Profiling, Screening, and Criminal Recruitment

• Published date: 01 December 2015
• Date: 01 December 2015
• DOI: http://doi.org/10.1111/jpet.12115
• Author: CHRISTOPHER COTTON,CHENG LI

PROFILING,SCREENING,AND CRIMINAL RECRUITMENT

CHRISTOPHER COTTON

Queen’s University

CHENG LI

Mississippi State University

Abstract

We model major criminal activity as a game in which a law

enforcement officer chooses the rate at which to screen dif-

ferent population groups, and a criminal organization (e.g.,

drug cartel, terrorist cell) chooses the observable character-

istics of its recruits. Our model best describes smuggling or

terrorism activities at borders, airports, and other security

checkpoints. The most effective law enforcement policy im-

poses only moderate restrictions on the officer’s ability to

profile. In contrast to models of decentralized crime, requir-

ing equal treatment never improves the effectiveness of law

enforcement.

1. Introduction

“In making routine or spontaneous law enforcement decisions,

such as ordinary traffic stops, federal law enforcement officers may

not use race or ethnicity to any degree, except that officers may rely

on race and ethnicity if a specific suspect description exists.”

Christopher Cotton, Department of Economics, Queen’s University, 94 University Av-

enue, Kingston, Ontario K7L 3N6, Canada (cotton@econ.queensu.ca). Cheng Li, Depart-

ment of Finance and Economics, Mississippi State University, Mississippi State, MS 39762,

(cheng.li@msstate.edu).

This paper has benefited greatly from comments by two excellent referees, Chris Parme-

ter, Mehdi Shadmehr, Dan Mantero, and Dave Kelly, as well as presentations at the

University of Miami, the 23rd International Conference on Game Theory at Stony Brook,

the 50th Anniversary Conference of the Public Choice Society, and the 1st International

Workshop on Economic Analysis of Institutions at Xiamen University.

Cotton is grateful for financial support provided through his position as the Jarislowsky-

Deutsch Chair in Economic and Financial Policy at Queen’s University.

Received October 15, 2013; Accepted October 17, 2013.

C2014 Wiley Periodicals, Inc.

Journal of Public Economic Theory, 17 (6), 2015, pp. 964–985.

964

Criminal Recruitment 965

“Given the incalculably high stakes involved in such investiga-

tions, federal law enforcement officers who are protecting national

security or preventing catastrophic events (as well as airport security

screeners) may consider race, ethnicity, alienage, and other relevant

factors.”

—U.S. Department of Justice fact sheet on racial profiling,

June 17, 2003

Profiling in law enforcement activities refers to the use of an individ-

ual’s race, ethnicity, or other observable characteristics by officers when

determining whether to stop, search, screen, or otherwise engage in law

enforcement. In 2003, the U.S. Department of Justice (DOJ) issued a policy

banning profiling during most routine federal law enforcement activities,

arguing that the ban improves the fairness and effectiveness of law enforce-

ment. At the same time, the DOJ made explicit exceptions for officers at

borders and airports, or otherwise involved with national security, arguing

that such exceptions are necessary to protect national security and prevent

catastrophic events. The national security exception has met with opposition

from a number of civil rights organizations, legislators, and other advocates

who argue that the ban on profiling should extend to all areas of law en-

forcement.1

The idea that banning profiling may improve the effectiveness of law

enforcement in certain settings is consistent with the academic literature.

Persico (2002) uses a model of vehicle search during traffic stops to consider

an agency problem that arises between law enforcement officers concerned

with catching existing criminal activity, and a society that is additionally con-

cerned with deterring crime. He shows how a ban on profiling may align

the search behavior of law enforcement with the deterrence objectives of

society, and may better minimize crime. Persico’s (2002) model involves de-

centralized crime, with many individuals independently choosing whether to

engage in criminal activity. It is an appropriate model for considering many

types of crime, but may be less appropriate for considering settings in which

criminal activity is centrally planned by criminal organizations.

At borders, airports, and other security checkpoints, law enforcement

is primarily concerned with preventing major crime such as smuggling and

terrorism. Such crime tends to be centrally planned by strategic criminal or-

ganizations (e.g., terrorist cells, drug cartels) which recruit agents to carry

contraband (e.g., drugs, weapons) through the security checkpoint.2By

1According to the American Civil Liberties Union (ACLU), “Law enforcement based on

general characteristics such as race, religion, and national origin, rather than on the ob-

servation of an individual’s behavior, is an inefficient and ineffective strategy for ensuring

public safety”(ACLU 2004, p. 2), and it continues to call for the DOJ to “close the loophole

for national security and border integrity investigation.”

2Consider the case of the September 11, 2001 terrorist attacks in the United States. These

attacks were planned and funded by al-Qaeda leadership including Osama bin Laden and