Homelessness Is a Housing Problem: How Structural Factors Explain U.S. Patterns

Homelessness Is a

Housing Problem

The publisher and the University of California Press Foundation gratefully acknowledge the generous support of the Anne G. Lipow Endowment Fund in Social Justice and Human Rights.

Homelessness Is a Housing Problem

How Structural Factors Explain U.S. Patterns

Gregg Colburn and Clayton Page Aldern

UNIVERSITY OF CALIFORNIA PRESS

University of California Press

Oakland, California

© 2022 by Gregg Colburn and Clayton Aldern

Library of Congress Cataloging-in-Publication Data

Names: Colburn, Gregg, 1972- author. | Aldern, Clayton Page, 1990– author.

Title: Homelessness is a housing problem : how structural factors explain U.S. patterns / Gregg Colburn and Clayton Page Aldern.

Description: Oakland, California : University of California Press, [2022] | Includes bibliographical references and index.

Identifiers: LCCN 2021037026 (print) | LCCN 2021037027 (ebook) | ISBN 9780520383760 (cloth) | ISBN 9780520383784 (paperback) | ISBN 9780520383791 (epub)

Subjects: LCSH: Homelessness—United States. | Metropolitan areas—Housing—Social aspects—United States. | Homelessness—Government policy—United States. | Homeless persons—Substance use—United States.

Classification: LCC HV4505 .C656 2022 (print) | LCC HV4505 (ebook) | DDC 362.5/920973—dc23

LC record available at https://lccn.loc.gov/2021037026

LC ebook record available at https://lccn.loc.gov/2021037027

Manufactured in the United States of America

31   30   29   28   27   26   25   24   23   22

10   9   8   7   6   5   4   3   2   1

To anyone who lives without stable housing—and to the policy makers, organizers, advocates, researchers, practitioners, and those with lived experience working to end homelessness

Contents

List of Figures and Tables

Acknowledgments

PART I. CRISIS

1. Baseline

2. Evidence

PART II. CAUSES

3. Individual

4. Landscape

5. Market

PART III. CONCLUSION

6. Typology

7. Response

Notes

Bibliography

Index

Figures and Tables

FIGURES

1. Public interest for five search terms

2. Ohio Continuum of Care map

3. Per capita rates of homelessness in select U.S. regions, 2007

4. Per capita rates of homelessness in select U.S. regions, 2019

5. Rank of indexed rates of homelessness, 2010 v. 2019

6. Median and variance of per capita homelessness, 2007–2019

7. Percent with income below poverty level versus PIT count (per capita)

8. Unemployment rate versus PIT count (per capita)

9. Rate of serious mental illness versus PIT count (per capita)

10. Rate of illicit drug use versus PIT count (per capita)

11. Rate of substance use disorder versus PIT count (per capita)

12. Percent Black/African Amrican versus PIT count (per capita)

13. January average temperature versus PIT count (per capita)

14. January average temperature versus indexed homelessness

15. January average temperature versus unsheltered count (per capita)

16. Benefit/rent ratio versus family PIT count (per capita)

17. Low-income migration rate versus PIT count (per capita)

18. Median gross rent as a percentage of household income versus PIT count (per capita)

19. First-quartile housing cost burden versus PIT count (per capita)

20. Median contract rent versus PIT count (per capita)

21. First-quartile rent versus PIT count (per capita)

22. Median rent versus rental vacancy rate

23. Rental vacancy rate versus PIT count (per capita)

24. First-quartile vacancy rate versus PIT count (per capita)

25. Change in population versus PIT count (per capita)

26. Population growth versus housing supply elasticity

TABLES

1. Summary of Vacant Units: San Francisco, 2017

2. A Typology of Cities

Acknowledgments

The motivation for this book stemmed from Gregg’s observation that the roots of the homelessness crisis in many cities in the United States were being misdiagnosed, often to frustrating and harmful ends. A public focus on mental health and drugs—certainly, important risk factors for homelessness—dominates narratives in our home in the Puget Sound region, thereby narrowing the policy conversation. In discussions with stakeholders throughout the region, it became apparent that an intense focus on behavioral health might be masking a more important root cause of this crisis: housing market conditions.

Gregg shared the early vision for this book in 2019 with colleagues throughout the Seattle region. One such conversation proved to be particularly valuable and changed the course of this book project. Rogers Weed, then board chair of Building Changes (a nonprofit focused on homelessness), introduced the two authors of this book. After a number of productive and engaging conversations, Clayton and Gregg decided to collaborate on this project. The two of us collaborated on the analysis, Gregg wrote the majority of the manuscript, and Clay applied his expertise in data visualization to the images found in this book. In a sure sign of a productive partnership, this final product is far better than any that would have resulted from a solitary effort.

Gregg is grateful to the many people who provided insight, wisdom, and support throughout this process. At the University of Washington, a wonderful community of scholars, including Arthur Acolin, Scott Allard, Kyle Crowder, Rachel Fyall, Rebecca Walter, and Thaïsa Way have supported my professional development and have provided valuable feedback at various stages of the writing process. In addition, scholars from around the country including Thomas Byrne, Brian McCabe, and Beth Shinn also provided feedback and guidance. Numerous members of the Seattle community served as sounding boards throughout the writing process; a special thank-you to Kollin Min who was a key supporter of this project. Finally, Gregg is also thankful for his wife, Jen, and children Grace and Grant for their patience and support. Much of this book was written as we quarantined together during the COVID-19 pandemic. I love you all and am grateful for the joy and laughter you bring to my life.

Clayton extends his deep gratitude to the advocates and researchers with and without the lived experience of homelessness who have worked to help him and others understand the structural roots of the crisis. To Jeff Rodgers, Tess Colby, Aras Jizan, Marc Dones, Anne Marie Edmunds, Valeri Knight, Sarah Appling, Claire Aylward Guilmette, Deborah L’Amoureux, Gerrit Nyland, Caroline Belleci, Geoff Campion, Vishesh Jain, Aman Sanghera, Annie Pennucci, Matt Lemon, Pear Moraras, Abby Schachter, Stephanie Roe, Jesse Jorstad, and Stephanie Patterson—thank you for your insight into (and leadership in) this space, and thanks for all you continue to teach us. Thanks to Jason Schumacher and Neal Myrick at the Tableau Foundation for your deep investments in data visualization and data literacy in homelessness policy and housing stability—and for letting Clay use your software for exploratory analysis and figure drafting. To Sara Curran, Tim Thomas, and Thaïsa Way at the University of Washington for the critical support via the Center for Studies in Demography and Ecology: Thank you. Gratitude to Whitney Henry-Lester for the late-night chats to these ends and to Lowell Wyse for the late-night chats to other ends. A near-impossible degree of gratitude to Anneka Olson, whose patience knows no bounds: Thank you for your keen editing brain, critical perspective, and warm presence. Henry the Cat and Maple the Cat also offered pivotal moral support.

The final chapter of this book incorporates the ideas of researchers, practitioners, and policy makers from around the country. We are grateful to those who shared their research and opinions about the strategies needed to end homelessness: Whitney Airgood-Obrycki, Thomas Byrne, Tess Colby, Dennis Culhane, Mary Cunningham, Conor Dougherty, Mark Ellerbrook, Katie Hong, Aras Jizan, Jill Khadduri, Margot Kushel, Kollin Min, Stephen Norman, Beth Shinn, and Dilip Wagle.

Finally, we thank Naomi Schneider, Summer Farah, and the rest of the University of California Press team for supporting this book project.

PART I

Crisis

CHAPTER ONE

Baseline

Homelessness occupies a prominent place in American political life. Although less than one-fifth of 1 percent of the U.S. population experiences homelessness on a given night in the country, the issue receives considerable attention from policy makers and the general public. This spotlight is striking given the scale of the homelessness crisis when compared to other prominent social problems. That fifth-of-a-percent figure translates to about five hundred sixty-eight thousand people. To be sure, this number should feel large and unacceptable. But on an absolute basis, for example, homelessness pales in comparison to the nation’s poverty crisis: Over thirty-four million Americans were living below the federal poverty line in 2019. Meanwhile, abundant evidence highlights the political preoccupation with homelessness. In 2020, a poll in Washington State revealed that voters ranked homelessness as the top priority for the state legislature—far above other common public concerns like transportation, the economy, the environment, and health care.¹ We observe a similar focus at the national level. As depicted in Figure 1, from January 2015 to January 2020, more people in the United States searched for the term homeless via Google than for inequality, racism, poverty, and climate change.²

Figure 1.  Public interest over time for five search terms. Data source: Google Trends

How might we explain this seemingly disproportionate interest in the issue of homelessness? Two potential explanations come immediately to mind. First, maybe this interest isn’t as disproportionate as it might initially appear. That is, maybe the numbers are wrong. Among astute observers, it is well understood that official point-in-time census estimates of homelessness underestimate the true size of the population experiencing homelessness on any given night.³ For example, the federal definition excludes many precariously housed individuals and families who might be living with a friend or temporarily living in a motel room. The more expansive definition of homelessness used by the U.S. Department of Education suggests a population of 1.35 million homeless students without counting their parents.⁴ Furthermore, across greater spans of time—say, lifetimes—roughly 5 percent of the population experiences homelessness at least once.⁵ In light of these figures, it is more accurate to consider homelessness as a problem that affects millions, rather than hundreds of thousands. But even the larger figure highlights the fact that only a small fraction of people living in poverty actually lose their housing.

More fundamentally, though, a second explanation for the intense interest in the topic may stem from the simple incongruity of a half million people living in shelters and on the street in the wealthiest country in the world. Reactions to this apparent paradox are diverse. For some, homelessness is a moral and political outrage indicting the capitalist system on which U.S. society is based; for others, homelessness is a scourge ruining the nation’s largest and most dynamic cities. Other observers reside somewhere in the middle of this spectrum. What is uncontroversial is that homelessness elicits strong and emotional responses from all corners of society. From the perspective of the public, the intense focus on homelessness requires—and demands—an explanation. There is a strong desire to understand the causes of homelessness and where to assign blame. This book is in part concerned with the question of blame.

In January 2020, just weeks before the outbreak of the coronavirus pandemic in the United States, a long-simmering debate about the origins of and responsibility for the homelessness crisis erupted in public. Members of the federal government, including President Trump, argued vocally that the high rates of homelessness in many U.S. cities were a function of the local failings of Democratic leadership and policies. Referencing Democratic House Speaker Nancy Pelosi, the president said, She ought to go home and take care of her District, where the homeless is all over the place, and the tents and the filth and the garbage is eroding right into the Pacific Ocean and into their beaches.⁶ In response to this finger-pointing, state and local policy makers—most notably California’s governor, Gavin Newsom, a Democrat—argued that a lack of federal assistance had starved local communities of sorely needed resources, and housing instability and homelessness had flourished in turn.

Certainly, some of this political jostling is a product of the polarized nature of U.S. politics in the 2020s. From voting rights to climate change—issues that would appear at face value to be resoundingly nonpartisan but which often provoke party-line votes—policy responses to (and public perception of) the issues of our time are characterized by tribalism. Tailored media narratives and the so-called filter bubbles of social media add fuel to the flame of confirmation bias. It’s harder than it should be to find fact-checked information, and it’s even harder to internalize narratives that run counter to our beliefs. In this respect, homelessness is no different: It tends to provoke hyper-partisan diagnoses and prescriptions. And as with most cases of hyper-partisanship, neither argument above—Trump’s nor Newsom’s—sufficiently explains the state of homelessness in the country. If inadequate federal support alone accounts for the crisis, why does the rate of homelessness vary so substantially across cities? Presumably, all cities would be equally starved of resources if federal retrenchment were the cause. Yet while some cities have seen rates of homelessness rise over the last ten years, many others have seen rates fall. And if Democratic mayors and governors are the problem, how can we account for the many cities and states with both Democratic leadership and policies and relatively low rates of homelessness? Unsurprisingly, the polarized plotlines are too simple, but they draw attention to essential questions about the nature and causes of the homelessness crisis.

As Ezra Klein writes in his recent book Why We’re Polarized, one of the other phenomena driving polarization in the country is a grafting of our political identities onto national (as opposed to local) politics.⁷ National politics, by definition, require a flattening of local variation—and in our de facto two-party system, with this flattening often comes a false dichotomization of many complex issues. This complicates the effort to respond to local issues that vary by geography—homelessness among them. In the United States, one of the most pressing and vexing questions about homelessness concerns the substantial variation in per capita rates of homelessness in cities across the country. Seattle and San Francisco, for example, have roughly four to five times the per capita homeless population of Chicago.⁸ The stark differences between seemingly vibrant and healthy cities invite us (and many others) to ask: Why is homelessness so bad in cities like Seattle and San Francisco? Is this a failure of individuals, politicians, markets, or other structural forces? An understanding of variation might help us unlock the drivers of this crisis.

Many of us have, for good reason, struggled to identify a credible explanation for this variation. Accounts of and references to homelessness on television, online, in newspapers, and in scholarly sources offer a long list of potential causes of the issue; among them addiction, mental illness, poverty, domestic violence, eviction, high housing costs, racial discrimination, unemployment, and many others. Reports based on interviews with people experiencing homelessness highlight a wide range of potential causes, as well. A recent report from Seattle/King County for example, noted the following self-reported causes of homelessness among respondents to the annual point-in-time homelessness census: job loss (24 percent of respondents), alcohol or drug use (16 percent), eviction (15 percent), divorce or separation (9 percent), rent increase (8 percent), argument with family or friend (7 percent), incarceration (6 percent), and family/domestic violence (6 percent).⁹ Confronted with the question of why some cities have far greater per capita rates of homelessness than others, a reasonable, logical reaction might be to assume that higher levels of homelessness stem from higher incidences of these self-reported causal factors in these cities. In this book, we examine this logic.

While perusing any list of potential causes of homelessness, one can generally break the ostensible explanations down into two overarching categories. Some causes are individual in nature, and some are structural. The bifurcation is consistent with decades of research on poverty and homelessness. On one side of the debate are those who argue that poverty and homelessness are the result of individual factors, that vulnerabilities related to housing instability are fueled by illness, mental condition, laziness, or poor decision-making, including—for these observers—excessive drug and alcohol use. And in the central downtowns of cities like Los Angeles, San Francisco, or Seattle, thousands of unsheltered people experiencing homelessness may indeed be suffering from a substance use disorder, mentally ill, and/or unemployed. Following this logic, it is the disproportionate presence of people with these vulnerabilities in certain cities that explains the substantial variation in per capita homelessness rates around the country. Whether born in or attracted to these cities, people comprise the homelessness crisis, and so homelessness is an individual problem. (It is not uncommon for some to argue that homelessness is exclusively an individual choice.) On the other side of the debate are those who argue that larger, structural forces, such as market conditions, housing costs, racism, discrimination, and inequality, causally explain the prevalence of homelessness. Under the structural explanation, homelessness is a consequence of broader and deeper societal factors driving people at the margins of society out of their housing.

Perhaps there is a middle road. The individual explanation is alluring—it’s individual people who lose their housing, after all. Surely there must be systematic factors at play, though; otherwise, how could we possibly account for the dramatically different rates of homelessness around the country? Even if you were entirely convinced of the individual explanation, you would have to acknowledge that some kind of systemic variation—some combination of environmental, political, economic, and demographic trends—characterizes different places. In 2019, less than 1 in 1,000 residents were unhoused in Alabama and Mississippi, while California and Oregon had over five times that rate. Why? Existing research provides a helpful roadmap to navigate the seemingly complex and, at times, contradictory evidence about the causal drivers of homelessness. Homelessness researcher Brendan O’Flaherty, for example, suggests that to generate causal explanations of homelessness, one must consider the interaction between individual characteristics and the context in which that person resides. Either explanation alone is insufficient to explain or predict individual homelessness. By extension, he argues that people who lose their housing are effectively the wrong people in the wrong place.¹⁰ This frame helps to provide a vantage point from which to consider the central question of this book: What explains the substantial regional variation in per capita homelessness rates in the United States?

To cut to the chase, the answer is on the cover of this book: Homelessness Is a Housing Problem. Regional variation in rates of homelessness can be explained by the costs and availability of housing. Housing market conditions explain why Seattle has four times the per capita homelessness of Cincinnati. Housing market conditions explain why high-poverty cities like Detroit and Cleveland have low rates of homelessness. Housing market conditions also explain why some growing cities, like Charlotte, North Carolina, are not characterized by the levels of homelessness that coastal boomtowns like Boston, Seattle, Portland, and San Francisco are. Variation in rates of homelessness is not driven by more of those people residing in one city than another. People with a variety of health and economic vulnerabilities live in every city and county in our sample; the difference is the local context in which they live. High rental costs and low vacancy rates create a challenging market for many residents in a city, and those challenges are compounded for people with low incomes and/or physical or mental health concerns.

•     •     •

According to estimates from the U.S. Department of Housing and Urban Development (HUD), at least 567,715 people experienced homelessness on a single night in 2019.¹¹ But this aggregate figure masks significant geographic variation in the distribution of per capita homelessness across the country. The metropolitan areas of New York, Los Angeles, Washington, D.C., San Francisco, Seattle, and Boston alone account for over 29 percent of the homeless population in the country, despite being home to only about 7 percent of the general population. Regardless of one’s view of the problem—and the political lens through which one considers homelessness—it is reasonable to wonder what it is about these cities that produces (or, according to some, attracts) such large and disproportionate populations of people experiencing homelessness. To explore this phenomenon, we shift the unit of analysis away from the individual and turn our attention to the metropolitan area. From this perspective, we are not interested in predicting whether a given person will experience homelessness or why someone lost their housing in the past; we are interested in understanding why, for example, the crisis is so much more extreme in Boston than in Cleveland. This analytic pivot does not preclude individual explanations for homelessness; instead, it clarifies the object in which we are interested: the city-to-city variation itself.

Understanding this variance is critical to formulating an appropriate policy response. In cities with substantial unhoused populations, it is common for rival political factions to blame one another for the crisis—a microcosm of the Trump–Newsom sparring cited above—and for the issue to devolve into a political hot potato. Often caught in the middle of this dispute are municipal leaders who are tasked with solving the problem (with resources that many consider to be inadequate). Societal cleavages emerge in which compassionate responses to homelessness—those that stress social service provision and respect for the dignity and rights of people experiencing homelessness—are criticized by community members who advocate a tougher response to the crisis. Proponents of the latter approach argue that overly permissive local policies have incubated an underlying problem, all while individual desperation facilitates property crime,