Selection and Serial Entrepreneurs

• Published date: 01 June 2013
• Author: Jing Chen
• DOI: http://doi.org/10.1111/jems.12016
• Date: 01 June 2013

Selection and Serial Entrepreneurs

JING CHEN

Department of Innovation and Organizational Economics

Copenhagen Business School

2000 Frederiksberg, Denmark

jc.ino@cbs.dk

There is substantial evidence that serial entrepreneurs outperform de novo entrepreneurs. But is

this positive association between prior experience and performance the result of learning by doing

or of selection on ability? This paper proposes a strategy that combines the fixed-effects model and

IV estimations to distinguish empirically selection effects from learning. Using panel data from

the NLSY79, I find that selection on ability is the more important determinant of serial business

formation and the early performance of new businesses. In contrast, the effects of learning by

doing are apparent only when the analysis focuses on founding new startups in sectors closely

related to entrepreneurs’ previous ventures.

1. Introduction

What motivates some individuals to become serial entrepreneurs, and how does this

influence their performance? Is serial entrepreneurship driven by taste or personal ac-

cidents of history, or are there more systematic predictors? Do agents become serial

entrepreneurs because their prior business failed, or because their prior business was a

success? Despite an emerging literature on serial entrepreneurship we continue to have

little confidence in our answers to these questions.

Early work, largely based on small-sample interviews with serial entrepreneurs,

suggest they are motivated by a variety of factors. Wright et al. (1997), for example,

identify half a dozen disparate motivations. Prominent among them, some start a new

venture because they want to explore new business opportunities, while others are

attempting to rebuild a failed business. Williams (2000) also identified disparate moti-

vations. While some interviewees were motivated by a desire to seize timely opportu-

nities, Williams also documents instances in which serial entrepreneurs enjoy starting

businesses because they can relate the products to their personal experiences.1

More recent work has exploited larger samples, focused on more objectively mea-

surable attributes of serial entrepreneurs, and provided performance comparisons with

first-time entrepreneurs (cf. Eesley and Roberts, 2006a;Stametal.,2006; Gompers et al.,

2006). These studies have yielded the following evidence:

rSerial entrepreneurs are more likely to have successfully sold their prior ventures

before starting a new business (Eesley and Roberts, 2006b; Stam et al., 2006).

I am grateful to Peter Thompson for advice and exceptional encouragement. I also thank Thomas Hell-

mann, Chuck Eesley, Serguey Braguinsky, Denis Gregoire, John Walsh, and participants at the 1st NBER

Entrepreneurship Research Boot Camp for helpful discussions on this research idea. Thoughtful comments

from two anonymous referees are truly appreciated.

1. Jeff Jacober, for example, started his first business when he was still in college, selling sportswear to

fraternities and sororities. After his brother was diagnosed with chronic kidney failure, Jacober found Ocean

Diagnostic Inc. which produced a home health test that helped people to self-diagnose various diseases.

C2013 Wiley Periodicals, Inc.

Journal of Economics & Management Strategy, Volume22, Number 2, Summer 2013, 281–311

282 Journal of Economics & Management Strategy

rEntrepreneurs whose business performs poorly are less likely to create a new business

(Stam et al., 2006).

rThe current firm has a higher chance of going public if the entrepreneur’s previous

venture was acquired (Eesley and Roberts, 2006a).

rEntrepreneurswho succeeded in prior business have a much higher chance of succeed-

ing in the current business, compared to first-time entrepreneurs and entrepreneurs

who previously failed (Gompers et al., 2006).

rAdditional evidence has also been provided by other studies that generally agree

on a positive relationship between previous entrepreneurial experience and current

business performance (Schilling et al., 2003; Delmar and Shane, 2006; Stuart and Abetti,

1990; Gompers et al., 2006, etc).

These findings are consistent with at least two distinct interpretations. Eesley and

Roberts (2006a), and Stam et al. (2006) interpret their results as evidence that learning

from prior founding experience has a positive impact on serial entrepreneurial per-

formance. Parker (2010) also agrees that knowledge acquired from previous business

experience helps entrepreneurs enhance their performance on subsequent ventures, al-

though the effect may be temporary. In contrast, Gompers et al. (2006,2010) emphasize

the importance of innate attributes/skills in the success of entrepreneurs’ next ventures.

Their results suggest that, instead of learning by doing, selection on innate ability of

entrepreneurs determines serial business formation and performance persistence, al-

though their findings are more likely to reflect venture capital firms learning about the

quality of entrepreneurs, rather than entrepreneurs learning about their own ability.

It has always been a challenge to empirically distinguish learning by doing from

learning about ability. Thompson (2010) develops a simple framework incorporating

these two types of learning, and highlights the difficulty in separating them in empirical

analysis. To my knowledge, only two studies have attempted to do so in the area of

occupational choice: one is by Farber (1994), who focuses on differences in the hazards

of job separation implied by the two models; the other is by Nagyp´

al (2007), who looks

at firm-specific price shocks and their distinctive impacts on employee turnover in the

two learning models [See Thompson (2010) for a more detailed discussion].

Identification is especially difficult in the study of serial entrepreneurship, as the

variable that often attracts our greatest interest is entrepreneurial experience. This vari-

able, usually proxied by business performance (e.g., Gompers et al., 2006) or founding

experience (e.g., Eesley and Roberts, 2006a), could easily capture both learning by doing

and learning about entrepreneurial ability. To decide which learning model accounts

more for the creation of serial entrepreneurs, this paper takes alternative approaches of

fixed effects and IV estimations to empirically separate one type of learning from the

other when entrepreneurial experience is revealed.

Compared to an emerging literature that emphasizes the effect of learning by

doing on entrepreneurial activities,2there has been little empirical work that specifically

examines the role of innate ability in serial entrepreneurship. However, the notion that

selection on ability is an important determinant of serial entrepreneurship builds on

reasonable theoreticalgrounds. According to Holmes and Schmitz’ (1990) seminal theory

of business transfer, selection on ability alone may serve as the mechanism that induces

business turnover. In their theory, people differ in their ability to respond to business

2. These studies focus on issues such as how serial entrepreneurs learn fromtheir successful or unsuccess-

ful prior experiences (Minniti and Bygrave 2001), or how experiential learning positively or negatively affects

their entrepreneurial performance (Politis 2005, Corbett 2005,P

´

astor, Taylor, and Veronesi 2009).