Group versus Individual Performance Pay in Relational Employment Contracts when Workers are Envious

• Date: 01 March 2015
• DOI: http://doi.org/10.1111/jems.12088
• Published date: 01 March 2015
• Author: Jenny Kragl

Group versus Individual Performance Pay in

Relational Employment Contracts when Workers are

Envious

JENNY KRAGL

EBS Universit¨

at f¨

ur Wirtschaft und Recht

EBS Business School

Department of Management & Economics

Rheingaustrasse 1, 65375 Oestrich-Winkel, Germany

jenny.kragl@ebs.edu

I compare group to individual performance pay when workers are envious and performance

is nonverifiable. Avoiding payoff inequity, the group reward scheme is optimal as long as the

firm faces no credibility problem. The individual reward scheme may, however, become superior

albeit introducing the prospect of unequal pay. This is due to two reasons: Group incentives are

relatively low-powered compared to individual incentives, requiring higher incentive pay and

impeding credibility of the firm. Moreover, with individual rewards, the firm benefits from the

incentive-strengthening effect of envy, allowing for yet smaller overall incentive pay and further

softening the credibility constraint. I also show that contracts combining both individual and

group rewards are often optimal, depending on the firm’s credibility problem. These contracts

include joint and relative performance pay schemes.

1. Introduction

Individual employment relationships are typically embedded in the larger framework

of the firm, thus in a social context where mutual comparisons may play a role. In fact,

empirical evidence from the lab and the field suggests that workers not only care about

their absolute but also their relative economic position.1For example, in a recent field

study, Card et al. (2012) find relative pay comparisons to significantly affect workers’

wellbeing. Accordingly, by now there is a growing literature linking standard incentive

theory and social preferences. Much of the work focuses on envy among co-workers

or inequity aversion as proposed by Fehr and Schmidt (1999) in environments with

verifiable measures of performance.2Typically,the workers’ social preference then comes

at a cost for the firm because the former need to be compensated for the expected

I would like to thank Dominique Demougin, Martina Gogova, Julia Schmid, Dorothee Schneider, Anja

Sch¨

ottner, two anonymous referees, and the assigned co-editor for helpful comments. Earlier versions of

this paper were presented at the Annual Meetings of the European Economic Association, the Southern

Economic Association, the “Verein f¨

ur Socialpolitik,” the “Kolloquium zur Personal¨

okonomie,” the German

Economic Association of Business Administration e.V.,and a conference on “Health, Happiness, Inequality.”

I am grateful to the seminar participants for helpful comments and discussions. Moreover, financial support

by the Deutsche Forschungsgemeinschaft (DFG) through the CRC 649 and Research Grant DE 1169 is gratefully

acknowledged.

1. See, e.g., Goranson and Berkowitz (1966), Berg et al. (1995), Fehr et al. (1998), Fehr and Schmidt (1999),

or Charness and Rabin (2002). For an overview of the extensive literature on the effects of relative pay

comparisons on well-being see, e.g., Clark et al. (2008) or Akerlof and Yellen(1990) and the references therein.

2. In fact, jealousy and envy are of particular relevance in the workplace (see, e.g., Vecchio (2000) and

the references therein). Also according to Fehr and Schmidt (1999), envy is the stronger social motivation as

comparedto compassion. Other notable contributions regarding the formalization of social preferences include

C2015 Wiley Periodicals, Inc.

Journal of Economics & Management Strategy, Volume24, Number 1, Spring 2015, 131–150

132 Journal of Economics & Management Strategy

disutility from payoff inequity. Consequently, many studies conclude that concerns for

equity, fairness, or envy could explain observed wage compressions.3In the present

paper, I demonstrate that the foregoing conclusion may no longer hold if not reverse

when incentives rely on nonverifiable performance.

Incentive contracts often involve a considerable degree of discretion (see, e.g.,

Gibbs, 2013). More specifically, employees are frequently motivated by incentives that

rely on subjective evaluations by supervisors and, hence, nonverifiable measures of

performance. For example, they can expect to obtain some reward such as a bonus pay-

ment, nonmonetary benefit, or promotion in case they perform (sufficiently) well. Such

incentive contracts cannot be enforced in court and must hence be self-enforcing. That

is, they are implemented as relational contracts in a repeated game between the firm

and its employees. The firm’s ability to sustain these contracts then depends decisively

on her reputation with respect to honoring the agreements (see, e.g., Baker et al., 1994).

Kragl and Schmid (2009) show that envy among workers may soften the firm’s reputa-

tion problem and thus enhance its ability to implement relational contracts. While the

foregoing paper focuses on individual incentives, in the present paper, I introduce in

a similar framework the possibility of group compensation and compare its advantage

with the individual incentive regime. It turns out that, in contrast to the situation with

verifiable performance, individual performance pay (or some combination thereof with

group incentives) often outperforms group performance pay, in particular if the firm’s

credibility problem is sufficiently severe.

More precisely, I find that group bonus contracts dominate individual bonus con-

tracts as long as the firm’s interest rate is sufficiently small, reflectinga large commitment

ability of the firm. In line with the situation under verifiable performance, the reason is

that the firm must compensate envious workers for the expected payoff inequity from

individual incentives while the group scheme resolves that problem. Yet, once the firm’s

interest rate is sufficiently large, the result may reverse. Specifically, the optimal group

reward then induces the firm to renege on its promise. Credibility thus requires re-

ducing the group bonus thereby inducing suboptimal effort levels and reduced profits.

In comparison, the individual bonus provides two benefits. First, group incentives are

relatively low-powered compared to individual incentives because a worker’s chance

to receive the former not only depends on his own action but also on the co-worker’s

effort. Hence, the group bonus must be larger than the respective individual bonus for

implementing a given level of effort. Second, with individual incentives, the firm can ex-

ploit the endogenous incentive-strengthening effect of envy which allows for lowering

the level of the reward. Both features facilitate credible commitment of the firm with an

individual incentive regime as compared to a group scheme. In an extension, I consider

contracts in which the firm can combine both individual and group incentives and show

that the intuition from above carries over to this case. In particular, joint performance

pay is often optimal if the firm’s credibility problem is moderate. Notably, if the firm’s

credibility problem is sufficiently severe, even relative performance pay may become

superior.

The present paper relates to the literature on relational contracts and that on incen-

tive contracting with social preferences. As to the latter,the majority of papers focuses on

Rabin (1993), Bolton and Ockenfels (2000), Dufwenberg and Kirchsteiger (2004), and Falk and Fischbacher

(2006).

3. For example, in a survey study,Bewley (1999) finds that internal pay structures aim at providing internal

pay equity. In more recent theoretical studies, Englmaier and Wambach (2010), Goel and Thakor (2006) and

Bartling (2011) show that inequity aversion or envy among agents may render team incentives optimal.