Technological Forecasting & Social Change 119 (2017) 294309

Contents lists available at ScienceDirect

Technological Forecasting & Social Change

The impact of Triple Helix agents on entrepreneurial innovations'

performance: An inside look at enterprises located in an
emerging economy
Maribel Guerrero a,, David Urbano b
a
Department of Strategy, Deusto Business School, University of Deusto, P. Mundaiz, 50 20012 Donostia-San Sebastin, Spain
b
Department of Business, Universitat Autnoma de Barcelona, Building B, 08193 Barcelona, Spain

a r t i c l e i n f o a b s t r a c t

Article history: During the past few decades, the congurations of new knowledge-intensive environments have required fertile
Received 29 January 2016 settings for innovative and entrepreneurial activities. In these environments, Triple Helix has been operational-
Received in revised form 10 June 2016 ized in different ways, spaces, and contexts where those agents are transforming their roles in the development
Accepted 13 June 2016
and strengthening of national innovation and entrepreneurial ecosystems. As a consequence, the phenomenon of
Available online 19 June 2016
entrepreneurial innovations emerged from enterprises with an entrepreneurial or high-growth orientation that
JEL classication:
collaborate with Triple Helix agents generating economic benets and spillover effects. In emerging economies,
L26 the available literature about innovation and entrepreneurship is limited to explore the determinants of innova-
I23 tion performance as well as innovation constrains. Based on this argument and diverging from prior research, this
A23 research tries to provide a better understanding about the inuence of Triple Helix agents on entrepreneurial in-
novations' performance of enterprises located in emerging economies. In particular, we analyze the effects pro-
Keywords: duced by the links of enterprises with other enterprises, universities and government on their innovation
University-enterprise-government performance (e.g., access to knowledge/technology, sources of funding, government subsidies), as well as, the
Triple Helix
moderation effects generated when those enterprises have a high-growth orientation (e.g., distinction of enter-
Innovation performance
prises that develop entrepreneurial innovations or traditional innovations). To achieve this aim we look inside at
High-growth orientation
Entrepreneurial innovation the case of Mexico because is an emerging economy that during the last two decades has facing a transition to a
Emerging economy knowledge-based economy. Using a cross-section dataset of 19,188 Mexican enterprises interviewed in the pe-
Mexico riod of 2006 to 2012, we tested our proposed conceptual model with a Tobit regression. Our study provides in-
teresting implications for the main actors involved in the Mexican Science, Technology and Innovation System,
as well as, contributes about the debate of the impact of enterprises-university-government linkages on entre-
preneurial innovations from diverse perspectives and research elds (e.g., open innovation, knowledge transfer,
high-growth entrepreneurship, academic entrepreneurship, public entrepreneurship).
2016 Elsevier Inc. All rights reserved.

1. Introduction entrepreneurially, etc.) in different spaces (e.g., global, national, region-

During the past few decades, the congurations of new knowledge-
intensive environments have required fertile settings for innovative and
entrepreneurial activities. Both types of activities play a crucial role in
the economy, and many studies have examined the factors that inu-
ence these activities. Despite innovation and entrepreneurship being
multidimensional processes, empirical studies continue to employ
models that presume that these phenomena occur at a single point in
time (McMullen and Dimov, 2013). Those facts explain why the Triple
Helix concept has been operationalized in different ways (e.g., with/
without government intervention, closed/opened, administrated/
Corresponding author.
E-mail addresses: maribel.guerrero@deusto.es (M. Guerrero), david.urbano@uab.cat
(D. Urbano).
al, local) and in different contexts (e.g., organizational, institutional,
technological, social, etc.). Because of this diversity, there has been
growing interest in the study of how Triple Helix organizations trans-
form their roles and practices in the development and strengthening
of national innovation and entrepreneurial ecosystems (Etzkowitz and
Leydesdorff, 2000; Leydesdorff and Ivanova, in press). Therefore,
how different agents operate, collaborate, make decisions, identify ben-
ets, or transform their roles is still an interesting research area
(Cunningham and Link, 2015; Cunningham et al., 2014) with special at-
tention in emerging economies where are introducing conventional
strategies during their transformation into knowledge economies
(Wright et al., 2005). In this regards, in both temporal and spatial con-
texts, entrepreneurial innovation is the result of a variety of elements
that compare the attributes of national innovation systems, entrepre-
neurship, contextual inuences and the main benets for the actors

http://dx.doi.org/10.1016/j.techfore.2016.06.015
0040-1625/ 2016 Elsevier Inc. All rights reserved.
 M. Guerrero, D. Urbano / Technological Forecasting & Social Change 119 (2017) 294309
involved in this process (Autio et al., 2014). However, only a few studies
have worked in the intersection of entrepreneurship and innovation lit-
erature (Zahra and Wright, 2011).
Applying these perspectives, this research focused on the inuence
of Triple Helix agents on the performance produced by entrepreneurial
innovations dened such as the economic benets obtained from inno-
vations by enterprises with an entrepreneurial orientation (i.e., Ireland
et al., 2009 operationalize this orientation in terms of risk-taking, pro-
activeness, and innovativeness). Given the difculties to operationalize
the entrepreneurial enterprises and based on the arguments exposed by
Autio et al. (2014), we adapted the high-growth orientation criteria
(Audretsch, 2012; OECD, 2010) introducing the basis of innovations'
performance (Klomp and van Leeuwen, 2001; Laursen and Salter,
2006; Stam and Elfring, 2008; Thornhill, 2006; Tsai and Wang, 2009;
Wiklund and Shepherd, 2005; Yamin and Otto, 2004; Zott and Amit,
2007) to identify enterprises that develop entrepreneurial innovations
and obtained benets from those type of innovations. More concretely,
we apply this distinction to enterprises that development innovations
and that have an orientation annualized to growth N 20% - in both in
terms of number of employees and in terms of turnover - per annum
over a three year period. In this scenario, micro-foundations of Triple
Helix organizations are explored through the main actors and actions
involved in the development of entrepreneurial innovations' perfor-
mance. Based on these arguments and diverging from prior research,
our study offers three contributions.
Firstly, linking innovation performance and the access of technology/
knowledge from R&D collaborations, traditional literature has largely fo-
cused on knowledge transfer and innovation, devoting particular atten-
tion to performance issues. It is widely agreed that innovation plays a
major role in enterprises that try to survive in the market, especially
by complementing their resources and capabilities required to improve
their performance (Ireland et al., 2009). Empirical studies found that in-
novation practices produced positive effect on enterprise performance
(Zott and Amit, 2007). However, research has been strictly focusing on
R&D cooperation with competing rms, and has paid little attention to
R&D collaboration with universities, or with rms that are not direct
competitors (Belderbos et al., 2004). Exploring R&D collaboration al-
lows understanding the role of internal/external agents involved the in-
novation process (Huizingh, 2011). In this point of view, Belderbos et al.
(2004) argue that any type of cooperation implies a decision to choose
the type of partner when the enterprise requires knowledge for the in-
novation process (scientic), or when exhibit faster technological and
product development (commercial or intrapreneurial) or when is
looking for a source of competitive advantage to have long lasting ef-
fects on rm performance (mixed). Even than a number of empirical
studies have found a positive impact of engaging in R&D cooperation
on innovation performance (Klomp and van Leeuwen, 2001; Faems
et al., 2005), others studies have obtained ambiguous results (Hall,
2005). An explanation is a substantial heterogeneity in the determi-
nants to establish R&D collaborations with different partners and their
impact on innovation performance. It could be related to the measures
used to analyze performance but also with the enterprise incentives to
cooperate such as a high-growth orientation (Van Leeuwen and
Klomp, 2001). However, the enterprises' decision about the selection
of a Triple Helix collaborator has been not explored in terms of the dis-
tinction of entrepreneurial innovations as well as in the context of
emerging economies. In this sense, this paper expects to contribute to
the open innovation literature debate about the necessity of exploring
the main effects produced by enterprises' selection of internal/external
collaborations for the development of entrepreneurial innovations
(Leydesdorff and Ivanova, in press), as well as, to the debate of the im-
plementation of new open innovation models to capture the benets
of collaborations among Triple Helix agents (Chesbrough, 2012).
Secondly, linking innovation performance and the access of funds from
R&D collaborations, nancial literature recognized a gap in the analysis
and use of internal funds for innovation beyond the favorable tax
295
treatment experimented in some many countries (Hall and Lerner,
2010). It provides an interesting opportunity to explore the role of Tri-
ple Helix agents that provides innovation funds. In this regards, the in-
novation literature also recognizes that the main innovation constrain
that faced enterprises located in emerging economies (e.g., Brazil,
Mexico, Central America) is associated to the access to nancial sources
based on the scant and insufcient bank loads or venture capital mar-
kets (Santiago et al., 2016). Moreover, a common practice in emerging
economies is that policymakers establish the requirement of the devel-
opment of enterprise-university partnership or commercial partnership
to receive some innovation subsidies (Cohen et al., 2002; Czarnitzki
et al., 2007). In this sense, this paper expects try to contribute to the -
nancial literature debate about the gap in the analysis and in the use of
internal/external source of funds for nancing entrepreneurial innova-
tions in emerging economies (Hall and Lerner, 2010; p. 40), as well as,
the impact of subsidies on enterprises with an entrepreneurial and in-
novative behavior and that supported their innovations by other
sources (Czarnitzki and Lopes-Bento, 2013; p.31). In this scenario,
universities play a relevant role such as partner that provides new tech-
nology/knowledge as well as the access to government subsidies
(Audretsch, 2014). However, the decision to select public/private inno-
vation funds will depends on the opportunity cost that implies each
fund for enterprise strategies (Hall, 2005; Moreno and Casillas, 2007).
In this regard, the literature about the impacts of universities in entre-
preneurial and innovation processes is limited. In this sense, this
paper expects to contribute to knowledge transfer literature debate
about to obtain a view or performance indicators of the impact of
university-enterprises collaboration on entrepreneurial innovations
(Grimaldi et al., 2011; p.1053; Guerrero et al., 2015).
Thirdly, linking innovation performance and the socio-economic con-
text, entrepreneurship literature also recognizes that any entrepreneur-
ial and innovative activity is inuenced by contextual conditions
(Thornton et al., 2011). In competitive and uncertain environments, in-
novation is a necessary feature for an enterprise to develop an organiza-
tional strategy, bring innovations to the market, satisfy customer needs,
and ensure survival (Boyd, 1991). We can observe an inverse relation-
ship because innovation contributes to regional development but at
the same time regional conditions affect the development of innovation.
According to this argument, the major policy challenge is an adequate
conguration of Innovation and Entrepreneurship Ecosystems (Autio
et al., 2014). However, the congurations and the benets among the
links of entrepreneurial and innovative enterprises with the other Triple
Helix agents are not well understood; particularly, in emerging econo-
mies that experiment complex socio-economic conditions (Edwards,
2001; Meyer et al., 2009; Wright et al., 2005) liked to the lack quality
of their institutions (Sobel, 2008). In this sense, this paper expects to
contribute to the entrepreneurship literature debate about the national
systems of innovation, entrepreneurship, and entrepreneurial innova-
tions; more concretely, about the impact of context for entrepreneurial
innovation and enterprise growth in emerging economies (Autio et al.,
2014; p. 1089 and p.1104).
Based on these arguments, the purpose of this research is to provide
a better understanding about the inuence of Triple Helix agents on en-
trepreneurial innovations' performance of enterprises located in emerg-
ing economies. More concretely, we analyze the effects produced by the
links of enterprises with other enterprises, universities and government
on their innovation performance (e.g., access to knowledge/technology,
sources of funding, government subsidies), as well as, the moderation
effects generated when those enterprises have a high-growth orienta-
tion (e.g., distinction of enterprises that develop entrepreneurial inno-
vations or traditional innovations). We selected the case of Mexico to
understand the ways, the spaces, and the contexts associated with
Triple Helix agents that affect entrepreneurial innovations' performance
of enterprises located in emerging economies. Authors such as
Hoskisson et al. (2000); Kuznetsov and Dahlman (2008) and Solleiro
and Castan (2005) argue that Mexico is facing a transition to a
296 M. Guerrero, D. Urbano / Technological Forecasting & Social Change 119 (2017) 294309
knowledge-based economy and has been classied by an emerging
economy during the last two decades. For one side, Mexico has evi-
denced positive economic signs such as a GDP growth rate of 30%, the
consolidation of N40 free trade agreements around the world, becoming
the major manufacturers of electronics sold in the United States
(Gobierno de los Estados Unidos Mexicanos, 2013). For other side,
Mexico also faced negative signs associated to crime rates that climbed
sharply (Heinle et al., 2015). Under these socioeconomic conditions,
Mexican government has developed/implemented strategies to rein-
force the National Science and Technology Law connecting Triple
Helix agents, encouraring a strong public/private funding for basic/ap-
plied research, technology and innovation, investing in qualied
human resources and infraestructure, and decentralizing scientic and
technological activities to regional governments (Diario Ocial, 2014).
To achieve our objective, using a cross-section dataset from the 2006,
2008, 2010 and 2012 Research and Technological Development Survey,
we tested our conceptual model with a Tobit regression in a sample of
19,188 Mexican enterprises. In this sense, our results provide interest-
ing implications for the main actors involved in the Mexican Science,
Technology and Innovation System, as well as, contribute to the debate
of these issues in several academic elds (e.g., innovation, nance, and
entrepreneurship).
This paper is organized as follows. Section 2 presents the litera-
ture review about innovation performance linked to Triple Helix's
agents. Also, this section presents the hypotheses and conceptual
model proposed. Section 3 describes the main elements that allow
understanding a National Innovation System of an emerging econo-
my. Section 4 integrates the methodological design, including the
data collection and statistical analysis. Section 5 evidenced the re-
sults obtained and Section 6 presents the discussion of those results
in the light of previous studies. Finally, Section 7 summarizes the
concluding remarks, limitations, implications, and avenues for fur-
ther research.
2. Linking triple helix agents and entrepreneurial innovations' per-
formance of enterprises located in emerging economies
Innovation literature suggests that an analysis of different types
of cooperation strategies should take into account the different pos-
sible aims of collaborative R&D efforts. By nature and more intensive
in emerging economies, enterprises face obstacles to innovation
such as resource scarcity (technological and human capabilities)
and a lack of industry/market information (sectorial analysis, mar-
keting and distribution strategies) that inuence on their innovation
performance (Santiago et al., 2016). In these scenarios, enterprises
try to create appropriate added value for their products, service, or
process through different types of collaborations (De Fuentes and
Dutrenit, 2012; De Fuentes and Dutrnit, 2016). More concretely, en-
terprises with an entrepreneurial or high-growth orientation make
efforts to develop collaborative agreements that allow them the ac-
cess to external knowledge and technological resources to generate
entrepreneurial innovations and obtain a higher performance
(Autio et al., 2014). However, each type of collaboration with Triple
Helix agents involves asymmetry of information, asset specicity,
opportunistic behaviors, and uncertainty in the appropriation of
the benets produced (Dutrnit et al., 2010). Therefore, enterprises
with an entrepreneurial or high-growth orientation will evaluate
different opportunities and risks associated with each type of collab-
oration to determine the collaboration's feasibility. In this section,
we analyze the effects produced by the links (e.g., access to knowl-
edge/technology, sources of funding, government subsidies) of en-
terprises with Triple Helix agents on their innovation performance,
as well as, the moderation effects generated when those enterprises
have a high-growth orientation (e.g., distinction of enterprises that
develop entrepreneurial innovations or traditional innovations).
2.1. Access to knowledge/technology
The selection of collaboration partners depends of enterprises' strat-
egies, their resources and their growth orientation. If the enterprise mo-
tivation is associated to access to knowledge and technologies, the
literature distinguish four types of collaboration with Triple Helix
agents. First, an intrapreneurial collaboration with their parent rm is
always accompanied by an organizational learning, an umbrella of tech-
nological and entrepreneurial competences (Drejer, 2000; Wiklund and
Shepherd, 2005). Based on the nature of this type of collaboration, even
that the evidence is limited in emerging economies, the positive effect
on innovation performance of the knowledge transfer from the parent
rms to their incumbent enterprises is similar in both advanced and
emerging economies (Klepper, 2001). Second, an enterprise collaborates
with other enterprise when both are looking for competitive advan-
tages across their industrial value chain and for obtaining commerciali-
zation benets with fast returns on investment (Chung et al., 2003; Tsai
and Wang, 2009). In the case of emerging economies, Crespi and Ziga
(2012) found that knowledge sharing through this type of collaboration
have positive effects on Mexican enterprises' innovation performance.
Third, an enterprise is more likely to choose universities partners
when are an important source of new technological knowledge for the
innovation process and when are involved in intensive sectors
(Audretsch, 2014). In the context of emerging economies, exacts studies
have evidenced the positive benets behind knowledge transfer be-
tween universities/research centers and enterprises (Dutrnit et al.,
2010); particularly, when those are related to long-term benets (De
Fuentes and Dutrenit, 2012). Fourth, mixed collaboration supposes a
great benet for enterprises, supporting new innovation patterns and
encouraging novel product innovations. Therefore, this type of collabo-
ration enhances innovation with a variety of knowledge, technological/
commercialization resources, and skill endowments (Yamin and Otto,
2004). In emerging economies, several linkages with universities/re-
search centers or the private sector reduces several constrains to per-
form innovation (Santiago et al., 2016).
Concerning entrepreneurial innovations, the involvement of multi-
actors across the innovation process produces higher effects on the per-
formance of enterprises with an entrepreneurial or high-growth orien-
tation (Autio et al., 2014). Even than empirical have found ambiguous
results on the effect produced by different cooperation types, the major-
ity of extant studies have found a positive impact of engaging in R&D co-
operation on innovation performance measured by sales of innovative
products (Delmar et al., 2003; Faems et al., 2005; Klomp and van
Leeuwen, 2001; Laursen and Salter, 2006; Thornhill, 2006). A possible
explanation about those ambiguous results may be partly attributed to
the difculties in allowing for an appropriate measure or to the unob-
served rm characteristics that impact on its incentives to cooperate
and their innovative output (Van Leeuwen and Klomp, 2001). In this
line, corporate entrepreneurship literature links these facts with an en-
trepreneurial orientation that is strongly related with entrepreneurial
innovations' performance (Stam and Elfring, 2008; Wiklund and
Shepherd, 2005). Entrepreneurship literature also emphasis that high-
growth (or entrepreneurial according to denition adopted in this
paper) enterprises, that has the primary goals of protability and
growth organically, manage strategic collaborative practices to create
socio-economic value nding unserved niches in the market (Friar
and Meyer, 2003) as well as focusing on radical entrepreneurial innova-
tion based on the co-creation with multiple agents (Autio et al., 2014).
Applying these ideas, the effect of knowledge transfer from Triple
Helix agents on innovation performance is moderated by enterprises
with high-growth orientation characterized by generate entrepreneur-
ial innovations. Based on these arguments, we hypothesize that:
Hypothesis 1. Enterprise's cooperation with Triple Helix agents to ac-
cess knowledge resources produces a positive effect on its innovation
performance.
 M. Guerrero, D. Urbano / Technological Forecasting & Social Change 119 (2017) 294309
Hypothesis 1a. The positive effect that produces the enterprise's coopera-
tion with Triple Helix agents to access knowledge resources on innovation
performance is reinforced when it has a high-growth orientation.
2.2. Access to funds for innovation activities
It is a widely held view that research and development (R&D) and
innovative activities are difcult to nance in a freely competitive
market place (it happens also in emerging economies). According
to Hall (2005), R&D has a number of characteristics that make it dif-
ferent from an ordinary investment. For example, it is characterized
by a higher percentage that is spending in wages of qualied em-
ployees. It affects the return rate from innovation as well as the de-
gree of uncertainty associated to innovation outputs/outcomes.
Under these conditions, enterprises need to rely on external sources
of innovation to overcome internal barriers and reducing their cost
of exploiting entrepreneurial innovations (Autio et al., 2014). Ac-
cording to Chesbrough (2012), adapting business model is a com-
mon way to connect value creation with an innovation strategy and
convert it into prots. Prior empirical studies found that a dynamic
business innovation model generates sustained performance (Zott
and Amit, 2007). In this sense, the collaboration with commercial
agents (e.g., other innovative enterprises) involved in the National
Innovation System provides alternatives to sharing the cost of inno-
vation and generating a positive impact on their innovation perfor-
mance (Landstrm et al., 2015).
Similarly, for entrepreneurial enterprises, the resource con-
straints and unpredictable market conditions create signicant chal-
lenges to access to external sources of funding for innovation
activities, therefore, these enterprises are supported under the um-
brella of their parents rms (Ireland et al., 2009; Stam and Elfring,
2008; Wiklund and Shepherd, 2005). Additionally, within the inno-
vation and entrepreneurial system, there are other private actors
like nancial organizations or venture capitalist that could provide
several alternatives for innovation funding. Traditionally, the access
to nancial organizations represents a higher cost/risk of nancing
new investment by debt than by internal sources such as reinvesting
earnings (Auerbach, 1984). It could be more complicated in emerg-
ing economies by the level of volatility and uncertainty (Hoskisson
et al., 2000; Wright et al., 2005) or by the obstacles that enterprises
address to access to credits within an embryonic nancial system
(Surez and Oliva, 2005). In these conditions, there is a lower prob-
ability that an entrepreneurial enterprise nancing its innovation
activities with debt (Moreno and Casillas, 2007). Based on these ar-
guments, retained earnings in the R&D investment decision takes
an important role for create value and positive signal of performance.
It is good reason to think that positive cash ow may be more
important for R&D than for ordinary investment (Himmelberg and
Petersen, 1994). In developed economies, entrepreneurial innova-
tion is based on a multilevel and multidimensional combination of
public/private sources of funds (Autio et al., 2014). Studies in emerg-
ing economies provide empirical insights into the obstacles to inno-
vation and the majority of them have an emphasis on the nancial
constraints that affect innovation in SMEs. For example, lvarez
and Crespi (2015) studied constraints on credit for investment in in-
novation and how it affects innovation performance of Chilean rms.
Similar than developing countries, poorly supportive nancial envi-
ronments reect structural constraints in local venture capital mar-
kets, and limited access to bank loans for technology-based SMEs
(Kuznetsov and Dahlman, 2008). Therefore, we argue that entrepre-
neurial enterprises are more likely to nancing their innovations by in-
ternal sources than external sources. Applying these ideas, the effect of
the access to funding from Triple Helix agents on innovation perfor-
mance is moderated by enterprises with high-growth orientation
297
characterized by generate entrepreneurial innovations. Based on these
arguments, we hypothesize that:
Hypothesis 2. The enterprise's access to funding from Triple Helix agents
produces a positive effect on its innovation performance.
Hypothesis 2a. The positive effect that produces the enterprise's access to
funding from Triple Helix agents on innovation performance is lower when
it has a high-growth orientation.
2.3. Access to government subsidies
According to the National Innovations Systems literature, govern-
ment is viewed as the dominant agent because it provided funds for uni-
versities and industries (McAdam et al., 2012) and thus orchestrated the
content and structure of the university business model (Miller et al.,
2014). However, current studies recognized that participation of gov-
ernment is more like other agent in Triple Helix instead of only as the
funder, as government now also promote regulations, programmes, ac-
tions, and participation in different stages of the university-enterprise-
government link (Crdenas et al., 2012; Padilla-Prez and Gaudin,
2014). Most governments also include a new variant: promoting and
ensuring the well-being of society and regional development through
the collaborations with universities and industries (Audretsch, 2014;
Guerrero et al., 2015; Kenney and Goe, 2004; Urbano and Guerrero,
2013; Wang and Altinay, 2012). In this perspective, numerous
policymakers have encouraged universities and research centers to
make their science and engineering more relevant to industry's needs
(Cohen et al., 2002, p. 2). For example, in many developed countries,
collaborative research is subsidized by public policy programmes that
provide resources for collaborative projects involving universities and
enterprises. Some examples are the framework programmes of the
European Commission (Caloghirou et al., 2001); the USA federal-
funded schemes such as the Advanced Technology Programme (Hall
et al., 2001); funding instruments provided by research councils, gov-
ernment departments in the UK (Howells et al., 2003); university-
industry projects within federal programmes in Germany (Almus and
Czarnitzki, 2003; Czarnitzki et al., 2007); and among some emerging
economies (Padilla-Prez and Gaudin, 2014). In this perspective, the ac-
cess to public subsidies allows enterprises to promote R&D activities
and develop cooperation with other agents (Cassiman and Veugelers,
2002).
A common practice observed in emerging economies is that govern-
ments require the development of enterprise-university partnerships or
commercial partnerships in order to give innovation subsidies (Cohen
et al., 2002). With this requirement, governments try to ensure knowl-
edge transfer and development of successful innovations (Boschma,
2005). In this scenario, public subsidies also help enterprises to nd
new partners, reach R&D cooperation agreements, and increment of in-
novation practices. Nevertheless, the performance measures do not nec-
essarily reect the whole ranges of anticipated benets (Perkmann and
Walsh, 2007). These practices have a certain degree of opportunistic
even than evidence suggests that these collaborations evolve over
time (van de Vrande et al., 2009). Empirical studies in European coun-
tries (Belgium and Germany) found that innovation subsidies do not
completely displace private expenditure on R&D (that is, they are
additional) but have a productive impact on enterprise performance
(Aerts and Czarnitzki, 2004; Aerts and Schmidt, 2008; Czarnitzki and
Lopes-Bento, 2013). In large Latin American economies, Hall and
Mafoli (2008) found similar set of results and reach a more nuanced
conclusion. In this perspective, we believe that enterprises with an en-
trepreneurial orientation located in emerging economies access to inno-
vation subsidies but with lower intensity/frequency than enterprises
without a high-growth orientation. The main explanation is associated
to the opportunity cost in terms of time/money that demands the ad-
ministrative governmental processes. However, when public funds
298 M. Guerrero, D. Urbano / Technological Forecasting & Social Change 119 (2017) 294309
seem insufcient to offset risk-aversion and induce investments in in-
novation; private funding is scant, difcult to access or excessively ex-
pensive (Dutrnit et al., 2010). Therefore, the contribution of Triple
Helix agents to access to government subsidies will be higher for
those enterprises with a non-high-growth orientation. Based on these
arguments, we hypothesize that:
Hypothesis 3. Enterprise's collaboration with Triple Helix agents to ac-
cess government subsidies produces a positive effect on its innovation
performance.
Hypothesis 3a. The positive effect that produces the enterprise's collabo-
ration with Triple Helix agents to access government subsidies on innova-
tion performance is reinforced when it has not a high-growth orientation.
2.4. The socio-economic context of emerging economies
According to the Organization for Economic Cooperation and Devel-
opment (OECD, 2012), macroeconomic conditions such as rise in unem-
ployment rate, limited access to nancing, a reduction in the demand
for products, and a decline in gross domestic product could affect
(positively or negatively) innovation performance and investments in
innovation mechanisms (reduced public nancing of R&D, a sluggish
evolution of demand). In this point of view, Thornton et al. (2011) ar-
gues that any innovative/entrepreneurial activity is constrained by
formal (regulations, norms, programmes, etc.) and informal factors
(culture, attitudes, etc.). Within each emerging economy, policy makers
usually try to translate successful formulas applied by developed econ-
omies, such as fostering entrepreneurial ecosystems to promote innova-
tion and entrepreneurship as the best transitional instrument (Cohen
et al., 2002; Edwards, 2001). However, the lack of quality of certain in-
stitutions inuences creative individuals/organizations to manipulate
political/legal conditions to be involved in unproductive entrepreneur-
ship (Sobel, 2008). For instance, organized crime participates actively
in this illegality, at times aided by intimidation, at times sustained by
Fig. 1. Proposed conceptual model. Source: Authors.
the interests of legitimate actors who play a de facto role as partners
(Kan, 2011). In these conditions, higher levels of criminality produces
a negative effect on any entrepreneurial innovation activity but it
could be higher for entrepreneurial enterprises that are the focus of ex-
tortion from the perspective of criminal groups (Smith, 2009). Based on
these arguments, we propose the following hypotheses:
Hypothesis 4. The socio-economic context of emerging economies pro-
duces a negative effect on innovation performance.
Hypothesis 4a. The negative effect that produces the socio-economic con-
text of emerging economies on innovation performance is reinforced when
enterprises have a high-growth orientation.
2.5. Proposed conceptual model
In summary, Fig. 1 shows the proposed conceptual model and hy-
potheses that try to provide a better understanding about the inuence
produced by the links (access to knowledge/technology, sources of
funding, government subsidies) of enterprises with Triple Helix Agents
(enterprise-university-government) on innovation performance in
emerging economies. In addition, we explored how this inuence is
moderated by the level of high-growth orientation of those enterprises
(Audretsch, 2012; OECD, 2010). Particularly, this moderation allows
doing the distinction between the enterprises that develop entrepre-
neurial innovations and the enterprises that develop traditional innova-
tions (Autio et al., 2014).
3. Looking inside the National Innovation System of an emerging
economy
3.1. Characterizing an emerging economy: Mexico
Emerging markets, also known as emerging or transitional econo-
mies, are assuming an increasingly prominent position in the world
 M. Guerrero, D. Urbano / Technological Forecasting & Social Change 119 (2017) 294309
economy (Hoskisson et al., 2000). These economies comprise countries1
with a rapid pace of development and government policies that favor
economic liberalization and the transition from centrally planned econ-
omies (Wright et al., 2005). For instance, there are nations that invest in
more productive capacity and adopt a free market or mixed economy to
move towards an innovative economy (Edwards, 2001; Meyer et al.,
2009). For this study, we selected Mexico as an example of an emerging
economy, as classied by the International Monetary Fund.2 Ofcially
called the United Mexican States, Mexico is a federal republic compris-
ing 31 regions and a capital that is the Federal District. It has a popula-
tion of 122 million with an average life expectancy of 77 years. Since
the rst editions of the Global Competitiveness Index, Mexico is classi-
ed as an efciency-driven (Porter and Schwab, 2008, p. 9). This means
that the country's main advantage comes from producing more ad-
vanced products and services highly efciently. Heavy investment in ef-
cient infrastructure, business-friendly government administration,
strong investment incentives, improving skills, and better access to in-
vestment capital allow for major improvements in productivity. The
trade of products and services and human capital movements between
countries has enormous effects on an economy's productivity and ef-
ciency, especially for efciency-driven countries such as Mexico. How-
ever, the challenge is to reinforce the ability to produce innovative
products and services at the global technology frontier using the most
advanced methods to become the dominant source of competitive ad-
vantage (Solleiro and Castan, 2005); in other words, a transformation
of an efciency-driven economy to an innovation-driven economy. In the
past two decades, Mexico has made progress towards achieving macro-
economic stability. For instance, Mexico also faced a process of political
transition from a one-party system that had dominated for 70 years
under the Partido Revolucionario Institucional (PRI) to a multiparty sys-
tem in 2000 (Bunker, 2013). As a result, Mexico had two Partido Accin
Nacional (PAN) administrations, under Vincente Fox (20002006) and
Felipe Caldern (20072012). Aligned with the purpose of this research,
we paid attention to the national development plan of the sexennial ad-
ministration of Felipe Caldern integrated by ve pillars: (i) law and se-
curity, (ii) competitive economy and generation of employment, (iii)
equal opportunity for priority groups (gender, ethnic, etc.), (iv) envi-
ronmental sustainability, and (v) effective democracy and responsible
foreign policy (Gobierno de los Estados Unidos Mexicanos, 2006).
More concretely, we focused on the strategies associated with pillar
two of this administration: creating a competitive economy and gener-
ating employment. To achieve this aim, the Ministry of Economy's
strategies were oriented towards reinforcing the nancial system
and implementing productivity and competitiveness programmes
based on technological innovation. It also aligned with the Ministry
of Communication's reinforcement of infrastructure (e.g., telecommuni-
cations, electricity, road/sea/air transport, etc.) and the Ministry of
Security's improving national security strategies.
At economic level, Fig. 2 shows the evolution of the Mexican GDP dur-
ing the Caldern administration in comparison with other Latin-
American emerging economies. Concretely, the World Bank Indicators3
show that Mexico's gross domestic product (GDP) was almost $0,967
trillion in 2006 and 1.186 trillion in 2012, less than its NAFTA partners:
1
According to Hoskisson et al. (2000), some countries identied as emerging or transi-
tional economies are (in alphabetical order): Albania, Argentina, Armenia, Azerbaijan,
Bangladesh, Belarus, Bosnia and Herzegovina, Botswana, Brazil, Bulgaria, Chile, China,
Colombia, Cote d'Ivoire, Croatia, Czech Republic, Ecuador, Egypt, Estonia, Georgia, Ghana,
Greece, Hungary, India, Indonesia, Israel, Jamaica, Jordan, Kazakhstan, Kenya, Korea,
Kyrgyzstan, Latvia, Lithuania, Macedonia, Malaysia, Mauritius, Mexico, Moldova,
Morocco, Nigeria, Pakistan, Peru, Philippines, Poland, Portugal, Romania, Russia, Saudi
Arabia, Slovakia, Slovenia, South Africa, Sri Lanka, Taiwan, Tajikistan, Thailand, Trinidad
and Tobago, Tunisia, Turkey, Turkmenistan, Ukraine, Uzbekistan, Venezuela, and
Zimbabwe.
2
For further details, visit the website [http://www.imf.org/external/pubs/ft/weo/2008/
01/weodata/groups.htm#oem], last access October 2015.
3
For further details, visit the website [http://data.worldbank.org/country/mexico], last
access October 2015.
299
the United States ($13 trillion to $16 trillion, respectively) and Canada
($1.310 trillion to $1.832 trillion, , respectively). Mexico's 200612
GDP growth rate was 30%. Regarding Mexican quality of life, the GDP
per capita was $8280 in 2006 and $9721 in 2012.
Regarding to the openness to international trade (Fig. 3), Mexico
started this process when entry to the General Agreement on Tariffs
and Trade (GATT) in 1986. As a result, Mexico established commercial
agreements to exchange capital, goods and services across N 45 interna-
tional borders.
United States, Canada and Mexico signed the North American Free
Trade Agreement (NAFTA) in 1993. After more than twenty years of
NAFTA, total trade among the three countries was six fold. During
19922013 periods, foreign direct investments (FDI) were from
United States, Netherland, Spain, Canada, Belgium, United Kingdom,
Switzerland, Germany, Japan, among others. Currently, Mexico exports
one-quarter of its oil and is one of the major manufacturers of electron-
ics sold in the United States. For instance, 83.6% of 2012 exportations
were responsible of Mexican manufacture as follows: 4% agro-
industry, 6% mining, 14% machinery, 23% electronics industry, 29% auto-
motive industry, and 24% others (Gobierno de los Estados Unidos
Mexicanos, 2013). Nowadays, Mexico has celebrated N40 free trade
agreements with different countries around the world.
Nevertheless, at social level, the growing criminal organizations in-
creased the violence/crime levels within Mexico (Kan, 2011). According
to Heinle et al. (2015, p. 2), at the end of the Caldern Administration,
the levels of violence in Mexico were lower than other Latin-American
countries such as Honduras, Venezuela, Belize, Salvador, Guatemala,
Colombia, Brazil, Dominican Republic, among others. However, after
Caldern's rst year, the number of intentional homicides registered
by the Mexico's National Institute of Statistics, Geography and Informa-
tion (INEGI) climbed sharply, with year-over-year increases of N 58% in
2008, 41% in 2009, 30% in 2010, and 5% in 2011. The number of inten-
tional homicides registered by the INEGI declined in the Caldern's
nal year (Fig. 4). In addition, a number of weaknesses continue affect-
ing the country in terms of a rigid labor market and poor educational
system, coupled with still-inequitable income distribution, social ten-
sions, rampant crime, and low levels of trust in politicians and a sense
of rigid reforms (Hausmann et al., 2009).
3.2. Describing the Mexican Science, Technology and Innovation System
Triple Helix organizations play a relevant role in a transition innova-
tion process of an emerging economy. Fig. 5 shows the main agents in-
volved in the Mexican Science, Technology and Innovation System
(Diario Ocial, 2014).
3.2.1. Laws, normative and the National Council for Science and Technology
(CONACYT)
Established in 1970, the National Council for Science and Tecnology
(CONACYT) is a descentalized public organization that is the responsible
for the dening, development, and implementation of the main scientif-
ic and technlogical policies. Based on the national priorities, CONACYT
promotes the reinforcement of human capital (scholarships and evalu-
ation of national researchers) and the development of scientic/techno-
logical activities (reseach funding, supporting innovative entreprises,
etc.) in strategic areas (e.g., communication, biotechnology, advanced
materials, manufacture design, socioeconomic development, social in-
novations). Since its foundation, to achieve its aim, CONACYT has collab-
orated with other government organizations such as Ministry of
Economy, Ministry of Education, Mexican Institute of Industrial Proper-
ty, Regional Councils of Science and Technology, among others. In this
process, CONACYT has experimented two reforms of the policies and
the establishement of the Science and Technology Law in 2002.
Concerning the 2002 Science and Technology Law, its main objective
is fostering scientic research, technological development, and innova-
tion. This Law promoted the creation of the Scientic and Technological
300 M. Guerrero, D. Urbano / Technological Forecasting & Social Change 119 (2017) 294309
Fig. 2. GDP in emerging Latin-American economies (Trillion of USD), 20022013. Source: World Bank Indicators3.
Fig. 3. Open market index (%), 19802012. Source: Gobierno de los Estados Unidos Mexicanos (2013, p. 23).
Advisory Forum integrated by representants from the government bod-
ies (Excecutive, Legislative and Judicial), industry, higher education sys-
tem, and society. The main purpose of this Forum was developing States
Laws of Science and Technology in each Mexican as well as the creation
of commissions of science and technology in state legislatures, and state
councils of science and technology. Based on on this Caldern
desentralization strategy, almost the 31 Mexican states have their own
Law on Science and Technology, with its Committee on Science and
Technology and the State Council of Science and Technology at 2013.
As a result, an increment of the State participation was produced during
20062012 and observed in the increment of researchers' member of
the national research system (56% to 60%), in the number of recognized
educational programmes (70% to 79%), and in the investment for basic/
applied research and innovation (89% to 91%) per region (CONACYT,
2011; Diario Ocial, 2014).
3.2.2. Government
In emerging economies, the government implements certain enter-
prises' innovation subsidies with a compulsory universities partnership
such as anstrategy to stimulate regional economic development and the
transition from an efciency economy to innovation economy (Cohen
et al., 2002). According to the Caldern's National Development Plan
(Gobierno de los Estados Unidos Mexicanos, 2006, pp. 108109), this
administration developed/implemented programmes to facilitate the
development of basic/applied research, technology and innovation:
(i) reinforce the implementation of the 2002 National Science and Tech-
nology Law establishing a link more strait among all agents involved in
the technological/innovation chain through the development of educa-
tional programmes and basic/applied research (universities and re-
search centers) aligned to the productive sectors (industry) that have
the greatest potential impact on wellbeing (society); (ii) encourage a
strong public/private funding for basic/applied research, technology
and innovation; (iii) assess the use of public resources to invest in
human capital in priority areas for the country; (iv) decentralization
of scientic and technological activities to contribute to regional devel-
opment and local needs; and (v) increase the investment in infrastruc-
ture. As a result, N 13 strategic programmes were implemented (OECD,
2010). According to the Diario Ocial (2014), Mexico spent 378,021
million pesos to develop scientic activities and technology during the
Caldern administation that represented around 37% of GDP. In general,
the distribution of expenditures for national science and technology
was 56.5% towards research and development of experimental activi-
ties, 23.9% towards education and training for scientists and technicians,
and 19.5% towards the services of science and technology. The main
funders have been the government (49.3%), private sector (44.5%),
and universities (6.2%).
3.2.3. Mexican industry
According to the 2004 Economic Census and the Mexican Business
Information System (SIEM),4 there are N 3 million economic units
4
The SIEM primarily comprises data of Mexican companies. It is regulated by the Law of
Business Chambers, its Confederations provide the legal framework, and the Ministry of
Economy denes the rules of operation annually.
 M. Guerrero, D. Urbano / Technological Forecasting & Social Change 119 (2017) 294309
Fig. 4. Homicide rate by Mexican administration, 19952013. Source: Adapted from Heinle et al. (2015, p.3).
(enterprises). By sector, about 12% are industries, 53% are commercial
providers, and 35% are service providers (CIDE and SEP, 2010). In gener-
al, small and medium enterprises (SMEs) represent N99% of all Mexican
rms. They have accounted for N 70% of all employment since 1993, and
they generate N 50% of gross domestic product (GDP) (Hausmann et al.,
2009). Mexico is also characterized as one of the world's most entrepre-
neurial countries in terms of the percentage of its population who has
started or is in the process of starting a new venture (Flores et al.,
2013). Yet there is evidence that Mexico is not friendly to entrepre-
neurs. It is estimated that between 60% and 90% of new ventures are
started in the informal sector. While job growth expectations and real-
izations arguably constitute the most visible medium-term impact of
entrepreneurship, innovative orientation impacts structural renewal
in the long-term. Mexico has made room for entrepreneurship, but it
does not seem to foster the kind of entrepreneurship required for eco-
nomic growth. In terms of innovation, only one-third of new entrepre-
neurs identied in the total entrepreneurial activity develop new
products or services for their customers, when in innovation-driven
economies it is almost one-half of new entrepreneurs (Flores et al.,
2013). In addition, innovation as an entrepreneurial strategy is almost
unknown for the majority of SMEs because, given their nature, they
are focused only on reducing costs and not on mechanisms for generat-
ing resources (Ruelas, 2004).
Fig. 5. Mexican Science, Technology and Innovation System: actors and outcomes. Source:
Diario Ocial (2014).
301
3.2.4. Higher educational system
Fig. 6 shows the main actors involved in the Mexican Educational
System managed by the Secretary of Public Education of Mexico (SEP).
Regarding universities, during the 20th century, the Mexican higher ed-
ucation system comprised 8 public universities, 5 private universities,
and 5 regional technological institutes. Traditionally, the organization
of higher education systems in Mexico, based on the national system,
was characterized by function through the mechanism of decision-
making, resulting in mainly public universities, which ultimately limit
the scope. Many of those universities found it difcult to obtain alterna-
tive sources of income in order to consolidate their academic and re-
search development projects. One of the main factors was the decline
in federal spending on higher education. As a consequence of some ed-
ucational reforms and institutional changes,5 new types of universities
emerged when the technological universities and decentralized organi-
zations of the federal government emerged. Therefore, unlike other
countries, currently Mexico has a broad and diverse higher education
system. According to the SEP (2012), the structure and organization of
higher education is: 218 Technological institutes that have as a priority
to train competent professionals and promote national development ac-
cording to the reality of each region; 104 technological universities offer
students who are completing upper secondary education intensive
training that allows them to join in a short time (after two years) pro-
ductive work or to continue studies at the undergraduate level in
other universities; 12 intercultural universities promote the training of
professionals committed to the economic, social, and cultural develop-
ment of indigenous peoples; 50 polytechnic universities are a set of pub-
lic universities committed to social and economic development with an
international projection, their mission is an integral training through
the generation, application, and dissemination of knowledge and cul-
ture; 9 federal public universities, in addition to the functions of teaching,
offer a broad spectrum of programmes and research projects (genera-
tion and innovative application of knowledge); 57 regional public uni-
versities were created by local congresses, under the legal concept of
decentralized public agencies to develop the functions of teaching, gen-
eration and innovative application of knowledge, and outreach and dis-
semination of culture; 422 higher teacher education universities are
responsible for training teachers of preschool, primary, and secondary
education; 538 private universities were created by certain groups of en-
trepreneurs or organizations with the objective to provide higher edu-
cation through teaching, research, and entrepreneurship activities;
5
The main educational reforms were developed in the 1980s and 1990s as a result of
the participation in the Noth American Free Trade Agreement signed with the USA and
Canada. It involved transforming the operating criteria organizations and changing the
subject of the certication and accreditation stands. The current government is also pro-
moting an educational reform oriented towards transforming the quality and the role of
universities in the society and economy.
302 M. Guerrero, D. Urbano / Technological Forecasting & Social Change 119 (2017) 294309

Fig. 6. Mexican educational system. Source: Adapted from SEP (2012).

and 6 public research centers have as their main objectives to dissemi-
nate in society science and technology. Promoted by the Law of Scince
and Technology, several units were created inside universities with
the aim to build and run projects on technological development and in-
novation and promote their links with the productive sectors and ser-
vice (similar to technology transfer ofces). Undoubtedly, it was other
result of the Caldern's desentralization strategy and also a great oppor-
tunity to make appropriate use of observed policy to maximize the ben-
ets in different Mexican companies to develop a virtuous cycle for the
generation of wealth.
4. Methodological design
4.1. Sample
Aligned with the purpose of this research, we paid attention on the
role of agents involved in the Mexican Science, Technology and
Innovation System on entrepreneurial innovations' performance in an
emerging economy (Mexico during the administration of Felipe
Caldern from 2006 to 2012). Similar than previous studies (Faems
et al., 2005), we used a cross-section dataset that contains 19,188
enterprise-level data about innovation activities collected in 2006,
2008, 2010 and 2012 by the Research and Technological Development
Survey (ESIDET, Encuesta sobre investigacin y desarrollo tecnolgico)
with the collaboration of the Council for Science and Technology
(CONACYT). According to the articles 37, 38 and 45 of the Mexican
Law on National Statistics System and Geographic Information, the
Mexican statistical data collected by the Mexican Institute of Statistics
(INEGI) is strictly condential and mandatory. Every two years, based
on the enterprises population identied in the economic censes, INEGI
applies the ESIDET survey to a probabilistic and stratied sample of eco-
nomic units with 20 or more employees distributed by productive sec-
tors (adopting the industrial classication adopted by the
Organization for Economic Cooperation and Development, OECD) and

Table 1
Descriptive statistics.

Variable Type N Mean Std. Min Max

Independent variable
Innsalep Percentage 19,118 8.32 2.49 0 100

Explanatory variables
Cooperation
Coop_intra Dichotomous 19,188 0.11 0.31 0 1
Coop_commercial Dichotomous 19,188 0.01 0.11 0 1
Coop_science Dichotomous 19,188 0.02 0.14 0 1
Coop_mixed Dichotomous 19,188 0.01 0.10 0 1
Financial resources
Own funds Dichotomous 16,128 0.19 0.40 0 1
Intra funds Dichotomous 16,128 0.02 0.13 0 1
Other enterprises funds Dichotomous 16,128 0.01 0.09 0 1
Bank funds Dichotomous 16,128 0.03 0.18 0 1
Government subsidies
Govsub_university Continue 19,188 0.10 0.48 0 7
Govsub_commercial Continue 19,188 0.28 0.96 0 15
Socioeconomic context
year_2006 Dichotomous 19,188 0.14 0.35 0 1
year_2008 Dichotomous 19,188 0.13 0.34 0 1
year_2010 Dichotomous 19,188 0.19 0.39 0 1
year_2012 Dichotomous 19,188 0.53 0.50 0 1

Control variables
Sizelog Continue 19,132 2.21 0.76 0 5.168933
Export Dichotomous 19,188 0.34 0.47 0 1
Intra_group Dichotomous 19,188 0.36 0.48 0 1
Technological sector Dichotomous 19,170 0.27 0.44 0 1
RTD_external Continue 19,188 455.43 17,609.96 0 2,240,000
RTD_internal Continue 19,188 4837.21 62,899.09 0 5,750,451

Moderated by
HGE_sales Dichotomous 19,188 0.21 0.41 0 1
HGE_employees Dichotomous 19,188 0.11 0.32 0 1
 M. Guerrero, D. Urbano / Technological Forecasting & Social Change 119 (2017) 294309
across the Mexican geography (covering the 32 States). With the sam-
ples estimated and response rates obtained those years, the INEGI as-
sumed a statistical error of b 5% or 8%. The owners/managers of the
company are jointly liable for the information provided in this survey
because they authorized or signed it. Based on this methodology design,
INEGI ensures that each economic unit that has answered the survey
was selecting without a previous distinction in the development of in-
novating or non-innovating activities. This avoids bias in the results ob-
tained with this dataset as suggested previous studies (Bayona et al.,
2001; Fritsch and Lukas, 2001).
4.2. Description of variables
Table 1 summarizes the description and statistics of the variables
used in this study. Based on previous studies (Delmar et al., 2003;
Klomp and van Leeuwen, 2001; Faems et al., 2005), the dependent vari-
able used to measure innovation performance was sales of innovative
products (Innsalep). More concretely, this variable represents the per-
centage of total sales related to the development of new/improved
products in the period t1 and t+1. In average, the percentage of sales
related to innovation of Mexican enterprises is around 8.32%.
To test our hypotheses about the links between enterprises and
other agents involved in the Mexican Science, Technology and Inno-
vation System, we introduced four explanatory variables. First, we
build four dichotomous variables associated to the types of coopera-
tion that developed enterprises to access to knowledge resources:
(i) cooperation with intrapreneurship agents such subsidiaries or cor-
porate units (Coop_intra); (ii) cooperation with commercial agents
like other enterprises (Coop_commercial); (iii) cooperation with scien-
tic agents like universities and research centers (Coop_science); and
(iv) simultaneous cooperation with diverse agents (Coop_mixed)
(Agarwal et al., 2004; Alcalde and Guerrero, 2016; Dutrnit et al.,
2010; Nieto and Santamaria, 2007; Klepper, 2001). These variables
take value 1 when developed one of those collaborations and 0 other-
wise. Second, we build four dichotomous variables associated to the
types of innovation funds that obtain each enterprise from Triple Helix
agents: (i) own funds that corresponding to reinvest earnings (Own
funds); (ii) funds from subsidiaries or partner enterprises (Intra
funds); funds from other enterprises (Commercial funds); and credits
obtained with nancial organizations (Bank funds) (Auerbach, 1984;
Autio et al., 2014; Himmelberg and Petersen, 1994). Third, we build
two continue variables associated to the number of government subsi-
dies that each enterprise had assessed with Triple Helix partners:
(i) access to public subsidy with a compulsory university partners to
support innovation practices (Govsub_universities); and (ii) access to
public subsidy with a compulsory commercial partners to support inno-
vation practices (Govsub_commercial)6 (Almus and Czarnitzki, 2003;
Dutrnit et al., 2010; Cohen et al., 2002; Czarnitzki et al., 2003;
Perkmann and Walsh, 2007). Four, given the INEGI's condentiality re-
strictions was not able to have access to the region where each
6
During 200620,012, the enterprises had access to several programmes (some of
them with a compulsory parnership with universities or with other enterprises). Admin-
istrated by CONACYT were: High Value Added Business with Knowledge and Entrepre-
neurs (AVANCE), Incentives for innovation (FONCYT), Technological Innovation of
Higher Value Added (INNOVAPYME), Development and Innovation of Technologies
(PROINNOVA),Technological Innovation and Competitiveness (INNOVATEC), Internation-
al Cooperation Funds (FONCICYT), Tax Incentives for Research and Technology Develop-
ment (ESTIMULOS FISCALES), Sector Funds (FONSEC), Mixed Funds (FOMIX),
Incorporation of Mexican Scientics and technologist in social sectors (IDEAS), Scholar-
ships inside enterprises (Estancias al interior de empresas), New fund for science and
technology (Nuevo fondo C&T). Administrated by Ministry of Economy were: Centre for
regional enterprise competitiveness (CETRO-CRECE), Fund for Micro, Small and Medium
Enterprises (FONDOPYME), Development of information and technological innovation
(PROSOFT 2.0), Development of Interactive Media (PROMEDIA), Fund for the promotion
of supply chain (FIDECAP), and Promotion of exportations (COMPEX). Administrated by
Ministry of Education were: Standards and Certication Council (CONOCER), among
others.
303
enterprise is located. As we explained in Figs. 2 and 3, the Caldern ad-
ministration was characterized by an interesting economic pattern but
also by an increment of organized crime. As context matters for enter-
prises innovation performance (Autio et al., 2014; Sobel, 2008), we in-
troduce four dichotomous proxies to capture the effect of socio-
economic conditions using the year when each enterprise was
interviewed (year_2006; year_2008; year_2010; year_2011).
Concerning control variables, almost all type of enterprises face sever-
al innovation challengessuch as nancial limitations, management is-
sues, risks, lack of experience, and erce competition forced by large
competitors(Aldrich and Auster, 1986; Thompson, 1999). Particularly,
Christensen (1997) recognize that the growth and survival of enter-
prises are associated with the characteristics of the entrepreneur
(education, age, motivation, etc.), the strategy (market, technological,
exporting, competition, etc.), and the structure (resources, age, size,
sector, location, ownership, etc.). Based on the data availability, this
study controlled at rm-level by certain strategical and structural char-
acteristics. Regarding to the structural characteristics, we included sev-
eral variables. First, the enterprise size was measured by the number of
employees and expressed in logarithms (Sizelog). There is a substantial
body of evidence suggesting scale economies in the innovation produc-
tion function (Acs and Audretsch, 1991; Stephan, 2012) and an in-
creased availability and access to critical resources (Gooding and
Wagner, 1985). Second, Galende and de la Fuente (2003) showed that
export and internationalization have a positive signicant effect on col-
laboration agreements. Therefore, we also controlled by the venture's
export activity (Exportations) measured with a dummy variable
where 1 represents the enterprises that have exported and 0 otherwise.
Third, Wadhwa et al. (2011) found a positive relationship with innova-
tion performance for those that are part of a company group with col-
laboration practices (Intra_group) measured such a binary variable
that takes value 1 if is part of a company and 0 if it only a single-unit en-
terprise. Four, we include sectorial controls to account for different pro-
pensities to innovate across technological sectors using a binary variable
(Technological_sector). Five, we controlled by the investing in research
and technological efforts on external (IDT_external) as well as internal
(IDT_external).
Finally, we built proxies to identify enterprises with a high-growth
orientation and moderate our analysis. According to the OECD (2007,
p. 61), a high-growth enterprise can be dened such as all enterprises
with average annualized growth N 20% per annum, over a three year pe-
riod should be considered as high-growth enterprises (both in terms of
employment - number of employees - and in terms of turnover). Based
on the nature our cross-section data and the available information col-
lected in the survey, we create a proxy adopting the criteria of average
annualized growth N 20% in the year previous of the year of the survey
(t 1) both in terms of sales (HGE_sales) and number of employees
(HGE_employees) (Audretsch, 2012; Moreno and Casillas, 2007).
Using those dichotomous variables, we create four entrepreneurial
growth proles: (Group 1) enterprises that in the period t1 do not re-
ported annualized growth N 20% in both of growth in sales and em-
ployees ( 20% employees and 20% sales); (Group 2) enterprises
that in the period t1 only reported annualized growth N 20% growth
in employees (+ 20% employees and 20% sales); (Group 3) enter-
prises that in the period t 1 only reported annualized growth N 20%
growth in sales (20% employees and +20% sales); and (Group 4) en-
terprises that in the period t1 do reported annualized growth N 20% in
both of growth in sales and employees (+ 20% employees and + 20%
sales). Even than it is a proxy, it helps us to associate the group 4 with
enterprises that develop innovation entrepreneurship such as was pro-
posed by Autio et al. (2014).
4.3. Statistical model
The dependent variable in the regression model is censored, since
the variable is the percentage of innovative sales and therefore by
304 M. Guerrero, D. Urbano / Technological Forecasting & Social Change 119 (2017) 294309

Table 2
Correlation matrix.

No Variable 1 2 3 4 5 6 7 8 9 10

1 Innsalep 1
2 Coop_intra 0.6279 1
3 Coop_commercial 0.1726 0.0392 1
4 Coop_science 0.2289 0.0488 0.0156 1
5 Coop_mixed 0.1907 0.0369 0.0118 0.0147 1
6 Own funds 0.6917 0.7254 0.2048 0.2856 0.2252 1
7 Intra funds 0.2021 0.1486 0.1620 0.0886 0.1105 0.2111 1
8 Other enterprises funds 0.1387 0.1027 0.0693 0.0280 0.0990 0.1496 0.1685 1
9 Bank funds 0.2718 0.2763 0.0931 0.0817 0.1135 0.3338 0.1065 0.1344 1
10 Govsub_university 0.2098 0.1254 0.0365 0.1847 0.1450 0.2789 0.1119 0.0802 0.1015 1
11 Govsub_commercial 0.1910 0.1625 0.0485 0.1062 0.0907 0.2321 0.0878 0.0490 0.1215 0.6046
12 year_2006 0.1389 0.1282 0.0360 0.0266 0.0391 0.0992 0.0323 0.0314 0.0307 0.0141
13 year_2008 0.0573 0.0732 0.0457 0.0119 0.0316 0.0406 0.0252 0.0188 0.0275 0.0251
14 year_2010 0.0481 0.0129 0.0007 0.0112 0.0046 0.3868 0.1276 0.0528 0.1685 0.0918
15 year_2012 0.1741 0.1497 0.0558 0.0355 0.0452 0.2637 0.0957 0.0600 0.1127 0.0798
16 Sizelog 0.0344 0.0157 0.0292 0.0248 0.0136 0.0032 0.0461 0.0153 0.0259 0.0098
17 Export 0.0035 0.0036 0.0012 0.0015 0.0011 0.0055 0.0015 0.0010 0.0021 0.0020
18 Intra_group 0.0092 0.0160 0.0400 0.0010 0.0096 0.0449 0.0885 0.0118 0.0369 0.0007
19 Technological sector 0.1045 0.1045 0.0107 0.0429 0.0407 0.1566 0.0615 0.0375 0.0594 0.0490
20 RTD_external 0.0172 0.0083 0.0108 0.0062 0.0038 0.0226 0.0318 0.0142 0.0040 0.0140
21 RTD_internal 0.1300 0.0798 0.0939 0.0406 0.0613 0.1394 0.1361 0.0197 0.0313 0.1074
Note: Level of statistical signicance: p 0.000.

denition ranges between 0 and 100. Accordingly, a Tobit regression
analysis is applied (Greene, 2003; Laursen and Salter, 2006). We devel-
oped Model 1 with the entire sample and Model 25 based on the high-
growth orientation estimated by the average annualized growth N20%
in the year previous of the year of the survey (t 1) both in terms of
sales and number of employees. Table 2 shows the collinearity statistics
and correlations of all variables. As a robustness test we applied the
Heckman test.
5. Results
Table 3 reports the results of the Tobit regression that explains the
innovative performance (Innsalep) across Mexican enterprises (Model
1) and the innovative performance by the level of annualized growth
N20% both in sales and employees (Models 2 to 5).
Concerning the cooperation of Mexican enterprises with Triple Helix
agents (enterprises-university-government) to access to knowledge
and technological sources, results from Model 1 strongly conrm the
contribution of R&D cooperation to innovation performance. Interest-
ingly, the cooperation with several agents involved in the Mexican
Technology and Innovation such as their parents enterprises (129.94;
p b 0.001), other enterprises (122.91; p b 0.001), universities and re-
search centers (126.02; p b 0.001), and with a combination of agents
(126.62; p b 0.001) is highly signicant and positive on innovation per-
formance. Regarding the cooperation by levels of high-growth orienta-
tion of Mexican enterprises, Model 25 also show a positive effect
produced by Triple Helix cooperation on innovation performance. How-
ever, the intensity of this positive effect varies across the high-growth
orientation of Mexican enterprises. Similar than Belderbos et al.
(2004) arguments, our results evidenced a substantial heterogeneity
in the determinants to establish R&D collaborations with different
partners. First, intra cooperation has higher impact on innovation perfor-
mance of enterprises without high-growth orientation than others
types of cooperation (128.28; p b 0.001). This evidences a pattern of
innovation performance based on the exchange of scientic and techno-
logical resources from parents' enterprises (Shrader and Simon, 1997).
Second, commercial cooperation has higher impact on the innovation
performance of enterprises that growth N 20% only in terms of em-
ployees (153.31; p b 0.001) or only in sales (140.68; p b 0.001) than
others types of cooperation. In this sense, a commercial cooperation al-
lows obtain knowledge regarding the industrial value chain and com-
mercial benets translated into sales of innovative products (Chung
et al., 2003; Damanpour, 1991). Third, science cooperation has higher
impact on the innovations performance of high-growth enterprises
than others types of cooperation (192.56; p b 0.001). It is not surprising
that the innovations performance of high-growth enterprises is strongly
supported by their collaboration with universities and public research
centers because they look for more advance technologies and disrupt
innovations (Audretsch, 2012; Delmar et al., 2003; Friar and Meyer,
2003). Compared with Model 1, Model 5 shows a reinforced effect on
innovation performance produced by multiple types of collaboration
among agents involved in the Mexican Science, Technology and Innova-
tion System.
Analyzing the access to fund for innovation from Triple Helix organi-
zation (enterprises-university-government), Model 1 shows the posi-
tive and signicant effect of both internal sources (own 82.61;
p b 0.001 and intra 9.88; p b 0.050) and external sources (other enter-
prises 21.01; p b 0.001 and banks not evidenced) on innovation perfor-
mance. By the level of high-growth orientation of Mexican enterprises,
Model 25 present the effects produced by each fund on innovation per-
formance. As well as cooperation, the effect of external funds varies by
the degree of high-growth orientation. First, the results strongly conrm
the contribution of own funds to innovation performance for all high-
growth proles. Second, intra funds have positive impact on innovation
performance of enterprises without a high-growth orientation (do not
growth N20% both in sales and employees in the previous year) than
others types of external funding (9.65; p b 0.100). Again, this prole is
more likely to operate under the umbrella of parents' enterprise
(Shrader and Simon, 1997). Third, credits from banks have a positive im-
pact on innovation performance of enterprises that growth N 20% only in
terms of employees in the previous year (26.80; p b 0.050) and funds
from other enterprise have a positive impact on innovation performance
of enterprises that growth N20% only in terms of sales in the previous
year (37.30; p b 0.050). Interestingly, these proles combine internal
and external sources even the complexity of the Mexican nancial mar-
ket (Wright et al., 2005). Four, funds from other enterprise have a positive
impact on the innovation performance of high-growth enterprises than
others types of cooperation (58.82; p b 0.050) but lower than the posi-
tive impact produced by the use of their own funds (79.50; p b 0.001).
These results are aligned with the main funders of the Mexican Scientif-
ic, Technological and Innovation System during the Caldern adminis-
tration: the government (49.3%), private sector (44.5%), and
universities (6.2%) (CONACYT, 2011).
Regarding the contribution of Triple Helix agents to access government
subsidies, Model 1 shows the positive effect of government subsidies
(obtained in collaboration with universities) on innovation
 M. Guerrero, D. Urbano / Technological Forecasting & Social Change 119 (2017) 294309 305

Table 2
Correlation matrix.

11 12 13 14 15 16 17 18 19 20 21

1
0.0791 1
0.1532 0.1600 1
0.0011 0.1979 0.1927 1
0.1608 0.4318 0.4204 0.5202 1
0.0018 0.1264 0.0041 0.1031 0.0594 1
0.0029 0.1478 0.0040 0.1005 0.0567 0.0294 1
0.0003 0.0707 0.0050 0.0526 0.0524 0.4762 0.0021 1
0.0621 0.2452 0.0096 0.1822 0.1216 0.0341 0.0045 0.1462 1
0.0085 0.0104 0.0101 0.0097 0.0065 0.0247 0.0003 0.0134 0.0014 1
0.0773 0.0067 0.0212 0.0151 0.0311 0.0779 0.0008 0.0432 0.0353 0.1447 1

performance (3.87; p b 0.100). We did not obtain evidence about the ef- subsidies by high-growth orientation, results only conrm a positive ef-
fect of government subsidies obtained in collaboration with other enter- fect on innovation performance of enterprises without a high-growth
prises. On the other hand, when we explored the effect of government orientation (5.15; p b 0.050). We did not nd enough evidence for the

Table 3
Tobit Regression, explaining innovative performance (Innsalep) across Mexican enterprises.

Dependent variable: Innsalep Mexican Enterprises based on the level of annualized growth N20% in t1 in both sales & employees

(1) (2) (3) (4) (5)

20% employees +20% employees 20% employees +20% employees

All enterprises 20% sales 20% sales +20% sales +20% sales

Coef. SE P N |t| Coef. SE P N |t| Coef. SE P N |t| Coef. SE P N |t| Coef. SE P N |t|

Collaboration
Coop_intra 129.94 4.14 128.38 4.97 119.61 18.47 130.83 9.47 131.95 15.73
Coop_commercial 122.91 5.88 114.19 6.96 153.31 27.28 140.68 14.77 127.80 20.50
Coop_science 126.02 5.97 118.92 7.28 79.97 24.09 139.81 13.40 192.56 25.65
Coop_mixed 126.62 6.17 121.37 7.54 139.29 27.92 133.67 13.59 129.61 22.37
Funds
Own funds 82.61 3.85 83.53 4.66 97.13 17.53 75.95 8.74 79.50 14.37
Intra funds 9.88 4.53 9.65 4.67 22.47 21.72 13.84 10.03 16.76 13.94
Other enterprise funds 21.01 6.35 12.74 7.96 3.03 22.18 37.30 15.45 58.82 20.82
Bank funds 5.25 3.32 4.56 4.12 26.80 13.00 2.45 7.97 4.59 10.22
Government subsidies
Govsub_university 3.87 1.01 5.15 2.40 10.62 9.67 0.47 0.50 7.36 7.74
Govsub_commercial 0.22 1.02 0.03 1.27 3.76 3.94 1.47 2.39 1.50 3.23
Socioeconomic context
year_2008 14.03 3.35 14.02 4.00 24.17 14.86 17.21 8.77 3.68 11.23
year_2010 16.83 3.79 14.65 4.54 53.82 18.75 23.41 9.47 19.61 13.95
year_2012 24.99 4.03 27.03 4.96 16.01 17.95 22.11 9.15 27.33 13.95
Control variables
Sizelog 6.73 1.63 6.18 1.95 5.75 7.45 6.58 4.12 9.46 6.49
Export 0.00 0.00 0.00 0.00 0.15 0.16 0.05 0.13 0.01 0.20
Intra_group 3.30 2.66 2.86 3.16 8.30 12.12 2.94 6.43 6.19 10.62
Technological sector 7.84 2.33 8.01 2.82 15.02 10.05 3.26 5.59 3.83 8.13
RTD_external 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
RTD_internal 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
_cons 129.52 5.62 129.37 6.79 154.55 27.02 124.80 13.06 119.88 20.50
/sigma 64.73 1.21 64.38 1.45 64.58 5.21 63.99 2.85 59.31 3.95
Left-censored 13,225 9748 704 2010 763
Uncensored 1843 1271 101 328 143
Right-censored 977 643 65 175 94
N 16,045 11,662 870 2513 1000
LR Chi2 11,361.97 7890.91 676.04 1899.00 927.00
Prob N Chi2
Pseudo R2 0.3233 0.3253 0.3401 0.3101 0.3412
Log likelihood 11,891.4 8183.9 655.9 2112.5 89.505

Note: Level of statistical signicance: p 0.000, p 0.05, p 0.10.

306 M. Guerrero, D. Urbano / Technological Forecasting & Social Change 119 (2017) 294309
Table 4
Testing hypotheses.
Inuence on innovation performance
Collaboration with Triple Helix agents to access knowledge resources
Moderated by high-growth orientation [+20% sales and +20% employees]
Access to innovation funds from Triple Helix agents
Moderated by high-growth orientation [+20% sales and +20% employees]
Access to government subsidies through Triple Helix agents
Moderated by high-growth orientation [20% sales and 20% employees]
Socio-economic context
Moderated by high-growth orientation [+20% sales and +20% employees]
rest of proles. Concerning the effect of Mexican socio-economic context
on the innovation performance, Figs. 2 and 3 presented interesting con-
trasts during the Caldern administration. Economically, Mexico
experimented positive signs in terms of GDP. Socially, Mexico evidenced
negative signs in terms of homicide rates. Taking as a reference 2006
(the rst year of this administration), Model 1 evidences a negative ef-
fect of the socio-economic context on innovation performance. The neg-
ative effect is highest for enterprises interviewed in 2012 (last year of
this administration). Interestingly, almost all models present a positive
effect on innovation performance for enterprises interviewed in 2010.
By proles, Model 5 evidences the negative effect on innovation perfor-
mance for high-growth enterprises interviewed in 2012 respect to 2006.
6. Discussion
The purpose of this research was to provide a better understanding
about the inuence of Triple Helix agents on entrepreneurial innova-
tions' performance of enterprises located in emerging economies. In
other words, in the context of an emerging economy, we analyze the
effects produced by the links of enterprises with other enterprises,
universities and government on their innovation performance
(e.g., collaboration, funding, subsidies, etc.), as well as, the moderation
of those effects generated by those enterprises have a high-growth ori-
entation (e.g., proxy used to differentiate entrepreneurial innovations
from traditional innovations).
Based on the tested set of hypotheses (Table 4),
Based on the evidence obtained about the inuence of access to
knowledge and technological sources from Helix agents (enterprises-
university-government) on Mexican enterprises' innovation perfor-
mance, we conrm the H1 and H1a. Firstly, our results are aligned to
extant empirical studies that found positive and signicant effect pro-
duced by cooperation (e.g., intra, commercial, scientic and mixed) on
innovation performance measured by sales of innovative products
(Alcalde and Guerrero, 2016; Klomp and van Leeuwen, 2001; Faems
et al., 2005; Santiago et al., 2016). Triple Helix collaborations allow en-
terprises have access to new/advanced knowledge and technological re-
sources required to explore new entrepreneurial opportunities as well
as developing incremental or radical innovations. In the same period
of analysis, promoted by the Ministry of Education, Crdenas et al.
(2012) administrated a survey to 351 hybrid units created within uni-
versities and public research center (similar than Technology Transfer
Ofces) across the country to explore their relationship with Mexican
enterprises. Their results evidenced that N 89% of those research and
technological units had a relationship with enterprises for human capi-
tal links (e.g., formation and labor insertion of university graduates) as
well as N 54% developed a relationship associated to research and inno-
vation projects, and only 36% provided technological services. Similarly,
Dutrnit et al. (2010) argue that the common channels of knowledge
transfer interaction between Mexican public research organizations
and enterprises are dened under the umbrella of collaboration to
joint R&D projects/contracts. Similarly than developed economies, in
emerging economies, the higher economic benets derived from this
H Theoretically (expected) Empirically (evidenced)
H1 Positive effect Supported
H1a Positive effect reinforced Supported
H2 Positive effect Supported
H2a Lower positive effect Supported
H3 Positive effect Supported
H3a Positive effect reinforced Supported
H4 Negative effect Supported
H4a Negative effect reinforced Supported
knowledge transfer relationship are usually observed in long-term (De
Fuentes and Dutrenit, 2012) and depends of the geographic proximity
(De Fuentes and Dutrnit, 2016). Secondly, paying attention on the re-
sults obtained by level of high-grow orientation, we found that the
group of enterprises with a strong orientation to 20% of growth in
both employees/sales are associated to enterprises that development
entrepreneurial innovations as suggested Autio et al. (2014). The litera-
ture about high-growth enterprises' orientation is limited or inexistent
in emerging economies. A plausive explanation is that Latin American
emerging economies have a limited number of ventures (e.g., with
high-growth orientation) because of the many restrictions that present
to create knowledge-based enterprises that producing higher perfor-
mance rates and spillover effects (Acs and Amors, 2008). Even than en-
terprises grow in many different ways and adopting several patterns of
growth (Delmar et al., 2003), the congurations and development of
enterprises that adopt growth patterns based on entrepreneurial inno-
vations depend of an adequate environmental with solid institutions
that support entrepreneurship and innovation as well as organizational
with entrepreneurial capabilities and behaviors. In the case of Mexico,
the consolidation of robust relationships among enterprises from sever-
al industries, universities/research centers and government faced
several challenges particularly associated with the lack of qualied
human resources, funds, and productivity (Crdenas et al., 2012;
Kuznetsov and Dahlman, 2008). CONACYT has implemented several
strategies to reinforce human capital and productivity but the results
will be reected in medium or long term. Our results provide some in-
sights about the importance of commercial and scientic collaborations
on innovation performance according to the degree of high-growth
orientation.
Based on the results obtained regarding the effect produced by the
access to fund for innovation from Triple Helix organization (enter-
prises-university-government) on enterprises' innovation performance,
we conrm our H2 and H2a. Any nancial decision represents a higher
cost/risk for enterprises and the main innovation constrain in emerging
economies (lvarez and Crespi, 2015; Crespi and Ziga, 2012;
Kuznetsov and Dahlman, 2008). Similar than previous studies (Hall
and Lerner, 2010), Mexican enterprises are more likely to use internal
innovation funds (e.g., retained earnings and parents' support) than ex-
ternal sources (e.g., other enterprises). However, in the case of enter-
prises with high-growth orientation, there are some Mexican
enterprises that develop their innovations and create value based on
funds provided other enterprises and exploring other business innova-
tion models (i.e., Chesbrough, 2012). An explanation of this innovation
funding pattern could be aligned with the expectation of obtaining
returns in long term and generate mechanism to face uncertainties or
assuming an affordable level of risk (De Fuentes and Dutrenit, 2012).
Moreover, our results do not provide strong evidence about the effect
of nancial institutions on innovation performance. It could be ex-
plained by the obstacles to access to credit in an embryonic Mexican -
nancial system (Surez and Oliva, 2005) or by the higher costs of
investing in technologies with long term return (Himmelberg and
Petersen, 1994).
 M. Guerrero, D. Urbano / Technological Forecasting & Social Change 119 (2017) 294309
Based on the results obtained regarding the contribution of Triple
Helix agents to access government subsidies, we conrm H3 and H3a. Dur-
ing the administration of Caldern, the majority of programmes associ-
ated to reinforce basic/applied research, technology and innovation
required an enterprise-university partnership. It is a common practice
implemented by the majority of governments that have encouraged
universities to make their science and engineering more relevant to
industry's needs (Cohen et al., 2002). Additionally, these results are
also aligned to extant empirical studies that found a productive impact
on rm performance (Aerts and Czarnitzki, 2004; Aerts and Schmidt,
2008; Czarnitzki and Lopes-Bento, 2013; Hall and Mafoli, 2008) as
well as other studies that recognize that the measures of performance
used do not necessarily reect those benets (Perkmann and Walsh,
2007). In this sense, even than Autio et al. (2014) argue that innovation
entrepreneurship used multiple and combined sources of funding (pub-
lic/private), enterprises with a high-growth orientation located in
emerging economies are more likely to use their own nancial re-
sources than government subsidies. A tentative explanation is the
opportunity/cost in terms of time/money that demands administra-
tively a government subsidy (Crdenas et al., 2012). Finally, based on
our results, we conrm our H4 and H4a. In this regard, we argue that
a context characterized by higher levels of criminality produce a neg-
ative effect on any entrepreneurial innovation activity and more on
those higher for entrepreneurial enterprises that will be the focus
of extortion of criminal organizations (Smith, 2009). In this context,
if movements from emerging economies are interesting to becoming
knowledge-based or entrepreneurial economies, they should adopt
an entrepreneurial and innovative behavior as well as recognizing
the uncertainty of the entrepreneurial/innovation process on the
promulgation of policies for nurturing environment within entrepre-
neurial innovations can spontaneously arise in both the private and
the public sectors (Audretsch and Link, 2016; Leyden, 2016; Leyden
and Link, 2015).
7. Conclusions
From this research, we can extract four relevant conclusions.
First, we identify that enterprise in emerging economies cooperates
with their parents (intrapreneurially), other enterprises involved in
the system (commercially), universities and research centers
(scientically) and also developed mixed cooperation. These types
of cooperation produce a positive effect on their innovation perfor-
mance and it is reinforced when the enterprise has a high-growth
orientation (Audretsch, 2012; Autio et al., 2014). Second, even than
diverse alternatives to access to innovation funds, enterprises ana-
lyzed are more likely to use internal sources for developing innova-
tion than external sources (e.g., other enterprises or banks). This
pattern is also observed in enterprises with a high-growth orienta-
tion but their innovation performance is affected positively when
they access to external fund from other enterprises (Himmelberg
and Petersen, 1994; Moreno and Casillas, 2007). Third, the positive
effect of government subsidies on innovation performance is only
evidenced when enterprises collaborate with universities
(Czarnitzki et al., 2007). However, this positive effect is only ob-
served in enterprises without a high-growth prole (Czarnitzki
and Lopes-Bento, 2013). Four, the socio-economic context of the pe-
riod analyzed has a negative impact on innovation performance of
enterprises interviewed.
The available literature about innovation and entrepreneurship in
emerging economies is limited to explore the determinants of innova-
tion performance as well as innovation constrains (lvarez and Crespi,
2015; Crespi and Ziga, 2012; De Fuentes et al., 2015; Kuznetsov and
Dahlman, 2008; Santiago et al., 2016). In this regard, the modest
academic contributions of this study are to: (i) contribute to the open
innovation literature debate about the necessity of exploring the main
effects produced by enterprises' selection of internal/external
307
collaborations for the development of entrepreneurial innovations
(Huizingh, 2011; p. 5; Leydesdorff and Ivanova, in press), as well as, to
the debate of open innovation models to capture the benets of collab-
orations among Triple Helix agents (Chesbrough, 2012); (ii) contribute
to the nancial literature debate about the gap in the analysis and in the
use of internal/external funds for nancing innovation (Hall and Lerner,
2010; p. 40), as well as, the impact of subsidies on enterprises with an
innovative behavior and that supported their innovations by other
sources (Czarnitzki and Lopes-Bento, 2013; p.31); (iii) contribute to
knowledge transfer and academic entrepreneurship literature debate
about a view or performance indicators of the impact of university-
enterprises collaboration on entrepreneurial innovations (Grimaldi
et al., 2011; p.1053), as well as, the entrepreneurial role of public sector
in the creation of a sustainable innovative and entrepreneurial economy
(Audretsch and Link, 2016; Leyden, 2016; Leyden and Link, 2015); and
(iv) contribute to the entrepreneurship literature debate about a com-
parison of national systems of innovation, entrepreneurship, and entre-
preneurial innovation; more concretely, about the impact of context for
entrepreneurial innovation and enterprise growth (Autio et al., 2014;
p. 1089 and p.1104). In conjunction, all those elements try to provide
evidence about the micro-foundations of Triple Helix in emerging
economies.
This study has several limitations that provide good opportunities
for future research. The rst limitation is the database used to analyze
this phenomenon. Given the INEGI's condentiality restrictions, we
did not have access to all variables of the survey (e.g., State or geograph-
ic location, age, and other variables), and their treatment/analysis was
limited to the statistical resources available in its installations. Future
studies could focus on reinforcing the statistical models by using other
techniques, such as multilevel analysis, that allow for the exploration
of regional impact on innovation performance using other variables
and databases (Aidis, 2005; Aidis et al., 2010). The second limitation is
the measure used to create our proles of high-growth orientation be-
cause each survey only collected information from t 1 and t+1
(Audretsch, 2012; OECD, 2010). Therefore, a natural extension of this
work could be building a panel data to explore these proles using
most adequate variables or exploring them via case studies. In this
sense, we can include other theoretical approaches (e.g., adopting Insti-
tutional Economic and Knowledge Spillover) to understand in depth the
micro-foundations of Mexican Triple Helix organizations as well as their
impact on regional development (Autio et al., 2014; Guerrero et al.,
2015; Wright et al., 2005). Another research venue could be an in-
depth analysis about the benets, productivity, and externalities pro-
duced by government innovation subsidies. For example, select a
group of enterprise-university partnerships that had to these kinds of
subsidies access (focus group) and other partnerships with similar char-
acteristics that had not access (control group). It allows us to provide a
better understanding about this phenomenon in a single or cross
emerging economies (Czarnitzki and Lopes-Bento, 2013; Hall, 2005).
Some implications emerge of this study for the main actors involved
in the Mexican Scientic, Technology and Innovation System. For policy
makers, this study exhibits key determinants of an enterprise's partici-
pation in collaborative innovation models. If policy makers expect a
transition towards a knowledge-based economy is still required to rein-
force innovation system as well as the entrepreneurial ecosystem
(investing market, educational system, labor market, security and stable
regulation for SMEs, among others). For enterprise managers, this study
offers insights about the best mechanisms and practices for innovation
practices with external public and private partners. For university man-
agers, the entrepreneurial university model is a good example in their
transformation process.
Acknowledgements
The authors acknowledge the support received by the Microdata
Team from the Mexican National Institute of Statistics and Geography
308 M. Guerrero, D. Urbano / Technological Forecasting & Social Change 119 (2017) 294309
(INEGI) to access the databases in its installations. Authors would like to
thank the anonymous reviewers for their valuable suggestions in improv-
ing the quality of this manuscript. Maribel Guerrero acknowledges the -
nancial support from Santander Universidades (Iberoamerica Scholarship
for Young Researchers) and David Urbano acknowledges the nancial
support from projects ECO2013-44027-P (Spanish Ministry of Economy
& Competitiveness) and 2014-SGR-1626 (Economy & Knowledge
DepartmentCatalan Government).
References
Acs, Z.J., Amors, J.E., 2008. Entrepreneurship and competitiveness dynamics in Latin
America. Small Bus. Econ. 31 (3), 305322.
Acs, Z.J., Audretsch, D.B., 1991. R&D, rm size and innovative activity. Innovation and
Technological Change: An International Comparison Vol. 98(2), pp. 451456.
Aerts, K., Czarnitzki, D., 2004. Using innovation survey data to evaluate RD policy: the case
of Belgium. ZEW-Centre for European Economic Research Discussion Paper, 04055.
Aerts, K., Schmidt, T., 2008. Two for the price of one?: Additionality effects of RD subsi-
dies: a comparison between Flanders and Germany. Res. Policy 375, 806822.
Agarwal, R., Echambadi, R., Franco, A., Sarkar, M., 2004. Knowledge transfer though inher-
itance: spin-out generation, development and survival. Acad. Manag. J. 47 (4),
501522.
Aidis, R., 2005. Institutional barriers to small- and medium-sized enterprise operations in
transition countries. Small Bus. Econ. 25 (4), 305317.
Aidis, R., Estrin, S., Mickiewicz, T., 2010. Institutions, nance and the level of development:
the impact on entrepreneurship in transition. Rev. Econ. Inst. 1, 126.
Alcalde, H., Guerrero, M., 2016. Open business models in entrepreneurial stages: evidence
from young Spanish rms during expansionary and recessionary periods. Int. Entrep.
Manag. J. 12 (2), 393413.
Aldrich, H.E., Auster, E.R., 1986. Even dwarfs started small: liabilities of age and size and
their strategic implications. Res. Organ. Behav. 8, 165198.
Almus, M., Czarnitzki, D., 2003. The effects of public R&D subsidies on rms' innovation
activities: the case of Eastern Germany. J. Bus. Econ. Stat. 21 (2), 226236.
lvarez, R., Crespi, G.A., 2015. Heterogeneous effects of nancial constraints on innova-
tion: evidence from Chile. Sci. Public Policy 42 (5), 711724.
Audretsch, D., 2012. Determinants of High-Growth Entrepreneurship Report Prepared for
the OECD/DBA International Workshop on High-Growth Firms: Local Policies and
Local Determinants Copenhagen, 28 March 2012.
Audretsch, D.B., Link, A.N., 2016. Essays in Public Sector Entrepreneurship. Springer.
Audretsch, D., 2014. From the entrepreneurial university to the university for the entre-
preneurial society. J. Technol. Transfer. 39 (3), 313321.
Auerbach, A.J., 1984. Taxes, rm nancial policy and the cost of capital: an empirical anal-
ysis. J. Public Econ. 231, 2757.
Autio, E., Kenney, M., Mustar, P., Siegel, D., Wright, M., 2014. Entrepreneurial innovation:
the importance of context. Res. Policy 43 (7), 10971108.
Bayona, C., Garca-Marco, T., Huerta, E., 2001. Firms' motivations for cooperative R&D: an
empirical analysis of Spanish rms. Res. Policy 30, 12891307.
Belderbos, R., Carree, M., Lokshin, B., 2004. Cooperative RD and rm performance. Res.
Policy 3310, 14771492.
Boschma, R., 2005. Proximity and innovation: a critical assessment. Reg. Stud. 39 (1),
6174.
Boyd, B.K., 1991. Strategic planning and nancial performance: a meta-analytic review.
J. Manag. Stud. 28 (4), 353374.
Bunker, R.J., 2013. Introduction: the Mexican cartelsorganized crime vs. criminal insur-
gency. Trends Organized Crime 16 (2), 129137.
Caloghirou, Y., Tsakanikas, A., Vonortas, N.S., 2001. University-industry cooperation in the
context of the European framework programmes. J. Technol. Transfer. 26 (12),
153161.
Crdenas, S., Cabrero, E., Arellano, D., 2012. The Difcult Linkage University-Enterprise in
Mexico: Towards the Construction of the Triple Helix? CIDE, Mxico
Cassiman, B., Veugelers, R., 2002. R&D cooperation and spillovers: some empirical evi-
dence from Belgium. Am. Econ. Rev. 92 (4), 11691184.
Centro de Investigacin y Docencia Econmica (CIDE), Secretaria de Educacin Pblica
(SEP), 2010e. Encuesta Nacional de Vinculacin Empresas, ENAVES. CIDE/SEP,
Mxico.
Chesbrough, H., 2012. Why companies should have open business models. MIT Sloan
Manag. Rev. 482, 2128.
Christensen, C.M., 1997. The Innovator's Dilemma: When New Technologies Cause Great
Firms to Fail. Harvard Business School Press, Boston.
Chung, J.W., Bae, Z.T., Kim, J.S., 2003. Changing patterns of technological cooperation ac-
tivities of innovative small rms along technological development stages in the Kore-
an telecommunication sector. Technovation 23 (2), 163173.
Cohen, W.M., Nelson, R.R., Walsh, J.P., 2002. Links and impacts: the inuence of public re-
search on industrial R&D. Manag. Sci. 48 (1), 123.
CONACYT, 2011. Informe General del Estado de la Ciencia y la Tecnologa 2010. Consejo
Nacional de Ciencia y Tecnologa, Mexico.
Crespi, G., Ziga, P., 2012. Innovation and productivity: evidence from six Latin American
countries. World Dev. 40 (2), 273290.
Cunningham, J.A., Link, A.N., 2015. Fostering university-industry R&D collaborations in
European Union countries. Int. Entrep. Manag. J. 11 (4), 849860.
Cunningham, J., O'Kane, C., O'Reilly, P., Mangematin, V., 2014. Managerial challenges of
publicly funded principal investigators. Int. J. Technol. Manag. 68 (3/4), 176202.
Czarnitzki, D., Ebersberger, B., Fier, A., 2007. The relationship between R&D collaboration,
subsidies and R&D performance: empirical evidence from Finland and Germany.
J. Appl. Econ. 22 (7), 13471366.
Czarnitzki, D., Lopes-Bento, C., 2013. Value for money? New microeconometric evidence
on public RD grants in Flanders. Res. Policy 421, 7689.
Damanpour, F., 1991. Organizational innovation: a meta-analysis of effects of determi-
nants and moderators. Acad. Manag. J. 34 (3), 555590.
De Fuentes, C., Dutrenit, G., 2012. Best channels of academiaindustry interaction for
long-term benet. Res. Policy 41 (9), 16661682.
De Fuentes, C., Dutrnit, G., 2016. Geographic proximity and universityindustry interac-
tion: the case of Mexico. J. Technol. Transf. 41 (2), 329348.
De Fuentes, C., Santiago, F., Dutrnit, G., Gras, N., 2015. Determinants of innovation and
productivity in the service sector in Mexico. Emerg. Markets Finance Trade 51 (3),
578592.
Delmar, F., Davidsson, P., Gartner, W.B., 2003. Arriving at the high-growth rm. J. Bus.
Ventur. 182, 189216.
Diario Ocial, 2014. Consejo Nacional de Ciencia y Tecnologa. Presidencia de la Repblica,
Estados Unidos Mexicanoshttp://www.derh.org.mx/21_ciencia_y_tecnologia.pdf
(last access October 2015).
Drejer, A., 2000. Organisational learning and competence development. Learn. Organ. 7
(4), 206220.
Dutrnit, G., De Fuentes, C., Torres, A., 2010. Channels of interaction between public re-
search organisations and industry and their benets: evidence from Mexico. Sci. Pub-
lic Policy 37 (7), 513526.
Edwards, S., 2001. Capital Mobility and Economic Performance: Are Emerging Economies
Different? (No. w8076). National Bureau of Economic Research.
Etzkowitz, H., Leydesdorff, L., 2000. The dynamics of innovation: from National Systems
and Mode 2 to a Triple Helix of universityindustrygovernment relations. Res. Pol-
icy 29 (2), 109123.
Faems, D., Van Looy, B., Debackere, K., 2005. Interorganizational collaboration and innova-
tion: toward a portfolio approach. J. Prod. Innov. Manag. 223, 238250.
Flores, M., Campos, M., Naranjo, E., Lucatero, I., Lopez, N., 2013. GEM Mexico 2013 Report.
Tecnolgico de Monterrey, Mexico.
Friar, J.H., Meyer, M.H., 2003. Entrepreneurship and start-ups in the Boston region: factors
differentiating high-growth ventures from micro-ventures. Small Bus. Econ. 212,
145152.
Fritsch, M., Lukas, R., 2001. Who cooperates on R&D? Res. Policy 30 (2), 297312.
Galende, J., de la Fuente, J.M., 2003. Internal factors determining a rm's innovative be-
haviour. Res. Policy 32 (5), 715736.
Gobierno de los Estados Unidos Mexicanos, 2006. Plan Nacional de Desarrollo 20072012.
Presidencia de la Repblica, Mexico (http://pnd.calderon.presidencia.gob.mx/pdf/
PND_2007-2012.pdf, last access October 2015).
Gobierno de los Estados Unidos Mexicanos, 2013. Plan Nacional de Desarrollo 20132018.
Presidencia de la Repblica, Mexico (http://pnd.gob.mx/wp-content/uploads/2013/
05/PND.pdf, last access October 2015).
Gooding, R.Z., Wagner III, J.A., 1985. A meta-analytic review of the relationship between
size and performance: the productivity and efciency of organizations and their sub-
units. Adm. Sci. Q. 462481.
Greene, W.H., 2003. Econometric Analysis. Prentice Hall, New York.
Grimaldi, R., Kenney, M., Siegel, D.S., Wright, M., 2011. 30 years after Bayh-Dole:
reassessing academic entrepreneurship. Res. Policy 40 (8), 10451057.
Guerrero, M., Cunningham, J.A., Urbano, D., 2015. Economic impact of entrepreneurial
universities' activities: an exploratory study of the United Kingdom. Res. Policy 44
(3), 748764.
Hall, B.H., 2005. The nancing of innovation. The Handbook of Technology and Innovation
Management, pp. 409430.
Hall, B.H., Lerner, J., 2010. The Financing of R&D and Innovation. Handbook of the Eco-
nomics of Innovation Vol. 1 pp. 609639.
Hall, B.H., Mafoli, A., 2008. Evaluating the impact of technology development funds in
emerging economies: evidence from Latin America. Eur. J. Dev. Res. 202, 172198.
Hall, B.H., Link, A., Scott, A., 2001. Barriers inhibiting industry from partnering with uni-
versities: evidence from the advanced technology program. J. Technol. Transfer. 26
(12), 8798.
Hausmann, R., Lozoya, E., Mia, I., 2009. The Mexico Competitiveness Report 2009. World
Economic Forum, Geneva.
Heinle, K., Molzahn, C., Shirk, D., 2015. Drug violence in Mexico: data and analysis through
2014. Special Report: Justice in Mexico Project. University of San Diego: Department
of Political Science & International Relations (www.justiceinmexico.org).
Himmelberg, C.P., Petersen, B.C., 1994. R D and internal nance: a panel study of small
rms in high-tech industries. Rev. Econ. Stat. 3851.
Hoskisson, R.E., Eden, L., Lau, C.M., Wright, M., 2000. Strategy in emerging economies.
Acad. Manag. J. 43 (3), 249267.
Howells, J., James, A., Malik, K., 2003. The sourcing of technological knowledge: distribut-
ed innovation processes and dynamic change. R&D Manag. 33 (4), 395409.
Huizingh, E.K., 2011. Open innovation: state of the art and future perspectives.
Technovation 31 (1), 29.
Ireland, R.D., Covin, J.G., Kuratko, D.F., 2009. Conceptualizing corporate entrepreneurship
strategy. Enterp. Theory Pract. 33 (1), 1946.
Kan, P.R., 2011. What we're getting wrong about Mexico. Parameters 41 (2), 3748.
Kenney, M., Goe, W.R., 2004. The role of social embeddedness in professorial entrepre-
neurship: A comparison of electrical engineering and computer science at UC Berke-
ley and Stanford. Res. Policy 33 (5), 691707.
Klepper, S., 2001. Employee start-ups in high-tech industries. Ind. Corp. Chang. 10,
639674.
Klomp, L., van Leeuwen, G., 2001. Linking innovation and rm performance: a new ap-
proach. Int. J. Econ. Bus. 8 (3), 343364.
 M. Guerrero, D. Urbano / Technological Forecasting & Social Change 119 (2017) 294309
Kuznetsov, Y., Dahlman, C., 2008. Mexico's Transition to a Knowledge-Based Economy:
Challenges and Opportunities. World Bank, Washington.
Landstrm, H., strm, F., Harirchi, G., 2015. Innovation and entrepreneurship studies:
one or two elds of research? Int. Entrep. Manag. J. 11 (3), 493509.
Laursen, K., Salter, A., 2006. Open for innovation: the role of openness in explaining inno-
vation performance among UK manufacturing rms. Strateg. Manag. J. 272, 131150.
Leyden, D.P., 2016. Public-sector entrepreneurship and the creation of a sustainable inno-
vative economy. Small Bus. Econ. 46 (4), 553564.
Leyden, D.P., Link, A.N., 2015. Public Sector Entrepreneurship: US Technology and Innova-
tion Policy. Oxford University Press, USA.
Leydesdorff, L., Ivanova, I.A., 2016. Open innovation and triple helix models of innovation:
can synergy in innovation systems be measured? J. Open Innov. (available at SSRN:
http://ssrn.com/abstract=2791914, in press).
McAdam, R., Miller, K., McAdam, M., Teague, S., 2012. The development of university tech-
nology transfer stakeholder relationships at a regional level: lessons for the future.
Technovation 32 (1), 5767.
McMullen, J.S., Dimov, D., 2013. Time and the entrepreneurial journey: the problems and
promise of studying entrepreneurship as a process. J. Manag. Stud. 50 (8),
14811512.
Meyer, K.E., Estrin, S., Bhaumik, S., Peng, M.W., 2009. Institutions, resources, and entry
strategies in emerging economies. Strateg. Manag. J. 30 (1), 6180.
Miller, K., McAdam, M., McAdam, R., 2014. The changing university business model: a
stakeholder perspective. R&D Manag. 44 (3), 265287.
Moreno, A.M., Casillas, J.C., 2007. High-growth SMEs versus non-high-growth SMEs: a dis-
criminant analysis. Entrep. Reg. Dev. 19 (1), 6988.
Nieto, M.J., Santamaria, L., 2007. The importance of diverse collaborative networks for the
novelty of product innovation. Technovation 27 (6), 367377.
OECD, 2007. Eurostat-OECD Manual on Business Demography Statistics. OECD, Paris.
OECD, 2010. High-Growth Enterprises: What Governments Can Do to Make a Difference,
OECD Studies on SMEs and Entrepreneurship. OECD Publishing http://dx.doi.org/10.
1787/9789264048782-en.
OECD, 2012. Science, Technology and Industry Outlook 2012. OECD, Paris.
Padilla-Prez, R., Gaudin, Y., 2014. Science, technology and innovation policies in small
and developing economies: the case of Central America. Res. Policy 43 (3), 749759.
Perkmann, M., Walsh, K., 2007. Universityindustry relationships and open innovation:
towards a research agenda. Int. J. Manag. Rev. 9 (4), 259280.
Porter, M., Schwab, K., 2008. Global Competitiveness Report 20082009. World Economic
Forum, Switzerland.
Ruelas, A., 2004. Innovar en Mercados Emergentes. Harv. Bus. Rev. 6271.
Santiago, F., De Fuentes, C., Dutrnit, G., Gras, N., 2016. What hinders innovation perfor-
mance of services and manufacturing rms in Mexico? Econ. Innov. New Technol.
http://dx.doi.org/10.1080/10438599.2016.1181297.
SEP, 2012. Mexican Educational System, 20112012 Indicators. Secretary of Public Educa-
tion of Mexico, Mexico.
Shrader, R.C., Simon, M., 1997. Corporate versus independent new ventures: resource,
strategy, and performance differences. J. Bus. Ventur. 12 (1), 4766.
Smith, R., 2009. Understanding entrepreneurial behaviour in organized criminals.
J. Enterprising Communities: People Places Glob. Econ. 33, 256268.
Sobel, R.S., 2008. Testing Baumol: institutional quality and the productivity of entrepre-
neurship. J. Bus. Ventur. 236, 641655.
Solleiro, J.L., Castan, R., 2005. Competitiveness and innovation systems: the challenges
for Mexico's insertion in the global context. Technovation 25 (9), 10591070.
309
Stam, W., Elfring, T., 2008. Entrepreneurial orientation and new venture performance: the
moderating role of intra- and extraindustry social capital. Acad. Manag. J. 511,
97111.
Stephan, P.E., 2012. How Economics Shapes Science. Harvard University Press, Cambridge,
MA, USA.
Surez, F.F., Oliva, R., 2005. Environmental change and organizational transformation. Ind.
Corp. Chang. 14, 10171041.
Thompson, J.L., 1999. A strategic perspective of entrepreneurship. Int. J. Entrep. Behav.
Res. 5 (6), 279296.
Thornhill, S., 2006. Knowledge, innovation and rm performance in high- and low-
technology regimes. J. Bus. Ventur. 215, 687703.
Thornton, P., Ribeiro-Soriano, D., Urbano, D., 2011. Socio-cultural factors and entrepre-
neurial activity: an overview. Int. Small Bus. J. 29 (2), 105118.
Tsai, K.H., Wang, J.C., 2009. External technology sourcing and innovation performance in
LMT sectors: an analysis based on the Taiwanese technological innovation survey.
Res. Policy 38 (3), 518526.
Urbano, D., Guerrero, M., 2013. Entrepreneurial universities socioeconomic impacts of ac-
ademic entrepreneurship in a European region. Econ. Dev. Q. 27 (1), 4055.
Van de Vrande, V., De Jong, J.P., Vanhaverbeke, W., Rochemont, M., 2009. Open innovation
in SMEs: trends, motives and management challenges. Technovation 29 (6),
423437.
van Leeuwen, G., Klomp, L., 2001. On the contribution of innovation to multi-factor pro-
ductivity growth. Paper Presented at the Eindhoven Centre for Innovation Studies
ECIS Conference, September 2023, 2001.
Wadhwa, A., Freitas, I.M.B., Sarkar, M.B., 2011. The paradox of being open: external tech-
nology sourcing and knowledge protection. In Paper Presented at the DIME Final
Conference vol. 6, p. 8.
Wang, C.L., Altinay, L., 2012. Social embeddedness, entrepreneurial orientation and rm
growth in ethnic minority small businesses in the UKInternational. Small Bus. J. 30
(1), 323.
Wiklund, J., Shepherd, D., 2005. Entrepreneurial orientation and small business perfor-
mance: a congurational approach. J. Bus. Ventur. 201, 7191.
Wright, M., Filatotchev, I., Hoskisson, R.E., Peng, M.W., 2005. Strategy research in emerg-
ing economies: challenging the conventional wisdom. J. Manag. Stud. 42 (1), 133.
Yamin, M., Otto, J., 2004. Patterns of knowledge ows and MNE innovative performance.
J. Int. Manag. 10 (2), 239258.
Zahra, S., Wright, M., 2011. Entrepreneurship's next act. Acad. Manag. Perspect. 25, 6783.
Zott, C., Amit, R., 2007. Business model design and the performance of entrepreneurial
rms. Organ. Sci. 18 (2), 181199.
Maribel Guerrero is a researcher in the Strategy Department at Deusto Business School,
University of Deusto. Her PhD was on Entrepreneurship and Small Business Management
at the Autonomous University of Barcelona (UAB), Spain. Her main research areas are re-
lated to intrapreneurship within public (entrepreneurial universities) and private organi-
zations (corporate entrepreneurship).
David Urbano is a Professor in Entrepreneurship of the Department of Business at
Universitat Autnoma de Barcelona (UAB Spain). His PhD was in Entrepreneurship
and Small Business Management (UAB and Vxj University, Sweden) and his research
is focused on the conditioning factors of entrepreneurship in different contexts using an
institutional approach.