CEO Overpayment and Dismissal: The Role of Attribution and Attention

• Author: Junxiong Fang,Lerong He
• DOI: http://doi.org/10.1111/corg.12129
• Date: 01 January 2016
• Published date: 01 January 2016

CEO Overpayment and Dismissal: The Role of

Attribution and Attention

Lerong He*and Junxiong Fang

ABSTRACT

Manuscript Type: Empirical

Research Question/Issue: This study investigates the moderating role of CEO overpayment on the relationship between f‌irm

performance and CEO dismissal. We also examine how contextual factors, includingcompensation disclosure regulation, f‌irm

index status,and f‌irm age, inf‌luence board attentionand attribution, and consequently affect the sensitivity of CEO dismissalto

f‌irm performance.

Research Findings/Insights: Using a sample of Chinese listed f‌irms between 2002 and 2011, we f‌ind that overpaid CEOs are

associated with a larger likelihood of dismissal in case of poor f‌irm performance compared to their underpaid counterparts.

In addition, CEO overpayment has a larger inf‌luence on the turnover-performance relationship when executive compensation

disclosure is mandatory, when a f‌irm is index-included, or younger.

Theoretical/Academic Implications: This study provides empirical support for attribution theory and the attention-based

view. Built on the concept of bounded rationality, it demonstrates that board sense-making and causal attributions affect the

CEO dismissal decision. Our study also sheds light on the inf‌luence of situational cues on shaping board attention and subse-

quent corporate governance decisions.

Practitioner/Pol icy Implications: This study offers insights to policymakers interested in enhancing the design of corporate

governance mechanisms by paying attention to cognitive processes in the boardroom.

Keywords: Corporate Governance, CEO Turnover, Executive Compensation, Pay Deviation, China

INTRODUCTION

Akey element of corporate governance is embedded in

the right of a board of directors to hire, compensate,

and f‌ire its top management. Specif‌ically, it includes an im-

plicit labor market agreement exercisable at the board’sdis-

cretion to replace incumbent managers when necessary. The

threat of termination thus provides top management with

powerful incentives to protect shareholder interests and mit-

igates agency problems caused by the separation of owner-

ship and control (Fama, 1980; Jensen & Meckling, 1976).

Extant empirical literature has documented a signif‌icant neg-

ative relationship between f‌irm performance and executive

turnover, that is, CEOs in poorly performing f‌irms are asso-

ciated with a larger likelihood of turnover (e.g.,Denis &

Kruse, 2000; Wagner, Pfeffer, & O’Reilly, 1984). Turnover-

performance sensitivity is also found to be moderated by

structural and contextual factors such as the power of

incumbent CEOs (Ocasio, 1994; Shen, Gentry, & Tosi,

2010), the origin of successor CEOs (Shen & Cannella,

2002; Zhang, 2008), the composition of the board of

directors (Weisbach, 1988), the effectiveness of corporate

takeover markets (Martin & McConnell, 1991), and the

inf‌luence of external stakeholders such as f‌inancial analysts

(Puffer & Weintrop, 1991; Wiersema & Zhang, 2011).

A premise underlying the negative relationship between

CEO turnover and f‌irm performance is that CEOs should be

responsiblefor poor f‌irm performance. However, CEO perfor-

mance evaluation is a complicated process surrounded by

uncertainty. According to Holmstrom (1982), assessments of

managerial capability are often noisy and diff‌icult because

f‌irm performance is not only driven by managerial decision

but is also subject to the inf‌luence of industry and macro-

environmental conditions that are often out of managerial

control. Therefore poor f‌irm performance may result from a

signif‌icant economic downturn rather than the fault of top

management. Similarly, good f‌irm performance could be a re-

sult of luck insteadof managerial efforts and skills (Garvey &

Milbourn, 2006) . The key issue underlying this ambiguity is the

degree to which top management should be held accountable

for f‌irm performance, which we argue is inf‌luenced by the

board’s perception and evaluation of managerial performance.

An important insight from attribution theory suggests that

evaluators may attribute a behavioral outcome to either inter-

nal personal factors caused by the person or external situa-

tional factors caused by the context (Heider, 1958; Kelley,

*Address for correspondence: Lerong He, School of Business Administration &

Economics, State University of New York at Brockport, Brockport, NY, USA. Phone:

1-585-395-5781; E-mail: lhe@brockport.edu

© 2015 JohnWiley & Sons Ltd

doi:10.1111/corg.12129

24

Corporate Governance: An International Review, 2016, 24(1): 24–41

1971; Liden & Mitchell, 1985). When the board (the evaluator)

attributes poor f‌irm performance to situational factors, the

board is unlikely to hold executives accountable for f‌irm per-

formance (Green& Liden, 1980). In contrast, when executives

are assumed to be the cause of poor f‌irm performance, they

are more likely to bear personal consequences such as dis-

missal (Haleblian & Rajagopalan, 2006). We thus suggest that

aboard’s interpretation and attribution of f‌irm performance

will affect subsequent CEO dismissal decisions. Specif‌ically,

we suggest that CEO compensation relative to peers,

i.e.,CEO overpayment or underpayment, serves as a judg-

mental anchor that ref‌lects a board’s initial expectation and

subsequent interpretation and evaluation of CEO perfor-

mance. Other things being equal, a board is more likely to

hold a CEO accountable for f‌irm performance when he/she

is overpaid.

The standard principal agency model interprets the

relationship between CEO pay and turnover as a result of

an optimal contracting process (Fama & Jensen, 1983; Jensen

& Meckling, 1976). Specif‌ically, Fama (1980)’s“ex post settling

up”argument views executive pay setting and CEO dis-

missal decisionsas multi-period phenomena. Due to informa-

tion asymmetry, a corporate board initially does not have

complete information on CEO quality. As managerial talent

is revealed over time, the board will update its prior knowl-

edge on CEO quality, adjust the executive compensation con-

tract, and make dismissal or retention decisions accordingly.

Using this concept, Wowak, Hambrick, and Henderson

(2011) examine the relationship between CEO overpayment

and turnover in a sampleof US f‌irms. They f‌ind that overpaid

CEOs are associated with a larger likelihood of turnover in

the case of poor f‌irm performance. They attribute this result

to the board’s adjustment of CEO employment contracts

based on f‌irm performanceand the CEO’s relative compensa-

tion to the market rate.

The “settlingup”theory is essentially a multi-period agency

model built on the notion of rational choice. Although it does

acknowledge incomplete information on CEO quality at the

time of hiring, the theory assumes the board will ultimately

obtain full information on CEO capability, and subsequently

implement an optimal employment contract by adjusting

CEO compensation or f‌iring incompetent CEOs. An implicit

assumption of this theory is that board members are able to

correctly distinguish f‌irm outcomes attributable to CEO ac-

tions from those caused by exogenous factors out of manage-

rial controland make optimal dismissal decisions.Apparently

there are instances where actions ofthe board are inconsistent

with actual CEO performance. As suggested by the seminal

work of Simon (1945, 1955), bounded rationality will con-

strain board decision making. Board members in reality only

possess limited ability to effectively process information and

solve complex problems such as CEO performance evalua-

tion. Prior research has indicated that bounded rationality

forces board members to apply heuristic principles or

cognitive shortcuts to simplify decision-making processes

(Hayward, Rindova, & Pollock, 2004). Such heuristic princi-

ples help “reduce the complex tasks of assessing probabilities

and predicting values to simpler judgmental operations”and

are particularly inf‌luential in situations where actions under

consideration are ambiguous (Tversky & Kahneman, 1974).

Wiersema and Zhang (2011), for example, document that

board members rely on investment analysts’assessment of

CEO quality when making CEO dismissal decisions. Graff‌in,

Boivie, and Carpenter (2013) suggest that the predecessor

CEO’s founder and star status as well as succeeding CEOs’

prior CEO experience are all important heuristics applied by

the board to assess early-stage CEO quality. Following this

line of literature, we view executive compensation relative to

peers as a decision-making heuristic that helps board mem-

bers to make inferencesabout CEO quality and their contribu-

tion to f‌irm performance,which will ultimately affectthe CEO

dismissal decision. Relying on insights from attribution the-

ory, we posit that CEO overpayment is more likely to trigger

internal attribution, thus resulting in a stronger link between

poor f‌irm performance and the likelihood of CEO dismissal.

This alternative explanation is not proposed as a replacement

for agency theory,but we believe that the CEO dismissaldeci-

sion cannot be rationalized by information asymmetry and

optimal contracting alone. Behavioral research oncausal attri-

bution, bounded rationality, and decision heuristics could

enrich arguments of the principal-agent theory and offer a

more psychologically based interpretation of the mechanism

underlining board decision making. Such a pluralistic

approach as suggested by O’Reilly and Main (2010) could

help provide a more f‌ine-grained understanding of board

decision making.

Bounded rationality may also lead to selective attention

when board members only pay attention to certain issues

but ignore others(Simon, 1945). Importantly, boardmembers’

focus of attentionis shaped by contextual cues they face and is

also constrained by a f‌irm’s structure, history, and social rela-

tionship (Ocasio,1997). Drawing on Ocasio (1997)’s attention-

based view, we then explicitlyinvestigate how contextual and

structural factors may shape board members’allocation of at-

tention and causalattribution, and subsequently inf‌luence the

sensitivity of CEO dismissal to f‌irm performance. By empha-

sizing behavioral processes in the boardroom, specif‌ically

the role of causal attribution and focusof attention in guiding

board decisions,this study provides a psychological explana-

tion of the relationship between executive compensation and

CEO turnover. It supplements optimal contracting theory

or managerial power theory employed by prior literature

(e.g.,Shen et al., 2010; Wowak et al., 2011) to gain more insight

into how social dynamics may inf‌luence the functioning of

boards of directors.

1

From this perspective, our research also adds on to a

growing streamof research that has demonstrated the impor-

tance of attribution in managerial or f‌irm decision outcomes

(e.g.,Clapham & Schwenk, 1991; Forbes & Milliken, 1999;

Ford, 1985; Sun & Shin, 2014). As Sun and Shin (2014) noted,

the majority of these studies have focused on management’s

causal attribution process while ignoring that of the board of

directors. We supplement these prior works by examining

how board causal attribution may affect the CEO dismissal

decision. Furthermore, we emphasize the importance of

contextual factors in shaping board members’attention and

perception,and ultimately affectingthe sensitivity of CEO dis-

missal to f‌irm performance. In thissense, we also complement

prior studies examining the role of board attention on moni-

toring and other strategic decisions by incorporating CEO

dismissal as a new outcome variable (Cho & Hambrick,

2006; Ocasio,1999; Tuggle,Sirmon, Reutzel, & Bierman, 2010).

25CEO OVERPAYMENT AND DISMISSAL

© 2015 JohnWiley & Sons Ltd Volume 24 Number 1 January 2016