Self‐managed work teams: An efficiency‐rationale for pay compression

• Author: Marc Möller,Nana Adrian
• DOI: http://doi.org/10.1111/jems.12339
• Published date: 01 April 2020
• Date: 01 April 2020

J Econ Manage Strat. 2020;29:315–334. wileyonlinelibrary.com/journal/jems © 2020 Wiley Periodicals, Inc.

|

315

Received: 17 January 2019

|

Revised: 29 November 2019

|

Accepted: 20 December 2019

DOI: 10.1111/jems.12339

ORIGINAL ARTICLE

Self‐managed work teams: An efficiency rationale for pay

compression

Nana Adrian |Marc Möller

Department of Economics, University of

Bern, Bern, Switzerland

Correspondence

Marc Möller, Department of Economics,

University of Bern, Bern, Switzerland.

Email: marc.moeller@vwi.unibe.ch

Abstract

This paper uncovers a novel mechanism through which pay dispersion can

have a negative effect on firm performance, even in the absence of equity or

fairness considerations. We use a stylized model of a self‐managed work team

to show that, when team‐work involves heterogeneous tasks, the provision of

incentives to exert effort conflicts with the provision of incentives to share

information relevant for decision‐making. Pay dispersion deteriorates

information sharing as it induces workers to conceal “bad news”to maintain

their coworkers motivation. The practical implications of our theory are that

team empowerment should go hand in hand with pay compression and that

empowerment should be avoided when team production involves strongly

heterogeneous tasks.

1|INTRODUCTION

The effect of pay dispersion on individual and organizational performance has been a topic of great interest in the

economics and management literature.

1

While incentive‐and tournament‐theories have advocated the positive

motivation‐and sorting‐aspects of pay dispersion (e.g., Lazear, 2000; Lazear & Rosen, 1981), equity and fairness

arguments have been employed to point out its potentially negative consequences for cooperation (e.g., Akerlof &

Yellen, 1990; Charness & Kuhn, 2007). Pfeffer and Langton's (1993) finding of the adverse effect of pay dispersion on

the productivity of university faculty has spurred empirical research into the relationship. However, results are a

rather inconclusive ranging from positive (e.g., Beaumont & Harris, 2003; Hibbs & Locking, 2000) to negative

(e.g., Bloom, 1999; Yanadori & Cui, 2013) and it has been noted that “a significant gap in our knowledge concerns the

underlying mechanisms or the mediators between pay dispersion and outcomes at the organizational, team, and

individual levels.”(Shaw, 2014, p. 538). This paper contributes to filling this gap by examining the consequences of pay

dispersion for the performance of self‐managed work teams.

Self‐managed work teams constitute an increasingly prevalent organizational form. For instance, Lazear and Shaw

(2007) document that in 1999, 72% of the U.S. firms employed self‐managed work teams, an increase from 27% in 1987.

Self‐managed work teams differ from standard teams in that they decide on a wide range of issues such as staffing,

scheduling, and budgeting, and therefore, enjoy a considerable degree of autonomy in “how to get the job done.”Firms

thus face the double challenge of providing self‐managed work teams not only with incentives to exert efforts but also

with incentives to share information relevant for decision‐making.

The central insight of our theory is that, in self‐managed work teams, pay dispersion may have a negative

effect on information sharing and hence decision‐making which counteracts, and potentially dominates, its

1

See Shaw (2014) for a recent survey.

positive effect on effort. Our theory thus identifies a novel mechanism, through which pay dispersion influences

organizational performance, and whose overall effect can be negative, even in the absence of equity or fairness

considerations. Besides its potential for informing empirical research, our theory might be of interest for

practitioners because the identified trade‐off has direct implications for organizational design variables, such as

team empowerment and task allocation.

The idea that pay dispersion may deteriorate information sharing in teams is in line with an influential study by

Beersma et al. (2003) for an experimental setting where team performance depends on both speed (effort) and accuracy

(information). The authors find that pay systems that reward all team members equally, have a negative effect on speed

but a positive effect on accuracy, “because they promote the diffusion of knowledge throughout a team”(Beersma

et al., 2003, p. 582). A direct implication of this effect is that in firms with self‐managed work teams the overall effect of

pay dispersion on performance should be less positive. Indeed, regressing plant‐level productivity on pay dispersion and

the extent of use of self‐managed work teams, Shaw, Gupta, and Delery (2002) find a significant, negative value for the

interaction coefficient. Although Shaw et al. (2002) consider an industry where the average use of self‐managed work

teams is rather low (Trevor, Reilly, & Gerhart, 2012), their results are at least indicative of the importance of the

mechanism identified by our theory.

In Section 2we propose a stylized model of a self‐managed work team by including a project‐selection stage into an

otherwise standard team production framework a la Holmström (1982). A competitive firm hires two workers to jointly

choose and execute one out of two mutually exclusive projects, Pand Q. Workers receive bonuses conditional on project

‐success which depends on project‐quality and the workers' noncontractible individual efforts. Project Qis of uncertain

quality and, although Q's quality is higher than P's from an ex ante perspective, each worker may privately and

independently receive “bad news”about Q. From an individual workers' viewpoint, information sharing is subject to a

trade‐off between adaptation and motivation: On the one hand, disclosure of bad news about Qleads to the adoption of

the (ex post) higher‐quality project P. On the other hand, concealment of bad news maintains the coworker's motivation

to exert high effort on project Q.

2

The trade‐off between adaptation and motivation extends from the individual to the organizational level when

workers are assigned to tasks (e.g., product‐engineering and product‐marketing) which differ in that effort spent on

one task is more decisive for project‐success than the effort spent on the other. In Section 3we consider as a

benchmark a firm with a standard team, where access to information concerning project‐quality and the authority to

choose the project are restricted to the firm's owner. We show that the efficiency losses from free‐riding are mini-

mized by offering a larger bonus to the worker assigned to the more decisive task.

3

In contrast, in Section 4we find

that, in a firm with a self‐managed work team, information sharing is optimized when a larger bonus is given to the

worker assigned to the less decisive task. Optimal incentives for information sharing turn out to be diametrically

opposed to optimal incentives for effort, because those tasks that are easiest to incentivize via a bonus are precisely

those that are easiest to manipulate via the concealment of bad news. As a consequence, the effect of pay dispersion

on performance becomes ambiguous: Pay dispersion increases efforts but may result in a break‐down of information

sharing and a reduction in the firm's quality of decision‐making. This finding contrasts with the common view that

team empowerment leads to greater organizational adaptability (Wageman, 1997) but is consistent with case studies

emphasizing the importance of communication within self‐managed work teams (e.g., Cummings, 2004; Griffin &

Hauser, 1992).

A practical implication of the above trade‐off is that, in self‐managed work teams, differences in compensation

across tasks should be compressed. The fact that corporate partnerships—an extreme example of self‐managed work

teams—have adopted a culture of “equal revenue‐sharing”can be taken as a sign that this recommendation tends to

match with observed practice. For example, Encinosa, Gaynor, and Rebitzer (2007) investigate medical practices and

find that 54% of partnerships consisting of three to five doctors share revenues equally, with equal sharing still playing

an important role (24%) in larger organizations (16–24 members).

4

2

In equilibrium the absence of bad news is understood as either no bad news having been received or bad news having been concealed, that is, information does not “unravel”as in Milgrom (1981)

and the absence of bad news signifies good news about Q.

3

This is in line with the free‐riding evidence provided by Gaynor, Rebitzer, and Taylor (2004). Investigating the cost‐saving efforts of physicians facing a group‐based incentive to meet a common

budgetary target, they find that physicians with a larger number of patients (whose cost‐saving efforts are more decisive for meeting the target) are promised a greater share of the bonus.

4

Farrell and Scotchmer (1988) discuss the practice, common within law firms, to equalize shares across partners of similar seniority. In many countries, equal revenue‐sharing constitutes the default

in the corporate definition of a partnership (e.g., United States Uniform Partnership Act §18a).

316

|

ADRIAN AND MÖLLER