Social Structure of Regional Entrepreneurship: The Impacts of Collective Action of Incumbents on De Novo Entrants

Abstract

The literature has posited that agglomeration economies and the formation of social relationships resulting from the geographic concentration of incumbents constitute the forces that “pull” new entrants into industry clusters. However, this proposition overlooks how the collective action of incumbents in pursuit of their own benefits affects new entrants. This study examines how business associations as collective action organizations established by incumbents to promote and safeguard group-wide interests contribute to de novo entrants. The empirical evidence from Canada’s telecommunication equipment manufacturing industry between 1995 and 2005 reveals that the prevalence of local business associations encourages de novo entrants. However, the impact is curvilinear such that excessive collective action on the part of local fellow incumbents can create a clubby environment and “push” new entrants away.

Full text

Social Structure of Regional
Entrepreneurship:
The Impacts of Collective
Action of Incumbents on
De Novo Entrants
Liang Wang
1
and Justin Tan
2,3
Abstract
The literature has posited that agglomeration economies and the formation of social relationships
resulting from the geographic concentration of incumbents constitute the forces that ‘‘pull’’ new
entrants into industry clusters. However, this proposition overlooks how the collective action of
incumbents in pursuit of their own benefits affects new entrants. This study examines how business
associations as collective action organizations established by incumbents to promote and safeguard
group-wide interests contribute to de novo entrants. The empirical evidence from Canada’s tele-
communication equipment manufacturing industry between 1995 and 2005 reveals that the preva-
lence of local business associations encourages de novo entrants. However, the impact is curvilinear
such that excessive collective action on the part of local fellow incumbents can create a clubby
environment and ‘‘push’’ new entrants away.
Keywords
new venture creation, collective action organizations, economic agglomeration, industry cluster,
business association, de novo entrants
‘‘Spatial clustering alone does not create mutually beneficial interdependencies.’’
Saxenian (1994, p. 161)
Why do entrepreneurs locate their new ventures around incumbents in certain regions despite
the fact that Internet technology and modern transportation have dramatically reduced com-
munication costs and supposedly made geography less relevant (Pe’er, Vertinsky, & King, 2008;
1
School of Management, University of San Francisco, San Francisco, CA, USA
2
Schulich School of Business, York University, Ontario, Canada
3
School of Management, Tianjin University, Nankai Qu, Tianjin Shi, China
Corresponding Author:
Justin Tan, Professor of Management, Newmont Chair in Business Strategy, Schulich School of Business, York University,
Toronto, ON, Canada M3J 1P3.
Email: jtan@schulich.yorku.ca
Article
Entrepreneurship Theory and Practice
2019, Vol. 43(5) 855 –879
© The Author(s) 2019
Article reuse guidelines:
sagepub. com/ journals- permissions
DOI: 10.1177/1042258717750861
journals. sagepub. com/ home/ etp

856 Entrepreneurship Theory and Practice 43(5)
Plummer & Pe’er, 2010; Qian, Acs, & Stough, 2013)? One can attribute the clustering of entre-
preneurial activities to the limited number of optimal locations for production (von Thu
¨nen,
1826; Weber, 1929) and/or to the agglomeration economies resulting from the clustering of
incumbents (Fujita & Thisse, 2002; Krugman, 1991; Marshall, 1920). One can also argue that
social interactions within a given spatial proximity facilitate the formation of social ties
(Sorenson & Stuart, 2001; Stuart & Sorenson, 2003) and the flow of knowledge (Rosenkopf
& Almeida, 2003; Tallman, Jenkins, Henry, & Pinch, 2004). The geographically bounded social
networks that are formed in an industry cluster—the ‘‘geographic concentrations of intercon-
nected companies and institutions’’ (Porter, 1998, p. 78)—help entrepreneurs to learn from
incumbents, recognize opportunities, and mobilize resources (Bhagavatula, Elfring, van
Tilburg, & van de Bunt, 2010; Spigel, 2015). As such, agglomeration economies and the
social relationships that form based on the geographic concentration of incumbents constitute
the forces that ‘‘pull’’ new entrants into clusters (Tan, 2006), as evidenced by empirical findings
of a higher rate of new venture creation within industry clusters (e.g., Sorenson & Audia, 2000;
Tan & Tan, 2017).
However, the above proposition of incumbents as the ‘‘pulling’’ factor within industry
clusters overlooks the existence of collective action organizations, that is, voluntary social
groups in an industry that are established by incumbents to promote and safeguard group-
wide interests (Knoke, 1988). As with other forms of social interactions, collective action
organizations foster the creation of social ties and can be viewed as exclusive local networks
that are nestled in macro network-based social structures. However, in contrast to other forms
of social interactions that do not share a common goal, collective action organizations facili-
tate collaboration among incumbent firms to make goods available to their members (Olson,
1965) such as information dissemination and collective representation (Maennig, O
¨lschla
¨ger,
& Schmidt-Trenz, 2015; Tura & Harmaakorpi, 2005). Although some goods are purely public
because they are available to both members and nonmembers, including new entrants (e.g.,
free industry information that is published online), some are quasi-public in nature because
they are only available to member firms. Additionally, collective action organizations estab-
lish and enforce coercive instruments such as industry standards, codes of conduct, and
quality control for the purpose of leveling collective sanctions for conduct that deviates
from the agreed-upon practices of incumbents (Bennett & Ramsden, 2007; Jones, Hesterly,
& Borgatti, 1997). While the benefits within a cluster ‘‘pull’’ new entrants into to a network,
beyond a certain optimal level, the costs resulting from conforming to the collective norm
outweigh the benefits and can ‘‘push’’ entrepreneurs away (Tan, 2006). This pushing effect
is especially discernable among entrepreneurs who aim to disrupt the status quo in an
industry cluster.
Given their distinct goal of promoting the group-wide interests of incumbent firms, col-
lective action organizations form a distinct type of social group within the network-based
social structure, and they should be treated as such. The failure to do so could result in the
literature promoting misleading policy recommendations—for example, more incumbent
social networking supports the creation of local entrepreneurial activities—regardless of the
nature of the network-based social structure (Lechner, Frankenberger, & Floyd, 2010). As has
been shown in the social-network literature, entrepreneurs can suffer from networking over-
load when too much time is spent networking (Steier & Greenwood, 2000). The performance
of a new venture can be dampened by being over-embedded in the network (Uzzi, 1996, 1997),
and entrepreneurs can find themselves losing flexibility if they become trapped in cohesive
networks (Gargiulo & Benassi, 2000; Maurer & Ebers, 2006). In line with the above, our study
attempts to provide more nuance to the proposition that social networking by incumbents is
conducive to entrepreneurship by asking the following question: How do collective action

857
Wang and Tan
organizations, as part of a geographically bounded and network-based social structure, affect de
novo entrants in a region?
For the purposes of this study, we focus on the collective action of incumbent firms
through business associations, and we operationalize collective action organizations as the
count of local business associations. Using this measure, a study of Canada’s telecommunica-
tion equipment manufacturing industry between 1995 and 2005 demonstrated that local
business associations encouraged de novo entrants, and that the impact was curvilinear.
The results confirmed that collective action organizations, as part of a network-based social
structure, generate societal externalities that complement the economic externalities of
agglomeration. However, more importantly, the curvilinear effect suggests that a place with
too many collective action organizations will likely come to resemble a clubby environment
that protects the interests of incumbents and builds barriers against newcomers. Thus, the
analysis revealed the complex ways in which the collective action of incumbents affects new
venture creation (Pyke, Becattini, & Sengenberger, 1990).
Theory and Hypotheses
Collective action refers to individuals with common interests in voluntary groups who act together
in coordination to further their common interests (Olson, 1965). Such actions can produce and
make nonexcludable goods available to members and sometimes to nonmembers. However, such
collective action creates the possibility of misconduct such as free-riding and shirking. The pre-
vention of these behaviors is a challenge because ‘‘unless there is coercion or some other special
device to make individuals act in their common interest, rational, self-interested individuals will
not act to achieve their common or group interest’’ (Olson, 1965, p. 2). Thus, organizing collective
action requires incentives to motivate rational and self-interested individual actors for the col-
lective good (Knoke, 1988), or it should be rooted in a social context that encourages individual
actors to participate and contribute (Gould, 1993; Wasko & Faraj, 2005; Wiertz & de Ruyter,
2007). While a few studies suggest that collective action may occur without intentional coordin-
ation (Peddibhotla & Subramani, 2007) or formal organization (Wilhoit & Kisselburgh, 2015),
the literature generally acknowledges that collective action is a process of organizing, and a
collective action organization can be the result of this process (Ostrom, 1990). While there is a
variety of other types of collective action organizations such as labor unions and recreational
associations (Knoke, 1988; Olson, 1965), the literature has recognized business associations as a
particular type of collective action organization that affects industry growth and regional com-
petitiveness (Battisti & Perry, 2015; Perez-Aleman, 2003; Schmitz, 1995, 1999) because it is
through business associations that incumbent firms organize collective action in pursuit of
their common interests. In this study of regional entrepreneurship, we examine how business
associations as collective action organizations contribute to de novo entrants.
The contexts in which incumbent organizations agglomerate and function create the locus
of entrepreneurial spawning (Tan & Tan, 2017). To entrepreneurs who are interested in
starting a de novo venture, a collective action organization represents a social group that is
made up of incumbents. From the social-network perspective, a collective action organization
that is formed by member firms in an industry represents an inter-firm subnetwork that is part
of the social structure in which economic activities are embedded (Granovetter, 1985; Uzzi,
1997). It differs from other parts of the social structure, which involve bilateral and/or more
informal interactions, in its use of multilateral agreements and formal governance processes
that are deliberately designed for the purpose of promoting common interests through col-
lective activities (Ostrom, 1990). Below, we propose two hypotheses for how collective action
organizations affect the regional variation of de novo ventures.

858 Entrepreneurship Theory and Practice 43(5)
Local Efficiency From the Collective Action of Incumbent Firms
In his seminal work, Schmitz (1995, 1999) suggested that deliberate and active joint action
(either the collective and/or multilateral action of firms that join forces or individual firms that
cooperate bilaterally) contributes to the growth of industry clusters. As such, collective action
is a source of collective efficiency, which refers to ‘‘the competitive advantage derived from
local external economies and joint action’’ (Schmitz, 1999, p. 466). Nadvi (1999) came to a
similar conclusion in a study of two local business associations in a Pakistani cluster of
surgical instrument manufacturers, where the associations organize collective activities such
as lobbying the government, disseminating industry-related information, promoting exports,
arbitrating disputes, and even facilitating negotiations with foreign buyers. In that case, the
collective action that was organized by local business associations created public (available to
everyone) and quasi-public (available to members only) goods in a region.
More generally, collective action organizations help individual actors to share information,
exchange ideas, and collaborate on issues of common interest and thus serve as an organized
platform of social interaction among people who may not otherwise have an opportunity to
connect (Amin, 1999). As such, collective action organizations perform an important facilitating
role in the creation of both weak ties and a network of collective trust (Bennett, 1998). Further,
collective action organizations as social groups are instrumental in the bridging (Burt, 1992) and
bonding of social capital (Coleman, 1990). Thus, they serve as catalysts for an ‘‘innovative milieu’’
or ‘‘the social and economic interactive relationships and networks of the actors within a spatially
defined area’’ (Maennig & O
¨lschla
¨ger, 2011, p. 442). For de novo entrants, an industry cluster with
a stronger endowment of collective action organizations offers channels that are designed to con-
nect people and diffuse information. Such a cluster allows entrepreneurs to better connect and
interact with incumbents and thus more effectively and successfully identify and commercialize
investment ideas (Acs, Audretsch, & Lehmann, 2013; Gilbert, McDougall, & Audretsch, 2008).
First, as a business association coordinates the activities of individual firms by collectively
informing and promoting an industry, it accelerates the dissemination of information on the
facts and trends of the industry (Rauch, van Doorn, & Hulsink, 2014) and thus helps entre-
preneurs to better identify business opportunities in the industry. Scott’s (1994) comparative
study of the gem and jewelry clusters of Los Angeles and Bangkok, for example, attributed
the greater dynamism of the Thai cluster to trade exhibitions and the international marketing
campaigns that were organized by the local business association to inform people about the
industry. Similarly, entrepreneurs can learn about an industry from the industry information
that is published by a business association. Sometimes the information becomes a public good
because it is made free and available to everyone who is interested in the industry.
Furthermore, faster information dissemination allows the crucial resources that are released
from failed ventures to be more efficiently redirected into new ventures for their best use
(Gilbert, 2012; Lee, Yamakawa, Peng, & Barney, 2011). Faster information dissemination
helps de novo entrants to discover entrepreneurial opportunities in the local business commu-
nity, and the more efficient allocation of resources helps them to realize those opportunities by
mobilizing the resources that have been released from failed incumbents.
Second, some regional resources such as skilled workers, which are intentionally developed
by a business association for the benefit of the incumbent firms, can also be utilized by the
entrepreneurs. For example, Scott (1994) also highlighted the role played by the training
programs organized by the local business association to the vibrant entrepreneurial ecosystem
in the gem and jewelry industry in Bangkok. In their study of the Ontario wine industry,
Massa, Helms, Voronov, and Wang (2017) documented how the local business association
took the lead in bringing the vinicultural practices of well-established wine regions into
Ontario and establishing the education institutions of wine production; and the resources

859
Wang and Tan
built from these collective activities (e.g., wine-making technologies and skillful wine makers)
contributed to the subsequent boom of the industry.
Third, the collective action of the incumbent firms in developing standards, rules, and
norms in an industry for their own benefit can result in reduced transaction costs for the
entrepreneurial activities of new entrants. Unlike informal interpersonal social ties, collective
action organizations must develop rules and norms to govern member behavior and to resolve
conflicts and disputes to ensure that their functioning is not hampered by inappropriate
behaviors such as free-riding, shirking, or the leaking of confidential information. Through
repeated practice, these rules and norms are developed and shared by local firms and the
community at large, which results in reduced transaction costs for inter-firm cooperation and
the promotion of localized trust (Raco, 1998). This type of local sociocultural environment
encourages entrepreneurial activities (MacLeod & Goodwin, 1999; Nadvi, 1999). For exam-
ple, Massa et al. (2017) revealed that the Ontario wine-making industry attracted steady new
entrants after the local business association established an appellation of wine origin system
and enforced standards of wine production in the region.
Fourth, collective action by local firms enhances a region’s legitimacy for businesses and
generates a positive public image for the region that attracts new entrants (McKendrick,
Jaffee, Carroll, & Khessina, 2003). For example, as documented in the description of
Italian ‘‘industrial districts’’ (Piore & Sabel, 1984), the functioning of local business
associations leads to identity creation, which makes a region more identifiable for potential
entrepreneurs. Considering the time limits and search costs that go into making location
decisions, an entrepreneur would naturally focus on the places that they already know, and
a better-recognized location is an easier choice.
It is worth noting that public (available to everyone) and quasi-public (available to members
only) goods that are produced by collective action complement the principles of agglomeration
economies (Fujita & Thisse, 2002; Krugman, 1991; Marshall, 1920) in explaining the spatial
variation of entrepreneurial activities (Armington & Acs, 2002). Economic agglomeration extern-
alities result from the co-location of existing firms (Acs & Varga, 2005; Fu, 2012; Hoover, 1948),
and it includes human capital and specialized suppliers (Ciccone & Hall, 1996; David &
Rosenbloom, 1990; Henderson, 2003; Rotemberg & Saloner, 2000) as well as knowledge spillover
(Acs, Armington, & Zhang, 2007; Gilbert et al., 2008; Qian & Acs, 2013). Firms that are clustered
by passive location enjoy an advantage over isolated firms (Chung & Kalnins, 2001; Cumming &
Johan, 2010; Jaffe, Trajtenberg, & Henderson, 1993; Kalnins & Chung, 2004) and thus lure
entrepreneurs to locate closer to the existing firms (Pe’er & Keil, 2013). In contrast, however,
collective action requires deliberate and active organization to develop collective efficiency
(Schmitz, 1999). Thus, the absence of collective action organizations in a region makes it unlikely
that the region will benefit from the public (available to everyone) and quasi-public (available to
members only) goods that are produced by collective action organizations. In addition, the
existence of collective action organizations can help a region to provide more public and
quasi-public goods to entrepreneurs. With this in mind, we predict the following:
Hypothesis 1: The greater the prevalence of collective action organizations in a region, ceteris pari-
bus, the more likely it is that de novo ventures will be created in that region.
Local Clubbiness From the Collective Action of Incumbent Firms
Thus far, our discussion suggests that collective action organizations contribute to new
venture creation in industry clusters by providing platforms for social interaction,

860 Entrepreneurship Theory and Practice 43(5)
the establishment of norms and rules, resource mobilization, legitimacy, and identity.
However, such a linear association is in direct contrast to Olson’s (1982) proposition that
collective action can contribute to economic decline and that collective action organizations
such as special interest groups might pursue their own interests at the expense of the greater
society and, in particular, newcomers who are not represented. From this perspective, col-
lective action organizations function as ‘‘distributional coalitions’’ (Olson, 1965, 1982) or
‘‘predatory lobbies’’ (Sabel, 1994) whose primary goal is to collectively lobby the government
and extract disproportionate economic benefits for their members that are unachievable in the
market. The impact on entrepreneurial activities in this case can be rather negative.
This paradox is echoed by Amin and Thrift (1994) who posit that a region with a ‘‘thick’’
social structure with active social groups provides shared rules, conventions, and knowledge
to boost local industry (Djelic, Nooteboom, & Whitley, 2005; Parker & Tamaschke, 2005);
however, when local social groups grow too strong, the place can become a ‘‘clubby’’ envir-
onment with high entry barriers that implicitly or explicitly exclude newcomers as outsiders
(Warf, 2001). As such, the impact of the thickness of social structures on new venture creation
can be curvilinear: The impact is positive only up to a certain point. For example, Manzetti
(1994, pp. 91–92) found a role for collective action organizations in facilitating collaboration
and disseminating technology; however, the study also found that the collective activities that
were organized by business associations in Argentina ‘‘increased in number across time
[emphasis added], and political life grew more divisive, and... economic growth turned into
prolonged stagnation.’’
For de novo entrants in particular, too much organized collective action by incumbents can
dampen entrepreneurial activities for the following reasons. First, as collective action organ-
izations increase in number over time to provide more platforms of social interaction, more
individual actors in the social structure will be connected with each other, which results in a
denser social network. The process of accelerated social connection will gradually lead to an
increasingly closed social network (Coleman, 1990) to the point that eventually everyone is
connected with one another. In an overly closed and cohesive social network, entrepreneurial
activities can be dampened due to over-embeddedness (Uzzi, 1996), flexibility can be reduced
as a result of being caught in such a network (Gargiulo & Benassi, 2000; Maurer & Ebers,
2006), and networking overload can be crippling for entrepreneurs when too much time is
spent on networking (Steier & Greenwood, 2000).
Second, incumbents who hold a more central position have more access to resources and
information in such a social network, and potential newcomers with fewer social connections
will be at a disadvantage. As a result, the creation of de novo ventures, in comparison with de
alio entrants who enjoy transfers of resources and capabilities from their parents (Pe’er et al.,
2008), can be dampened because of distortion in the supply of economic resources. This is
consistent with the proposition that collective action organizations are special interest groups
with disproportionate organizational power to facilitate collective actions. Such dispropor-
tionate power includes possible collusion to advance the interests of group members at the
expense of nonmembers who are not well organized (Olson, 1965, 1982). Such imbalances
in power can lead to reductions in efficiency and in the economic growth of a society,
particularly societies with weak governance (Clague, 1997).
Third, too much formal organization will result in too many norms and rules for a poten-
tial entrepreneur to follow. This can become cumbersome for entrepreneurs because they must
become acquainted with the local norms and rules, learn the local ‘‘language,’’ and build a
legitimate identity before they can join collective action organizations, obtain access to quasi-
public and member-only goods, and effectively explore business opportunities. Faulconbridge
(2007) noted how new advertising and law entities in London and New York often could not

861
Wang and Tan
develop identities as legitimate participants through either formal meetings or informal cir-
cumstances. While a key function of collective action organizations is to develop norms and
rules for the group, too much organization can result in a ‘‘clubby’’ environment in which
there are too many norms and rules, which raises entry barriers to newcomers (Warf, 2001).
Additionally, it is necessary to enforce group standards of practice and codes of conduct
for finished products or production processes and to punish violators through various instru-
ments such as revoked membership. Access to local resources, such as investment capital, may
thus be denied if standards are not met. In this sense, industry standards that are endorsed by
incumbent firms through collective action can prevent or limit de novo entrants when the
standards function as a screening mechanism. Following the above line of reasoning, we
expect the following relationship:
Hypothesis 2: The positive impact of collective action organizations on de novo venture creation is
curvilinear.
Methods
In the context of analyzing firms in an industry as individual actors, it is widely acknowledged
that collective action organizations are embodied in business associations (Battisti & Perry,
2015; Knoke, 1988; Olson, 1965; Perez-Aleman, 2003; Schmitz, 1995, 1999). Business associ-
ations, which are also known as employer’s associations, trade associations, or business inter-
est groups, are ‘‘collective bodies that are intermediary between individual business action and
state action’’ (Bennett, 1998, p. 244). The functions of business associations include repre-
senting common interests and providing legitimacy, developing standards and enforcing codes
of conduct, formulating common objectives and mediating conflicts, disseminating informa-
tion, and facilitating social interactions and bridging otherwise disconnected entities (Bennett
& Ramsden, 2007; Dalziel, 2006; Streeck & Schmitter, 1985). Following the literature, we
collected data on business associations to examine the hypotheses.
Research Setting
This study focuses on Canada’s telecommunication equipment industry (i.e., Telephone and
Telegraph Apparatus, SIC code 3661) across the dotcom bubble (1995–2005). As a major
component of the information technology sector, the telecommunication equipment manu-
facturing industry has observed great technological uncertainty and regulatory changes in the
past two decades. Such a challenging external environment has created a fertile background
against which firms have actively engaged in inter-organizational collaboration and collective
activities on issues of common interest (Amesse, Latour, Rebolledo, & Se
´guin-Dulude, 2004).
Starting from the mid-1990s, the telecommunication equipment industry saw a need for
collaboration, as the pace of technological change was enormously accelerated by the intro-
duction of the Internet. Traditional telecommunication equipment manufacturers, whose key
businesses were to serve wired or wireless telephone networks, suddenly had to address com-
puter networks in the emerging ‘‘information society’’ (Abramson & Raboy, 1999). At the
same time, the global telecommunication equipment market opened up due to deregulation,
and a large number of new players entered the industry to explore the previously heavily
regulated market (Amesse et al., 2004). Both the incumbents and the new entrants shared a
common interest: the creation of novel technological solutions and networks (Godoe, 2000).
As a result, the industry witnessed the emergence of various forms of regional industry

862 Entrepreneurship Theory and Practice 43(5)
collaboration (e.g., business associations, technology consortia, and professional societies) for
the purpose of coordinating technological and market development (Hawkins, 1999).
This trend of inter-firm collaboration and collective industry engagement was witnessed in
the Canadian telecommunication equipment industry, often at the regional or local levels.
Canada has an approximately century-long history of manufacturing telecommunication
equipment. The Canada–U.S. Free Trade Agreement, which was set in place in 1989, both
opened up the U.S. market and exposed Canadian telecommunication equipment manufac-
turers to competition from the south (Globerman & Booth, 1989). The need for ‘‘collabor-
ating in order to compete’’ in the industry triggered local economic and industrial adjustment,
to different extents, across Canada. For example, the center of Canada’s information tech-
nology sector, which is generally situated surrounding the city of Waterloo, Ontario—now
popularly known as ‘‘Canada’s Technology Triangle’’—experienced a transformational devel-
opment of collaborative and associative organizations beginning in the early 1990s (Leibovitz,
2003). A Canadian news report described the seemingly successful experience as ‘‘one example
of how people at the local level—in business, government, education, social agencies and
unions—helped this region make the transition from old industrial Ontario... to a new
knowledge-based one’’ (Crane, 1997). Such a context, where firms were actively engaged in
collaborative and collective activities on issues of common interest at a regional level, provides
a decent empirical setting to examine collective action organizations.
We also situated the study across the dotcom bubble (1995–2005). The dotcom bubble
(sometimes called the IT bubble) refers to the speculative bubble that developed and then
burst because of advances in the information technology sector. The Canadian telecommu-
nication equipment manufacturing industry witnessed soaring demand during the bubble as
service providers launched ambitious plans to build the next generation of network infra-
structure. Indeed, the manufacturing of information technology equipment and components
doubled between 1997 and 2000 (Statistics Canada, 2001). In 2000, the sector grew by 21%
even though the overall gross domestic product of Canada increased by only 4.5%. However,
when the bubble burst in 2000, demand dramatically declined, and global investments in IT
infrastructure waned. The industry, which had depended heavily on foreign markets—more
than 80% of information and communication technology products that were manufactured in
Canada were exported—was dealt its first blow when manufacturing in this sector crashed in
the latter part of 2000 (Statistics Canada, 2003). After the burst of the dotcom bubble, exports
by Canada’s telecommunication manufacturing industry did not reach half their peak value
over the next 5 years. The turbulence that was experienced by this industry at this time makes
it an ideal context to test our hypotheses (Wang, 2017).
Data Collection
We collected three sets of data to represent the history of Canada’s telecom equipment
manufacturing industry during the dotcom bubble period (1995–2005). First, we obtained a
listing of telecom equipment manufacturing firms from Scott’s Corporate Directory (1995–
2005), which was first launched in 1957 and is the most complete and comprehensive source of
data on Canadian manufacturers. For each manufacturer, it lists an array of basic informa-
tion, including production location, years of founding and dissolution, estimated sales, and
headquarters. For companies that operate in more than one location, each establishment is
reported separately, an approach that is consistent with the definitions of both Statistics
Canada and the U.S. Census Bureau, which refer to an establishment as a single physical
location where manufacturing is performed. However, given the purpose of this study of
analyzing new venture creation rather than the growth of existing ventures, we removed all

863
Wang and Tan
of the establishments that were a subsidiary of a parent company (i.e., those with headquar-
ters as reported in Scott’s Corporate Directory). This is consistent with the literature (Pe’er
et al., 2008). By doing so, we excluded parent company ventures and focused exclusively on de
novo new ventures, which accounted for approximately 90% of all newly created establish-
ments in the database. The national list represented a compilation of data from Scott’s
Ontario Manufacturers, Quebec Manufacturers, Atlantic Industrial, and Western Industrial
lists. The Atlantic portion covered New Brunswick, Prince Edward Island, Nova Scotia, and
Newfoundland and Labrador. The Western list consisted of British Columbia, Alberta,
Saskatchewan, and Manitoba. Other Canadian territories were not included because they
hosted virtually no telecom equipment manufacturers.
Second, to control for the potential impact of location characteristics on firm-founding, we
collected municipal-level information about each place where a manufacturer was located from
the annual publication of Financial Post Markets: Canadian Demographics, which reports demo-
graphics, income, household expenses, education, and occupations by major groups. Consistent
with the methodology of Statistics Canada and the Financial Post Markets: Canadian
Demographics, we used municipalities to define ‘‘place’’ for our research. A municipality is a
city, town, or census agglomeration, and it is the Canadian equivalent of a metropolitan statistical
area, which is widely used as the unit of analysis in the agglomeration research (e.g., Canina, Enz,
& Harrison, 2005). In total, 91 Canadian municipalities were included in our research, and
information about each municipality helped to control for municipal-level variance.
Third, the collective action organizations were measured by business associations in a given
municipality. Detailed information about the business associations came from Associations
Canada: The Directory of Associations in Canada (1995–2005). This directory is the most
extensive list of Canadian associations available; it lists business associations by subject in
relation to a generic field of interest. There was a dramatic development in information and
communication technology in the telecom equipment manufacturing industry during our
study period, so we included the associations that are listed under the following three subject
categories: computers, information technology, and telephones and telecommunications.
We supplemented and cross-referenced this dataset with information from Scott’s
Directories (2000–2002). If these sources diverged, we granted privilege to the Associations
Canada database, which provides a more comprehensive listing.
Measures
Collective action organizations. We measured this independent variable according to the count of
business associations in each municipality, following the existing literature (Battisti & Perry,
2015; Knoke, 1988; Olson, 1965; Perez-Aleman, 2003; Schmitz, 1995, 1999). In the
Associations Canada (1995–2005) database, an association is ‘‘a voluntary nongovernmental,
nonprofit organization composed of personal and/or institutional members, with or without a
federal or provincial charter, formed for some particular purpose or to advance a common
cause, especially of a public nature.’’ For example, the profile of the Information Technology
Association of Canada’s Vancouver Division describes its goal as ‘‘to provide leadership on
issues that affect the growth and profitability of the information technology industry.’’ The
assumption is that increased numbers of the business associations of an industry in a location
should result in increased levels of organized collective action of the incumbent firms, mutual
awareness of involvement in a common agenda, and structured coalition patterns. It is also
worth noting that the Associations Canada (1995–2005) database includes all business and
trade groups, including groups that establish and enforce industry standards that are endorsed
by the incumbent firms in the industry and thus raises entry barriers for new entrants.

864 Entrepreneurship Theory and Practice 43(5)
New venture creation. We measured this dependent variable as the count of de novo new telecom
equipment manufacturing firms that are established in a municipality in a given year. The key
resources for new venture creation, such as labor supply, are unevenly distributed in Canada,
so we used ‘‘municipality’’ (i.e., city, town, or census agglomeration) as the unit of analysis,
which is consistent with the previous geographic concentration research that has used geo-
graphically meaningful areas, such as areas that are designated by postal code (Stuart &
Sorenson, 2003) or state (Sorenson & Audia, 2000), as the unit of analysis. Following the
previous research methods (e.g., Sorenson, 2005; Stuart & Sorenson, 2003; Wang, Madhok, &
Li, 2014), we only included municipalities that hosted telecom equipment manufacturing
operations during the investigation period. In total, the analysis included 292 founding
events in 910 yearly observations, spread over 91 municipalities.
Control variables. We added a ‘‘spatial lag’’ term to account for potential spatial dependence
(Plummer, 2010). To test the impacts of business associations on firm-founding, we controlled
for other variables that might have offered alternative explanations. At the municipality level,
we included a comprehensive set of control variables. The estimated gross domestic income of
each municipality controlled for the size of the municipality. Household expenses other than
food, shelter, and education helped us to control for the demand for telecommunication
products in the local market, as firms may have co-located closer to their customers
(Rosenthal & Strange, 2003). Different types of municipalities—cities, towns, or census
agglomerations—have different advantages and disadvantages for manufacturing activities,
so we added two municipality-type dummies. Education, or the percentage of the population
with college or higher degrees in science and engineering, was included to control for the
impact of knowledge endowment on the founding of high-tech telecom manufacturers (Miller
& Acs, 2013). Research centers, which refers to the number of wireless communications
research centers in a municipality as recorded by Industry Canada, was included to control
for the impact of government spending and research-based institutions on new venture cre-
ation (Gilbert & Campbell, 2015; Plummer & Gilbert, 2015). French population, which refers
to the percentage of French-speakers in a municipality, was added to control for the political
dynamics in Canada in the form of the impact of the Quebec sovereignty movement on the
economic development of many Canadian municipalities. We included cultural diversity, or
the percentage of the population that uses languages other than English or French, to control
for the possible impacts of cultural diversity and openness on new venture creation (Acs &
Megyesi, 2009). We added firm exits, which refers to the number of firms that exited the
industry in each municipality, to control for the impact of economic turbulence and industry
consolidation (Pe’er & Vertinsky, 2008). Lastly, we followed previous empirical analyses of
agglomeration (Ellison & Glaeser, 1997; Ellison, Glaeser, & Kerr, 2010) and measured the
agglomeration of labor as the sum of the existing employees of all of the telecom manufac-
turers in a municipality.
At the provincial level, the corporate tax rate of each province was included to control for the
impact of tax structures on investment decisions. We also added the unemployment rate for
each province to further control for the impact of macroeconomic upswings and downswings.
At the industry level, we borrowed from the density dependence model in the population
ecology literature (Carroll & Hannan, 1989) and the national density and squared term of the
number of telecom equipment manufacturers in Canada to control for ecological dynamics.
We included exports, which refers to the dollar amount of the industry’s overseas sales, to
control for the product or technology life cycle that can affect new entrants. We also included
dotcom bubble as a dummy variable that was coded as 1 if the year was after 2000 to control
for the impact of economic downturns after the burst of the dotcom bubble.

865
Wang and Tan
All of the independent and control variables were lagged 1 year to account for the time that
is needed for any new venture creation decision, which thus reduces concerns about the
temporality of the data, reverse causality, and simultaneity. This followed the empirical
studies with similar data structures (e.g., Yang, Phelps, & Steensma, 2010).
Estimation methods. Poisson regression was chosen as an appropriate method because with a
count variable for the dependent variable, it can take nonnegative integer values. With
Poisson regression as the starting point, we adopted negative binomial regression as the
modeling strategy. The Poisson regression assumes that the dependent variable has a
Poisson distribution with equal conditional variance and mean. However, the Poisson vari-
ance assumption is violated by the data for which the variance is greater than the mean (i.e., in
situations of over-dispersion). Following the previous research (e.g., Simons & Ingram, 2003),
we chose a negative binominal regression model to account for over-dispersion.
To properly analyze the panel data, we used Allison and Waterman’s (2002) unconditional
estimation of fixed-effects negative binomial models by including dummy variables for all
municipalities and all years. This followed the empirical studies with the same data structure
(Yang et al., 2010). The year dummies controlled for unobserved systematic period effects,
and the municipality dummies controlled for unobserved and temporally stable municipality
differences in firm-founding. We also employed the more conventional conditional maximum
likelihood estimation procedure that was developed by Hausman, Hall, and Griliches (1984),
and we obtained consistent results. We choose to report the results of Allison and Waterman’s
(2002) method because the Hausman et al. (1984) method does not qualify as a true fixed-
effects method as it does not control for unchanging covariates.
Another empirical issue of our analysis is the possible existence of spatial dependence
(Anselin, 1988) because firm-founding in one location might be a function of firm-founding
in nearby locations. This can be a serious problem because it violates the assumption of the
regression analysis that the observations of the variables are not spatially correlated. In a
pioneering effect, Plummer (2010) demonstrated how spatial dependence is especially prob-
lematic for the study of entrepreneurial activities and outlined the econometric techniques that
are needed to address the problem. We followed the procedure as suggested by Plummer
(2010) and employed two methods to account for the potential existence of spatial depend-
ence. First, we calculated a ‘‘spatial lag’’ term and included it as an additional control variable
in the regression models, following several previous studies (Acs, Plummer, & Sutter, 2009;
Acs & Plummer, 2005; Plummer & Acs, 2014; Plummer & Gilbert, 2015). For each observa-
tion (i.e., each municipality in a given year in our study), the ‘‘spatial lag’’ term was the
weighted average of the dependent variable (i.e., firm-founding) observed over the neighbor-
ing municipalities, and this controlled for any possible spillover of new venture creation across
geographical boundaries (Anselin, 2001). Second, we applied the Driscoll–Kraay estimator in
an additional regression analysis to correct for both spatial and serial correlation in the
regression residuals (Driscoll & Kraay, 1998; Hoechle, 2007; Plummer & Gilbert, 2015).
Third, we used robust errors and clustered the standard errors by Canadian province to con-
sider the possibility that new venture creation in one province across different municipalities is
spatially dependent.
Finally, we addressed the issue of simultaneity bias. It was possible that we might encoun-
ter a confounding variable in the regression analysis because of the possibility that the high
degree of agglomeration of labor could result in a high number of business associations in a
municipality. According to our theory, collective action organizations encourage new venture
creation up to a tipping point, and the created new ventures contribute to the agglomeration
of labor in the industry and the creation of more business associations. Our regression models

866 Entrepreneurship Theory and Practice 43(5)
would thus be subject to simultaneity bias. In consideration of this issue, we followed the
practice of Plummer and Acs (2014) and ran additional analyses and used the three-state least
squares (3SLS) estimator to simultaneously estimate collective action organizations and new
venture creation models. The results that were obtained from the different regression models
were consistent with one another.
Results
In Table 1, we report the means, standard deviations, and correlations for all of the study
variables. To test for the presence of multicollinearity, we used ordinary least squares to
calculate the variance inflation factors (VIF) for all of the control variables and independent
variables. All of the VIF values were within the acceptable range with a mean value of 2.96;
thus, the regression models were free of significant multicollinearity concerns (Meyers, Gamst,
& Guarino, 2006).
Table 2 presents the results of three different regression methods: the unconditional fixed-
effects negative binominal analysis (Models 1 and 2), the 3SLS (Models 3 and 4), and the
Driscoll–Kraay estimators (Models 5 and 6). For each method, we first introduced all of the
control variables in the first model (i.e., Models 1, 3, and 5) and then added collective action
organizations and its squared term in the second model (i.e., Models 2, 4, and 6).
The regression results lent support to the hypotheses. Models 1 and 2 reported the
results of the unconditional fixed-effects negative binomial regression analysis. In Model 1,
we found a positive impact of agglomeration of labor (�¼0.53,p<.01), which supported
the existing studies (e.g., Sorenson & Audia, 2000) in finding that the higher the degree of
agglomeration, the more new firms were founded in that place. We also found a positive
impact of firm exits (�¼0.23, p<.001) with more exits associated with more founding.
Considering the market turbulence that shook many well-established firms during the
dotcom bubble, the finding suggests that the exit of incumbents released resources that
could have been used to start new businesses, which is consistent with the prediction that
business associations facilitate efficient resource allocation. Model 2 added the independent
variable of collective action organizations and its squared term. We found a positive first-
order effect with an opposite sign for the squared term (�¼1.10, p<.001; �¼�0.20,
p<.001, respectively). The positive first-order and negative second-order effects suggested
an inverted U-shaped impact (i.e., the prevalence of local business associations at lower
ranges related positively to firm-founding; however, beyond a certain point, the impact
diminished). The results demonstrated a curvilinear impact and offered strong, clear sup-
port for Hypotheses 1 and 2.
Models 3 and 4 reported the results of the 3SLS estimator. The regression analysis simul-
taneously estimated new venture creation and collective action organizations. We also
included year dummies and municipality dummies to control for unobserved temporal and
panel heterogeneity. The results in Model 3, which included all of the control variables, did
not offer support to the prediction that the agglomeration of labor leads to the establishment
of business associations, although population seems to be a significant predictor. Model 4
added the independent variable and its squared term, and it showed consistent results (col-
lective action organizations, �¼2.61, p<.001; its squared term, �¼�0.34, p<.001).
Models 5 and 6 reported the results of the Driscoll–Kraay estimator. We found consistent
results (i.e., collective action organizations had a positive first-order effect) with the opposite
for the squared term (�¼0.62, p<.001; �¼�0.09, p<.001, respectively). The consistent
results that were obtained from the three different regression methods demonstrated strong
support for our hypotheses.

867
Wang and Tan
Table 1. Descriptive Statistics and Correlations (N¼910).
Variable Mean SD (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18)
(1) New venture creation 0.32 0.74
(2) Gross domestic income ($10 billion) 0.37 0.49 .33
(3) Household expenses/1,000 8.78 1.86 .05 .12
(4) City 0.69 0.46 .10 .27 �.13
(5) Town 0.11 0.32 .02 �.10 .42 �.53
(6) Education 0.03 0.01 .09 .16 .44 .18 .11
(7) Research center 0.26 0.69 .11 .61 �.13 .22 �.14 .08
(8) French population 24.42 35.98 �.11 �.23 �.43 .12 �.20 �.07 �.07
(9) Cultural diversity 12.77 11.50 .22 .43 .24 .29 �.04 .21 .15 �.23
(10) Firm exits 0.37 0.85 .46 .40 .05 .12 .00 .14 .16 �.11 .23
(11) Agglomeration of labor/1,000 0.23 0.72 .37 .23 .04 .09 �.06 .21 .08 �.06 .13 .26
(12) Corporate tax rate 0.11 0.01 �.03 �.11 .04 .09 .01 .02 .08 .03 �.16 �.06 �.02
(13) Unemployment rate 8.37 2.19 �.07 �.29 �.34 .18 �.17 �.12 �.09 .42 �.15 �.11 �.07 .27
(14) Density 144.7 15.44 .05 �.07 �.29 .01 �.01 �.25 .00 .02 �.20 .00 .07 .00 .25
(15) Density
2
1.00 0.82 .01 .05 .18 .00 .00 .11 .00 �.01 .05 .04 .00 .00 �.03 �.46
(16) Exports (billion $) 5.81 2.49 .00 .01 .05 .00 .00 �.04 .00 .00 �.11 .02 .01 .00 �.16 .29 �.17
(17) Dotcom bubble 0.60 0.49 �.03 .08 .29 .00 .01 .23 .00 �.02 .08 .02 �.06 .00 �.41 �.69 .11 .28
(18) Collective action organizations 0.70 2.25 .21 .62 .02 .15 �.03 .15 .64 �.13 .27 .24 .14 .01 �.10 .02 �.01 .00 �.02
(19) Collective action organizations
2
1.00 5.47 .07 .43 �.01 .10 �.04 .06 .53 �.07 .17 .09 .04 .02 �.06 .02 .00 .00 �.03 .90
Note. All correlations with absolute values greater than .065 are significant at p<.05.

868 Entrepreneurship Theory and Practice 43(5)
Table 2. Regression Analysis of New Venture Creation of Canada’s Telecom Equipment Manufacturers.
Model 1
Baseline
Model 2
H1 & H2
Model 3
Baseline
Model 4
H1 & H2
Model 5
Baseline
Model 6
H1 & H2
Variable Fixed effects negative binomial Three-stage least squares Driscoll–Kraay
New venture creation
Intercept 1.31 (2.15) 2.04 (2.50) 0.19 (1.43) 1.27 (1.79) 0.07 (2.04) 0.68 (2.02)
Founding spatial lag 0.10 (0.11) 0.09 (0.12) 0.03 (0.04) 0.03 (0.04) 0.03 (0.03) 0.03 (0.03)
Gross domestic income ($ 10 billion) 0.25 (0.39) 0.05 (0.36) 0.23 (0.15) 0.71 (0.24)
**
0.28 (0.07)
***
0.08 (0.16)
Household expenses/1,000 0.00 (0.02) 0.00 (0.03) 0.01 (0.03) 0.01 (0.03) 0.03 (0.02) 0.02 (0.02)
City 0.13 (0.70) 0.85 (0.70) 0.76 (0.36)
*
0.84 (0.49) 0.76 (0.56) 0.38 (0.44)
Town 1.24 (0.95) 1.38 (0.47)
***
0.30 (0.46) 0.16 (0.51) 0.32 (0.44) 0.23 (0.34)
Education 7.23 (5.81) 9.49 (7.37) 1.43 (3.28) 2.90 (3.62) 0.16 (1.12) 0.44 (0.79)
Research center 0.26 (0.22) 0.20 (0.18) 0.26 (0.11)
*
0.21 (0.17) 0.27 (0.11)
*
0.09 (0.13)
French population 0.00 (0.01) 0.00 (0.00) 0.01 (0.00) 0.00 (0.00) 0.00 (0.01) 0.00 (0.01)
Cultural diversity 0.03 (0.03) 0.02 (0.02) 0.01 (0.01)
*
0.00 (0.01) 0.01 (0.01) 0.01 (0.01)
Firm exits 0.23 (0.04)
***
0.20 (0.04)
***
0.23 (0.03)
***
0.13 (0.04)
***
0.24 (0.10)
*
0.22 (0.10)
*
Agglomeration of labor/1,000 0.53 (0.18)
**
0.51 (0.18)
**
0.30 (0.04)
***
0.27 (0.04)
***
0.30 (0.10)
**
0.29 (0.10)
**
Corporate tax rate 4.84 (31.80) 31.42 (29.62) 5.12 (13.50) 29.71 (17.09) 3.16 (14.28) 6.93 (15.28)
Unemployment rate 0.02 (0.12) 0.00 (0.10) 0.03 (0.03) 0.07 (0.04) 0.00 (0.03) 0.01 (0.02)
Density 0.98 (0.96) 0.80 (0.88) 0.10 (0.04)
*
0.11 (0.05)
*
0.10 (0.03)
**
0.09 (0.03)
**
Density
2
0.73 (0.71) 0.61 (0.65) 0.08 (0.05) 0.10 (0.05) 0.04 (0.02)
*
0.05 (0.02)
**
Exports (billion) 0.26(0.08)
***
0.27(0.13)
*
0.01(0.01) 0.02 (0.01) 0.01(0.01)
**
0.01(0.01)
Dotcom bubble 0.08 (0.40) 0.07 (0.36) Omitted Omitted 0.11 (0.07) 0.08 (0.06)
(continued)

869
Wang and Tan
Table 2. Continued
Model 1
Baseline
Model 2
H1 & H2
Model 3
Baseline
Model 4
H1 & H2
Model 5
Baseline
Model 6
H1 & H2
Variable Fixed effects negative binomial Three-stage least squares Driscoll–Kraay
Year dummies Included Included Included Included
Municipality dummies Included Included Included Included
Year effects Included Included
Municipality fixed effects Included Included
Collective action organizations 1.10 (0.29)
***
2.61 (0.48)
***
0.62 (0.17)
***
Collective action organizations
2
�0.20 (0.04)
***
�0.34 (0.06)
***
�0.09 (0.02)
***
Collective action organizations
Intercept �0.47 (0.03)
***
�0.46 (0.03)
***
Agglomeration of labor/1,000 0.00 (0.05) 0.03 (0.05)
Population 0.00 (0.00)
***
0.00 (0.00)
***
New venture creation 0.01 (0.08) �0.10 (0.08)
Log Pseudolikelihood �522.14 �516.67
R
2
0.37 0.24 0.20 0.22
Total observation 910 910 910 910 910 910
Note. In total, 292 new ventures were created across 91 municipalities. Robust errors clustered by province are in parentheses. H1¼Hypothesis 1; H2 ¼Hypothesis 2.
*p<.05. **p<.01. ***p<.001.

870 Entrepreneurship Theory and Practice 43(5)
While the information in Table 2 is informative, it remains somewhat limited. To further
demonstrate the curvilinear effect, we illustrated the effect in Figure 1. The inverted U-shaped
curve of the fitted value clearly shows a tipping point. The relationship starts out positive
before the point and becomes negative after that. The results that are shown in Figure 1
illustrate this curvilinear relationship between collective action organizations and new venture
creation.
Discussion and Conclusion
Entrepreneurs establish their start-ups in certain regions and around certain types of incum-
bent firms to improve their chances of survival and growth. Consequently, added insights
about the context and the intrinsic nature in which a collection of incumbent firms attracts or
spawns entrepreneurial founding and performance are tremendously relevant for academic
researchers, policy makers, and corporate decision makers. Social scientists are paying
increasing attention to the sociological explanation of this interdisciplinary phenomenon, in
addition to the century-old theory of economic agglomeration (Marshall, 1920). Given that
intentional collaboration is not needed for the principle of agglomeration economies to be
effective (Gordon & McCann, 2000), the agglomeration theorization of the spatial variation
of entrepreneurial activities implicitly treats individuals and organizations as atomistic enti-
ties, and it thus overlooks the social structure in which economic activities are embedded and
governed (Granovetter, 1985). In contrast, sociological theorization sheds light on the role of
network-based social structures in the geographic concentration of entrepreneurial activities
(Sorenson, 2005; Stuart & Sorenson, 2003). It posits that geographically bounded social net-
works that are formed within an industry cluster help entrepreneurs to identify opportunities
-1 012
New venture creation (count of entry)
0 5 10 15 20
Collective action organizations (count of business associations)
95% Confidence interval Fitted values
Figure 1. The predicted new venture creation with 95% confidence interval.

871
Wang and Tan
and mobilize resources and thus reinforce the clustering of entrepreneurial activities
(Sorenson & Audia, 2000). However, a literature review of this line of research shows that
sociological theorization will remain incomplete unless it goes beyond its fundamental yet
simplistic proposition on the benefits of social networks for entrepreneurs and separately
examines the diverse functions and impacts of the different types of social networks that
form the different fabrics of regional social structures (McCann & Folta, 2008). Without
this more precise examination, the literature risks implying an oversimplified linear associ-
ation between network-based social structures and new venture creation, regardless of the
distinct natures of different social networks.
As a departure from the sociological theorization of the geographic concentration of eco-
nomic activities, we join the debate by offering more fine-tuned insights regarding the research
question: How do collective action organizations as part of a geographically bounded and net-
work-based social structure affect de novo entrants in a region? Following the literature
(Battisti & Perry, 2015; Knoke, 1988; Olson, 1965; Perez-Aleman, 2003; Schmitz, 1995,
1999), our study operationalized collective action organizations as business associations.
The earlier empirical studies of the impacts of business associations on economic growth
generated mixed evidence (Curran & Blackburn, 1994; Foreman-Peck, Makepeace, &
Morgan, 2006; Houghton, Smith, & Hood, 2009; Sievers & Maennig, 2006), which reflects
the divergent views of collective action organizations from network facilitators (Bennett,
1998; Bennett & Ramsden, 2007; Dalziel, 2006; Maennig & O
¨lschla
¨ger, 2011) to distributional
coalitions (Manzetti, 1994; Olson, 1965, 1982). With the aim of shedding light on the incon-
clusive empirical results and divergent theoretical views of collective action organizations, this
study yielded intriguing empirical results by using data from Canada’s telecommunication
equipment manufacturing industry between the years 1995 and 2005.
Among the significant insights from the study, we found empirical evidence that economic
agglomeration contributed positively to new venture creation, which was consistent with the
existing literature. More importantly, we found strong evidence that collective action organ-
izations affected new venture creation. Consistent with the previous studies (Dalziel, 2006;
Houghton et al., 2009; Maennig & O
¨lschla
¨ger, 2011), we found that local business associ-
ations encouraged firm-founding as they helped to establish a social platform for new venture
creation and economic growth. In addition, we found the positive relationship to be curvi-
linear. In line with the previous studies that revealed the possible negative impacts of business
associations (Foreman-Peck et al., 2006; Manzetti, 1994; Sievers & Maennig, 2006), this
finding suggests that a region with too many collective action organizations comes to resemble
a clubby environment that erects high entry barriers against newcomers, perhaps for the
purpose of protecting the interests of the incumbents.
Theoretical Contributions and Practical Implications
As one of the earliest efforts to examine how collective action organizations create a ‘‘pulling’’
effect (Tan, 2006) in new venture creation, this research contributes to the entrepreneurship
research in general (Busenitz, Plummer, Klotz, Shahzad, & Rhoads, 2014) by offering fine-
tuned insights and empirical evidence on the role of local collective organizations. The empirical
studies have generated abundant qualitative evidence on the role of organized collective action in
building industry clusters (e.g., American Electronics Association in the building of Silicon
Valley; see (Saxenian, 1994)) and sporadic quantitative evidence of the impacts of business asso-
ciations on a variety of dependent variables that range from strategic competence (Houghton
et al., 2009) and innovation performance (Dalziel, 2006) to firm growth (Foreman-Peck et al.,
2006) and economic development (Beugelsdijk & Van Schaik, 2005);however, a direct assessment

872 Entrepreneurship Theory and Practice 43(5)
of the impacts of collective action organizations on new venture creation has been missing. The
pioneering study of Maennig and O
¨lschla
¨ger (2011) partially filled the gap by reporting a positive
relationship between business associations and regional start-up rates; however, the relationship
was not at the 95% significance level, and the implications of the findings were limited due to the
study’s cross-sectional design. By being the first to reveal quantitative evidence from longitudinal
data, our study advances the understanding of how collective action that is organized by business
associations affects the location choice of entrepreneurial activities.
More specifically, the positive and curvilinear impact of business associations on new venture
creation reconciles the theoretical debate over whether collective action organizations are network
facilitators (Bennett, 1998; Bennett & Ramsden, 2007; Dalziel, 2006; Maennig & O
¨lschla
¨ger, 2011)
or distributional coalitions (Manzetti, 1994; Olson, 1965, 1982). The curvilinear relationship that
is revealed in this study suggests that both perspectives can be valid and that the impacts of
collective action are essentially a double-edged sword. Collective actions that are organized by
business associations do facilitate networking, the development of trust, and collaboration and
collective representation; thus, they help to build industry clusters with more social capital, lower
transaction costs, faster knowledge dissemination, a more visible regional identity, and conse-
quently more entrepreneurial activities. However, the functions of business associations are dir-
ected toward the representation of the existing members and the creation of collective industry
standards, norms, and rules, which leads to an overly dense social structure, which becomes costly
for new entrants. Such a social structure favors the status quo and is no longer conducive to
entrepreneurship. Instead, it becomes an entry barrier for entrepreneurs and for de novo entrants in
particular because they may disrupt the status quo in the industry cluster. As a result, at some
point, the inhibiting ‘‘push’’ force overwhelms the positive ‘‘pull’’ factor that has attracted new
entrants and nurtured entrepreneurial growth in a cluster (Tan, 2006), and the facilitating benefits
of local collective action organizations start to decline. To validate our findings, we presented the
empirical results to some of the external constituents who were also the primary source of our data.
These individuals confirmed our hypotheses, and one of them made the following remark: ‘‘I can
see how too many rule-setting organizations in one location could complicate entrepreneurship
and dampen creativity, being more of a hindrance than a help.’’ The implication for policy makers
and industry leaders is thus readily apparent as such insights may offer actionable guidance as they
attempt to emulate the winning attributes and avoid potential hindrance in clusters.
Additionally, the findings contribute to the sociological theorization of economic activities by
directly measuring collective action organizations, which are a key social structure element, thus
confirming that social structure is relevant in the regional variation of entrepreneurial activities
(Sorenson & Audia, 2000). As the literature suggests, the persistence of geographic industry
concentration can be a result of either agglomeration externalities that help to maintain the
competitiveness of existing firms or the social structure of industry clusters that generates and
attracts new business ventures. Social connections in industry clusters provide entrepreneurs,
particularly spinoffs (Klepper & Sleeper, 2005), with opportunities to observe how local firms
respond to customer needs and then adapt and alter the strategies that are used by existing firms
(Almeida & Kogut, 1997; Audretsch, 1998; Audretsch, 2003; Saxenian, 1994). However, the
literature of sociological theorization of industry clusters has not directly and separately assessed
the different types of social networks with divergent objectives and functions. Consistent with
economic reasoning, our study finds that the agglomeration of labor has a positive impact on the
founding of new ventures; and, after controlling for agglomeration economics, we found that
collective action organizations have a positive impact on new venture creation. Thus, our study
confirms the impacts of both agglomeration and social structure, and it empirically demonstrates
that social structure (i.e., collective action organizations) complements economic externalities by
contributing to entrepreneurial activities. Finally, the curvilinear relationship that was found is

873
Wang and Tan
consistent with the research on the dark side of social capital (Lechner et al., 2010). In sum, these
nuanced findings reveal a complexity in the functions and impacts of social structure on entre-
preneurial activities that has not yet been addressed in the literature.
Implications for Future Research
This study examines how the collective action of incumbent firms affects de novo entrants. Our
research strategy was grounded by Saxenian’s (1994) suggestion to examine local industrial struc-
ture to understand why more new ventures emerge in some places but not in others. In particular,
a place with a ‘‘thick’’ social structure ‘‘ranging from strong local institutional presence through
to the strength of shared rules, conventions, and knowledge’’ (Amin & Thrift, 1994, p. 2) has the
potential to shape local industry (Djelic et al., 2005; Parker & Tamaschke, 2005). For the pur-
poses of this study, we focus on the collective action of incumbent firms through business asso-
ciations, and we operationalize collective action organizations as the count of local business
associations. By doing so, the study does not account for the other types of collective action
organizations, such as labor unions and recreational associations (Knoke, 1988; Olson, 1965).
However, these groups can also potentially facilitate networking and knowledge flow for entre-
preneurs to identify opportunities and mobilize resources. Considering the similarities and dif-
ferences, it is interesting to note whether the curvilinear impact of business associations applies to
the other collective action organizations. More generally, future research should attempt to
capture how other types of social networks within industry clusters play roles in new venture
creation. A more comprehensive and multilevel research design (e.g., Tan, Zhang, & Wang, 2015)
that incorporates firm-level, industry/cluster-level, and regional-level interactions may offer more
accurate account and fine-tuned insights about this important yet underexplored issue.
Finally, in response to recent developments, future research efforts are called for to exam-
ine why the agglomeration of incumbents may have beneficial effects on attracting new
entrants and increasing founding rates but detrimental effects on subsequent venture survival
(Tan & Tan, 2017). Our finding of a positive relationship between firm exits and new venture
creation (see Table 2) provides an opportunity for further study. While it indirectly confirms
our theory that collective action organizations facilitate the efficient allocation of resources in
the industry (such that the resources that are released from failed ventures can be quickly
distributed to their next best use in a new venture), and we ran additional regression analyses
to test the impact of business associations on firm exits, the result was not significant, and we
were not able to draw a definitive conclusion. Future research on the functions of collective
action organizations in terms of both new venture creation and firm exits is thus needed. In
sum, by offering preliminary findings on an issue of increasing importance, our research aims
to raise scholarly interest and to provoke future debate.
Acknowledgments
We thank ETP Editor James Fiet and anonymous reviewers for comments and suggestions.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or
publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or
publication of this article: This research was in part supported by a Social Science and Humanities

874 Entrepreneurship Theory and Practice 43(5)
Research Council of Canada research grant and National Natural Science Foundation of China research
grants (71472131, 71732005, and 71401183).
References
Abramson, B. D., & Raboy, M. (1999). Policy globalization and the ‘‘information society’’: A view from
Canada. Telecommunications Policy,23, 775–791.
Acs, Z. J., Armington, C., & Zhang, T. (2007). The determinants of new-firm survival across regional econo-
mies: The role of human capital stock and knowledge spillover. Papers in Regional Science,86, 367–391.
Acs, Z. J., Audretsch, D. B., & Lehmann, E. E. (2013). The knowledge spillover theory of entrepre-
neurship. Small Business Economics,41, 757–774.
Acs, Z. J., & Megyesi, M. I. (2009). Creativity and industrial cities: A case study of Baltimore.
Entrepreneurship & Regional Development,21, 421–439.
Acs, Z. J., & Plummer, L. A. (2005). Penetrating the ‘‘knowledge filter’’ in regional economies. Annals of
Regional Science,39, 439–456.
Acs, Z. J., Plummer, L. A., & Sutter, R. (2009). Penetrating the knowledge filter in ‘‘rust belt’’ econo-
mies. Annals of Regional Science,43(4), 989–1012.
Acs, Z. J., & Varga, A. (2005). Entrepreneurship, agglomeration and technological change. Small
Business Economics,24, 323–334.
Allison, P. D., & Waterman, R. P. (2002). Fixed-effects negative binomial regression models.
Sociological Methodology,32, 247–265.
Almeida, P., & Kogut, B. (1997). The exploration of technological diversity and the geographic local-
ization of innovation. Small Business Economics,9, 21–31.
Amesse, F., Latour, R., Rebolledo, C., & Se
´guin-Dulude, L. (2004). The telecommunications
equipment industry in the 1990s: From alliances to mergers and acquisitions. Technovation,24, 885–897.
Amin, A. (1999). An institutionalist perspective on regional economic development. International
Journal of Urban and Regional Research,23, 365–378.
Amin, A., & Thrift, N. J. (1994). Globalization, institutions, and regional development in Europe. Oxford:
Oxford University Press.
Anselin, L. (1988). Spatial econometrics: Methods and models. Dordrecht: Kluwer Academic Publishers.
Anselin, L. (2001). Spatial econometrics. In B. Baltagi (Ed.), A companion to theoretical econometrics
(pp. 310–330). Oxford: Basil Blackwell.
Armington, C., & Acs, Z. J. (2002). The determinants of regional variation in new firm formation.
Regional Studies,36, 33–45.
Associations Canada. (1995–2005). Financial post markets: Canadian demographics. Toronto: Financial Post.
Audretsch, B. (1998). Agglomeration and the location of innovative activity. Oxford Review of Economic
Policy,14, 18–29.
Audretsch, D. B. (2003). Managing knowledge spillovers: The role of geographic proximity. In J. Baum
& O. Sorenson (Eds.), Geography and strategy: Advances in strategic management (pp. 23–48).
Oxford: JAI Press.
Battisti, M., & Perry, M. (2015). Small enterprise affiliations to business associations and the collective
action problem revisited. Small Business Economics,44, 559–576.
Bennett, R. J. (1998). Business associations and their potential contribution to the competitiveness of
SMEs. Entrepreneurship & Regional Development,10, 243–260.
Bennett, R. J., & Ramsden, M. (2007). The contribution of business associations to SMEs: Strategy,
bundling or reassurance?. International Small Business Journal,25, 49–76.
Beugelsdijk, S., & Van Schaik, T. (2005). Differences in social capital between 54 Western European
regions. Regional Studies,39, 1053–1064.
Bhagavatula, S., Elfring, T., van Tilburg, A., & van de Bunt, G. G. (2010). How social and human
capital influence opportunity recognition and resource mobilization in India’s handloom industry.
Journal of Business Venturing,25, 245–260.
Burt, R. S. (1992). Structural holes: The social structure of competition. Cambridge, MA: Harvard
University Press.

875
Wang and Tan
Busenitz, L. W., Plummer, L. A., Klotz, A. C., Shahzad, A., & Rhoads, K. (2014). Entrepreneurship research
(1985–2009) and the emergence of opportunities. Entrepreneurship Theory and Practice,38, 981–1000.
Canina, L., Enz, C. A., & Harrison, J. S. (2005). Agglomeration efects and strategic orientations:
Evidence from the U.S. lodging industry. Academy of Management Journal,48, 565–581.
Carroll, G. R., & Hannan, M. T. (1989). Density dependence in the evolution of populations of news-
paper organizations. American Sociological Review,54, 524–541.
Chung, W., & Kalnins, A. (2001). Agglomeration effects and performance: A test of the Texas lodging
industry. Strategic Management Journal,22, 969–988.
Ciccone, A., & Hall, R. E. (1996). Productivity and the density of economic activity. American Economic
Review,86, 54–70.
Clague, C. (1997). Institutions and economic development: Growth and governance in less-developed and
post-socialist countries. Baltimore, MD: Johns Hopkins University Press.
Coleman, J. S. (1990). Foundations of social theory. Cambridge, MA: Harvard University Press.
Crane, D. (1997, September 16). Technology triangle: A model for job and wealth creation. Toronto
Star, p. D2.
Cumming, D., & Johan, S. (2010). The differential impact of the internet on spurring regional entre-
preneurship. Entrepreneurship Theory and Practice,34, 857–883.
Curran, J., & Blackburn, R. (1994). Small firms and local economic networks: The death of the local
economy? London: Paul Chapman.
Dalziel, M. (2006). The impact of industry associations: Evidence from statistics Canada data.
Innovation,8, 296–306.
David, P. A., & Rosenbloom, J. L. (1990). Marshallian factor market externalities and the dynamics of
industrial localization. Journal of Urban Economics,28, 349–370.
Djelic, M.-L., Nooteboom, B., & Whitley, R. (2005). Introduction: Dynamics of interaction between
institutions, markets and organizations. Organization Studies,26, 1733–1741.
Driscoll, J. C., & Kraay, A. C. (1998). Consistent covariance matrix estimation with spatially dependent
panel data. Review of Economics and Statistics,80, 549–560.
Ellison, G., & Glaeser, E. L. (1997). Geographic concentration in U.S. manufacturing industries: A
dartboard approach. Journal of Political Economy,105, 889–927.
Ellison, G., Glaeser, E. L., & Kerr, W. R. (2010). What causes industry agglomeration? Evidence from
coagglomeration patterns. American Economic Review,100, 1195–1213.
Faulconbridge, J. R. (2007). Exploring the role of professional associations in collective learning in
London and New York’s advertising and law professional-service-firm clusters. Environment and
Planning A,39, 965–984.
Foreman-Peck, J., Makepeace, G., & Morgan, B. (2006). Growth and profitability of small and medium-
sized enterprises: Some Welsh evidence. Regional Studies,40, 307–319.
Fu, X. (2012). Foreign direct investment and managerial knowledge spillovers through the diffusion of
management practices. Journal of Management Studies,49, 970–999.
Fujita, M., & Thisse, J. F. (2002). Economics of agglomeration: Cities, industrial location, and regional
growth. Cambridge, MA: Cambridge University Press.
Gargiulo, M., & Benassi, M. (2000). Trapped in your own net? Network cohesion, structural holes, and
the adaptation of social capital. Organization Science,11, 183–196.
Granovetter, M. (1985). Economic action and social structure: The problem of embeddedness. American
Journal of Sociology,91, 481–510.
Gilbert, B. A. (2012). Creative destruction: Identifying its geographic origins. Research Policy,41,
734–742.
Gilbert, B. A., & Campbell, J. T. (2015). The geographic origins of radical technological paradigms: A
configurational study. Research Policy,44, 311–327.
Gilbert, B. A., McDougall, P. P., & Audretsch, D. B. (2008). Clusters, knowledge spillovers and new
venture performance: An empirical examination. Journal of Business Venturing,23, 405–422.
Globerman, S., & Booth, P. (1989). The Canada-US free trade agreement and the telecommunications
industry. Telecommunications Policy,13, 319–328.

876 Entrepreneurship Theory and Practice 43(5)
Godoe, H. (2000). Innovation regimes, R&D and radical innovations in telecommunications. Research
Policy,29, 1033–1046.
Gordon, I. R., & McCann, P. (2000). Industrial clusters: Complexes, agglomeration and/or social net-
works? Urban Studies,37, 513–532.
Gould, R. V. (1993). Collective action and network structure. American Sociological Review,58, 182–196.
Hausman, J., Hall, B. H., & Griliches, Z. (1984). Econometric models for count data with an application
to the patents-R & D relationship. Econometrica,52, 909–938.
Hawkins, R. (1999). The rise of consortia in the information and communication technology industries:
Emerging implications for policy. Telecommunications Policy,23, 159–173.
Henderson, J. V. (2003). Marshall’s scale economies. Journal of Urban Economics,53, 1–28.
Hoechle, D. (2007). Robust standard errors for panel regressions with cross-sectional dependence. Stata
Journal,7, 281–312.
Hoover, E. M. (1948). The location of economic activity. New York, NY: McGraw-Hill.
Houghton, S. M., Smith, A. D., & Hood, J. N. (2009). The influence of social capital on strategic choice:
An examination of the effects of external and internal network relationships on strategic complexity.
Journal of Business Research,62, 1255–1261.
Jaffe, A. B., Trajtenberg, M., & Henderson, R. (1993). Geographic localization of knowledge spillovers
as evidenced by patent citations. Quarterly Journal of Economics,108, 577–598.
Jones, C., Hesterly, W. S., & Borgatti, S. P. (1997). A general theory of network governance: Exchange
conditions and social mechanisms. Academy of Management Review,22, 911–945.
Kalnins, A., & Chung, W. (2004). Resource-seeking agglomeration: A study of market entry in the
lodging industry. Strategic Management Journal,25, 689–699.
Klepper, S., & Sleeper, S. (2005). Entry by spinoffs. Management Science,51, 1291–1306.
Knoke, D. (1988). Incentives in collective action organizations. American Sociological Review,53,
311–329.
Krugman, P. (1991). Geography and trade. Cambridge, MA: MIT Press.
Lechner, C., Frankenberger, K., & Floyd, S. W. (2010). Task contingencies in the curvilinear relation-
ships between intergroup networks and initiative performance. Academy of Management Journal,53,
865–889.
Lee, S.-H., Yamakawa, Y., Peng, M. W., & Barney, J. B. (2011). How do bankruptcy laws affect
entrepreneurship development around the world? Journal of Business Venturing,26, 505–520.
Leibovitz, J. (2003). Institutional barriers to associative city-region governance: The politics of institu-
tion-building and economic governance in ‘‘Canada’s technology triangle.’’ Urban Studies,40,
2613–2642.
MacLeod, G., & Goodwin, M. (1999). Space, scale and state strategy: Rethinking urban and regional
governance. Progress in Human Geography,23, 503–527.
Maennig, W., & O
¨lschla
¨ger, M. (2011). Innovative milieux and regional competitiveness: The role of
associations and chambers of commerce and industry in Germany. Regional Studies,45, 441–452.
Maennig, W., O
¨lschla
¨ger, M., & Schmidt-Trenz, H.-J. (2015). Organisations and regional innovative
capability: The case of the chambers of commerce and industry in Germany. Environment and
Planning C: Government and Policy,33, 811–827.
Manzetti, L. (1994). Institutional decay and distributional coalitions in developing countries: The argen-
tine riddle reconsidered. Studies in Comparative International Development,29, 82–114.
Marshall, A. (1920). Principles of economics. London: Macmillan.
Massa, F. G., Helms, W. S., Voronov, M., & Wang, L. (2017). Emotions uncorked: Inspiring evangelism
for the emerging practice of cool-climate winemaking in Ontario. Academy of Management Journal,
60, 461–499.
Maurer, I., & Ebers, M. (2006). Dynamics of social capital and their performance implications: Lessons
from biotechnology start-ups. Administrative Science Quarterly,51, 262–292.
McCann, B. T., & Folta, T. B. (2008). Location matters: Where we have been and where we might go in
agglomeration research. Journal of Management,34, 532–565.

877
Wang and Tan
McKendrick, D. G., Jaffee, J., Carroll, G. R., & Khessina, O. M. (2003). In the bud? Disk
array producers as a (possibly) emergent organizational form. Administrative Science Quarterly,
48, 60–93.
Meyers, L. S., Gamst, G., & Guarino, A. J. (2006). Applied multivariate research: Design and interpret-
ation. Thousand Oaks, CA: Sage Publications.
Miller, D. J., & Acs, Z. J. (2013). Technology commercialization on campus: Twentieth century frame-
works and twenty-first century blind spots. Annals of Regional Science,50, 407–423.
Nadvi, K. (1999). The cutting edge: Collective efficiency and international competitiveness in Pakistan.
Oxford Development Studies,27, 81–107.
Olson, M. (1965). The logic of collective action: Public goods and the theory of groups. Cambridge, MA:
Harvard University Press.
Olson, M. (1982). The rise and decline of nations. New Haven, CT: Yale University Press.
Ostrom, E. (1990). Governing the commons: The evolution of institutions for collective action. Cambridge,
MA: Cambridge University Press.
Parker, R., & Tamaschke, L. (2005). Explaining regional departures from national patterns of industry
specialization: Regional institutions, policies and state coordination. Organization Studies,26,
1787–1807.
Peddibhotla, N. B., & Subramani, M. R. (2007). Contributing to public document repositories: A critical
mass theory perspective. Organization Studies,28, 327–346.
Pe’er, A., & Keil, T. (2013). Are all startups affected similarly by clusters? Agglomeration, competition,
firm heterogeneity, and survival. Journal of Business Venturing,28, 354–372.
Pe’er, A., & Vertinsky, I. (2008). Firm exits as a determinant of new entry: Is there evidence of local
creative destruction? Journal of Business Venturing,23, 280–306.
Pe’er, A., Vertinsky, I., & King, A. (2008). Who enters, where and why? The influence of capabilities and
initial resource endowments on the location choices of de novo enterprises. Strategic Organization,6,
119–149.
Perez-Aleman, P. (2003). A learning-centered view of business associations: Building business-govern-
ment relations for development. Business and Politics,5, 193–213.
Piore, M. J., & Sabel, C. F. (1984). The second industrial divide: Possibilities for prosperity. New York,
NY: Basic Books.
Plummer, L. A. (2010). Spatial dependence in entrepreneurship research: Challenges and methods.
Organizational Research Methods,13, 146–175.
Plummer, L. A., & Acs, Z. J. (2014). Localized competition in the knowledge spillover theory of entre-
preneurship. Journal of Business Venturing,29, 121–136.
Plummer, L. A., & Gilbert, B. A. (2015). The effect of defense agency funding of university research on
regional new venture creation. Strategic Entrepreneurship Journal,9, 136–152.
Plummer, L. A., & Pe’er, A. (2010). The geography of entrepreneurship. In Z. J. Acs & D.
B. Audretsch (Eds.), Handbook of entrepreneurship research (pp. 519–556). New York, NY:
Springer-Verlag.
Porter, M. E. (1998). Clusters and the new economics of competition. Harvard Business Review,76,
77–90.
Pyke, F., Becattini, G., & Sengenberger, W. (1990). Industrial districts and inter-firm cooperation in Italy.
Geneva: International Institute for Labor Studies.
Qian, H., & Acs, Z. J. (2013). An absorptive capacity theory of knowledge spillover entrepreneurship.
Small Business Economics,40, 185–197.
Qian, H., Acs, Z. J., & Stough, R. R. (2013). Regional systems of entrepreneurship: The nexus of human
capital, knowledge and new firm formation. Journal of Economic Geography,13, 559–587.
Raco, M. (1998). Assessing ‘‘institutional thickness’’ in the local context: A comparison of Cardiff and
Sheffield. Environment and Planning A,30, 975–996.
Rauch, A., van Doorn, R., & Hulsink, W. (2014). A qualitative approach to evidence-based entrepre-
neurship: Theoretical considerations and an example involving business clusters. Entrepreneurship
Theory and Practice,38, 333–368.

878 Entrepreneurship Theory and Practice 43(5)
Rosenkopf, L., & Almeida, P. (2003). Overcoming local search through alliances and mobility.
Management Science,49, 751–766.
Rosenthal, S. S., & Strange, W. C. (2003). Geography, industrial organization, and agglomeration.
Review of Economics and Statistics,85, 377–393.
Rotemberg, J. J., & Saloner, G. (2000). Competition and human capital accumulation: A theory of
interregional specialization and trade. Regional Science and Urban Economics,30, 373–404.
Sabel, C. (1994). Learning by monitoring: The institutions of economic development. In N. Smelser &
R. Swedberg (Eds.), The handbook of economic sociology (pp. 137–165). Princeton, NJ: Princeton
University Press.
Saxenian, A. (1994). Regional advantage: Culture and competition in Silicon Valley and Route 128.
Cambridge, MA: Harvard University Press.
Schmitz, H. (1995). Collective efficiency: Growth path for small-scale industry. Journal of Development
Studies,31, 529–566.
Schmitz, H. (1999). Collective efficiency and increasing returns. Cambridge Journal of Economics,23,
465–483.
Scott, A. J. (1994). Variations on the theme of agglomeration and growth: The gem and jewelry industry
in Los Angeles and Bangkok. Geoforum,25, 249–263.
Sievers, T., & Maennig, W. (2006). Die rolle des dritten sektors als determinante im nationalen und
internationalen standortwettbewerb. In H. J. Schmidt-Trenz & R. Stober (Eds.), Jahrbuch recht und
okonomik des dritten sektors 2005/2006 (pp. 276–292). Baden-Baden: Nomos.
Simons, T., & Ingram, P. (2003). Enemies of the state: The interdependence of institutional forms and
the ecology of the Kibbutz, 1910–1997. Administrative Science Quarterly,48, 592–621.
Sorenson, O. (2005). Social networks and the persistence of clusters: Evidence from the computer work-
station industry. In S. Breschi & F. Malerba (Eds.), Clusters, networks, and innovation (pp. 297–316).
Oxford: Oxford University Press.
Sorenson, O., & Audia, P. G. (2000). The social structure of entrepreneurial activity: Geographic con-
centration of footwear production in the United States, 1940–1989. American Journal of Sociology,
106, 424–462.
Sorenson, O., & Stuart, T. E. (2001). Syndication networks and the spatial distribution of venture capital
investments. American Journal of Sociology,106, 1546–1588.
Spigel, B. (2015). The relational organization of entrepreneurial ecosystems. Entrepreneurship Theory
and Practice,41, 49–72.
Statistics Canada. (2001). Information and communication technologies in Canada. Ottawa: Statistics
Canada.
Statistics Canada. (2003). Canada’s journey to an information society. In: Compendium publication on
information and communications technologies (ICTs) in Canada (pp. 3–48). Ottawa: Statistics
Canada.
Steier, L., & Greenwood, R. (2000). Entrepreneurship and the evolution of angel financial networks.
Organization Studies,21, 163–192.
Streeck, W., & Schmitter, P. C. (1985). Private interest government: Beyond market and state. London:
Sage Publications.
Stuart, T., & Sorenson, O. (2003). The geography of opportunity: Spatial heterogeneity in founding rates
and the performance of biotechnology firms. Research Policy,32, 229–253.
Tallman, S., Jenkins, M., Henry, N., & Pinch, S. (2004). Knowledge, clusters, and competitive advan-
tage. Academy of Management Review,29, 258–271.
Tan, D., & Tan, J. (2017). Far from the tree? Do private entrepreneurs agglomerate around public sector
incumbents during economic transition? Organization Science,28, 113–132.
Tan, J. (2006). Growth of industry clusters and innovation: Lessons from Beijing Zhongguancun science
park. Journal of Business Venturing,21, 827–850.
Tan, J., Zhang, H., & Wang, L. (2015). Network closure or structural hole? The conditioning effects of
network-level social capital on innovation performance. Entrepreneurship Theory and Practice,39,
1189–1212.

879
Wang and Tan
Tura, T., & Harmaakorpi, V. (2005). Social capital in building regional innovative capability. Regional
Studies,39, 1111–1125.
Uzzi, B. (1996). The sources and consequences of embeddedness for the economic performance of
organizations: The network effect. American Sociological Review,61, 674–698.
Uzzi, B. (1997). Social structure and competition in interfirm networks: The paradox of embeddedness.
Administrative Science Quarterly,42, 35–67.
von Thu
¨nen, J. H. (1826). The isolated state. Oxford: Pergamon Press.
Wang, L. (2017). Time and space in business: Dynamic geographic concentration and localized industry
life cycle. Journal of Strategy and Management,10, 1–28.
Wang, L., Madhok, A., & Li, S. X. (2014). Agglomeration and clustering over the industry life cycle:
Toward a dynamic model of geographic concentration. Strategic Management Journal,35, 995–1012.
Warf, B. (2001). Global dimensions of U.S. legal services. Professional Geographer,53, 398–406.
Wasko, M. M., & Faraj, S. (2005). Why should I share? Examining social capital and knowledge con-
tribution in electronic networks of practice. MIS Quarterly,29, 35–57.
Weber, A. (1929). Theory of the location of industries. Chicago, IL: University of Chicago Press.
Wiertz, C., & de Ruyter, K. (2007). Beyond the call of duty: Why customers contribute to firm-hosted
commercial online communities. Organization Studies,28, 347–376.
Wilhoit, E. D., & Kisselburgh, L. G. (2015). Collective action without organization: The material con-
stitution of bike commuters as collective. Organization Studies,36, 573–592.
Yang, H., Phelps, C., & Steensma, H. K. (2010). Learning from what others have learned from you: The
effects of knowledge spillovers on originating firms. Academy of Management Journal,53, 371–389.
Author Biographies
Liang Wang is an associate professor of entrepreneurship, innovation and strategy in the
School of Management at University of San Francisco. He received his PhD from the
Schulich School of Business at York University. His research interests include firm location
strategy, local institutions and regional innovation, with a focus on China’s innovation
ecosystem.
Justin Tan is a Professor and the Newmont Chair in Business Strategy at the Schulich School
of Business at York University. His research in strategy, entrepreneurship and innovation has
been supported by grants from the Ford Foundation, Social Science and Humanities
Research Council of Canada, and National Science Foundation of China, among others.