Bridging the Gap: Evidence from Externally Hired CEOs

• Author: SARAH L. C. ZECHMAN,JONATHAN L. ROGERS,CALEB RAWSON,YONCA ERTIMUR
• DOI: http://doi.org/10.1111/1475-679X.12200
• Date: 01 May 2018
• Published date: 01 May 2018

DOI: 10.1111/1475-679X.12200

Journal of Accounting Research

Vol. 56 No. 2 May 2018

Printed in U.S.A.

Bridging the Gap: Evidence from

Externally Hired CEOs

YONCA ERTIMUR,

∗CALEB RAWSON,

∗

JONATHAN L. ROGERS,

∗AND SARAH L. C. ZECHMAN ∗

Received 1 November 2015; accepted 20 November 2017

ABSTRACT

We investigate executive employment gaps (hereafter, gaps) between the ap-

pointment of an external CEO at a public firm and the individual’s prior

executive position at a public company. These gaps cannot be reliably ob-

tained from common databases. We hand-collect data for externally hired

CEOs at public companies from 1992 to 2014. These CEOs represent approx-

imately 40% of the 5,095 CEO successions and have a mean gap of 1.9 years.

The gap increases to 3.2 years for the subset of new hires with a gap. We hy-

pothesize that labor market frictions and executive skill sets contribute to the

existence and length of these gaps. Using theories from labor economics, we

predict (equilibrium) associations between two measures of “fit” (executive

∗University of Colorado, Boulder.

Accepted by Christian Leuz. This paper is the final Registered Report resulting from

the Registration-Based Editorial Process (REP) implemented by JAR for its 2017 conference;

details of the process are available here: https://research.chicagobooth.edu/arc/journal-of-

accounting-research/2017-registered-reports. The accepted proposal and an Online Ap-

pendix for this report are available here: https://research.chicagobooth.edu/arc/journal-of-

accounting-research/online-supplements. We thank an anonymous referee and workshop

participants at the University of Oregon, the Leeds School of Business Summer Brownbag

Series, and the 2017 Journal of Accounting Research conference for their comments and sug-

gestions. We also thank Leah Baer, Paige Patrick, Andrea Pawliczek, and Nikki Skinner for

programming assistance, and Michael Arney, Nicole Buggy,Kevin Choe, Brett Eliasen, Andre

Kaiser, Ramona Kimmett, Colin Mierau, Madelyn Mohr, Addison Overholt, Pinran Pan, and

Shazad Sahak for data collection. Yonca Ertimur gratefully acknowledges financial support

provided by the KPMG Accounting Faculty Fellowship; Caleb Rawson gratefully acknowledges

financial support provided by the Gerald Hart Research Fellowship; and Sarah Zechman grate-

fully acknowledges financial support provided by the EKS&H Faculty Fellowship.

521

Copyright C, University of Chicago on behalf of the Accounting Research Center,2018

522 Y.ERTIMUR,C.RAWSON,J.L.ROGERS,AND S.L.C.ZECHMAN

compensation and long-term match quality) and gaps (both existence and

length). Finally, we provide descriptive evidence on what executives do (e.g.,

sit on boards, work for private consulting companies, or consume leisure)

during their gaps. This project was subject to and published through a reg-

istered report process. Any tests that were not included in the accepted pro-

posal are marked as unplanned analyses.

JEL codes: J64; M12; M40; M41

Keywords: corporate governance; CEO turnover; CEO compensation; ex-

ecutive labor market; employment gaps

1. Introduction

We investigate executive employment gaps (hereafter, gaps) between the

appointment of an external CEO at a public firm and the individual’s

prior position at a public company. Specifically, by augmenting several com-

mon databases with extensive hand-collected data on 2,036 externally hired

CEOs, we examine the impact of labor market frictions and executive skill

sets on the existence and length of gaps. We also examine whether the ex-

istence and length of gaps are associated with executive compensation and

long-term match quality (two measures of executive–firm “fit”).

The proportion of external CEO hires has been steadily increasing over

the past decades (Vancil [1987], Parrino [1997], Huson, Parrino, and

Starks [2001], Frydman [2007], Murphy and Z´

abojn´

ık [2007]). This trend,

combined with labor market frictions such as noncompete constraints

(NCCs; Garmaise [2011]), suggests that transitions into CEO positions are

frequently preceded by an employment gap.1Supporting this conjecture,

we find employment gaps to be prevalent among external CEO hires in

our sample; 57.5% of the CEOs experience a gap of at least 30 days and

the average gap lasts 1,174 days (3.2 years).2This seems economically large

considering average CEO tenure between 1992 and 2007 is only seven years

(Kaplan and Minton [2012]). To the extent that executives do not find

supplemental income sources during the gap, a gap of 3.2 years represents

$18.7 million foregone compensation for an average CEO at an S&P 1500

firm over the 1992–2014 period (calculation based on total compensation,

TDC1, in Execucomp).

1Descriptive statistics in Fee and Hadlock [2003] suggest that some of their “nonraided”

sample has gaps. In table 3, they split their sample between 101 “raid” observations, defined as

“cases where the individual jumped immediately from a prior public employer to the new posi-

tion” (p. 1333), and 31 “nonraid” observations. Anecdotal evidence (Lublin [2010a], Feintzeig

[2014]) also points to the salience of executive employment gaps.

2We recognize that minor timing differences (due to physical moves, vacations, etc.) can

arise even when a new CEO accepts the position before leaving another firm. Unfortunately,

offer acceptance is not necessarily observable. As a result, we classify employment gaps of less

than 30 days as “no gap” or zero. Given the arbitrary nature of this cutoff, we report results in

online appendix B using a 60- and a 90-day cutoff.

BRIDGING THE GAP:EVIDENCE FROM EXTERNALLY HIRED CEOS 523

Understanding executive employment gaps is important for several rea-

sons. First, the prevalence and length of gaps suggest significant frictions

in the CEO labor market. Analysis of gaps, therefore, can shed light on

the existence and nature of these frictions. Second, gaps can affect exec-

utives’ skills as well as the quality of the CEO–firm match. For example,

sitting out of the job market can result in lost wages and deterioration of

executives’ skills. At the same time, by allowing executives to recharge and

invest in the search process, gaps can result in better CEO–firm matches.

Third, gaps can have broader implications for firms, given CEOs’ role as

key decision makers in firms (Bertrand [2009]) and evidence that individ-

ual CEOs affect firm outcomes (e.g., Hambrick and Mason [1984], Johnson

et al. [1985], Hayes and Schaefer [1999], Bertrand and Schoar [2003]).

To shed light on the role of labor market frictions, we consider the ef-

fect of institutional arrangements (specifically, NCCs) as well as executive

skill sets on the existence and length of gaps.3Noncompete agreements

can mitigate potential damages caused by departing employees (Garmaise

[2011], Bishara, Martin, and Thomas [2015]), but are not always enforce-

able (e.g., in California), so simply signing one does not necessarily con-

strain behavior. Thus, we view an executive to be constrained when he/she

signs an agreement and that agreement’s enforceability is relatively high.

We use the term NCC to describe noncompete agreements (noncompetes)

combined with a high level of enforceability.

As for executive skill sets, the increasing importance of general manage-

rial skills and the decrease in the importance of firm- or industry-specific

expertise is arguably the most salient trend related to CEOs’ educational

and professional backgrounds over recent decades (Frydman [2007], Mur-

phy and Z´

abojn´

ık [2007], Bertand [2009], Cust´

odio, Ferreira, and Matos

[2013]). Therefore, we begin by examining generalist versus specialist ex-

ecutives and the differential role that NCCs likely play for these two execu-

tive types with respect to gaps. We hypothesize that NCCs are less binding

for generalists. Specifically, we predict that NCCs affect the existence and

length of specialists’ gaps more than those of generalists.

Next, we turn our attention to the relation between gaps and match qual-

ity. Gaps exist for a variety of reasons (e.g., illness, poor performance). Po-

tential employers likely view some of these reasons as innocuous and others

as worrisome. Conventional wisdom and the signaling models in the rank-

and-file literature suggest gaps are red flags to prospective employers re-

sulting in fewer interviews, fewer offers, and lower pay (e.g., Kroft, Lange,

and Notowidigdo [2013]). Our setting differs from the rank-and-file setting

3Labor economics literature examines “contemporaneous unemployment spells” among

rank-and-file employees, which is analogous to our notion of gaps (section 2 briefly discusses

this literature). Yet, the insights from this research may not apply to executive gaps. For exam-

ple, with respect to determinants and with the aim to inform the policy debate on unemploy-

ment levels, labor economists typically focus on the effects of unemployment insurance on the

duration of unemployment spells, an issue that is not relevant to the executive labor market.