Bifurcated Effects of Place‐of‐Origin Diversity on Individual and Team Performance: Evidence from Ten Seasons of German Soccer

• Date: 01 October 2017
• Author: Avner Ben‐Ner,John‐Gabriel Licht,Jin Park
• DOI: http://doi.org/10.1111/irel.12188
• Published date: 01 October 2017

Bifurcated Effects of Place-of-Origin Diversity

on Individual and Team Performance: Evidence

from Ten Seasons of German Soccer*

AVNER BEN-NER, JOHN-GABRIEL LICHT, and JIN PARK

How does diversity affect performance? We develop a theoretical framework in

which diversity of prevention employees (protecting the organization from harm)

and of promotion employees (advancing positive outcomes) have different effects

on individual and organizational performance. We use data for twenty-eight soc-

cer teams and 1723 players that played in 6120 games during ten seasons and

find that diversity of defensive (prevention) players has a positive effect on player

and team performance whereas the opposite holds for offensive (promotion) play-

ers. Joint tenure of offensive players tends to amplify these effects.

Introduction

In recent decades, there has been a massive movement of people across

national boundaries. For example, in 2010, almost 40 million out of 309 mil-

lion people in the United States were born in another country; 10.5 of 82 mil-

lion in Germany and 7.5 of 62 million in the UK were born in another

country (www.OECD.StatExtracts). As a result, in numerous organizations,

people of different ethnic, national, religious, cultural, and linguistic back-

grounds work side by side. People from different places bring skills and ideas

not found in homogeneous workplaces, but working in diverse teams may

involve favoritism and discrimination that results in less productive

*The authors’affiliations are, respectively, University of Minnesota-Twin Cities, Minneapolis, Minnesota.

E-mail: benne001@umn.edu; St. Norbert College, De Pere, Wisconsin. E-mail: jglicht@gmail.com; and

Zayed University, United Arab Emirates. E-mail: jin.park.s@gmail.com. The authors are grateful for the

helpful comments to referees and many colleagues, including Christine Bishop, John Budd, Colleen Manch-

ester, Andreas Ortmann, Ariane Szafarz, Enno Siemsen, Aaron Sojourner, Daniel Sokol, Migiwa Tanaka,

Andy Van de Ven, Joel Waldfogel; participants at seminars at the Carlson School of Management and Law

School at the University of Minnesota, School of Labor and Employment Relations, University of Illinois at

Urbana-Champaign, School of Management and Labor Relations, Rutgers University, Economics Depart-

ment, Universit

e Catholique de Louvain and Economics Department, University of Siena; and participants at

conferences, including SABE/ICABEEP 2014, THEEM University of Konstanz 2014, ISNIE Florence 2013,

IIOC Chicago 2014, IMBESS Oxford 2014, Academy of Management Philadelphia 2014, and ASSA Boston

2015. Eliot Ben-Ner and John Souter gave extremely helpful feedback and advice on how players in differ-

ent positions operate and interact with each other on the soccer field.

INDUSTRIAL RELATIONS, Vol. 56, No. 4 (October 2017). ©2017 Regents of the University of California

Published by Wiley Periodicals, Inc., 350 Main Street, Malden, MA 02148, USA, and 9600 Garsington

Road, Oxford, OX4 2DQ, UK.

555

collaboration as well as practical problems of communication. Hence, diversity

in organizations does not have a uniformly net positive or negative effect on

organizational outcomes; empirical studies have found just that, at multiple

levels of analysis (for reviews, see Alesina and La Ferrara 2005; Horwitz and

Horwitz 2007; Joshi and Roh 2009; van Knippenberg and Schippers 2007).

This raises the possibility that some factors moderate the diversity–perfor-

mance relationship (Joshi and Roh 2009); a variety of factors have been stud-

ied in the literature. Garnero, Kampelmann, and Rycx (2014) carefully

reviewed the literature on the demographic diversity–performance relationship

in several disciplines, identified various factors that influence this relationship,

and evaluated the empirical strategies employed in various studies.

In this paper we propose that the organizational role of a group of employ-

ees influences how the group’s diversity affects performance. Specifically, we

distinguish between organizational subgroups or departments that focus on the

promotion of organizational goals (e.g., sales, marketing, communication, and

research and development) and subgroups that focus on prevention of harm to

the organization (e.g., human resources, legal services, purchasing, and quality

control); a third group mediates between promotion and prevention (e.g., oper-

ations and logistics).

Prevention and promotion tasks may attract different types of individuals,

and present them with different incentives. Prevention employees, who face

external threats, are inclined to unite, whereas promotion employees may tend

to promote themselves. Moreover, it is easier to identify individual contribu-

tions in promotion-focused roles, hence individual incentives may be stronger

than in prevention roles. Altruistic sentiments and trust, which are important

for collaboration, are weakened by favoritism towards in-group members and

by discrimination against out-group members associated with diversity. These

differences between the two roles suggest that collaboration will decline faster

with increased diversity in promotion than in prevention.

Does collaboration improve over time? Diverse people may learn to be more

accepting and trusting over time and therefore collaborate better. However,

relationships that start off poorly may suffer from a vicious cycle of negative

reciprocity, so improvements over time, if any, may be lesser in promotion

than prevention.

The diversity–performance nexus has been studied mostly at the group or

organization level. However, if performance of subgroups is affected by diver-

sity in opposite directions, then looking at aggregate performance may be mis-

leading because of offsetting effects. For example, if diversity has negative

performance effects in promotion and positive effects in prevention, the diver-

sity effects at the entire group or organization level may be positive, negative,

or nil, depending on the size of the subgroups and the magnitude of the

556 / AVNER BEN-NER,JOHN-GABRIEL LICHT AND JIN PARK

effects. It is therefore important to disaggregate the group or organization into

subgroups (as Carton and Cummings [2012] emphasized).

The effects of diversity have been studied at the group or organization level,

emphasizing consequences of interactions among individuals. These interac-

tions also affect, of course, the performance of individuals. In terms of the liter-

ature on peer effects (e.g., Mas and Moretti 2009) the question then is how the

characteristics of an individual’s coworkers affect the individual’s performance.

To investigate empirically the diversity–performance relationship in different

organizational roles and the impact of joint tenure on this relationship one needs

detailed longitudinal data on diversity, performance, and tenure by organizational

roles (job classifications, departments, or functions in organizations). The obser-

vations need to be longitudinal in order to allow the observation of an individual

in different situations and to isolate the effect of group composition from other

effects. Such data are generally unavailable. One exception is professional team

sports, in which detailed information is collected consistently over time. Indeed,

there have been several studies of the effect of diversity on performance in sports

(e.g., Haas and N€

uesch [2012] for soccer, Timmerman [2000] for basketball, and

Kahane, Longley, and Simmons [2013] for hockey).

1

Crucially, in certain team

sports, like soccer, there are well-defined roles that map clearly onto the promo-

tion and promotion roles. This distinction has not been exploited in the literature.

We study all the games for ten seasons of the foremost German soccer league,

Bundesliga. In soccer, prevention employees are the “defenders,”promotion

employees are the “forwards,”and “midfielders”are employees in the mediating

role.

2

Thanks to a unique and rich dataset, we are able to evaluate not only organi-

zation-level (soccer team) performance relative to organization-level diversity but

also organization-level performance relative to diversity in subgroups. Further-

more, we can look at performance indicators that link more strongly to a particu-

lar subgroup (goals scored and goals conceded). Importantly, we have measures

of individual performance, which are measured independently of the team-level

outcomes but are correlated with it (such as the quality of passes, tackles, and

shots, the relative importance of which varies by organizational role). We are also

able to assess the effect of joint tenure in the promotion and prevention subgroups

on the diversity–performance relationship. Unlike other studies, we control for

player, subgroup, and team skills, which obviously affect performance.

1

Kahn (2000); Katz (2001); Day, Gordon, and Fink (2012); and others argue that sports teams are use-

ful for the study of various organizational phenomena; several authors have done so using data on soccer

teams (e.g., Apesteguia and Palacios-Huerta 2010; Grund 2012; Bucciol, Foss, and Piovesan 2014).

2

There is some variation in what different positions are called. In our classification, defense includes

the titles: goal keeper, center back, sweeper, right and left back, and right and left wing backs. The midfield-

ers include various midfielder titles: defensive, center, attacking, wide, and so on. The forward group

includes titles such as forward (right, center, and left), striker, and winger.

Bifurcated Effects of Diversity on Performance / 557