Director overconfidence

• Author: Shawn Mobbs,Randy Beavers
• Date: 01 June 2020
• DOI: http://doi.org/10.1111/fima.12268
• Published date: 01 June 2020

DOI: 10.1111/fima.12268

ORIGINAL ARTICLE

Director overconfidence

Randy Beavers1Shawn Mobbs2

1School of Business, Government,and

Economics, Seattle PacificUniversity, Seattle,

WA,USA

2Culverhouse College of Business, The

University of Alabama, Tuscaloosa,AL, USA

Correspondence

RandyBeavers, 3307 Third Avenue West, Suite

201,Seattle, WA 98119-1950.

Email:reb@spu.edu

Abstract

We examine overconfident chief executive officer (CEO) directors

and find they attend more board meetings, are more likely to serve

on the nominating or the compensation committee, have more inde-

pendent directorships, and foster higher attendance rateson boards.

Boards with overconfidentdirectors are more likely to appoint a bet-

ter prepared and more reputable CEO following a turnover. These

newly appointed CEOs are also more likely to be overconfident.This

evidence indicates overconfident CEO directors exhibit significant

influence on the board and over the firm's CEO selection.

1INTRODUCTION

The existing literatureregarding independent chief executive officer (CEO) directors primarily focuses on their differ-

encesfrom other directors, but variation among CEO directorshas not been explored (Fahlenbrach, Low, & Stulz, 2010;

Fahlenbrach, Minton, & Pan, 2011; Fracassi& Tate, 2012; Hallock, 1997). This is important, as there can be significant

differences among CEO directors. Recent research revealsdifferences in CEO overconfidence are associated with sig-

nificant differences in CEO decision making.1,2 Because directors are appointed for their decision management skills

(Fama, 1980; Fama & Jensen, 1983), differences in overconfidence among CEOs can lead to important differences in

CEO directors. Yet, no empirical study has analyzed overconfident CEO directors and their actions on the boards of

other firms.

In this study, we seek to understand how the actions of CEO directors differ based on their overconfidence and

whether their presence influences the board's CEO selection decisions. We find evidence that overconfident CEO

directors are more engaging board members relativeto other CEO directors and their presence fosters higher meeting

attendance rates among the other directors on the same board. In addition, boards with overconfident CEO directors

are more likely to appoint a better prepared and more reputable CEO following a turnover.These newly appointed

CEOs are also more likely to be overconfident. These findings represent important new insights into board decision

making by revealing the active and influential role played by overconfident CEO directors on corporateboards. For

c

2019 Financial Management Association International

1Wefollow Malmendier and Tate(2005) in defining confidence versus optimism. Confidence refers to one's own abilities and personal outcomes. Optimism

refersto outcomes beyond one's control. Overestimation of these outcomes leads to overconfidence and overoptimism, respectively.

2Forexample, differences in CEO overconfidence affect decisions related to investment (Malmendier & Tate,2005), equity issuance (Malmendier, Tate,& Yan,

2011),acquisition (Malmendier & Tate, 2008), innovation (Hirshleifer,Low, & Teoh, 2012), investmentpolicies (Banerjee et al., 2014a, 2014b), and accounting

practices(Burg, Pierk, & Scheinert, 2013).

Financial Management. 2020;49:389–421. wileyonlinelibrary.com/journal/fima 389

390 BEAVERSAND MOBBS

example,their influence on CEO selections reveals one channel by which overconfident executives may havean advan-

tage in CEO tournaments (Banerjee, Humphery-Jenner,& Nanda, 2014b).

To separate CEO directors by overconfidence, we begin by identifying all of the independent directors in the

Risk Metrics director database from 1996 to 2011 and find approximately 16% are CEO directors. Then, we

identify those CEO directors who are overconfident following Malmendier and Tate (2005) and define a board

as being overconfident if at least one independent director is an overconfident CEO director. About 12% of

the independent directors are overconfident CEO directors and about 33% of the firms have an overconfident

board.

We begin our analysis by examining director levelactions. Specifically, we examine board meeting attendance and

committee membership as measures of a director's levelof effort and activity in a given directorship (Adams & Ferreira,

2009; Masulis & Mobbs, 2014). We find that overconfident CEO directors miss fewer meetings than other directors

and exhibit a significantly greater likelihood of serving on the nominating and compensation committees relative to

other CEO directors. These differences are also economically significant. For example, overconfident CEO directors

exhibit an almost 40% lower absentee rate when compared to other CEO directors. These results are consistent with

vast research in the field of psychology that suggests overconfident individuals behavedifferently. For instance, over-

confidence is associated with an individual's belief that they are better than the average person (Svenson, 1981) and

have a greater sense of and need for being in control (Weinstein, 1980). Relatedly,Zarnoth and Sniezek (1997) find

individuals with greater confidence are more active decision makers in a group. Our findings that overconfident CEO

directors are more involved on the boards where they serve and that they likely havea more influential role on the

board are consistent with these prior findings.

We also find that meeting attendance is higher among all directors on the board when an overconfidentCEO direc-

tor is present and that overconfident CEO directors serve on significantly more boards than do other CEO directors,

which is consistent with the greater demand for their services. The result that overconfidence can facilitate better

board functionality further echoes prior findings in the psychologyliterature. According to Busenitz and Barney (1997,

p. 10), “Overconfidence maybe p articularlybeneficial i n implementing a specific decision and persuading others to be

enthusiastic about it as well.” Further, anynegative aspects associated with overconfidence are mitigated by serving

with other directors in a group setting (Moore, 1977; Sniezek, 1992).

Next, we examine an important board-level decision, the selection of a CEO, which is arguably the most impor-

tant and involved decision boards must make (Adams, Hermalin, & Weisbach, 2010; Dahya, McConnell, & Travlos,

2002; Denis, Denis, & Sarin, 1997; Denis, Denis, & Walker,2015; Fee & Hadlock, 2003; Hermalin & Weisbach, 2003;

Huson, Malatesta, & Parrino, 2004; Huson, Parrino, & Starks, 2001; Mace, 1971; Mobbs, 2013; Mobbs & Raheja,

2012; Parrino, 1997; Shleifer & Vishny, 1997; Vancil, 1987; Warner, Watts, & Wruck, 1988; Weisbach, 1988;

Yermack,1996). Variation in overconfidence among directors can have important implications for this decision. More-

over,because overconfident individuals tend to make the decision for a group, overconfident directors will have a sig-

nificant influence on the board, which makes understanding their decision process even more important. We know

from the psychology literature that: 1) overconfident individuals have more certainty about their decisions and, as

such, expect their actions to produce success (Miller & Ross, 1975) and 2) overconfident managers quickly implement

decisions, despite their reluctance to incorporate or gather new information (Goel & Hakor, 2008). We find over-

confident boards select CEOs who are more likely to be an insider, have prior experience as a CEO, and who hold

more directorships relative to CEOs chosen by nonoverconfident boards. Each of these characteristics is associated

with more existing information and less uncertainty about the new CEO's ability and also with overconfident CEO

directors being more inclined to select a CEO who has an established reputation. Interestingly, we also find over-

confident boards are more likely to select an overconfident CEO. In fact, in our sample of new CEO appointments,

an overconfident board is 5.7% to 9.5% more likely to appoint an overconfident CEO relative to other similar, but

nonoverconfident boards. This suggests that overconfidentdirectors are likely to select overconfident candidates like

themselves.

Since firms may appoint overconfident CEO directors because they need to hire an experienced or overconfident

CEO, there is an endogenous selection concern. We address this concern in two ways. First, we utilize the 2002

BEAVERSAND MOBBS 391

exchange listing requirement changes that forced many firms to appoint additional independent directors. In doing

so, some firms appointed overconfident CEO directors to their board that otherwise would not havedone so. Utilizing

this regulatory shock, we examine the heterogeneous effects of this shock to the composition of overconfident CEO

directors on the board. Using this analysis, we continue to find support for overconfident boards selecting a CEO who

has prior experience and who is also overconfident.

In our second approach, we rely on overconfidence as being a trait that maynot be known by the board at the time

of the director's nomination. In fact, Malmendier and Tate(2005, p. 2664) state overconfidence is “harder to identify

ex-ante.” Therefore, we focus only on the subset of overconfident CEO directors who do not reveal themselves as

overconfident in their CEO role until after theyare appointed as a CEO director. Using this subset, we continue to find

support for overconfident boards selecting a CEO who has prior experience,is a firm insider, and who is overconfident.

These findings mitigate concerns of endogenous selection as a reason for our results and provide suggestive evidence

of the CEO hiring preferences held by overconfident CEO directors.

In summary,our findings suggest that director overconfidence is an important director trait that is associated with

significantly different director activity and board decisions. This represents an important extension to the corporate

finance literature. Beginning with Bertrand and Schoar (2003) and, more recently, Schoar and Zuo (2017), research

has shown that individual CEO traits or styles are important factors in firm decision making. Relatedly,the literature

on corporate directors has begun to examineindividual director characteristics beyond their classifications as insiders

(Masulis & Mobbs, 2011; Mobbs, 2013) or outsiders (Defond, Hann, & Hu, 2005; Fich & Shivdasani, 2006; Hwang &

Kim, 2009). Our findings reveal that the trait of overconfidence separatesdirectors based on their personal decision-

making style. This is important because two equally independent directors may makevery different decisions based on

their differences in overconfidence.

By studying overconfident CEO directors, we expandthe literature on the role of overconfidence. Starting with the

groundbreaking work of Malmendier and Tate(2005, 2008), many studies have explored overconfidenceand its effect

on CEO behavior with the firm (Gervais, Heaton, & Odean, 2011; Hirshleifer,Low, & Teoh, 2012; Malmendier,Tate, &

Yan, 2011). However, no empirical study has analyzed overconfident CEO directors and their actions on the boards

of other firms. One challenge with existing studies of executive overconfidence is that anyactions in their firm that

serve to identify executivesas overconfident could also be associated with observed outcomes in that firm. For exam-

ple, Malmendier and Tate(2005) put great effort into ruling out the possibility that executive information advantage,

rather than overconfidence, is driving their findings. In our study,this is less of a concern. By studying the role of over-

confidence in directors, when overconfidence is identified by actions in a firm other than the one where they serve

as a director, we diminish anyconcerns that director actions attributed to overconfidence are driven by any informa-

tional advantage in the executive'shome firm. Thus, our evidence that overconfident CEO directors are important and

influential board members is an important extension of the literatureon overconfidence.

We also contribute to the literature on CEO directors. Previous studies focus on how CEO directors are different

from non-CEO directors (Hallock, 1997; Fahlenbrach et al., 2011; Fracassi& Tate, 2012). Our study is the first to high-

light important differences among CEO directors based on whether or not they are overconfident.

Finally, these findings contribute to the large psychology literature on overconfidence.3Our director levelanal-

ysis of overconfident corporate directors contributes to the literature regarding how overconfidence affects indi-

vidual actions (Alicke, 1985; Alicke, Klotz, Breitenbecher,Yurak, & Vredenburg, 1995). Our board level analysis con-

tributes to the literature examining how overconfidence affects individuals in a group setting (Russo & Schoemaker,

1992).

The remainder of the paper is organized as follows. In Section 2, we describe the sample data and methodology

employed in the analysis. Section 3 presents and discusses the results of our main empirical tests, while Section 4 pro-

vides our conclusions.

3Thepsychology literature measures overconfidence as extreme outliers from results of surveys about ratings, questions, and decisions in experimental set-

tings.In the finance literature, we are able to identify overconfidence based on actual decisions about stock and option purchases and exercises.