The Concentrated Geography of Eviction

• Author: Matthew Desmond,Devin Q. Rutan
• Published date: 01 January 2021
• DOI: 10.1177/0002716221991458
• Date: 01 January 2021
• Subject Matter: Structural Forces

64 ANNALS, AAPSS, 693, January 2021

DOI: 10.1177/0002716221991458

The

Concentrated

Geography of

Eviction

By

DEVIN Q. RUTAN

and

MATTHEW DESMOND

991458ANN THE ANNALS OF THE AMERICAN ACADEMYTHE CONCENTRATED GEOGRAPHY OF EVICTION

research-article2021

Preventing eviction is a tractable, efficient way to

reduce homelessness. Doing so requires understanding

the precise geography of eviction. Drawing on more

than 660,000 eviction records across seventeen cities,

this study finds the geography of evictions to be durable

across time. Rather than occurring when the status quo

is disrupted, through gentrification or other modes of

neighborhood change, eviction is itself the status quo in

some pockets of American cities. The study shows that

a few buildings are responsible for an outsized share of

cities’ eviction rates. Focusing on three cities—

Cleveland, Ohio; Fayetteville, North Carolina; and

Tucson, Arizona—it finds that the one hundred most-

evicting parcels account for more than one in six evic-

tions in Cleveland and two in five evictions in

Fayetteville and Tucson. Policy-makers looking to pre-

vent homelessness can use the diagnostic tools devel-

oped in this study to precisely target high-evicting

neighborhoods and buildings.

Keywords: homelessness; eviction; durable inequality;

displacement

Because eviction is a direct cause of home-

lessness, upstream interventions that pre-

vent displacement would effectively reduce

homelessness. However, developing such inter-

ventions requires identifying the precise geog-

raphy of displacement, an area of inquiry that

researchers have so far neglected. Two conflict-

ing bodies of work suggest evictions may occur

in neighborhoods sporadically, during periods

of transformation; or chronically, as a charac-

teristic of concentrated, durable poverty.

Understanding which view is most accurate will

inform policies that address homelessness.

Devin Q. Rutan is a graduate student in sociology and

social policy at Princeton University. His interests

include housing, material hardship, and the elderly.

Matthew Desmond is the Maurice P. During Professor

of Sociology at Princeton University and the principal

investigator of The Eviction Lab.

Correspondence: drutan@princeton.edu

THE CONCENTRATED GEOGRAPHY OF EVICTION 65

Drawing on more than 660,000 residential evictions in seventeen midsized

American cities over the course of a decade, we show that the distribution of

evictions is durable: a neighborhood’s contribution to the citywide eviction rate is

highly correlated with its contribution a decade earlier. We extend this analysis to

evaluate the distribution of evictions at the level of the building, focusing on

three cities. We observe that evictions are heavily concentrated among a small

subset of properties within each city. The one hundred most-evicting parcels

account for more than one in six evictions in Cleveland, Ohio; and more than two

in five evictions in Fayetteville, North Carolina, and Tucson, Arizona.

These findings have substantial implications for the design and implementa-

tion of policies and services meant to prevent families from becoming homeless.

By showing that every year a handful of the same buildings produce outsized

proportions of evictions in a city, this study provides a diagnostic tool that local

officials can use to efficiently prevent displacement before it happens.

Interventions that target property owners and tenants of the highest-evicting

buildings could be more cost-efficient and effective than general, nontargeted

interventions that neglect to focus on the communities and addresses where dis-

placement is most acute.

Eviction and Homelessness

Recent government estimates suggest that roughly 568,000 people were homeless

on a single night in 2019 (U.S. HUD 2019). This estimate is not a sufficient rep-

resentation of the overall prevalence of homelessness because substantial majori-

ties of the homeless population experience only brief periods of transitional

homelessness (Kuhn and Culhane 1998; Phelan and Link 1999). Lifetime esti-

mates suggest that between 5 and 6 percent of Americans have been homeless at

some point in their lives (Shelton etal. 2009). Black and Latinx Americans, as well

as those with limited incomes, experience homelessness much more frequently

than their peers (Fusaro, Levy, and Shaefer 2018; see Olivet etal., this volume).

Housing markets structure the prevalence and risk of homelessness. Multiple

studies have demonstrated that homelessness is more common when affordable

housing is scarce (Lee, Tyler, and Wright 2010). For instance, one study found

that urban families with infants who experience a severe illness were far more

likely to experience homelessness if they lived in cities with higher housing costs

NOTE: The JPB, Gates, and Ford Foundations, as well as the Chan Zuckerberg Initiative,

funded this research. The Eunice Kennedy Shriver National Institute of Child Health &

Human Development of the National Institutes of Health under Award Number P2CHD047879

supported the research reported in this publication. The content is solely the responsibility of

the authors and does not necessarily represent the official views of the National Institutes of

Health. We thank the editors and reviewers for their thoughtful comments. We are also grate-

ful to participants at the author’s conference and members of the Eviction Lab who offered

insightful suggestions that helped to refine this study. We thank Urban Footprint; Cumberland

County, NC; Cuyahoga County, OH; and Pima County, AZ, for providing data that made this

study possible.