Relational contracts with subjective peer evaluations

• Author: Joyee Deb,Jin Li,Arijit Mukherjee
• Date: 01 February 2016
• DOI: http://doi.org/10.1111/1756-2171.12116
• Published date: 01 February 2016

RAND Journal of Economics

Vol.47, No. 1, Spring 2016

pp. 3–28

Relational contracts with subjective peer

evaluations

Joyee Deb∗

Jin Li∗∗

and

Arijit Mukherjee∗∗∗

We study optimal dynamic contracting for a firm with multiple workers where compensation is

based on public performance signals and privately reported peer evaluations. We show that if

evaluation and effort provision are done by different workers (e.g., consider supervisor-agent

hierarchy), first-best can be achieved even in a static setting. However, if each worker both

exerts effort and reports peer evaluations (e.g., consider team setting), effort incentives cannot

be decoupled from truth-telling incentives. This makes the optimal static contract inefficient.

Relational contractsbased on public signals increase efficiency. Interestingly,the optimal contract

may ignore signals that are informative about effort.

1. Introduction

In modern labor markets, most workers perform jobs where objectiveperformance measures

are hard to obtain (Prendargast, 1999). Consequently, firms often rely on subjective performance

measures to provide work incentives. For instance, a firm may use measures such as leadership

skills, entrepreneurial drive, or client satisfaction scores. Some of these subjective measures may

be publicly observable. It is well known that such publicly observable measures, even if they are

subjective and imperfect, can be used as an input into compensation structure to sharpen worker

incentives.

∗YaleUniversity; joyee.deb@yale.edu.

∗∗Nor thwesternUniversity; jin-li@kellogg.northwestern.edu.

∗∗∗Michigan State University; arijit@msu.edu.

For their helpful comments and suggestions, we would like to thank Chris Ahlin, Ricardo Alonso, Susan Athey, Heski

Bar-Isaac, Anthony Creane, Carl Davidson,Wouter Dessein, William Fuchs, Marina Halac, Jin-Hyuk Kim, VijayKrishna,

Jim Malcomson, Bentley MacLeod, NikoMatouschek, Paul Milg rom, and seminaraudiences at the 2012 Asian Meeting

of the Econometric Society,Bocconi University, Columbia Strategy Conference, IUPUI, Michigan State University,2013

Midwest Economics Association Meetings, NBER Organizational Economics Meeting (December 2012), New York

University (Stern School of Business), University of Hong Kong, and Yale University. We are also grateful to David

Martimort (the editor) and two anonymous referees for their valuable feedback.

C2016, The RAND Corporation. 3

4/THE RAND JOURNAL OF ECONOMICS

However, there may be other valuable performance-related information that is not publicly

observable, but rather diffused within the organization. This is becoming increasingly the case

with the prevalence of team-based organizations coupled with increased complexity of tasks and

decentralization of authority (see, e.g., Che and Yoo, 2001). In such settings, coworkers mayhave

the most information about an individual’s performance and contribution toward the overall team

outcome (Fedor, Bettenhausen, and Davis,1999; May and Gueldenzoph, 2006). However, it may

be difficult for the organization to elicit and use this information as it is privately observed by

workers and inherently subjective.

Indeed, firms often institute systematic processes to elicit subjective evaluation of a worker’s

performance from his coworkers. Forexample, in a typical organizational hierarchy, the supervi-

sor is often in charge of evaluating her subordinates’ performance and providing this information

to the owners of the firm (see Tirole, 1986). In a team-production environment, the use of 360-

evaluations is commonplace. Under such evaluations,the firm seeks a worker’s performance eval-

uation from several people whoworked closely with him during the evaluation period,ir respective

of their relative position in the organizational hierarchy. As May and Gueldenzoph (2006) note,

increasingly “companies are turning to 360-degree multirater feedback and intragroup peer evalu-

ation systems for the purpose of managing performance and determining compensation rewards.”

An estimated 90% of Fortune 1000 firms have implemented some form of multisource assessment

that includes peer evaluations (Edwards and Ewen, 1996). Such feedback is typically privately

observed by the firm to ensure anonymity of the evaluator and to encourage candid reporting.

It is well documented that firms often combine publicly observed subjective measures and

subjective private performance evaluation in their compensation policies. For example, Field

(2006) presents a Harvard Business Review (HBR) case study of a mutual fund company in

which the performance evaluation system for portfolio managers involvespaying bonuses where

60% of the bonus is determined by the financial performance of the fund they direct and 40%

determined by the quality of teamwork, which is assessed through structured feedback (gathered

from team members and analyzed by top managers). Another HBR case study by Rose and Sesia

(2010) on the investment bank Credit Suisse highlights a similar compensation structure.1,2

The central objective of this article is to explain how a firm should optimally use public

subjective measures along with dispersed private subjective measures to offer sharper work

incentives. When the performance measure is public (but nevertheless, subjective), a common

channel for incentive provision is a relational contract: the firm offers the workers a discretionary

bonus based on their publicly observed performance, and lives up to its promised payments as

reneging may result in future retaliation by the workers.3However, offering performance pay

based on private evaluationis fraught with a unique set of challenges. First, the firm must provide

the workers the right incentives to report evaluations of their coworkers truthfully. Moreover,

as the subjective evaluations are reported privately to the firm, the firm itself must have the

incentive to report the true evaluations back to the workers. It turns out that truthful revelation

of private subjective information mayimpose an additional agency cost as the workers’ incentive

for exerting effort is interlinked with their incentive for truthful reporting.

We study a setting in which a firm can use publicly observable and privately reported

performance measures and analyze how incentivesbased on private information can be combined

1In addition to the financial service sector, similar compensation policy is also documented in other industries, such

as consumer goods. Simons and Kindred (2012) report a case study on Henkel, a manufacturer of personal care products,

where workers’compensation includes a significant variable pay (bonus) component that depends on overall organization

performance, team results, and the team leaders’ subjective evaluations on the workers’individual contributions.

2Note that some aspects of the firm’sperfor mance (e.g., sales volume),are verifiable, and can, in principle, affect

a worker’s compensation through explicit contracts. However, the firms often refrain from using such explicit contracts

as these verifiable measures could be misaligned with the firm’slong-ter m goals. Instead, relational contracts are often

used to link pay to performance measures that may be nonverifiablebut better reflect the fir m’s business objectives (see

Baker, Gibbons, and Murphy,1994).

3See, for instance, Bull (1987), Levin (2003), and Malcomson (2013) for a survey.

C

The RAND Corporation 2016.