Prior ties and the limits of peer effects on startup team performance

• Author: Rembrand Koning,Sharique Hasan
• Date: 01 September 2019
• DOI: http://doi.org/10.1002/smj.3032
• Published date: 01 September 2019

RESEARCH ARTICLE

Prior ties and the limits of peer effects on startup

team performance

Sharique Hasan

1

| Rembrand Koning

2

1

Strategy Department, Duke University

Fuqua School of Business, Durham, North

Carolina

2

Strategy Unit, Harvard Business School,

Boston, Massachusetts

Correspondence

Rembrand Koning, Harvard Business

School, Boston, MA.

Email: rem@hbs.edu

Funding information

Ewing Marion Kauffman Foundation;

Stanford Graduate School of Business;

Stanford SEED Center; Indhraprasta

Institute of Information Technology; The

Indian Software Product Industry

Roundtable (iSPIRT)

Research Summary:We conduct a field experiment at an

entrepreneurship bootcamp to investigate whether interac-

tion with proximate peers shapes a nascent startup team's

performance. We find that teams whose members lack

prior ties to others at the bootcamp experience peer effects

that influence the quality of their product prototypes. A

1-SD increase in the performance of proximate teams is

related to a two-thirds SD improvement for a focal team.

In contrast, we find that teams whose members have many

prior ties interact less frequently with proximate peers, and

thus their performance is unaffected by nearby teams. Our

findings highlight how prior social connections, which are

often a source of knowledge and influence, can limit new

interactions and thus the ability of organizations to lever-

age peer effects to improve the performance of their

members.

Managerial Summary:Researchers and policymakers

believe that accelerators, incubators, and bootcamps help

entrepreneurial ecosystems spur innovation and drive

startup growth. The effectiveness of these organizations,

in large part, depends on the new social interactions fos-

tered among colocated entrepreneurs. Yet, little evidence

exists about the extent to which such interactions actually

lead to spillovers. We ran a controlled experiment at a

startup bootcamp to investigate when entrepreneurs were

most affected by their colocated peers. Not everyone

benefited. We found that entrepreneurs with many prior

ties to others at the bootcamp made fewer new connec-

tions, especially to neighboring peers, and thus did not

Received: 25 September 2017 Revised: 11 January 2019 Accepted: 31 January 2019 Published on: 23 April 2019

DOI: 10.1002/smj.3032

1394 © 2019 John Wiley & Sons, Ltd. wileyonlinelibrary.com/journal/smj Strat Mgmt J. 2019;40:1394–1416.

experience significant spillovers. In contrast, those without

prior connections experienced the greatest spillovers

because they interacted frequently with people on nearby

colocated teams. Our findings highlight how organizations

like incubators and bootcamps, designed to foster new

connections, might sometimes just reinforce old networks.

KEYWORDS

accelerators and incubators, entrepreneurship, entrepreneurship

training, field experiment, peer effects

1|INTRODUCTION

Regions from New Delhi, India, to Durham, North Carolina, are attempting to build their own Sili-

con Valleys—ecosystems propelled by nascent startups developing innovative products

(Bresnahan & Gambardella, 2004; Kenney, 2000). Researchers and policymakers have long attrib-

uted the success of such ecosystems to the informal networks between entrepreneurs (Saxenian,

1990; Sorenson, 2005) and more recently to the peer learning facilitated by institutions like accelera-

tors and incubators (Hochberg, 2016). Indeed, regions hope that these networks and organizations

will spark powerful peer effects that will improve the quality of their startups (Hochberg &

Fehder, 2015).

The idea that peer effects drive performance is not new. Such effects have been demonstrated in

diverse contexts including retail (Mas & Moretti, 2009; Chan, Li, & Pierce, 2014a), finance (Hwang,

Liberti, & Sturgess, 2018), and scientific innovation (Azoulay, Graff-Zivin, & Wang, 2010; Catalini,

2017; Oettl, 2012). Peers share knowledge, provide help, and set performance expectations

(e.g., Herbst & Mas, 2015; Housman & Minor, 2016; Mas & Moretti, 2009). Similarly, much of the

value of startup incubators and accelerators is derived from the new social interactions they engineer

between nascent entrepreneurs (Chatterji, Delecourt, Hasan, & Koning, 2018; Hallen, Bingham, &

Cohen, 2014; Hasan & Koning, 2018). However, when researchers have tried to engineer peer effects

in other contexts, they have often failed (Carrell, Sacerdote, & West, 2013). Individuals appear to

exercise considerable discretion in choosing whom they interact with, preferring to form ties to peers

with similar ability (Carrell et al., 2013), ethnicity (Hjort, 2014), or class (Kato & Shu, 2016). This

self-selection of connections after peer assignment appears to limit the ability of organizations to

design effective social interventions and therefore encourage peer learning.

Research has also suggested that a critical constraint on the formation of new connections is the

prior ties of individuals (Gargiulo & Benassi, 2000; Granovetter, 1973; Kim, Oh, & Swaminathan,

2006). Within and outside organizations, individuals have ties to existing collaborators, advisers, and

friends (Krackhardt & Hanson, 1993; Lincoln & Miller, 1979). While extensive research shows that

these ties are rich conduits of information and support (e.g., Podolny & Baron, 1997; Sparrowe,

Liden, Wayne, & Kraimer, 2001), other work suggests that prior ties may reduce how much attention

someone devotes to cultivating new interactions with peers—even those who are easily accessible

(Kim et al., 2006). Prior connections also entail social commitments and obligations of time and

attention (Granovetter, 1973). Given this latter view, we might expect the prior networks and ties of

individuals to limit an organization's ability to engineer peer effects.

HASAN AND KONING 1395