Should the police give priority to violence within criminal organizations? A personnel economics perspective

• Published date: 01 December 2023
• Author: C. Bravard,J. Durieu,J. Kamphorst,S. Roché,S. Sémirat
• Date: 01 December 2023
• DOI: http://doi.org/10.1111/jpet.12666

Received: 16 February 2023

|

Accepted: 31 August 2023

DOI: 10.1111/jpet.12666

ORIGINAL ARTICLE

Should the police give priority to violence

within criminal organizations? A personnel

economics perspective

C. Bravard

1

|J. Durieu

1

|J. Kamphorst

2

|S. Roché

1

|

S. Sémirat

1

1

CNRS, UMR CREG‐GAEL‐PACTE,

Sciences‐Po Grenoble, Université

Grenoble‐Alpes, Grenoble, France

2

Netherlands Authority for Consumers

and Markets (ACM), Erasmus School of

Economics, and Tinbergen Institute,

Amsterdam, The Netherlands

Correspondence

C. Bravard, CNRS, UMR GAEL,

Université Grenoble‐Alpes, Grenoble,

France.

Email: christophe.bravard@gmail.com

and christophe.bravard@univ-grenoble-

alpes.fr

Abstract

Even among themselves, criminals are not seen as

trustworthy. Consequently, a criminal organization

needs to incentivize its members, either by threats of

violence or by rewarding good behavior. The cost of

using violence depends on the resources police

allocate to investigating intraorganizational vio-

lence. This means that the police may affect the

choice of an incentive scheme by the criminal

organization. The design of the optimal strategy for

crime control has to take this into account. We

develop a model of an infinitely repeated criminal

labor market where (i) a criminal organization hires

and incentivizes members, and (ii) peripheral crime

(crime outside the criminal organization) is a

stepping stone to a career in organized crime. We

establish that there are two possible optimal strate-

gies for the police. (i) There are situations in which

the optimal strategy for the police is to use all of

their resources to decrease the efficiency of crim-

inals. (ii) In other situations, the optimal strategy for

the police is to spend the minimum amount of

resources to ensure that the criminal organization

J Public Econ Theory. 2023;25:1361–1393. wileyonlinelibrary.com/journal/jpet

|

1361

This is an open access article under the terms of the Creative Commons Attribution‐NonCommercial License, which permits use,

distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes.

© 2023 The Authors. Journal of Public Economic Theory published by Wiley Periodicals LLC.

cannot punish disloyal criminals, and spend the rest

of their resources to decrease the efficiency of

criminals.

KEYWORDS

incentives mechanisms, organized crime, police policy

They always fudge. They figure they're out doing the job, who wants to give up half

of what they get to somebody that's not even there? …Now the thing is, it's a

dangerous game, because if you get caught, you're liable to get whacked—

killed. Joseph D. Pistone (better known as Donnie Brasco) on criminals of the

American Mafia (Pistone & Woodley, 1987)

1|INTRODUCTION

Many countries are struggling with increases in organized crime.

1

For many criminal

organizations, the threat and use of violence are important tools, both externally and internally.

This raises the question of whether public authorities should care about intraorganizational

violence. What is the problem if one criminal kills another? In other words, should the police

focus on investigating murders committed within criminal organizations?

In this paper, we use standard insights from personnel economics to show that focusing on

cases of intraorganizational violence can reduce organized crime in equilibrium. However, this

is not without its costs. Surprisingly, suppressing organized crime in this way may attract more

people to crime.

Criminal organizations face the problem of maintaining internal discipline. Even

among themselves, criminals have a reputation for not being trustworthy (see, e.g.,

Campana, 2016; Campana & Varèse, 2013; von Lampe, 2016, p. 119; as well as the quote

from Pistone at the beginning). Because criminal organizations have good reasons not to

use receipts, there are ample opportunities to steal from the organization. Instead, many

criminal organizations rely on fear to discipline their members. Threats of physical

punishment, death, or harming a member's loved ones are not uncommon.

2

However,

maintaining discipline through fear requires that the threat of punishment be credible. In

this paper, we explore the advantages and disadvantages for the police of limiting the

1

A 2011 report by the US National Security Council states: “Transnational organized crime (TOC) poses a significant

and growing threat to national and international security”(https://obamawhitehouse.archives.gov/administration/eop/

nsc/transnational-crime/threat)

2

For instance, Hopkins et al. (2013) provide evidence that homicide and violence are used in retaliation/punishment for

theft from the group. According to Campana (2016) and Campana and Varèse (2013) there is typically little trust among

criminals which is not backed up by some type of threat. Finally, the UNODC (2014, pp. 35–42) reports the use of

violence for internal discipline for some criminal organizations.

1362

|

BRAVARD ET AL.

criminal organization's options for maintaining internal discipline by increasing the costs

of intragang punishments. In particular, we consider an infinitely repeated labor market

where a crime boss hires and incentivizes members for her organization (organized

criminals). Nonhired individuals can choose an occupation as an upstanding citizen or as a

criminal outside the organization (peripheral criminal). Then organized criminals choose

whether to be disloyal, that is, steal from the boss. If an organized criminal is disloyal, the

boss faces the choice of whether to punish him or not. If she does, she faces a cost. Note

that the boss may not incur this cost directly. However, the boss does need to com-

pensate the members who do incur that cost. Consequently, modeling this as a cost directly

to the boss simplifies the analysis without affecting the results. This cost increases with the

resources police allocate to detecting these punishments. If instead the boss does not

punish, she loses credibility and is replaced by a new boss. Therefore, the boss needs to

induce loyalty. She can do so by punishing disloyalty, by rewarding loyalty,

3

or both.

Our model has six main assumptions. First, there is just one criminal organization. This

allows us to focus on the incentives for organized criminals, rather than on the competition

between criminal organizations. Second, the organization consists of a hierarchy of a boss and

organized criminals, with the boss as the residual claimant.

4

Third, people die. This may be due

to punishment or due to natural causes. Consequently, the criminal organization needs to look

for new members periodically. Fourth, the criminal organization seeks new members among

peripheral criminals before it invites upstanding citizens.

5

Fifth, peripheral criminals compete

with each other, but not with the criminal organization. Although this assumption is made

primarily to simplify the analysis, there is evidence that the criminal activities of peripheral

criminals and organized crime do not overlap much (see, e.g., Paoli, 2002,2013). In Section 6,

we discuss how this affects our results. Finally, no crime boss will hire any worker who has a

history of disloyalty.

6

We find three equilibria in which the organization is formed.

7

Each of these equilibria is

associated with a specific mechanism for controlling members of the organization. We find that

the incentives to induce loyalty depend on the cost of punishing disloyalty to the boss and the

amount organized criminals can gain by being disloyal (the “strength of the temptation”). If

both the cost of punishment and the strength of the temptation are sufficiently small, the boss

will punish disloyalty. Wages compensate the criminals for the opportunity costs of not

working as an upstanding citizen, and for the expected costs of being caught and convicted.

8

The expected current and lifetime payoffs are the same for individuals in all three types of

employment.

3

Chalfin and McCrary (2017), among others, show that criminals do react to monetary incentives. Levitt and Vankatesh

(2000) show that gang members are also likely to be incentivized by the prospect of future payoffs.

4

According to Schloenhardt (1999): “the profit of organized criminal activities go to people who stand back and are not

directly involved in committing the crime.”In our model, these people are captured by a single agent: the boss.

5

There is evidence that past criminal acts are used as signals within the criminal community. For instance, Gambetta

(2009, p. 16) writes: “villains can certainly ask a potential partner or recruit to give them evidence of having committed

crimes.”This assumption simplifies the analysis, but is not necessary by itself. Our analysis requires only that past

disloyalty reduces the chance of being hired.

6

This too relates to trust. See Gambetta (2009) and von Lampe and Johansen (2004).

7

We assume that if neither reward nor punishment is used, there will be no incentive to form the organization because

all organized criminals will become disloyal.

8

Carvalho and Soares (2016) show that members of a criminal organization do earn higher wages, yet also face

additional risks.

BRAVARD ET AL.

|

1363