New Firm Performance and the Replacement of Founder‐CEOs

• Author: Peter Thompson,Jing Chen
• Date: 01 September 2015
• DOI: http://doi.org/10.1002/sej.1203
• Published date: 01 September 2015

NEW FIRM PERFORMANCE AND THE

REPLACEMENT OF FOUNDER-CEOs

JING CHEN1* and PETER THOMPSON2

1Department of Innovation and Organizational Economics, Copenhagen Busi-

ness School, Copenhagen, Denmark

2Scheller College of Business, Georgia Institute of Technology, Atlanta, Georgia,

U.S.A.

We study the causes and consequences of the replacement of founder-CEOs in a sample of

4,172 Danish start-ups. We propose that founder-CEO replacement is driven in part by

mismatches between business quality and founder ability. Our framework suggests that

replacements are more likely among the worst- and best-performing firms, with low (high)-

ability founders replaced by manager with higher (lower) ability. Replacement is not unam-

biguously associated with better subsequent performance. Firms that replaced the founder

were much morelikely to fail, but the surviving firms among them grew considerably faster. Our

empirical results are consistent with these proposed predictions. Copyright © 2015 Strategic

Management Society

INTRODUCTION

Why do entrepreneurs cede operating control to spe-

cialist managers? What factors influence the likeli-

hood of founder-manager succession? What

implications does the departure of the founder have

for subsequent business performance? And what

happens to founders after they have left? Answers

to these questions not only have pertinent implica-

tions for practitioners, but they also have the poten-

tial to advance academic research on entrepreneurial

motivation and its relationship with business perfor-

mance. This article examines the causes and conse-

quences of founder-CEO replacement among new

businesses that are generally small, almost invariably

not financed by venture capital, and generally not at

risk of an IPO event. We know surprisingly little

about founder turnover among such firms.1

Founder turnover events have been studied exten-

sively in other types of firms. A small literature that

has focused on high-growth and VC-backed firms 2

finds that a founder is most likely to be replaced by

a manager after a period of rapid expansion (Boeker

and Karichalil 2002) or upon attainment of critical

milestones, such as completion of product develop-

ment or securing new rounds of outside funding

(Wasserman 2003). More important, perhaps, is that

founder turnover is typically motivated by an expec-

tation among stakeholders that the firm is likely to

undergo an episode of transformational growth in the

Keywords: entrepreneurship; business creation; founder turn-

over; CEO turnover

*Correspondence to: Jing Chen, Department of Innovation and

Organizational Economics, Copenhagen Business School,

Kilevej 14, 2000 Frederiksberg, Denmark. E-mail: jc.ino@

cbs.dk.

1For convenience, we will use the term ‘founder turnover’ to

describe an event where a founder who had been managing the

firm as his/her primary occupation rescinds operating control

while maintaining ownership. Replacements will be referred to

as ‘managers.’

2Of course, executive turnover in established, typically public,

firms has been studied extensively for many years (e.g., Hofer

1980, Chen and Hambrick, 2012). There is also an active lit-

erature on succession in family-owned businesses, but the con-

cerns of this literature are rather different from those in which

executive control is passed outside the founder’s family. See

Handler (1994) for a review.

bs_bs_banner

Strategic Entrepreneurship Journal

Strat. Entrepreneurship J., 9: 243–262 (2015)

Published online in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/sej.1203

Copyright © 2015 Strategic Management Society

near future. Indeed, a portion of the literature explic-

itly samples on such expectations by studying only

firms that are preparing for IPOs (Nelson, 2003; Jain

and Tabak, 2008; Pollock, Fund, and Baker, 2009).

The extant research lacks findings that we can

confidently extrapolate to the typical small and non-

VC-backed businesses that are predominant in every

economy. First, founders of VC-financed firms

appear to be replaced at markedly higher rates than

firms that are not subject to significant control by

outside equity investors (Hellmann and Puri, 2002;

Boeker and Wiltbank, 2005). This difference, as

Hellmann and Puri (2002) have argued, highlights

the role of VC investors in professionalizing top

management teams at entrepreneurial firms. But it

also implies that the replacement decisions at

VC-financed firms are often reached based on inves-

tors’ beliefs about what is best for the businesses,

rather than on the motives of entrepreneurs. Second,

founder turnover at VC-backed firms are often

accompanied by a phenomenon termed ‘the paradox

of entrepreneurial success’ (Wasserman 2003). That

is, the more quickly entrepreneurs become success-

ful, the more likely they are to be replaced by pro-

fessional CEOs. However, the majority of small

businesses created in the economy are unlikely to

undergo transformations of the sort that appear to

motivate many founder turnover events at

VC-financed firms. Third, VC-financed firms are not

random draws from the firm population—they are

more likely to grow substantially, they are better

financed, and their founders and employees, on

average, have more human capital. Correlations

between variables estimated from a sample drawn

from the upper tail of the quality distribution may

not hold in other segments of the population, espe-

cially when there are possibilities for nonmonotonic

relationships in the population as a whole. Lastly, it

is well known that VC-financed firms account for a

tiny fraction of business launches,3but they play

a much larger role in economic growth, employ-

ment creation, and innovation. As a result, academic

effort to understand the dynamics of control in

VC-financed firms is appropriately disproportionate

to their numbers. Nonetheless, given their numerical

importance, we should also be concerned with the

control dynamics of the typical small business. Pat-

terns of founder turnover among representative busi-

nesses can also provide insights into the forces

governing business transfers.4They are also key to

understanding the expected opportunity costs of

business creation, as well as the dual phenomenas of

serial and portfolio entrepreneurs.

Although the primary contribution of this article is

intended to be empirical, we need a theoretical

framework to guide our empirical analysis and facili-

tate interpretation of the results. Among VC-backed

firms, the goal of founder turnover is generally held

to be resolving mismatches between founders’

managerial skills and the growth expectation of firms

in the near future. We work within this framework

but, because the firms in our sample do not typically

undergo transformational growth, we focus on

current rather than anticipated mismatches. We

suppose that firm performance is a function of the

ability of the founder and the quality of the business

idea. While a founder may be aware of his/her

ability, the quality of the business is not fully

revealed until after it is founded. Ability and quality

are complements in production so that there can

arise mismatches of sufficient magnitude to justify

replacing the founder with a manager who is better

suited to the needs of the business. This framework

yields two types of founder turnover: those in which

a high ability founder hires a less able manager with

a lower opportunity cost of time, and those in which

a low ability founder hires a more capable, but more

expensive, manager. Under conditions we explore in

the next section, turnover may be more likely in the

tails of both the founder ability and firm perfor-

mance distributions. Moreover, turnoverin the lower

tail of the firm performance distribution tends to be

associated with subsequent improvements in firm

performance among surviving firms, while the rela-

tionship is a priori ambiguous in the upper tail.

For our empirical analysis, we construct a sample

of 4,172 Danish businesses created in 1999 and

2000. Compared to previous studies that are con-

fined to small samples within high-tech sectors,5our

sample has a broad coverage of industries. Only 18

percent of businesses included are classified as

3For example, in the United States, 626,771 new businesses

were created in 2008, only 3,276 of which received VC financ-

ing (Bureau of Labor Statistics, 2014; NVCA 2009).

4Founders may sell businesses because they are poorly

matched to the needs of the businesses or because they have

reasons to cash out their equity positions. Looking at founder

turnover in businesses that are not transferred allows us to gain

insights about the first of these two motivations.

5For example, the sample constructed by Wasserman (2003)

consists of 202 Internet firms. Pollock et al. (2009) also study

340 Internet start-ups financed by VCs. Hellmann and Puri

(2002) focus on a sample of 173 high-tech start-ups located in

Silicon Valley.

244 J. Chen and P. Thompson

Copyright © 2015 Strategic Management Society Strat. Entrepreneurship J.,9: 243–262 (2015)

DOI: 10.1002/sej