Legal institution and the evolution of moral conduct

• DOI: http://doi.org/10.1111/jpet.12301
• Date: 01 October 2018
• Published date: 01 October 2018

725

Journal of Public Economic Theory. 2018;20:725–741. wileyonlinelibrary.com/journal/jpet © 2018 Wiley Periodicals, Inc.

Received: 21 June 2017 Accepted: 25 March 2018

DOI: 10.1111/jpet.12301

ARTICLE

Legal institution and the evolution of

moral conduct

Jon C. Thompson1Jiabin Wu2

1AtkinsonGraduate School of Management,

WillametteUniversity

2Departmentof Economics, University of Oregon

JonC. Thompson, Atkinson Graduate School of

Management,Willamette University, Salem, OR

97302(jcthomps@willamette.edu).

JiabinWu, Department of Economics,

Universityof Oregon, Eugene, OR 97403

(jwu5@uoregon.edu).

This paper argues that to better understand the relationship

between a legal institution and moral conduct, two features of the

legalinstitution need to be jointly considered: It is costly and endoge-

nouslydetermined. We propose a simple cultural transmission model

in the context of a legal institution. The strength of the legal institu-

tion is determined by majority voting and the cost of implementing

the legal institution is shared by the citizens. We find that the legal

institution can either “crowd in” or “crowdout” moral conduct in the

long run depending on the cost of implementation as well as the ini-

tial distribution of moral conduct in a society.

1INTRODUCTION

Formalinstitutions (such as a legal institution or framework) and people's intrinsic values (such as trust and moral con-

duct) are both shown to be crucial for economic development.1However, a large empirical and experimentallitera-

ture demonstrates that the interaction between formal institutions and intrinsic values is complicated (see Bowles

and Polanía-Reyes, 2012, for an excellentsurvey). For example, Titmuss (1970), Falkinger, Fehr,Gächter, and Winter-

Ebmer (2000), Gneezy and Rustichini (2000), Frey and Jegen (2001), Heyman and Ariely (2004), Irlenbusch and Sli-

wka (2005), Reeson and Tisdell (2008), and Li, Xiao, Houser,and Montague (2009), among many others, find evidence

of crowding out phenomena, in which formal institutions are counterproductive in fostering intrinsic values. On the

other hand, Grosjean (2011), Guiso, Sapienza, and Zingales (2008a), Tabellini (2010), Cassa, d'Adda, and Grosjean

(2014)find evidence of crowding in phenomena, in which formal institutions complement intrinsic values in encouraging

cooperation.

Correspondingly, many works provide theoretical explanations for these two types of phenomena. Forexample,

Frey (1997), Bar-Gill and Fershtman (2005), Güth and Ockenfels (2005), Bénabou and Tirole (2006), Sobel (2009),

Agion, Algan, Cahuc, and Shleifer (2010), Dufwenberg, Heidhues, Kirchsteiger, Riedel, and Sobel (2011), Hwang and

Bowles (2011), and Acemoglu and Jackson (2017) demonstrate various channels leading to crowding out phenomena.

Tabellini(2008) and Guiso, Sapienza, and Zingales (2008b), on the other hand, model crowding in phenomena. In addi-

tion, Huck (1998), Bohnet, Frey,and Huck (2001), and Jackson (2011) argue that crowding out or crowding in depend

on the strength of the formal institutions in a nonmonotonic way.

1For example, see Acemoglu, Johnson, and Robinson (2001, 2002, 2006) and Acemoglu and Robinson (2012) for a discussion on the former and Guiso,

Sapienza,and Zingales (2006) for a discussion on the latter.

2THOMPSON ANDWU

726

T

Most of these works treat formal institutions as exogenously given or determined by social planners. We instead

follow Tabellini (2008), who considers a model in which the strength of the legal institution is endogenously deter-

mined by the citizens through majority voting.However, Tabellini(2008) assumes that implementing a legal institution

is costless. Such a simplification ignores the fact that a legal institution voted on through a democratic process serves

as a public good for the citizens and the cost of implementing the legal institution is paid by the citizens. We believe

that incorporating the cost is of vital importance because it gives an additionally extrinsicincentive to the citizens and

therefore potentially alters the trajectory of the evolutionof intrinsic values.

Yetto our knowledge no paper considers how an endogenous and costly legal institution shapes intrinsic values. We

attempt to answer that question in this paper.

Todo this we propose a simple cultural transmission model in the context of a legal institution. Suppose that there

are two types of citizens in the population, moral and individualistic, who are randomly matched in pairs to play some

identical prisoner's dilemma format game with two actions: cooperate or defect. The moral type citizens are assumed

to have moral conduct and they enjoy an extrautility term by choosing to cooperate (as in Tabellini, 2008), while the

individualistic type citizens are only concerned about their material payoffs.

In each generation, the citizens first jointly determine whether to implement a legal institution as well as the

strength of the legal institution through majority voting. The legal institution we model follows Tabellini(2008). It can

detect a pair of citizens with one choosing to cooperateand the other choosing to defect. If detected, the citizen choos-

ing to defect pays a fine and the citizen choosing to cooperate receives an award equal in magnitude to the fine. The

detection rate measures the extent to which a legal institution can protect the citizens from unilateral defection. In

addition, increasing the detection rate increases the cost citizens need to payfor the legal institution.

After voting for the legal institution all citizens are randomly matched to play the prisoner's dilemma format game

with the potential consequences based on the legal institution they haveimplemented.2

Each citizen then becomes a parent and gives birth to a child. Children's types are malleable. Assume that parents

care about their children's utility in an imperfect fashion and they are motivated to exerteffort to inculcate their own

typesto their children. The higher the effort exerted, the higher the probability that a child adopts the type from his/her

parent. This assumption is called imperfect empathy in the seminal cultural transmission model by Bisin and Verdier

(2001). If a citizen fails to inculcate their own type to their child, the child randomlyfinds a role model in the population.

Given the three components of the model, we derive a cultural evolutionary dynamic and we find that the predic-

tions of the dynamic depends on the cost of implementing a legal institution.3

First, when the cost of implementation is sufficiently low, we find that in most circumstances a strict legal institu-

tion is implemented, in which the detection rate is sufficiently high to induce cooperation from everycitizen. However,

given that all citizens cooperate, from each citizen's own point of view the moral type citizens do equally well as the

individualistic type citizens. Therefore, in the cultural transmission process, neither type of citizen has an incentive to

exerteffort on inculcation. In this case, the corresponding population states are unchanging.

Our first result is related to some important contributions in the literature. Sobel (2009) and Dufwenberg et al.

(2011) find that in market settings citizens having social preferences cannot be distinguished from individualistic type

citizens. As argued by Bénabou and Tirole (2006), intrinsic incentives can be crowded out because extrinsicincentives

make a person's good conduct a less convincing signal of their intrinsic incentives, resulting in observers interpreting

the good conduct as merely individualistically motivated. In our model, a strict legal institution homogenizes every

citizen's behavior leading to a situation where the moral type citizens cannot be distinguished from the individualis-

tic type citizens.4Yet, moral conduct is not crowded out through the cultural transmissionprocess. Citizens'behav-

ior being indistinguishable is not necessary an adverse phenomenon from a societal point of view because everyone

2Notethat when the citizens vote, they anticipate their actions in the prisoner's dilemma.

3The cost of implementing a legal institution can be affected byvarious factors including population density andthe specific technologies used for detecting

illegalbehavior in certain economic activities. In addition, as argued by Besley and Persson (2009), it can be dependent on the legal origin of a country: the cost

ofprotecting property rights may be lower under common law than under civil law.

4Thisstands in contrast to Tabellini(2008) because our model presents the possibility of perfect enforcement, which creates a situation in which bot htypesof

citizensact identically. In Tabellini(2008) the moral and the individualistic type citizens always act differently regardless of the strength of the legal institution.

T

c

w

d

a

t

f

t

i

m

d

m

l

e

i

s

a

i

s

w

c

i

m

r

c

i

n

w

m

t

t

c

2

2

I

1

w

i

d

o