Relative Performance Evaluation and Competitive Aggressiveness

• Published date: 01 December 2022
• Author: CHRISTOPH FEICHTER,FRANK MOERS,OSCAR TIMMERMANS
• Date: 01 December 2022
• DOI: http://doi.org/10.1111/1475-679X.12431

DOI: 10.1111/1475-679X.12431

Journal of Accounting Research

Vol. 60 No. 5 December 2022

Printed in U.S.A.

Relative Performance Evaluation

and Competitive Aggressiveness

CHRISTOPH FEICHTER,∗FRANK MOERS,†

AND OSCAR TIMMERMANS‡

Received 19 June 2020; accepted 13 March 2022

ABSTRACT

We examine the relation between incentive plans based on relative perfor-

mance and competitive aggressiveness. Using data on executive incentive-

compensation contracts in large U.S. firms, we find a positive association be-

tween competitive aggressiveness and peer group overlap—that is, the extent

to which two firms select each other as peers in these incentive plans. Our

findings indicate that managers of such firms take more frequent as well as

more complex competitive actions, relative to managers evaluated on relative

performance without peer group overlap. Moreover, we show that these com-

petitive tactics are more pronounced when managers compete against: (1)

peers with similar grant sizes, (2) peers on similar performance metrics, and

(3) peers in the same industry. Collectively, our findings provide evidence

∗Vienna University of Economics and Business; †Maastricht University; ‡London School of

Economics

Accepted by Regina Wittenberg Moerman. We gratefully acknowledge comments from

Martin Artz, Iwan Bos, Isabella Grabner, Melissa Martin, Jae Yong Shin, Aner Zhou, and an

anonymous referee. We further thank workshop participants at Alliance Manchester Business

School, Maastricht University, Open University of the Netherlands, Seoul National Univer-

sity, Vienna University of Economics and Business, and conference participants at the 11th

Conference on New Directions in Management Accounting, the 2nd Swiss Winter Account-

ing Conference, and the 2020 Management Accounting Section Midyear Meeting for their

valuable comments. We also thank Brian Connelly for sharing his code to classify competi-

tive actions into types. Finally, we thank The Wharton School for access to ISS Incentive Lab

and RavenPack News Analytics. An online appendix to this paper can be downloaded at http:

//research.chicagobooth.edu/arc/journal-of-accounting- research/online-supplements.

1859

© 2022 The Authors. Journal of Accounting Research published by Wiley Periodicals LLC on behalf of The

Chookaszian Accounting Research Center at the University of Chicago Booth School of Business.

This is an open access article under the terms of the Creative Commons

Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium,

provided the original work is properly cited, the use is non-commercial and no modifications or

adaptations are made.

1860 c. feichter, f. moers, and o. timmermans

on how widely used incentive-compensation practices relate to strategic firm

decisions.

JEL codes: D22, J33, J41, L1, M4

Keywords: relative performance evaluation; peer group overlap; competi-

tive aggressiveness; strategic interaction; collusion

1. Introduction

The usage of incentive plans based on relative performance in large U.S.

firms has grown from 22% to 67% from 2006 to 2019 (e.g., Meridian Com-

pensation Partners LLC. [2019], Equilar [2020]). Given the ubiquity of rel-

ative performance plans, it is of paramount interest to investors, regulators,

and practitioners to understand how these incentive plans affect firm deci-

sions. However, empirical evidence on the implications of relative perfor-

mance plans for firm decisions is limited. We address this important void

by examining how these incentive-compensation practices relate to com-

petitive actions that aim to directly challenge rivals. In particular, we relate

the extent to which firms select each other as peers to their competitive

aggressiveness.

A key purpose of relative performance evaluation (hereafter “RPE”) is to

improve risk-sharing between the principal and the agent by benchmarking

the agent’s performance against peers that are affected by common shocks.

This allows the principal to provide more efficient incentives (e.g., Holm-

ström [1982]). However, basing pay on relative performance at the same

time puts the agent in direct competition to its peers. An agent can respond

to this competition in two ways, that is, (1) sabotage and (2) collusion (e.g.,

Gibbons and Murphy [1990]). In terms of competitive aggressiveness, sab-

otage implies more competitive aggressiveness, whereas collusion implies

less competitive aggressiveness. This suggests that RPE incentive plans can

increase or decrease competitive aggressiveness. Below we explain the ra-

tionale for each scenario in detail.

The reason why RPE incentive plans can increase competitive aggressive-

ness is fairly intuitive. The manager of the focal firm has an incentive to

gain an advantage over competitors and improve the firm’s relative posi-

tion, which he/she can achieve by engaging in competitive actions. This

intuition is formalized in a simple theoretical framework developed by Ag-

garwal and Samwick [1999]. This framework predicts that, in a setting with

two competing agents, the incentives to act aggressively are greatest if both

agents are evaluated based on own- and peer performance, because both

agents then have an incentive to outperform each other. Such “reciprocity”

is not necessarily present in relative performance plans of large U.S. firms—

a focal firm’s peers need not use RPE themselves or, if they use RPE, they

need not select the focal firm as their own peer. Throughout the paper,

we refer to such an overlapping peer relationship as “peer group overlap.”

There are no economic forces that either prohibit or require firms to select

rpe and competitive aggressiveness 1861

each other as peers, and as such there is variation in the degree to which

competing firms are evaluated based on each other’s performance. As a re-

sult, if relative performance plans increase competitive aggressiveness, this

increase is proportional to the extent to which two firms select each other

as peers in these incentive plans.

The reason why RPE incentive plans can decrease competitive aggressive-

ness is also intuitive but less straightforward. It is typically the case that all

firms would be better off when none of them are competitively aggressive

compared to when all of them are (e.g., Aggarwal and Samwick [1999]). In

the presence of RPE, a necessary condition for the latter is a commitment

to abstain from being competitively aggressive. Such commitment can be

created through collusion. If firms using RPE incentive plans collude, they

are less competitively aggressive than firms that do not use these incentive

plans. Moreover, if relative performance plans decrease competitive aggres-

siveness, then this decrease is also proportional to the extent to which two

firms select each other as peers in these incentive plans. Because both sce-

narios are plausible, the goal of this paper is to test which scenario plays

out in practice. Strictly speaking, our question is therefore an empirical

question.

We empirically examine the relation between peer group overlap in rela-

tive performance plans and competitive aggressiveness using a sample of

355 unique U.S. firms (1,623 firm-years) over the period 2006 through

2017. We identify peer group overlap in managers’ incentive-compensation

contracts based on the Compensation Discussion and Analysis (CD&A) sec-

tion of the proxy statement. The average peer group overlap is 15%, which

implies that approximately one in seven firm-peer relationships is a recip-

rocal peer relationship. To measure competitive aggressiveness, we follow

a rich literature on competitive actions in strategic management, and use

structured content analysis of competitive actions identified by news events.

Examples of such actions are the introduction of new products in an at-

tempt to steal market share from a peer, the launch of a new marketing

campaign, price cuts, and the initiation of a joint venture. Our first mea-

sure of competitive aggressiveness is the firm’s action volume, which cap-

tures the number of competitive actions for a given period. On average,

our sample firms take about 33 actions per year. Our second measure is the

firm’s action complexity. This measure embraces the idea that not all actions

are identical—it exploits variation in firms’ competitive repertoires across

multiple action types, such as, new products, pricing, marketing, and joint

ventures. Our sample firms have, on average, considerable variation in their

action repertoires across action types. A key advantage of using these two

measures is that they allow us to capture a broad and comprehensive set

of relevant and impactful competitive actions. Nevertheless, to ensure that

our inferences are not unique to these specific measures of competitive ag-

gressiveness, we triangulate and confirm that our results are robust to using

accounting-based input (i.e., advertisement expenditures) and output (i.e.,

operating margins) measures of competitive aggressiveness.