The 2020 De-Policing: An Empirical Analysis

• Published date: 01 September 2024
• DOI: http://doi.org/10.1177/10986111231218438
• Author: Dae-Young Kim
• Date: 01 September 2024

Article

Police Quarterly

2024, Vol. 27(3) 380–402

© The Author(s) 2023

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/10986111231218438

journals.sagepub.com/home/pqx

The 2020 De-Policing: An

Empirical Analysis

Dae-Young Kim

1



Abstract

In 2020, police activities decreased substantially across large U.S. cities in the wake of

the COVID-19 pandemic and the killing of George Floyd by a police officer. Less well

understood are when and where the de-policing phenomenon took place. Using NYC

panel data from 2017 to 2022 at the census tract level, the study found significant

declines in proactive policing, immediately subsequent to the stay-at-home order and/

or death of George Floyd. However, all police activities began increasing during the

summer of 2020 and ultimately returned to the pre-intervention level and afterwards

above it in early 2022. In addition, there is evidence that both the pandemic and BLM

protests interact with neighborhood factors in affecting police activities, but not in the

same direction. The results are robust across a range of model specifications. Finally,

research and policy implications are discussed.

Keywords

COVID-19, de-policing, George Floyd, law enforcement, proactive policing

Introduction

In 2020, the COVID-19 pandemic resulted in substantial changes across criminal

justice agencies and practices (Jennings & Perez, 2020;National Commission on

COVID-19 and Criminal Justice, 2020). A range of criminal justice agencies reduced

in-person contact and scaled back their operations to avoid the risk of contagion, illness,

1

State University of New York Buffalo State University, Buffalo, NY, USA

Corresponding Author:

Dae-Young Kim, Criminal Justice Department, State University of New York Buffalo State University, 1300

Elmwood Ave, Buffalo, NY 14222, USA.

Email: kimd@buffalostate.edu

and death. For example, courts postponed or canceled trials and other court pro-

ceedings, which resulted in a backlog of court cases (Godfrey et al., 2022;Jurva, 2021).

There were substantial declines in jail and prison populations due to early releases and

reduced admissions (Carson et al., 2022;Lum et al., 2022). In addition, video con-

ferencing and other remote supervision technologies replaced in-person contact and

became an essential element of probation and community corrections to curve the

spread of the coronavirus (for Chile, Galleguillos et al., 2022; for U.S., Viglione et al.,

2020;Schwalbe & Koetzle, 2021).

Policing was not an exception to changes brought by the coronavirus. The pandemic

interfered with the daily routines of police officers, which led to substantial declines in

law enforcement activities across large cities within the U.S. (Lum et al., 2022). To

make matters more complicated, the murder of George Floyd by a police officer and

ensuing protests significantly disrupted law enforcement functions and disengaged

officers from active police work (Cassell, 2020). This phenomenon is called de-

policing. It refers to the police withdrawal from active policing as a result of public

criticism and scrutiny or other external factors. De-policing might take place at both the

individual and organizational levels, which is where and why the 2020 de-policing

phenomenon got more complicated than previous ones. That proactive policing de-

creased in 2020 is apparently clear (Cassell, 2020). However, what caused that fall

remains disputable. Despite the importance of the topic, not enough research has been

conducted on the 2020 de-policing phenomenon.

Using NYC census tract data from 2017 to 2022, the present study examines

whether the 2020 de-policing phenomenon, as measured by pedestrian stops, frisks,

searches, and arrests, was associated with the COVID-19 pandemic, the Black Lives

Matter (BLM) protests, or both. It also examines where it occurred, specifically whether

it took place across the city or only in its particular segments. The results show an

immediate decline in pedestrian stops, frisks, and searches following the stay-at-home

order (SAHO), while no significant change in arrests occurred during the same period.

In addition, all police activities significantly decreased immediately after the death of

George Floyd. Further, there is evidence that both the pandemic and BLM protests

interact with neighborhood factors in affecting police activities, but not in the same

direction. The post-SAHO decline in police stops is concentrated in census tracts with

low levels of African American populations. In contrast, during the post-BLM era,

census tracts with high levels of concentrated disadvantage and African American

populations experienced a greater decline in police stops, frisks, and searches. Finally,

the study offers implications for research and policy.

Literature Review

An Overview of De-Policing Research

The term de-policing indicates the disengagement of police officers from proactive

law enforcement under increased external criticism and scrutiny, which might be

Kim 381