Strengthening the Rule of Law Through Community Policing: Evidence From Liberia

• Published date: 01 March 2025
• DOI: http://doi.org/10.1177/00104140241252090
• Author: Benjamin S. Morse
• Date: 01 March 2025

Article

Comparative Political Studies

2025, Vol. 58(3) 608–642

© The Author(s) 2024

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/00104140241252090

journals.sagepub.com/home/cps

Strengthening the Rule of

Law Through

Community Policing:

Evidence From Liberia

Benjamin S. Morse

1



Abstract

How to improve security and strengthen the rule of law in fragile states?

Community policing programs have long been at the forefront of policy-

makers’efforts to address this challenge. These programs tend to be more

expansive than those found in developed countries, focusing not only on

building trust through meetings and foot patrols, but also on eliciting ‘co-

production’from communities to supplement scarce police capacity and

provide alternatives to vigilantism. I partnered with the Liberian National

Police (LNP) to experimentally evaluate the effectiveness of this approach in

Monrovia, Liberia, one of sub-Saharan Africa’s most crime-ridden cities.

Drawing on a large-scale resident survey and administrative crime data, I

f‌ind that the program improved relations between police and citizens,

strengthened social norms against vigilantism, and mobilized communities to

participate in the police’s“Watch Forum”initiative by forming and sustaining

local security groups designed to facilitate cooperation with police. These

changes were accompanied by a roughly 40% reduction in the incidence of

mob violence. Despite these improvements, the program did not reduce the

overall incident of crime, improve perceptions of security, or increase crime

reporting.

1

Social Impact Inc, Arlington, VA, USA

Corresponding Author:

Benjamin S. Morse, Technical Director, Social Impact Inc, 4201 Wilson Blvd, Arlington, VA

222013382, USA.

Email: bensmorse@alum.mit.edu

Data Availability Statement included at the end of the article.

Keywords

African politics, conf‌lict processes, experimental research, quantitative

methods, state building

Introduction

Insecurity due to crime and violence is pervasive in many of the world’s

weakest states (Mack et al., 2013). To address this problem, donors and aid

organizations invest millions of dollars annually in assistance programs

designed to strengthen the capacity of state security forces and legitimate their

authority in the eyes of citizens. This two-pronged approach has long

dominated the practice of security sector reform, and is motivated by the idea

that institutional capacity and citizen cooperation are “close complements”in

the production of security and therefore must be built in tandem if countries

are to escape fragility and violence (Baranyi et al., 2011).

Yet in recent years a growing number of scholars have begun to question

whether this narrow, state-centric approach to security sector reform is op-

timal, and in particular, whether it adequately addresses short term security

needs (Baker, 2009;Chirayath et al., 2005;Wisler & Onwudiwe, 2008). At the

core of these concerns is a fundamental problem of sequencing: building

strong, effective security forces takes decades; in the interim, encouraging

citizens to rely on these institutions even as they remain fundamentally

untrustworthy and unreliable is unlikely to lead to meaningful improvements

in security, and could potentially backf‌ire if engaging with corrupt and in-

competent security forces serves to harden citizens’negative perceptions.

Traditional, state-centric models of security reform also struggle to provide

a viable alternative to extrajudicial practices such as mob violence, lynching,

and vigilantism. Instead, they address these problems with outreach and

awareness programs designed to foster norms that support the rule of law and

reject extrajudicial practices (Carothers, 2006). Yet because these interven-

tions do little to address the def‌iciencies of state security forces and little to

reduce citizens’security concerns, reliance on extrajudicial mechanisms tends

to persist.

If building effective, legitimate security institutions takes decades, how can

policymakers improve security and reduce reliance on extrajudicial practices

in the short term? In this paper, I partner with the Liberian National Police

(LNP) to test an approach rooted in “multidimensional”models of security

reform, which call for policymakers to recognize that multiple non-state actors

often play a role in security provision in countries with severe and deeply-

rooted capacity def‌icits (Baker & Scheye, 2007;OECD, 2008). In these

settings, “f‌irst-best”solutions founded on the Weberian ideal of the state as the

sole legitimate provider of security are not always possible; instead, the best

Morse 609

way to “respond to the short-term needs of enhanced security and justice

service delivery, while also building the medium-term needs of state capacity”

may be for security forces to develop “productive partnerships”with local

communities and non-state security actors (OECD, 2008, p. 7).

The challenge, of course, is to do so in a way “that builds on local networks

and institutions without encouraging vigilantism”(Rose-Ackerman, 2004,

p. 187, emphasis mine). Liberia’s model of community policing responds to

this challenge by expanding the scope of traditional, ‘Weberian-style’

community policing programs, to focus not only on encouraging reliance

on the police, but also on providing communities with a lawful, human rights-

respecting avenue through which they may ‘coproduce’certain security

functions in coordination with the police.

Coproductive models of community policing are increasingly common in

fragile states, and often receive considerable support from aid agencies and

international donors (Dinnen & Peake, 2013;Kagoro, 2019;Wisler &

Onwudiwe, 2008). But there is little rigorous experimental evidence to in-

form whether these programs are effective. To address this gap, I partnered

with the Liberian National Police to experimentally evaluate the effectiveness

of their community policing program in Monrovia, one of sub-Saharan Af-

rica’s most crime-ridden cities and a place where authorities have long

struggled to combat vigilantism and mob violence (Downie, 2013;Zanker,

2017). Across nine of Monrovia’s ten police zones, police commanders

identif‌ied a sample of 93 communities, 45 of which were randomly selected to

received the program for a period of 10 months, from February to

November 2018.

Using survey data, off‌icial crime reports, and qualitative f‌ield reports from

research assistants assigned to shadow the police for the duration of the study,

I provide four sets of results. First, I f‌ind that the program improved relations

between police and citizens, making residents of treatment communities more

likely to view the police was capable and well-intentioned, and more likely to

personally know individual off‌icers. Second, and more signif‌icantly, the

program mobilized communities to more actively organize, support, and

sustain local security groups. While such groups were common in control

communities as well, groups in treatment communities were more likely to be

registered with the police’s Watch Forum initiative and were sustained by

greater contributions of time and effort from residents.

A central concern for policymakers is whether coproduction models of

community policing can re-orient communities’efforts to self-provide se-

curity towards more rights-respecting practices. My third set of f‌indings

strongly suggests they can. In addition to being more trusting of police,

residents of treatment communities were more knowledgeable about the rules

and guidelines governing local security groups (especially with regards to

suspects’rights) and less willing to condone mob violence. Finally, and

610 Comparative Political Studies 58(3)