The Effects of Academic Incubators on University Innovation

• DOI: http://doi.org/10.1002/sej.1242
• Published date: 01 June 2017
• Author: Christos Kolympiris,Peter G. Klein
• Date: 01 June 2017

THE EFFECTS OF ACADEMIC INCUBATORS ON

UNIVERSITY INNOVATION

CHRISTOS KOLYMPIRIS

1

*and PETER G. KLEIN

2

1

School of Management, University of Bath, Bath, U.K.

2

Hankamer School of Business, Baylor University, Waco, Texas, U.S.A.

Research summary: In this article, we analyze the impact of academic incubators on

the quality of innovations produced by U.S. research-intensive academic institutions.

We show that establishing a university-affiliated incubator is followed by a reduction

in the quality of university innovations. The conclusion holds when we control for the

endogeneity of the decision to establish an incubator using the presence of incubators

at peer institutions as an instrument. We also document a reduction in licensing

income following the establishment of an incubator. The results suggest that university

incubators compete for resources with technology transfer offices and other campus

programs and activities, such that the useful outputs they generate can be partially off-

set by reductions in innovation elsewhere.

Managerial summary: Do university incubators drain resources from other university

efforts to generate innovations with commercial relevance? Our analysis suggests that

they do: after research-intensive U.S. universities establish incubators, the quality of

university innovations, which we measure with patents, drops. This finding has immedi-

ate implications for practice, as it suggests that the benefits and costs of incubation

should not be analyzed in isolation. Rather, the effects of incubators extend to the over-

all innovation performance of the university. It follows that measuring the net eco-

nomic effect of incubators is challenging because besides the effects on innovation

efforts, the presence of an incubator may attract particular kinds of faculty and stu-

dents, enhance the prestige of the university, generate economic multiplier effects, and

benefit the community as a whole.Copyright © 2016 Strategic Management Society.

INTRODUCTION

Universities are increasingly tasked with fostering

entrepreneurship and innovation, encouraged to gener-

ate revenues from research produced on campus, and

contribute to (local) economic growth (Etzkowitz,

1998, 2002; Etzkowitz et al., 2000; Goldstein and

Renault, 2004). This view of the entrepreneurial uni-

versity reflects two recent trends. First, universities are

increasingly patenting research with commercial

potential and subsequently seeking to increase their

licensing revenues (Bulut and Moschini, 2009; Hen-

derson, Jaffe, and Trajtenberg, 1998). At the same

time, universities are creating incubator facilities to

assist faculty members, university graduates, commu-

nity members, or other parties to start new firms that

not only contribute to local economic growth, but also

generate income for the university, which often holds

equity positions in the incubator’stenantfirms.

Establishing university incubators and increasing

university patenting reflect similar underlying pres-

sures: both are driven by reductions in public fund-

ing for academia and increasing pressures for

public accountability. Moreover, the resources and

capabilities used to support start-ups and to

Keywords: incubators; patents; innovation; forward citations;

licensing

*Correspondence to: Christos Kolympiris, School of Manage-

ment, University of Bath, Room 3.33, Building 8 West,

Quarry Rd, Bath BA2 7AY, U.K. E-mail:

c.kolympiris@bath.ac.uk.

Copyright © 2016 Strategic Management Society

Strategic Entrepreneurship Journal

Strategic Entrepreneurship Journal, 11: 145–170 (2017)

Published online 30 January 2017 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/sej.1242

generate patented inventions are largely shared;

maintaining these two activities simultaneously

involves leveraging the same academic knowledge

and talent, devoting dedicated personnel for patent-

issuing procedures and auxiliary services to start-

ups, and directing significant investments for

research equipment that can be used not only by

university faculty and staff, but also by incubator

tenants. By extension, the overlap of goals and

resources between university patenting and incuba-

tors suggests that decisions to increase university

revenue and contribute to innovation and local eco-

nomic growth through the twin channels of patent-

ing and incubator activities are connected. This

observation calls for reflection upon the basic, yet

unexplored, question of how each channel affects

the other. In this article, we address that question

by examining empirically whether the quality and

economic value of university innovation efforts are

influenced by the creation of incubator facilities at

research-intensive U.S. universities.

Theoretically, the creation of incubation facil-

ities can improve the quality and economic value

of university patents by facilitating knowledge

flows between academic inventors and market

participants—knowledge that can not only help

university patents articulate the commercial value

of their inventions, but also help generate ideas to

university inventors that lead to valuable patents.

Moreover, assuming that industry–academia collab-

oration often yields superior outcomes, incubators

can lead to higher-quality patents if incubator

tenants collaborate with university inventors. But,

the presence of an incubator can reduce the quality

of university patents if auxiliary incubator services

and patenting activities compete for the same

scarce university resources such as funds and dedi-

cated personnel. Similarly, the average quality of

university patents may fall once an incubator is in

place if the university’s overall focus and associ-

ated investments and resources shift toward, say,

start-ups over high-quality patenting. Our research

aims to see which effect outweighs the other.

We must keep in mind, however, that the deci-

sion to establish an incubator can be endogenous;

if incubators are followed by increases in patent

quality, this could indicate that universities with

good projects in the pipeline, and the prospect of

high-quality patents down the road, choose to

establish an incubator, even though there is no

direct effect of incubators on patent quality.

Likewise, a decline in patent quality following the

establishment of an incubator could indicate that

the university expects patent quality to decrease

and establishes an incubator as an alternative

mechanisms for generating revenue and fulfilling

its entrepreneurial mission.

Theorizing about the connection between incu-

bators and patent quality and empirically testing

that connection have not, as far as we are aware,

been addressed in previous work. We also add to

the literature on the quality of university patenting

which, in addition to insightful, mainly descriptive

historical accounts (Henderson et al., 1998), has

focused primarily on the effects of regulatory inter-

ventions such as the Bayh-Dole Act and the impact

of university experience and other university-

specific features (e.g., Mowery, Sampat, and

Ziedonis, 2002; Mowery and Ziedonis, 2002;

Owen-Smith and Powell, 2003; Sampat, Mowery,

and Ziedonis, 2003).

Our empirical work follows convention in approx-

imating quality. First, we proxy for the scientificand

economic value of a patent by recording the number

of times a given patent is cited by subsequent patents

(forward citations), adjusted for the age of the patent

(e.g., Harhoff et al., 1999; Lerner, 1994). Using a

large sample of university patents, we then run a

series of regressions comparing patent quality before

and after the university establishes an incubator, con-

trolling for patent-, university-, and time-specific

characteristics that may affect patent quality. To miti-

gate the aforementioned endogeneity, we also run

instrumental variables regressions; our identification

strategy builds on the insight that universities com-

pete with each other and tend to imitate their peer

institutions, particularly those that are geographically

close (Rey, 2001). Hence, we use the presence of

incubators at similar, nearby (and potentially compet-

itor) universities as an instrument for the focal uni-

versity’s decision to establish an incubator. As an

additional robustness check, we change the unit of

analysis to the university and estimate how the estab-

lishment of incubators affects licensing income, a pri-

mary goal of university patenting. We describe this

exercise in more detail later.

To build our dataset, we collect information on

all 55,919 patents granted from 1969 to 2012 to

U.S.-based universities that were members of the

Association of American Universities (AAU) as of

the end of 2012. These universities are research-

intensive, they patent extensively, and those that

146 C. Kolympiris and P. G. Klein

Copyright © 2016 Strategic Management Society Strategic Entrepreneurship Journal, 11: 145–170 (2017)

DOI: 10.1002/sej