information about homelessness in the U.S., shares insights from research on the connection between homelessness and housing supply, and identifies commonsense, evidence-based solutions for addressing homelessness while ensuring community safety and vibrancy.
The Basics of Homelessness in the U.S.
The U.S. Department of Housing and Urban Development’s (HUD) annual Point-in-Time (PIT) and count, presented as part of the agency’s Annual Homelessness Assessment Report, is the primary source of nationwide data on homelessness.1 It is a count of sheltered and unsheltered people experiencing homelessness on a single night in January, performed across the country by Continuums of Care (CoCs)—regional or local planning bodies composed of nonprofit organizations and government entities that coordinate support services for homeless populations in a specific geographic area. HUD requires CoCs maintain a Homeless Management Information System (HMIS) to collect data on the provision of housing and services to homeless individuals and households at risk of homelessness.
The 2022 PIT count found:i
- Approximately 582,500 people were unhoused in January 2022.
- 60% of those people were sheltered in locations such as emergency shelters, safe havens, or transitional housing programs, while the remaining 40% were unsheltered—i.e., living on the street, in abandoned buildings, or in other places unsuitable for human habitation.
- Homelessness increased slightly from 2020-2022, with the increase driven by a growing unsheltered population.
- 76% of all people experiencing homelessness were adults 25 or older, 17% were under the age of 18, and 7% were young adults ages 18 to 24.
- Six of every 10 people experiencing homelessness were men or boys, while 38% were women or girls, and less than 1% identified as transgender, did not identify as singularly female or male, or were questioning their gender identity.
- People who identify as transgender, gender non-confirming, or questioning exhibit high rates of unsheltered homelessness.
- 50% of people experiencing homelessness identified as White; 37% identified as Black; 3.4% identified as American Indian, Alaska Native, or Indigenous; 1.8% identified as Native Hawaiian or Pacific Islander; 1.4% identified as Asian; and 6.1% identified as multiple races. 24% identified as ethnically Hispanic or Latino.
- Black and Indigenous people are overrepresented among the homeless population compared to their share of the U.S. population.
The PIT count also shines a light on the geographic distribution of homelessness:
- Homelessness is concentrated in urban jurisdictions, with half (50.3%) of unhoused people living in the country’s 50 largest cities.
- More than half of all people experiencing homelessness in the country live in four states: California (30%), New York (13%), Florida (5%), and Washington (4%).
- California has the highest rate of homelessness and accounted for half of all unsheltered people in the country (115,491 people).
One noteworthy statistic offers a glimmer of hope: The number of unhoused veterans declined by 11% between 2020 and 2022. This improvement continues a trend in veterans’ homelessness reduction, with the number of veterans experiencing homelessness dropping by nearly 55% since 2009.
BPC previously wrote about how the reduction in veteran homelessness provides a pathway to addressing homelessness more broadly. Since 2009, interventions like the Housing First-focused HUD-Veterans Affairs Supportive Housing (HUD-VASH) program and Supportive Services for Veteran Families (SSVF), which offers rapid re-housing and prevention support, have increased the resources available for unhoused veterans.
Types of Homelessness
Researchers make a distinction between sheltered and unsheltered homelessness. Sheltered homeless populations have a temporary place to stay, such as short-term shelters or supportive housing facilities, while unsheltered populations live and sleep outdoors or in places unfit for human habitation.
The share of sheltered and unsheltered populations varies greatly across jurisdictions, often shaping local views on homelessness. For example, New York City had almost as many unhoused people as Los Angeles in 2022, but its share of unsheltered homelessness was much smaller (17% vs. 81.6%). This is in part due to New York City’s right-to-shelter law, which led the city to build the largest shelter system in the country.ii
Scholars and policymakers also differentiate between transitional, episodic, and chronic homelessness:iii
- Transitional homelessness describes people who are briefly unhoused and interact with a homelessness response system for a short period. Most people experiencing homelessness at a given moment fall under this category.iv
- Episodic homelessness involves frequent entries and exits from homelessness, such as people who frequently sleep in shelters.
- Chronic homelessness describes populations who remain unhoused for periods of a year or longer. Nearly 30% of unhoused persons in 2022 had chronic patterns of homelessness, and there has been a steady rise in the number of individuals experiencing chronic homelessness since 2016.v
Responses to Homelessness
The homelessness response system is complex and varies across jurisdictions and CoCs. There are six main types of support for households experiencing homelessness, divided into two categories: short-term shelter and permanent housing:2
- Emergency shelters: facilities providing temporary shelter for people experiencing homelessness. Some shelters focus on serving specific subpopulations, such as domestic abuse victims.
- Safe havens: places providing private or semi-private temporary shelter and services to unhoused people with severe mental illness.
- Transitional housing: temporary accommodations meant to help people experiencing chronic homelessness progress towards permanent housing by offering support services—such as for substance use disorders or mental health.
- Rapid re-housing: A low-barrier-to-entry intervention designed for families and individuals who do not need significant or ongoing support to exit homelessness. Rapid re-housing programs assist recently unhoused people in finding secure rental housing in the private market. It also includes limited financial assistance to cover moving expenses and case management to assist the household in remaining stably housed.
- Permanent supportive housing (PSH): An intervention providing safe and stable housing for unhoused people with chronic issues such as severe mental illness or substance use disorders, alongside voluntary support services. PSH operates on a Housing First basis, providing housing and voluntary services without requiring sobriety or the completion of transitional programs.
- Other permanent housing: Initiatives intended for individuals or families transitioning out of homelessness, but not restricted to people with disabilities or suffering from mental health issues or substance use disorders.
CoCs report the number of beds available across all programs in a Housing Inventory Count at the same time they conduct PIT counts.
Homelessness and Individual Characteristics
Poverty, unemployment, domestic violence, mental health issues, and substance use disorders may increase the probability someone will experience homelessness at some point in their life.
Recent research shows between 25-40% of individual unhoused people (i.e., not part of a family unit) have a substance use disorder, with around a quarter of unhoused people experiencing some form of mental illness.vi The chronically homeless population is also more likely to use illicit drugs, consume high levels of alcohol, and suffer from severe mental health conditions compared to those with stable housing.vii
Personal attributes that make individuals susceptible to discrimination—such as race, ethnicity, sexual orientation, gender identity, and disability status—also increase the possibility a particular person may become homeless.viii
However, the growing consensus among researchers is that individual attributes and circumstances (sometimes referred to as precipitants of homelessness) do not drive overall rates of homelessness. While they may make individuals more likely to experience homelessness, they do not explain why some places experience a greater incidence of homelessness than others.ix
The concentration of homelessness in specific places isn’t caused by the prevalence of poverty, unemployment, or other socioeconomic conditions. Cities with very high rates of poverty and unemployment, such as Cleveland or Baltimore, have some of the lowest per capita rates of homelessness in the country.x This trend holds for drug use as well. For example, while West Virginia has an extremely high drug overdose mortality rate compared to other states, it also maintained one of the lowest homelessness rates in the country.
Other variables beyond individual characteristics seem to drive the prevalence of homelessness in the places where it is most common.
Homelessness and Housing Supply
As Gregg Colburn and Clayton Page Aldern write in their book Homelessness is a Housing Problem, cities with large homeless populations tend to share specific characteristics:
- They have experienced significant population growth but have not added enough housing to meet the new demand, and
- The increasing demand for housing amidst a constrained supply of homes reduces vacancy rates and increases rents, squeezing low-income and vulnerable populations out of the housing market and into homelessness.3
Their main insight—that the difficulties in building housing in certain places drive the higher incidence of homelessness relative to other jurisdictions—ties the question of homelessness to the broader crisis of housing affordability across the country. As communities add jobs and experience economic growth, more people seek to live in them. In the absence of a robust housing market response to this inflow of new people, population growth will lead to an increase in rents across all levels of the market. This dynamic interacts with the individual characteristics laid out in the prior section to exacerbate the risk of homelessness for certain groups. In other words, people whose circumstances or attributes make them more likely to fall into homelessness are particularly vulnerable to the impact of these housing supply constraints.
According to an estimate by Rosen Consulting, the U.S. underbuilt housing by 5.5 million and 6.8 million homes over the last 20 years.xi Such an acute housing shortage impacts low-income households the hardest: For every 100 low-income households (earning less than 80% of the area median income), just 55 homes were affordable and available. For very-low-income households (earning less than 50% of the area median income), there were just 36 homes.
Research by Zillow shows homelessness increases at a faster rate in places where people spend 32% or more of their income on housing on average, a signal that higher rents seem to drive increases in homelessness.xii
While there is no single fix-all solution for homelessness, policies that address the shortage of affordable homes and provide demand-side supports to households at risk of eviction can go a long way toward preventing individuals and families from becoming homeless. Advocates and other policy experts have suggested these elements as part of an effective federal response to homelessness:
- Codifying emergency rental assistance programs: Emergency rental assistance, such as the funding provided by the CARES Act in response to COVID-19, was widely noted to have helped reduce the number of people in need of shelter during the pandemic. Other supports, such as cash payments to households, expanded unemployment benefits, and an expanded childcare credit, also helped at-risk families avoid homelessness.xiii
- Expanding the Housing Choice Voucher program: Housing Choice Vouchers are the federal government’s main tool for helping low-income households maintain their housing stability. While there is ample research evidence the program prevents homelessness, it has failed to keep up with growing needs.xiv BPC has previously recommended expanding access to housing vouchers to more extremely low-income households (those with incomes at or below 30% of area median income) to provide crucial support to the individuals and families most at risk of falling into homelessness.xv
- Eliminating barriers to building affordable housing: Cities with persistent homelessness tend to be places where it is difficult to add housing units in response to growing demand due to regulatory barriers such as restrictive zoning and land use policies. The federal government can incentivize state and local governments to reduce these barriers, as well as strengthen the Low-Income Housing Tax Credit to leverage private investment to add much-needed affordable housing supply.xvi
- Expanding permanent supportive housing: Randomized control trials of Housing First permanent supportive housing interventions, such as the Pathways to Housing program, found that participants tended to exit homelessness faster and were more likely to remain housed than those who participated in other kinds of programs.xvii While there are some limitations to modalities of permanent supportive housing, including mixed evidence on whether it improves health outcomes, the Housing First approach is demonstrably successful in getting unhoused people into long-term homes.xviii
The academic and policy literature shows that while personal circumstances and attributes can increase individuals’ risk of experiencing homelessness, they do not explain why certain places have a higher prevalence of homelessness than others. People struggle with drug dependency, poverty, and mental illness across the country, but unlike these social problems, homelessness is mostly concentrated in specific cities and regions.
Rather, current research paints a portrait of homelessness as a function of housing market dynamics. The cities facing the most significant problems in addressing homelessness in the United States, such as San Francisco, Los Angeles, and Boston, seem to share a set of structural dynamics: They have experienced significant population growth but have not been able to keep up housing construction to meet the new demand.
Helping Americans experiencing homelessness find a safe, stable place to call home has long been a bipartisan, interagency, and intergovernmental priority. Yet with limited resources and an increasingly acute shortage of safe, decent, and affordable homes around the country, homelessness is on the rise. With a greater appreciation for the relationship between housing supply constraints and homelessness trends, federal policymakers can advance a holistic national response that includes support for both researched-backed homelessness interventions and measures to address the urgent need for more affordable housing.
1 Some organizations argue that the HUD PIT count underestimates homelessness, as it doesn’t capture people who are recently homeless and staying in supportive housing. Since the count happens in January, it can also miss people using their limited resources to stay housed only during the coldest months in the year.
2 CoCs decide how to categorize their programs when reporting data to HUD for annual counts.
3 Colburn and Page Aldern’s conclusions are the result of a study of homelessness across the 35 largest metropolitan statistical areas (MSAs) in the United States.
i U.S. Department of Housing and Urban Development, “The 2022 Annual Homelessness Assessment Report to Congress,” December 2022. Available at: https://www.huduser.gov/portal/sites/default/files/pdf/2022-AHAR-Part-1.pdf.
iii R. Kuhn and D.P. Culhane, “Applying Cluster Analysis to Test a Typology of Homelessness by Pattern of Shelter Utilization: Results from the Analysis of Administrative Data,” American Journal of Community Psychology, April 1, 1998. Available at: https://doi.org/10.1023/A:1022176402357.
iv National Coalition for the Homeless, “Homelessness in America,” 2022. Available at: https://nationalhomeless.org/about-homelessness/
v U.S. Department of Housing and Urban Development, “The 2022 Annual Homelessness Assessment Report to Congress,” December 2022. Available at: https://www.huduser.gov/portal/sites/default/files/pdf/2022-AHAR-Part-1.pdf.
vi S. Fazel, V. Khosla et al., “The Prevalence of Mental Disorders among the Homeless in Western Countries: Systematic Review and Meta-Regression Analysis,” PLoS Medicine 5, no. 12. 2008. ; Shinn and Khadduri, In the Midst of Plenty: Homelessness and What to Do About It, January 2020.
vii S. Allgood and R. Warren, “The Duration of Homelessness: Evidence from a National Survey,” Journal of Housing Economics 12, No. 4: 273–90, December 2003. Available at: http://dx.doi.org/10.1016/j.jhe.2003.09.001.
viii J. Olivet, C. Wilkey et al., “Racial Inequity and Homelessness: Findings from the SPARC Study,” The American Academy of Political and Social Science, April 2, 2021. Available at: https://doi.org/10.1177/0002716221991040.
ix G. Colburn and C.P. Aldern, Homelessness is a Housing Problem, 2022; J.D. Fargo, E. Munley et al, ”Community-Level Characteristics Associated with Variations in Rates of Homelessness Among Families and Single Adults,” American Journal of Public Health, 2013. Available at: http://works.bepress.com/dennis_culhane/132/; J. Demsas, “The Obvious Answer to Homelessness,” The Atlantic, December 12, 2022. Available at: https://www.theatlantic.com/magazine/archive/2023/01/homelessness-affordable-housing-crisis-democrats-causes/672224/.
x G. Colburn and C.P. Aldern, Homelessness is a Housing Problem, 2022.
xi K.T. Rosen, D. Bank et al., “Housing Is Critical Infrastructure: Social and Economic Benefits of Building More Housing,” National Association of Realtors, June 2021. Available at: https://www.nar.realtor/advocacy/housing-is-critical-infrastructure
xii C. Glynn, T. Byrne and D. P. Culhane, “Inflection Points in Community-Level Homeless Rates,” The Annals of Applied Statistics Vol. 15 Iss. 2, 2021. Available at: http://works.bepress.com/dennis_culhane/259/
xiii Bipartisan Policy Center, “Providing Stable, Healthy, and Affordable Rental Housing Through the COVID-19 Crisis,” September 2020. Available at: https://bipartisanpolicy.org/report/stable-healthy-and-affordable-housing-covid19/. Bipartisan Policy Center, “Emergency Rental Assistance: Challenges to Efficient Distribution and Potential Solutions,” September 2021. Available at: https://bipartisanpolicy.org/blog/emergency-rental-assistance-challenges/.
xiv M. Shinn and J. Khadduri, In the Midst of Plenty: Homelessness and What to Do About It, 2020; A. Oliva, ”Why Expanding Housing Choice Vouchers Is Essential to Ending Homelessness,” Center on Budget and Policy Priorities, June 9, 2021. Available at: https://www.cbpp.org/research/housing/why-expanding-housing-choice-vouchers-is-essential-to-ending-homelessness
xv BPC, “A Bold, Bipartisan Response to the Housing Affordability Crisis,” October 2022. Available at: https://bipartisanpolicy.org/blog/bipartisan-response-housing-crisis/. BPC Housing Commission, “Housing America’s Future: New Directions for National Policy,” February 2013. Available at: https://bipartisanpolicy.org/report/housing-americas-future-new-directions-national-policy/. BPC, “Forging an Enduring Bipartisan Consensus on Affordable Rental Housing,” February 2017. Available at: https://bipartisanpolicy.org/download/?file=/wp-content/uploads/2019/03/BPC-Housing-Rental-Housing.pdf.
xvii J. Tsai, “Is the Housing First Model Effective? Different Evidence for Different Outcomes,” American Journal of Public Health, 110(9), September 2020. Available at 10.2105/AJPH.2020.305835
xviii National Academies of Sciences, Engineering, and Medicine, Permanent Supportive Housing: Evaluating the Evidence for Improving Health Outcomes Among People Experiencing Chronic Homelessness, The National Academies Press, 2018. Available at: https://doi.org/10.17226/25133
Support Research Like This
With your support, BPC can continue to fund important research like this by combining the best ideas from both parties to promote health, security, and opportunity for all Americans.Donate Now