Punishing the Foreigner: Implicit Discrimination in the Premier League Based on Oppositional Identity*
| Date | 01 February 2013 |
| Author | Edoardo Gallo,J. James Reade,Thomas Grund |
| Published date | 01 February 2013 |
| DOI | http://doi.org/10.1111/j.1468-0084.2012.00725.x |
136
©Blackwell Publishing Ltd and the Department of Economics, University of Oxford 2012. Published by Blackwell Publishing Ltd,
9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA.
OXFORD BULLETIN OF ECONOMICS AND STATISTICS, 75, 1 (2013) 0305-9049
doi: 10.1111/j.1468-0084.2012.00725.x
Punishing the Foreigner: Implicit Discrimination in
the Premier League Based on Oppositional IdentityÅ
Edoardo Gallo†, Thomas Grund‡ and J. James Reade§
†Department of Economics, University of Cambridge and Queens’ College,
Cambridge CB3 9ET, UK (e-mail: edo@econ.cam.ac.uk)
‡Department of Sociology, University of Oxford, Nuffield College, New Road, Oxford OX1 1NF,
UK (e-mail: thomas.u.grund@gmail.com)
§Department of Economics, University of Birmingham, JG Smith Building,
Birmingham B15 2TT, UK (e-mail: j.j.reade@bham.ac.uk)
Abstract
We present the first empirical study to reveal the presence of implicit discrimination
in a non-experimental setting. By using a large dataset of in-match data in the English
Premier League, we show that white referees award significantly more yellow cards against
non-white players of oppositional identity. We argue that this is the result of implicit
discrimination by showing that this discriminatory behaviour: (i) increases in how rushed
the referee is before making a decision, and (ii) it increases in the level of ambiguity of
the decision. The variation in (i) and (ii) cannot be explained by any form of conscious
discrimination such as taste-based or statistical discrimination. Moreover, we show that
oppositional identity players do not differ in their behaviour from other players along
several dimensions related to aggressiveness and style of play providing further evidence
that this is not statistical discrimination.
I. Introduction
Implicit discrimination is a form of discrimination based on the unconscious mental asso-
ciation between members of a social group and a given (usually negative) attribute. It has
received a lot of attention from psychologists, while economists have tended to focus on
conscious forms of discrimination such as statistical discrimination or conscious taste-
based discrimination. Due to its nature, the evidence in support of the existence of implicit
discrimination has so far come exclusively from laboratory or field experiments. Most
ÅThe authors thank Neli Demireva for very insightful comments and suggestions, Markus Eberhardt for invaluable
technical assistance, and three anonymous referees for their helpful suggestions. Edoardo Gallo would like to thank
the George Webb Medley Endowment Fund and Nuffield College forfinancial support. Thomas Grund would like
to thank the Department of Sociology at Oxford and Nuffield College for financial support. All remaining errors are
our own.
JEL Classification numbers: L83, J71.
Implicit discrimination in the Premier League based on oppositional identity 137
commonly, implicit discrimination is assessed with implicit association tests (IAT), in
which participants are asked to relate names, words, or images to previously derived cat-
egories (e.g. pleasant).
This paper is the first study to provide empirical evidence of the existence of implicit
discrimination in a non-experimental setting. Following Bertrand, Chugh and Mullaina-
than (2005), we identify the discrimination that we observe as implicit discrimination by
showing that it is increasing in the level of ambiguity of the decision and it is decreasing
in the amount of time the discriminator has to take a decision. Complementing earlier
results with empirically derived findings adds to the existing literature on discrimination.
Furthermore, our study shows how large datasets can be used in an innovative way to test
the external validity of experimental results.
Discrimination is an important phenomenon which may lead to significantly worse
labour-market outcomes for the targeted individuals. However, it has proven to be very
elusive to systematic empirical studies. The major hurdle is the difficulty in collecting
comprehensive data on individuals’ decisions, while controlling for ambiguity and indi-
viduals’ heterogeneity. Early studies on discrimination rely on anthropological and ethno-
graphical case-studies that provide only anecdotal evidence. More recent studies rely on
experimental designs to overcome these difficulties.
In this paper, we exploit a large dataset on football matches in the English Premier
League (EPL) to systematically investigate discrimination by referees against players
belonging to specific social groups. The analysis of 1,050,411 in-match events allows
us to investigate all the decisions made by referees to sanction players, while taking
into account very detailed information on players’ actions on the field. The football
context is ideal for our purposes because the incentives of all individuals involved in
the game are well-defined and unambiguous, and there are clear rules that govern players’
actions.
We focus on referees’disciplinary sanctions against players of oppositional identity to
the referees. The concept of oppositional identity comes from Akerlof and Kranton (2000),
who argue that the strongest determinant of the formation of an oppositional identity is
difference in socio-economic opportunities. All the referees in our sample were born in the
United Kingdom and they have been living in the United Kingdom throughout their life.
We de fine oppositional identity players as players who are foreign,non-white and who
are from the same background as the most sizable minorities in the United Kingdom. This
definition fits the United Kingdom context in which oppositional identity is not defined
exclusively along racial lines, as in the United States.1
We provide evidence that referees award a significantly higher number of yellow cards
against oppositional identity players. Specifically, a conservative estimate is that an oppo-
sitional identity player is 15% more likely to receive a booking than a white player. This
result holds after controlling for several player, team, referee and match characteristics.
Moreover, there is no significant difference in the probability of receiving a booking for
other social groups defined along purely racial (i.e. Asians/Blacks/Whites), nationality
(i.e. UK/foreign) or linguistic criteria (i.e. English/non-English native speakers), which
indicates that our definition of the discriminated social group captures the nature of oppo-
1See, for example Berthoud (2010).
©Blackwell Publishing Ltd and the Department of Economics, University of Oxford 2012
To continue reading
Request your trial