Crafting messages to fight dishonesty: A field investigation of the effects of social norms and watching eye cues on fare evasion
Organizational Behavior and Human Decision Processes XXXXXXXXXX–19
Available online 10 November 2019
XXXXXXXXXX/© 2019 Elsevier Inc. All rights reserved.
Crafting messages to fight dishonesty: A field investigation of the effects of
social norms and watching eye cues on fare evasion
Shahar Ayal a, Jérémy Celse b,*, Guy Hochman a
a Baruch Ivcher School of Psychology, Interdisciplinary Center (IDC) Herzliya, Israel
Burgundy School of Business, CEREN EA7477, Université de Bourgogne Franche-Comté, Dijon, France
A R T I C L E I N F O
Jel codes:
C93
K42
B41
Keywords:
Fare evasion
Field experiment
Unethical behavior
Public transportation
Behavioral ethics
A B S T R A C T
The impact of watching eyes cues and descriptive social norm messages on fare evasion was studied in two
experiments that were conducted in two railway stations in France. In Study 1, a natural field experiment,
passengers were exposed for a two-week period to either a control eye-cues poster or to an experimental eye-cues
with a social norm messaging campaign. In Study 2, an artefactual experiment in the field, participants in the
experimental train station were asked to participate in a lying task before and after they were exposed to the
messaging campaign. The results from both studies suggest that although watching eye cues alone are not
effective in a crowded train station, exposing passengers to watching eye cues together with a descriptive social
norm messaging campaign reduced the fare evasion rates observed by standard inspection operations, and
eliminated lying behavior measured by the die-under-cup paradigm. These results strengthen the external validity
of laboratory experiments that have documented small scale cheating behavior and highlight the potential
enefits of internal enforcement techniques to fight dishonesty in the field. In addition, they stress the advantage
of combining visibility cues and social norms when orienting people toward more moral behavior.
1. Introduction
Small scale cheating, fraud, racketeering, and other forms of un-
ethical behavior on the part of ordinary people are among the greatest
personal and societal challenges of our time (Ariely, 2012; Bazerman &
Te
unsel, XXXXXXXXXXOn the individual level, the misdeeds of everyday
people appear relatively small and isolated. However, because so many
people are engaged in these misbehaviors, when considered cumula-
tively, they cause considerable societal damage and add up to millions of
dollars (Ariely, 2012; Barkan, Ayal, & Ariely, XXXXXXXXXXFor instance, tax
evasion in the US is estimated to be close to $500 billion a year (Cebula
& Feige, 2012) and roughly €80 billion a year in France.1 Insurance
fraud costs over $80 billion annually, and these costs are passed on to
customers (Wa
en & Schweitzer, in press). In addition to their direct
costs, research in behavioral ethics shows that dishonesty on the part of
ordinary people chips away trust, encourage negative social norms, and
increases the prevalence and spread of other unethical behaviors
(Kirchler, Hoelzl, & Wahl, 2008; Welsh, Ordóñez, Snyder, & Christian,
2015). All this underscores the need to find effective ways to regulate
and reduce small-scale unethical behavior.
Fare evasion (or fare dodging) is one example of unethicality and is
defined as the unlawful use of transit facilities by riding without paying
the applicable fare (Bureau of Transportation Statistics).2 Fare evasion
consists of travelling on public transportation without purchasing a
ticket or without validating one’s ticket in the machine. According to the
few studies that have recently examined fare evasion, it constitutes a
major source of loss of revenue for transport operators (Clarke, Contre,
& Petrossian, 2010; Cu
ie & Delbosc, 2017; Delbosc & Cu
ie, 2016;
Guarda, Galilea, Handy, Muñoz, & Ortúzar, XXXXXXXXXXWeerman (2007)
eported that half of all students in the Western u
anized part of the
Netherlands admitted to fare dodging at least once during the preceding
school year. In France, fare evasion rates are estimated - depending on
the type of transportation – to be between 5% and 25%, and costs op-
erators about €600 million annually. This equals the cost of 2390 busses
or the operating expenses for the public transit facilities of Lyon, Mar-
seille, Toulouse, Lille, and Bordeaux combined. In addition to its direct
costs, the costs associated with combatting fare evasion in France (such
as hiring ticket inspectors or investing in new technologies such as ticket
* Co
esponding author.
E-mail addresses: XXXXXXXXXX (S. Ayal), XXXXXXXXXX (J. Celse), XXXXXXXXXX (G. Hochman).
1 Solidaires-Finances Publiques, 2012.
2 The terms fare evasion and fare dodging are used here interchangeably.
Contents lists available at ScienceDirect
Organizational Behavior and Human Decision Processes
journal homepage: www.elsevier.com/locate/obhdp
https:
doi.org/10.1016/j.obhdp XXXXXXXXXX
Received 31 May 2018; Received in revised form 1 October 2019; Accepted 15 October 2019
mailto: XXXXXXXXXX
mailto: XXXXXXXXXX
mailto: XXXXXXXXXX
www.sciencedirect.com/science/journal/ XXXXXXXXXX
https:
www.elsevier.com/locate/obhdp
https:
doi.org/10.1016/j.obhdp XXXXXXXXXX
https:
doi.org/10.1016/j.obhdp XXXXXXXXXX
https:
doi.org/10.1016/j.obhdp XXXXXXXXXX
http:
crossmark.crossref.org/dialog/?doi=10.1016/j.obhdp XXXXXXXXXX&domain=pdf
Organizational Behavior and Human Decision Processes XXXXXXXXXX–19
10
gates) represents roughly €195 million annually. In addition, findings
indicate that fare evasion rates are related to criminal offenses such as
vandalism or assaults against fellow passengers or staff, and measures to
combat fare evasion have led to a significant decrease in these criminal
activities (Smith & Clarke, XXXXXXXXXXStudies have also shown that
increased rates of fare evasion generate perceptions of mistrust and
insecurity, thus increasing passengers’ psychological discomfort and
discouraging them from using public transportation (Bijleveld, 2007;
Killias, Scheidegger, & Nordenson, 2009; Reddy, Kuhls, & Lu, 2011).
Traditional forms of regulation such as the standard economic
approach to cu
ing fare evasion rely on external enforcement and
dete
ence. This approach tends to view passengers (fare evaders
included) as rational agents who aim at maximizing their utility by
weighting the benefits of travelling for free against the costs of being
caught (Becker, 1968; Boyd, Martini, Rickard, & Russell, 1989; Hirsch,
Bottoms, Burney, & Wikström, XXXXXXXXXXBased on this assumption, fare
evasion can be reduced by (1) increasing the likelihood of being caught
(e.g., more ticket inspections, physical platform access ba
iers, video
surveillance); and (2) increasing the severity of the punishment by
imposing heavier sanctions. In fact, transportation companies in France,
for example, have increased ticket inspection rates as well as the number
of fines (e.g., in 2016, 97% of all transportation companies increased
their fines), and even curtailed the presumed advantages of fare evasion
(e.g., shorter period of ticket validity). Studies that have investigated
external enforcement interventions stress the effectiveness of increasing
the probability of being caught, and indicate that higher inspection rates
tend to discourage fare evasion (e.g., Delbosc & Cu
ie, 2016; Koore-
man, 1993).
Nevertheless, recent studies have highlighted the limitations of
external enforcement (e.g., Ayal, Gino, Barkan, & Ariely, 2015; Hoch-
man & Erev, 2013; Thaler & Ganser, XXXXXXXXXXMore specifically, in-
spections are highly costly and it is virtually impossible to monitor the
actions of all passengers at all times. Furthermore, external punishment
and rewards can encroach on internal motivation (Ryan & Deci, 2000)
and untether people from their own moral compass (Ariely, 2008;
Bazerman & Gino, XXXXXXXXXXResearch on external inspections has reported
mixed outcomes. Killias et al XXXXXXXXXXfor instance examined the effects of
increasing the probability of inspections in Zurich’s subu
an transport
system and found that fare evasion was negatively co
elated with in-
spection rates but in a non-linear fashion. Similarly, Guarda et al. (2016)
showed there was an optimal inspection rate above which inspection
costs exceeded the benefits. Other studies have reported no co
elation
etween ticket inspection rates and fare evasion (Clarke et al., XXXXXXXXXXIn
terms of the severity of punishment; i.e., the sanction incu
ed if caught,
Guarda et al XXXXXXXXXXfound that inspection strategies can be cost-
effective even when evaders are not given a fine, and Bijleveld (2007)
showed surprisingly that increased severity of punishment led to higher
ates of fare evasion.
Here, we put forward two techniques of internal enforcement as an
alternative approach to cu
fare evasion rates, and describe how they
were tested in the field. The basic theoretical premise behind our
approach is that people value honesty and morality (Chaiken, Giner-
Sorolla, & Chen, 1996; Greenwald, 1980; Sanitioso, Kunda, & Fong,
1990), and strive to maintain a positive self-image as moral actors
(Duval & Wicklund, 1972; Rosenberg, XXXXXXXXXXAt the same time, when
there is an opportunity to benefit from immoral behavior, most people
choose to do so, at least to a certain extent (Gino, Ayal, & Ariely, 2009;
Hochman, Glöckner, Fiedler, & Ayal, 2016; Mazar, Amir, & Ariely,
2008; Ordóñez & Welsh, 2015; Ordóñez, Schweitzer, Galinsky, &
Bazerman, 2009; Shalvi, Handgraaf, & De Dreu, 2011a). This constant
conflict between people’s unethical behavior and their desire to main-
tain a moral and positive self-image creates a form of psychological
tension du
ed Ethical Dissonance (Ayal & Gino, 2011; Barkan et al.,
2015). This tension may act as a moral gate-keeper that enables people
to uphold their moral standards (Barkan et al., XXXXXXXXXXHowever, since
people find creative ways to dismiss this dissonance, in many situations
which pose a conflict between the desire to increase personal profit by
cheating and the desire to maintain a positive self, cheating behavior
prevails.
A growing body of research illustrates this dynamic, and shows that
people take advantage of specific contextual factors and employ self-
serving biases and justifications that enable them to benefit from
dishonesty without negatively updating their moral ledger (Ayal,
Hochman, & Ariely, 2016; Gino & Pierce, 2010; Shalvi, Gino, Barkan, &
Ayal, XXXXXXXXXXFor instance, Gino, Ayal, and Ariely XXXXXXXXXXfound that
when people’s dishonesty benefits others, they are more likely to view
dishonesty as morally acceptable, and thus cheat more and feel less
guilty about it. In a similar vein, cheating level increases when people
are confronted with more ambiguous situations (Hochman et al., 2016;
Shalvi, Dana, Handgraaf, & De Dreu, 2011b), or when they are exposed
to wealth cues such as stacks of dollars (Gino & Pierce, XXXXXXXXXXThus, a
promising avenue of internal enforcement stems from the fact that the
same factors that facilitate wrongdoing by reducing ethical dissonance
can be harnessed to design policies and interventions that stress the
dissonance and increase moral behavior. Specifically as regards socially
desired behaviors, previous research points at two main factors that may
help to design social environments that encourage compliance and
moral behavior: informing people of the desired social norms and what
others are doing (e.g., Bicchieri, 2005; Cialdini, Kallgren, & Reno,
1991), and giving people the feeling that they are observed by others (e.
g., Ernest-Jones, Nettle, & Bateson, 2011; Nettle, Nott, & Bateson,
2012).
The first factor, which is the main focus of our investigation, is social
proof (Cialdini et al., 1991; Reno, Cialdini, & Kallgren, 1993), a simple
principle which argues that knowing what others are doing (descriptive
norms) or what others think we should be doing (injunctive norms) exert
enormous influence on behavior (Bicchieri & Dimant, in press; Cialdini
& Trost, XXXXXXXXXXIn one classic natural field experiment, Cialdini et al.
(2006) documented the impact of descriptive social norm messages
about the theft of petrified wood in a US National Park and indicated
that a positive framing (e.g., percentage of people who do not steal) was
more effective than negative framing (e.g., percentage of people who
steal) in reducing theft. In a like manner, people are more likely to use
the stairs rather than the elevator if a sign informs them that most people
use the stairs than if the sign informs them about the health benefits of
using the stairs (Burger & Shelton, XXXXXXXXXXHouten and Nau (1981)
showed that informing drivers about the percentage of drivers who kept
within the speed limit was ten times more effective than extra police
patrols in reducing unsafe driving and speeding on the highway.
More recently, social norms have been used to successfully influence
taxpayers to pay their taxes on