Content uploaded by Seidali Kurtmollaiev
Author content
All content in this area was uploaded by Seidali Kurtmollaiev on Feb 15, 2023
Content may be subject to copyright.
1
A Behavioral Perspective on Open Innovation: A Defense Mechanism or an
Offensive Weapon?
Seidali Kurtmollaiev
Center for Service Innovation—Department of Strategy and Management
NHH - Norwegian School of Economics
Helleveien 30, N-5045
Bergen; Norway
Seidali.Kurtmollaiev@nhh.no
Keld Laursen
Department of Organizational Economics and Innovation
Copenhagen Business School; Kilevej 14a
2000 Frederiksberg; Denmark
kl.ino@cbs.dk
and
Center for Service Innovation—Department of Strategy and Management
Norwegian School of Economics
Helleveien 30, N-5045
Bergen; Norway
Per Egil Pedersen
Center for Service Innovation—Department of Strategy and Management
NHH - Norwegian School of Economics
Helleveien 30, N-5045
Bergen; Norway
Per.Pedersen@nhh.no
Kurtmollaiev, S., Laursen, K., & Pedersen, P. E. (2014). A Behavioral Perspective on Open
Innovation: A Defense Mechanism or an Offensive Weapon? Academy of Management
Proceedings, Vol. 2014, No. 1, p. 11665, doi.org/10.5465/ambpp.2014.11665abstract
Abstract. Drawing on the behavioral theory of the firm, we examine what drives the
implementation of a broad open innovation model entailing inbound and outbound elements. We
find two main categories of reasons—resource constraints on innovation, and growth
aspirations—representing respectively reactive and proactive approaches, which have different
origins. We argue that implementation of an open innovation model results from the interplay
between innovation resource constraints and growth aspirations. The central idea proposed in
this paper is that a combination of financial, knowledge-related, and market-centered resource
constraints leads the firm to adopt an innovation-based growth strategy, which in turn induces it
to adopt an open innovation model. We apply structural equation modeling to a representative
sample of 3,461 firms drawn from the Norwegian R&D and Innovation Survey conducted in
2008. We find support for the idea that the relationship between resource constraints and
implementation of an open innovation model is partially mediated by the firm’s growth strategy.
Keywords. Open innovation, behavioral theory, resource constraints, growth strategy.
2
ABehavioralPerspectiveonOpenInnovation:ADefenseMechanism
oranOffensiveWeapon?
Abstract
Drawingonthebehavioraltheoryofthefirm,weexaminewhatdrivestheimplementationofa
broadopeninnovationmodelentailinginboundandoutboundelements.Wefindtwomain
categoriesofreasons—resourceconstraintsoninnovation,andgrowthaspirations—
representingrespectivelyreactiveandproactiveapproaches,whichhavedifferentorigins.We
arguethatimplementationofanopeninnovationmodelresultsfromtheinterplaybetween
innovationresourceconstraintsandgrowthaspirations.Thecentralideaproposedinthispaper
isthatacombinationoffinancial,knowledge‐related,andmarket‐centeredresourceconstraints
leadsthefirmtoadoptaninnovation‐basedgrowthstrategy,whichinturninducesittoadopt
anopeninnovationmodel.Weapplystructuralequationmodelingtoarepresentativesampleof
3,461firmsdrawnfromtheNorwegianR&DandInnovationSurveyconductedin2008.Wefind
supportfortheideathattherelationshipbetweenresourceconstraintsandimplementationof
anopeninnovationmodelispartiallymediatedbythefirm’sgrowthstrategy.
Keywords
Openinnovation,behavioraltheory,resourceconstraints,growthstrategy
Introduction
Innovation practice in firms rarely relies solely on own resources; instead, it depends on a
combination of the own resources and firm-external resources (Garud, Gehman, & Kumaraswamy,
2011; Kyriakopoulos, 2011; Laursen & Salter, 2006; Oliver, 2001; Powell, Koput, & Smith-Doerr,
1996). It has been known for some time that firms seldom innovate alone. For instance, Rosenberg
(1963) describes the collective process of inventive activity in the machine tool industry in 1840-
1910, and Allan (1983) describes a similar process in the context of the English iron industry in
the second part of the 19th century. Allan describes how firms made their new plant designs
available to competitors, which allowed the latter to integrate these innovations in the new and
improved facilities they built. In addition, the early innovation literature demonstrates that
successful innovators—compared to unsuccessful innovators—paid more attention to user needs,
3
and made more use of outside technology and scientific advice (Rothwell, Freeman, Jervis,
Robertson, & Townsend, 1974).
A more recent example of the innovation process conceptualized as open and distributed, is
Henry Chesbrough’s (2003) open innovation model. Recent changes to managerial practice related
to innovation have led to firms “increasingly applying inflows and outflows of knowledge to
accelerate internal innovation, and expand the markets for external use of innovation”, i.e. to firms
increasingly adopting open innovation models (Chesbrough, 2006). Chesbrough (2003) proposes
the stylized argument that this shift represents a move from “closed” to “open innovation”, where
firms draw knowledge from a range of external actors to develop and commercialize new
technology. According to Chesbrough, the emergence of this shift process at this point in time is
related to certain “erosion factors” which imply a decreasing possibility to appropriate the benefits
deriving from corporate research and development (R&D) investments. These erosion factors
include in particular greater labor mobility, increasingly strong “software movements”, greater
division of labor, and—at least in some countries—more availability of venture capital. Inspired
by the concept of open innovation, more and more researchers are studying innovation as the
outcome of an interactive process between the firm and its external environment (for overviews of
this literature, see Dahlander & Gann, 2010; West & Bogers, 2014).
However, while the open innovation literature yields some very important insights into the
effects of various types of open innovation practices, little research has been conducted on the
strategic reasons related to firms’ adoption of an open innovation model to various degrees within
the same context (for exceptions, see Drechsler & Natter, 2012; Laursen & Salter, 2014; van de
Vrande, de Jong, Vanhaverbeke, & de Rochemont, 2009). Chesbrough’s generic erosion factors
explain why open innovation practices in general are becoming more widespread but do not
4
explain firm-level heterogeneity in their adoption. In particular, it is unclear whether open
innovation is driven by the need to compensate for a lack of resources, or is a means to stimulate
innovation-based growth. This is the question addressed in this paper. Although the first obvious
answer might be that both motivations apply, on closer inspection, it can be seen that they may be
conflicting. The first category of motives represents a reactive approach to open innovation, while
the second represents a proactive approach.
To the best of our knowledge, this is the first study to investigate this issue. We draw on the
behavioral theory of the firm, and the problem solving perspective of the knowledge-based theory
of the firm in order to theorize about how and why innovation-based growth strategies and resource
constraints trigger firms’ adoption of a broad open innovation model. We employ Huizingh’s
(2011) distinction between defensive motives (coping with challenges) and offensive motives
(market-related motives) for adopting an open innovation model. We argue that implementation
of an open innovation model is the outcome of the interplay between resource constraints for
innovation and growth aspirations. The principal notion supporting our behavioral approach is that
the combination of financial, knowledge-related, and market-centered resource constraints leads
to innovation-based growth aspirations (growth strategy) which in turn induce (resource
constrained) firms to adopt an open innovation model.
We use a sample of 3,461 Norwegian firms drawn from the 2008 Community Innovation
Survey (CIS 2008) for the empirical analysis. We estimate a structural equation model (SEM), in
which the obstacles to innovation (financial, knowledge-related, and market related) have both
direct and indirect effects on the adoption of an open innovation model. The indirect effect works
through the influence that the obstacles (or constraints) to innovation have on the degree to which
firms aspire to an innovation-based growth strategy—which in turn should affect the degree to
5
which firms adopt a broad set of open innovation practices. The results show that the relationship
between resource constraints and the implementation of an open innovation model is partially
mediated by the firm’s growth strategy. In other words, open innovation is primarily a tool for
achieving growth goals through innovation, rather than a direct response to the factors hampering
the innovation process.
Our study is novel first in providing a theoretical and empirical understanding of the
mechanisms that can lead firms to adopt open innovation strategies. In particular we propose a
behavioral theory of how the innovation-based growth aspirations and the constraints to innovation
that firms face, together trigger their adoption of a broad set of open innovation practices. In this
context van de Vrande et al. (2009) demonstrate that medium-sized firms, on average are more
heavily involved in open innovation than smaller companies. They find also that small and medium
sized enterprises (SMEs) pursue open innovation strategies primarily for market-related motives
such as meeting customer demands, or keeping abreast of competitors. Garriga et al. (2013) show
that an abundance of knowledge in the external environment combined with internal constraints to
innovation can affect the degree of firm openness in relation to the depth and breadth of external
knowledge search. Drechsler and Natter (2012) show that combining depth and breadth of external
foreign and domestic knowledge search is a function of the innovation strategy, the scarcity of
own resources, the appropriability regime under which the firm operates, and the market dynamics.
Laursen and Salter (2014) show that breadth of both search and collaboration for innovation is tied
closely linking to the firm’s appropriability strategy. However, van de Vrande et al. (2009) is based
on descriptive evidence, not statistical analysis. Garriga et al.’s (2013) and Drechsler and Natter’s
(2012) contributions, which include innovation obstacles, are the most closely related to ours, and
Drechsler and Natter (2012), moreover, also include an explicit measure of innovation strategy.
6
However, neither of these studies investigates the broad adoption of open innovation practices;
rather they focus on inbound open innovation in terms of external search. We use a broader
measure that includes both inbound and outbound open innovation (Chesbrough, 2003; Dahlander
& Gann, 2010). Also Garriga et al. and Drechsler and Natter provide relatively linear stories of
how the independent variables increase the use of external search whereas our study looks at the
degree to which the link between innovation obstacles and the adoption of open innovation
practices is mediated by the importance of the firm’s innovation-based growth strategy.
The second novelty is the way of analyzing complex constructs and relationships emerging in
the innovation process based on categorical indicators with the use of structural equation
modeling’s method of weighted least squares. This study is a very rare example of innovation data
used to model complex constructs and therefore provides an empirical contribution to the
innovation literature.
TheoreticalBackground
We have noted above that open innovation “involves the use of purposive inflows and outflows of
knowledge to accelerate internal innovation, and expand the markets for external use of innovation,
respectively” (Chesbrough, 2006: 1). This definition makes it clear that the concept of open
innovation has a number of different dimensions. An organization can open up its innovation
process via inbound open innovation (incoming knowledge from external sources) and outbound
open innovation (outgoing knowledge from the organization), each of them being either pecuniary
or non-pecuniary in nature (Dahlander & Gann, 2010). Open innovation can also result in closed
or open product outcomes (Huizingh, 2011). So far, research has focused mostly on the links
between the organization and external sources of innovation (e.g., Brusoni, Prencipe, & Pavitt,
2001; Chesbrough & Crowther, 2006; Laursen & Salter, 2006; Leiponen & Helfat, 2010) and the
7
characteristics of the open innovation process in different organizations (e.g., Bianchi, Cavaliere,
Chiaroni, Frattini, & Chiesa, 2011; Chesbrough, 2007; Dodgson, Gann, & Salter, 2006; Henkel,
2006). In this paper, we are proposing a behavioral theory of why firms adopt an open innovation
model.
The behavioral theory of the firm (Cyert & March, 1963; Gavetti, Greve, Levinthal, & Ocasio,
2012; Tippmann, Mangematin, & Scott, 2013) posits that one of the types of search that firms
undertake is “problemistic search” (i.e. search triggered by a problem). Problemistic search is
prompted by managers recognizing that organizational performance, which is in part a function of
past performance (and/or competitors’ performance), is below aspiration levels. In addition to
(poor) economic performance driving firms to search for new solutions (e.g. by adopting an open
innovation model) according to the problem-solving perspective of the firm (Laursen & Salter,
2006; Nickerson & Zenger, 2004), lack of knowledge inside the firm can be an important
motivation for external search. In sum, when the organization is under pressure, it can increase its
search for innovation if decision makers judge that upgrading their technology and product
portfolio could solve performance problems (Argote & Greve, 2007; Greve, 2003; Greve &
Taylor, 2000).
A problem related to empirical analysis in line with the behavioral approach is how managers’
perceived aspiration levels should be calculated since this central variable is not directly
observable. Solutions include relying on the organization’s past performance (Greve, 1998; Wezel
& Saka-Helmhout, 2006), measuring the past performance of a relevant reference group of other
organizations (e.g. Audia, Locke, & Smith, 2000; Greve, 2008), or taking the weighted average of
these two measures (e.g., Greve, 2003). The assumption is that these measures give an indication
of the focal firm’s aspirations; however, actual aspirations are unobserved. In this paper, we take
8
a different empirical approach based on the idea that managers’ perceived financial, knowledge-
related and market-centered resource constraints regarding innovation contain information about
the focal firm’s aspiration level. That is, a high level of perceived constraints is a manifestation of
high organizational aspirations related to innovation which in turn is related to an innovation-based
growth strategy. This strategy increases the probability that the focal firm subsequently will adopt
an open innovation model.
Figure 1 provides a visualization of our theoretical model. Although one of the novelties of the
present paper is the notion that the effect of perceived constraints to innovation on the
implementation of the broad open innovation model is mediated by an innovation-based growth
strategy, following Drechsler and Natter (2012) and Garriga et al. (2013), we acknowledge the
possibility of a direct effect of these perceived constraints on the implementation of an open
innovation model. We include this effect in our model and hypothesize about it (in Hypothesis 1)
from a behavioral perspective.
Figure 1. Theoretical model
Market
Stagnation
Financial
Shortage
Open
Innovation
Innovation-
Based
Growth
Strategy
H1a
+
H1c
+
H1b
+
H2a
+
H2c
+
Lack of
Knowledge
H3
+
H2b
+
9
HypothesesDevelopment
Openinnovationasaresponsetoperceivedproblems
As mentioned above, the idea of innovation as a response to organizational problems is not new.
For example, Greve (2003) and Vissa et al. (2010) argue and corroborate empirically that
organizations innovate to improve poor performance by engaging in problemistic search. In the
context of financial performance, firms with declining profits can embark on R&D races to restore
profitability (Antonelli, 1989; Kamien & Schwarz, 1982). In extending Bolton’s (1993) work,
Greve (2003) suggests that economic performance below aspiration levels not only motivates
managers to search for solutions, it also makes them more likely to accept risky solutions such as
additional R&D spending. In the context of open innovation, a risky solution might become
acceptable in order to open up the innovation process. Open innovation is considered a risky and
rather costly strategy for two related reasons. One is that working with external knowledge sources
is costly since all sources can be considered search channels that might lead to recombination and
innovation. According to Laursen and Salter (2006: 133), these external knowledge sources should
be considered “a separate search space, encompassing different institutional norms, habits, and
rules, often requiring different organizational practices in order to render the search processes
effective within the particular knowledge domain.” Thus, there is a cost connected to each search
channel that the firm engages with, and at the same time, there is no guarantee that the effort will
result in a successful innovation. The second reason is that external engagement carries the
possibility of knowledge leakages which firms have to take (costly) action to prevent, by
implementing appropriability strategies (Cassiman & Veugelers, 2002; Laursen & Salter, 2014;
Lavie, 2006). Nevertheless, some leakage of knowledge is inevitable (Levin, Klevorick, Nelson,
& Winter, 1987; Mansfield, Schwartz, & Wagner, 1981; Roper, Vahter, & Love, 2013).
10
On the benefits side, resource-constrained organizations can save on costs (despite the fact—as
pointed out above—that external knowledge sourcing and exchange is not a free endeavor) by
making actors external to the firm contribute to the innovations of the focal firm, and through
access to a broad range of ideas for innovation. External actors are often willing to contribute to
the focal firm’s innovation processes because this allows them to benefit from the resulting
innovation in some way (von Hippel, 1988). Arguably, the breadth of the ideas available from
external channels and the cost-saving aspect, are at the core of conceptualizations of the benefits
of open innovation since this notion was proposed (Chesbrough, 2003). In other words, lack of
financial resources will induce firms to perform problemistic search to find cheaper ways to
innovate. In this case, inbound open innovation may be the more attractive route. In the case of
outbound open innovation, lack of financial resources will force managers to consider whether the
focal firm has valuable intellectual property (IP) that could be sold or licensed out to generate
revenue to finance the firm’s operations. Every penny counts in this case: The luxury to be sure
that new competitors will not emerge as a consequence of traded IP may be unaffordable for
financially constrained firms. In other words, financially constrained firms will be more likely to
bring their IP to the market. Considering the arguments related to both inbound and outbound open
innovation, we posit that:
Hypothesis 1a. Lack of finance to support the organization’s own innovation activity
has a positive effect on opening up its innovation process.
The firm may not only be constrained economically by a lack of financial resources, it may also
be constrained by the characteristics of the given main market (stagnant or growing) in which it
operates. In line with the behavioral approach, we propose that if firms find themselves caught in
markets that offer few opportunities, they will be motivated to search for solutions to resolve the
11
stagnation problem and to allow them to utilize their existing technological and organizational
resources while diversifying their product portfolios towards more dynamic markets (Dougherty,
1992). One tactic allowing firms to make such a move would be an innovator strategy. Given that
firms in this situation are resource constrained, and given that they may lack knowledge of the new
context, they may try to engage in inbound open innovation. Similarly, in order to gain the
necessary pecuniary and non-pecuniary (inbound knowledge in the exchange for outbound
knowledge) resources to enable a diversification strategy, firms can sell or license out some of
their IP (outbound open innovation). In sum:
Hypothesis 1b. Operating in markets that are characterized by stagnation has a positive
effect on opening up the organization’s innovation process.
Near to medium term economic concerns may not be the only constraints leading firms to seek
a new business model in the context of innovation, the problem-oriented nature of the
technological search process may also induce the focal firm to perform more search, i.e. lack of
firm-internal knowledge triggers more internal and external search. Extending the insights
provided by evolutionary economics (Nelson & Winter, 1982), and proposing a problem-solving
perspective on a knowledge-based theory of the firm, Nickerson and Zenger (2004: 618-619)
advocate that:
…if a firm is to develop unique knowledge or a unique new capability through any
manner other than luck, it must identify a valuable problem and conduct an efficient
solution search. Valuable solutions deliver value to the firm, either through
enhancement or development of a product or service or by reducing the cost of
production or delivery.
12
In the particular context of innovation, Rosenberg (1969) pointed out that bottlenecks and
imbalances in technological progress induce and shape the direction of technological search. He
demonstrated that technological problems and related lack of knowledge often spur firms’
technological search efforts in the innovation process. Some innovation related search may be
internal (Chen & Miller, 2007; Greve, 2003; Katila & Ahuja, 2002; Yayavaram & Ahuja, 2008),
some may involve external organizations (Laursen & Salter, 2006; Leiponen & Helfat, 2010;
Rothaermel & Alexandre, 2009; Shan, Walker, & Kogut, 1994). In many cases the knowledge
required to solve a new problem does not coincide with the organization’s current knowledge base,
and the process goes beyond the organization’s boundaries in a search for complementary
knowledge (Jeppesen & Laursen, 2009; Laursen & Salter, 2006; Postrel, 2002). There is an
abundance of external knowledge for potential problem solutions available compared to the
knowledge that exists in individual organizations; in addition, external knowledge sources may
offer a different perspective on a given problem (Jeppesen & Lakhani, 2010; von Hippel, 1994).
Although lack of internal knowledge is most relevant in the context of inbound open innovation,
it may also motivate engagement in outbound pecuniary forms of open innovation since inbound
open innovation (whether pecuniary or non-pecuniary) is costly (Laursen & Salter, 2006) and may
require the firm to raise resources by selling parts of its intellectual assets. Moreover, the use and/or
purchase of external knowledge is often linked directly to sharing and selling knowledge as part
of formal and informal knowledge exchange between firms (see, Gambardella & Hall, 2006; von
Hippel, 1987). In sum, these arguments lead us to hypothesize:
Hypothesis 1c. Lack of relevant knowledge for the organization’s own innovation
activity has a positive effect on opening up the organization’s innovation process.
Constraintsinducinganinnovation‐basedgrowthstrategy
13
Strategy also includes alignment of the firm to the opportunities and barriers in its environment
(Chesbrough & Appleyard, 2007). In line with the behavioral theory of firm growth (Audia &
Greve, 2006; Desai, 2008; Greve, 2008), we assume that the decision to adopt open innovation
depends on the organizational strategy, which in turn is affected by problems related to the
organization’s environment. The central notion in the behavioral approach to firm growth is that
when the firm’s resources do not match its aspiration levels there will be a stronger incentive to
adopt a growth-based strategy. Greve (2008: 479) notes that:
…although managers have aspiration levels for performance goals as well as for size
goals, there is less justification for positing that an organization pursues performance
goals more weakly when it is above its aspiration level for performance. Instead,
managers will be motivated to lock in high organizational performance by avoiding
risky actions.
Since organizational size and hence growth are not a “part of the organizational core” (Greve,
2008: 479), if the organization is not facing serious problems, managers will not have a strong
incentive to attempt a growth strategy. Certainly, organizational size becomes subject to satisficing
behavior because efficiency and legitimacy as opposed to performance measures such as
profitability and stock market value, do not justify a very large size. In the context of knowledge
and innovation, we argue that when organizations perceive severe financial, knowledge-related, or
market-centered resource constraints, they are more likely to opt for an innovation-based growth
strategy. These perceived constraints which demonstrate to top-managers that the organization
does not meet aspiration levels, make it more likely they will accept risky solutions to problems,
which includes pursuing an innovation-based growth strategy. We hypothesize that:
14
Hypothesis 2a. An organization’s tendency to pursue an innovation-based growth
strategy is positively affected by lack of financial resources.
Hypothesis 2b. An organization’s tendency to pursue an innovation-based growth
strategy is positively affected by market stagnation.
Hypothesis 2c. An organization’s tendency to pursue an innovation-based growth
strategy is positively affected by lack of knowledge.
Openinnovationasameanstoanend
In line with the literature on open innovation (Drechsler & Natter, 2012; Garriga et al., 2013), the
behavioral theory of the firm, and the problem-solving perspectives, we have argued that economic
and knowledge-related constraints push firms to open up their innovation processes. However, we
would suggest that, openness to innovation also requires a deliberate growth-oriented strategy.
One way to achieve growth and profits is to adopt a deliberate strategy focusing on innovation
(Bierly & Chakrabarti, 1996). Product innovation is an important way to achieve growth through
either increased market share based on superior products or entry into new markets (Banbury &
Mitchell, 1995; Blundell, Griffiths, & Reenen, 1999). Inbound open innovation is clearly a way to
strengthen firms’ innovative output (Chesbrough, 2003; Laursen & Salter, 2006; Love, Roper, &
Vahter, 2013) while, in the context of outbound open innovation, revenue growth can be achieved
by selling and licensing out technologies as long as this does not threaten the firm’s own product
market (Arora & Ceccagnoli, 2006; Fosfuri, 2006; Gambardella & Giarratana, 2013). In sum, we
hypothesize that:
Hypothesis 3. The pursuit of an innovation-based growth strategy is positively related
to opening up the organization’s innovation process.
15
EmpiricalAnalysis
Thedata
The empirical study is based on the Norwegian R&D and Innovation Survey carried out by
Statistics Norway, as part of the CIS 2008. The CIS survey method and questions are based on
Oslo Manual (1st edition) published in 1992 by Eurostat/OECD. Laursen and Salter (2006) point
out that CIS surveys are “subject-oriented” and draw on a long tradition of research on innovation.
CIS data have been used in numerous studies because of their extensively tested interpretability,
reliability, and validity.
The Norwegian R&D and Innovation Survey consists of two parts. The first asks about R&D
activities (in 2008), described as “creative work undertaken on a systematic basis in order to
increase knowledge and the use of this knowledge to devise new applications” (Wilhelmsen, 2011:
109). The second part asks about the scope, objectives, hindrances, types, sources, etc., of the
innovation activities carried out between 2006 and 2008. It is assumed that innovation is “based
on [the] results of either technology development, new combinations of existing technology or
exploitation of other knowledge acquired by the enterprise” (Wilhelmsen, 2011: 113). The survey
covers the manufacturing, mining, agricultural, fishing, oil/gas production industries and most
service sectors. It includes firms with five or more employees. In June 2009 the questionnaire was
sent to 6,225 enterprises (from the population of 17,273 firms); the response rate was 97%. The
survey was mandatory under the Norwegian Statistics Act, so enterprises were contacted to remind
them to respond, and two additional reminders were sent before violators were fined. A high
response rate is commonly acknowledged to avoid non-response bias (Armstrong & Overton,
1977: 396). The sample is representative of the population in the surveyed industries and across
Norwegian geographical regions. The responses were scrutinized to weed out obvious mistakes,
16
and after the first registration of data, approximately 40 mechanical controls were carried out. The
responses from previous surveys and structural statistics were used as a reference during data
revision (Wilhelmsen, 2011).
To study the relationships between barriers to innovation, firm strategy and open innovation, a
subsample (4,858 firms) of the survey firms is used in this paper, including enterprises in all the
manufacturing and service industries surveyed. Standard Industrial Classification (NACE) codes
are shown in Appendix 1. Also, similar to Tether (2002) and Drechsler and Natter(2012), we
restricted our analysis to enterprises that engaged in or showed an interest in innovation activity
by including only firms that answered “yes” to at least one of the questions concerning either
innovation barriers, innovation goals or any type of innovation activity (selling or buying
knowledge, engaging in innovation cooperation, having ongoing or abandoned innovation
activities, or introducing any type of innovation). This reduced the sample size to 3,627 firms, and
after listwise deletion of incomplete data, we obtained a final sample of 3,461 firms.
Measuresandtools
Previous studies based on CIS data (e.g., Drechsler & Natter, 2012; Garriga et al., 2013; Laursen
& Salter, 2006; Tether, 2002), mostly use traditional multivariate techniques (mainly regression
analysis). However, traditional regression analysis has been criticized in cases where models
include complex constructs (latent variables) that cannot be measured directly, and variables that
are simultaneously outcomes and predictors (Bartholomew, Steele, Moustaki , & Galbraith, 2008;
Byrne, 1998; Kline, 2011). To test our hypotheses, we apply SEM using LISREL 8.71 in this
study. Since all the indicators are either dichotomous or ordinals, they do not follow a normal
distribution, and according to Joreskog (2002: 1), do not “have origins or units of measurements”,
so their “means, variances and covariances… had no meaning”. Thus we applied the weighted
17
least squares (WLS) estimation, which is an appropriate method for the analysis of ordinal
variables. The estimation required the calculation of polychoric correlations and their asymptotic
covariance matrix (Flora & Curran, 2004; Joreskog, 2002; Kline, 2011; Stolzenberg, 2003).
Following the Oslo Manual (2005), Segarra-Blasco et al. (2008) and D’Este et al. (2012), three
constructs representing exogenous latent variables can be formulated as follows: 1) Financial
Shortage is measured by three indicators representing high innovation cost and lack of financial
resources inside/outside the organization; 2) Lack of Knowledge is measured by four observed
variables indicating lack of market and/or technological information and problems with personnel
and/or partners; 3) Market Stagnation is measured by three indicators reflecting market domination
and uncertain or insufficient demand. A detailed description of all the constructs and indicators is
contained in Appendix 2.
The firm’s tendency to pursue an Innovation-based Growth Strategy is measured by the four
indicators related to growth goals (Bridge & Dodds, 1978; Kumar, 2010; van de Vrande et al.,
2009) of increasing the range of products or services, increasing market share, entering new
markets, and increasing the capacity of production or services provision. This construct is used as
both an outcome and a predictor. Open Innovation is measured by three observed variables that
indicate innovation activities associated with external relationships (selling R&D services to
external organizations, buying R&D services from others, and the variable reflecting breadth of
cooperation with other enterprises introduced by Laursen & Salter, 2006).
The polychoric correlations of the indicators are presented in Appendix 3. We assess the
measurement model with a WLS confirmatory factor analysis in LISREL. The minimum fit-
function χ2 is 728.95 and the degrees of freedom of the model are 109, so in terms of exact fit, the
model could be rejected although the high value of this test is very likely to be explained by the
18
large size of the sample and non-normality of ordinal data (Bartholomew et al., 2008; Hu, Bentler,
& Kano, 1992; Joreskog, 2004; Kline, 2011). However, fit statistics indicate a very good fit
(RMSEA = 0.041, CFI = 0.98, RFI = 0.97, SRMR = 0.07, NNFI = 0.97, IFI = 0.98), with RMSEA
less than 0.05, CFI, RFI, NNFI, IFI and GFI greater than 0.9 and SRMR below the cutoff value of
0.08 (Bartholomew et al., 2008). Thus, we can conclude that the model fit is acceptable. Factor
loadings for the measures are shown in Appendix 4. All hypothesized factor loadings are
statistically significant, and the constructs have acceptable levels of composite reliability and
average variances extracted (AVE), indicating convergent validity.
Regarding discriminant validity, we found a slightly higher correlation between Lack of
Knowledge and Market Stagnation than the square root of the corresponding AVE. Although this
issue could be entirely resolved by removing items 11 and 17, we believe these items represent
theoretically important dimensions of the underlying concepts and thus chose to retain them. We
conducted an additional test by collapsing Lack of Knowledge and Market Stagnation into one
construct; this model has a significantly worse fit than the hypothesized model (Δχ2 [Δd.f.] =
119.98 [4]), and the new construct does not make theoretical sense (Farrell, 2010). Accounting
further for the high composite reliabilities (0.94 for Lack of Knowledge and 0.90 for Market
Stagnation) and large sample size, we can conclude that this minor issue is unlikely to affect the
results of the subsequent analyses (see Grewal, Cote, & Baumgartner, 2004).
Research involving cross-sectional survey data such as those used in this study, is susceptible
to common method bias (CMB). To help reduce the effects of consistency artifacts, the CIS
questionnaire design includes different types of responses (in line with the recommendations made
in Salancik & Pfeffer, 1977). For example, response types included Likert scales, yes/no answers,
indications of percentages, and absolute numbers. Nevertheless, CMB might still be an issue. To
19
check this, first, we conducted a Harman’s one-factor test on the items included in our model.
Since we identified multiple factors and since the primary factor accounts for less than the majority
of the variance (the first factor accounted for 30% of the variance), this test does not indicate
potential problems associated with CMB (Podsakoff & Organ, 1986). Second, we employed the
single factor procedure based on confirmatory factor analyses (see, Podsakoff, MacKenzie, Lee,
& Podsakoff, 2003), to further test for potential CMB problems. Specifically, we examined the fit
of the single factor model in which all items are loaded onto one factor. The reasoning behind the
single factor procedure is that if CMB is responsible for most of the co-variation among the
constructs, confirmatory factor analysis should show that a single factor model fits the data. Fit
statistics for the single factor model are: χ2=2358.56, RMSEA=0.74, SRMR=0.25, CFI = 0.91,
RFI = 0.90, NNFI = 0.90, IFI = 0.91. These statistics show that the model is not very representative
of the data. Finally, we created a measurement model in LISREL, with a common latent factor
connected to all observed items (a single unmeasured latent method factor). The solution was non-
admissible (indicating a poor model), but when we drop the admissibility check, we find no major
(0.2 criterion) differences in the standardized regression weights of this model and the
corresponding standardized regression weights of the model without the common latent factor
(Podsakoff et al., 2003). While these statistical tests do not totally eliminate potential CMB, they
provide evidence that inter-item correlations are not driven by it to any important extent.
Results
Parametertests.Figure 2 shows the parameter estimates and fit statistics for the structural model
obtained from the LISREL WLS analysis. The coefficients of determination (squared multiple
correlations) are 0.25 and 0.29 respectively for Open Innovation and Innovation-based Growth
Strategy.
20
According to Figure 2, as expected, Financial Shortage positively influences Open Innovation
implementation (H1a). However, the relation between Open Innovation and Lack of Knowledge
constraint is insignificant (p>0.1) (H1c). Also, in contrast to our expectations, Market Stagnation
and Open Innovation (H1b) have a negative coefficient. Financial Shortage and Lack of
Knowledge are positively and significantly related to an Innovation-based Growth Strategy. This
finding is in line with hypotheses H2a and H2c. However, contrary to our expectations (H2b),
Market Stagnation is negatively associated with an innovation-based Growth Strategy. In line with
our expectations, a strategy of Innovation-based Growth appears to be positively related to Open
Innovation (H3).
Figure 2. Parameter Estimates for the Structural Model
Notes: χ
2
= 728.95; df = 109; RMSEA = 0.041; CFI = 0.98; RFI = 0.97; SRMR = 0.070; NNFI = 0.97; IFI = 0.98;
**
p<0.001;
*
p<0.05
Market
Stagnation
Financial
Shortage
Open
Innovation
Innovation-
Based
Growth
Strategy
H1a
0.23
**
H1c
0.21 n.s
H1b
-0.16
**
H2a
0.18
**
H2c
0.52
**
Lack of
Knowledge
H3
0.40
**
H2b
-0.35
*
21
Mediationtest. The test for mediating effects of the growth strategy is based on a two-level
procedure described by Holmbeck(1997). Although Innovation-based Growth Strategy and Open
Innovation are both dependent variables in the model, to explain the mediation test, Innovation-
based Growth Strategy is described here as a mediator, and Open Innovation is the dependent
variable.
The first step in the procedure includes assessing the fit of the direct effect model. Here,
Holmbeck (1997) follows the logic of Baron and Kenny (1986), who suggest estimating the
equations under three conditions to check the significance of linkages in the models with mediating
variables (assuming acceptable fit): 1) the independent variable must significantly affect the
dependent variable (in the direct effect model); 2) the independent variable must significantly
affect the mediator; 3) the mediator must significantly affect the dependent variable. If there is
mediation, the effect of the independent variable on the dependent variable will be less under the
third condition compared to the second condition. The test of the direct effect model (without the
Innovation-based Growth Strategy variable) is presented in Table 1. The values for the two other
conditions are depicted in Figure 2. The coefficient of determination is 0.14.
Table 1. Mediation tests
Parameter estimates and fit indices
H1a 0.38**
H1b -0.41**
H1c 0.34*
Fit indices χ2 = 371.44
df = 59
RMSEA = 0.039
CFI = 0.98
RFI = 0.98
SRMR = 0.053
NNFI = 0.98
IFI = 0.98
Note: **p<0.001; *p<0.01
22
The direct effect model is found to fit the data, so the second step in the estimation of mediating
effects according to Holmbeck (1997), is to compare the fit of the model under two conditions: 1)
when the direct path between the independent and the dependent variables is constrained to zero;
and 2) when this path is not constrained. If there is mediation, adding the direct path will not
improve the model fit (or, in our case, deleting the direct paths will not worsen the fit). Figure 3
shows the parameter estimates and fit statistics for the model without a direct path between Lack
of Knowledge and Open Innovation. The difference in minimum fit-function χ
2
is 3.79 (
df=1),
which is less than 3.84 (p<0.05) and thus statistically insignificant. This indicates that Innovation-
based Growth Strategy fully mediates the relationship between Lack of Knowledge and Open
Innovation.
Figure 3. Parameter Estimates for the Final Structural Model
Notes: χ
2
= 732.74; df = 110; RMSEA = 0.040; CFI = 0.98; RFI = 0.97; SRMR = 0.070; NNFI = 0.97; IFI = 0.98;
**
p<0.001;
*
p<0.05
Market
Stagnation
Financial
Shortage
Open
Innovation
Innovation-
Based
Growth
Strategy
H1a
0.28
**
H1b
-0.16
**
H2a
0.18
**
H2c
0.52
**
Lack of
Knowledge
H3
0.42
**
H2b
-0.20
*
23
The paths between Financial Shortage and Open Innovation, and between Market Stagnation
and Open Innovation do not become insignificant after adding Innovation-based Growth Strategy
to the model, but the magnitude of these relationships is dramatically reduced. This indicates that
Innovation-based Growth Strategy partially mediates the effects of Financial Shortage and Market
Stagnation on Open Innovation (Holmbeck, 1997; Iacobucci, Neela S., & D., 2007).
Alternativeexplanations
To test whether our results might be driven by alternative explanations, we carry out a number
robustness checks. For reasons of space, the results are not reported here but are available on
request. These alternative explanations include the fact that some firms in our sample are more
knowledge intensive than others, that they are of unequal size, and that they come from different
industrial contexts. They may also have different appropriability strategies, and provided that firms
with a stronger emphasis on appropriability may be much better prepared for open innovation
(Laursen & Salter, 2014), the firm’s appropriability strategy could be a confounding factor in
explaining implementation of an open innovation model (and could also explain in part the degree
to which firms follow an innovation-based growth strategy).
First, we conduct group analyses for different groups of firms, two-by-two,1 comparing (i)
smaller firms to large firms; and (ii) service firms to manufacturing firms. Overall, for each
comparison, the model holds for both groups separately (although inevitably some significance is
lost in each case). This suggests that the results are not driven by differences in firm size, or
whether the firm’s primary activity can be categorized as services or manufacturing.
In addition to the group analysis, we carried out three single-equation ordinary least square
regressions in Stata, all including a set of control variables to test the seven relations we
1 Group analysis bears some resemblance to including control variables in a regression analysis, because it tests for
interaction effects of a group variable on the model parameters.
24
hypothesized. Our controls include variables reflecting R&D intensity, firm size, emphasis on
appropriability strategy (following Laursen & Salter, 2014, we include involving patents,
registration of design, trademarks, secrecy, lead time, and complexity as the items in firms’ overall
appropriability strategy), and 38 industry dummies. In the first regression, we explain the
Innovation-based Growth Strategy by the three constraint variables plus the controls. In the second
regression, we explain implementation of a broad Open Innovation model by Innovation-based
Growth Strategy and the controls. In the third regression, we explain Open Innovation by the
Innovation-based Growth Strategy, the three constraint variables, and the control variables.
Although the control variables all help to explain the dependent variables, in all cases, in terms
of signs and significance, the SEM results and our hypothesized results are almost identical. The
only important difference is that in the SEM, Market Stagnation has a positive (and significant)
sign in its relation to the Innovation-based Growth Strategy. However, given that this result does
not change regardless of whether we include the controls in the regression analysis, for two reasons
we are more confident about the negative and significant result emerging from the SEM exercise
(despite the result contradicting our expectations). First, multicollinearity is explicitly modeled in
the SEM case, but is assumed not to be present in the regression case. Second, simultaneous
modeling of all the relationships is appropriately performed in a single model in the SEM case,
while single-equation regressions ignore these “cross-relationship” effects. Explicit simultaneous
modeling of all the relationships in a single model typically alters the sizes but not usually the
signs of the coefficient (Wooldridge, 2002).
Another concern is over whether the constraints or obstacles to innovation measure “real”
resource constraints to innovation or if they represent the opportunities available to each firm. A
potential problem is that firms with more resources, which perform broader technological search,
25
and have generally better opportunities, will likely perceive more innovation constraints compared
to firms with few resources, which engage in few search activities and have few opportunities.
D’Este et al. (2012) refer to the first type of constraints as “deterring barriers” and to the second
as “revealed barriers” (see also, Blanchard, Huiban, Musolesi, & Sevestre, 2013, for a discussion
and an in-depth analysis of this problem). Clearly, the theory we propose in this paper is related to
“real” resource constraints. To deal with this problem, we perform an additional set of regressions,
explaining the three constraint variables related to R&D intensity, firm size and industry affiliation
(38 industries). From these OLS regressions we extract the residuals, thereby removing the
variation in the three constraint variables that is related to differences in resources due to firm size,
R&D-based search activities, and the different opportunities offered by each industry. In the
subsequent regressions, we use the “cleaned” variables (residuals) in lieu of the three original
constraint variables. The results show that although the coefficients of the constraint variables are
marginally smaller, they retain their signs and significance compared to the results obtained using
the original variables.
ConclusionandDiscussion
We set out to study firm-level heterogeneity in the adoption of an open innovation model, and to
add to our understanding of why there is variation in the degree to which firms implement a broad
open innovation model involving both inbound and outbound aspects. An important descriptive
finding is that inbound and outbound aspects can be considered to be part of a single construct. In
other words, although the concept of open innovation has several dimensions (see the
Introduction), the degree of openness in the firm’s innovation process is a latent firm trait, in
relation to both inbound and outbound components. This is in line with Chesbrough’s (2003) idea
that open innovation can be central to the firm’s business model.
26
Our findings demonstrate that implementation of a broad open innovation model is the outcome
of the interplay between resource constraints for innovation and aspirations regarding an
innovation-based growth strategy. The principal finding is that a combination of financial and
knowledge-related resource constraints leads firms to adopt an innovation-based growth strategy,
which in turn induces firms to adopt a broad open innovation model. We also hypothesized about
the direct effects of innovation constraints for the implementation of a broad open innovation
model. We found overall empirical support for our behavioral model of open innovation adoption.
However, the market stagnation variable works opposite to our expectations: Market stagnation
does not induce an innovation-based growth strategy; rather it seems to be associated with lack of
such a strategy. Moreover, market stagnation seems also to be linked to reduced rather than
increased adoption of open innovation.
To account for this result, we speculate that the implications for the firm of market stagnation
are more severe than the other types of resource constraints. Financial and knowledge-related
constraints tend to spark search for innovative solutions, while market stagnation implies that firms
need to diversify their product offerings in order to pursue a growth strategy and to renew their
current product portfolio within the given market. Diversifying into new product markets is more
challenging and may require not only new products but also new complementary assets (Dosi,
Faillo, & Marengo, 2008; Teece, 1986). A central problem in this context is that senior
management attention can be a scarce resource that must be competed for (Ocasio, 1997). In
situations of serious constraint such as a stagnating market, managers may need to focus on solving
the immediate problems related to those constraints, which will reduce their focus on
implementation of a growth strategy or a broad open innovation model. There is an additional
explanation based on the theory of collective invention (Allen, 1983). Allen (1983: 20) suggests
27
that “if the product demand and supply curves are sufficiently elastic … intra-marginal rents will
consequently rise when the industry shifts to a free information regime.” Firms facing the implied
situation of nascent growth, typically characterized by negative responses to the items for our
measure of market stagnation, will benefit more from open innovation. Thus, a positive
relationship between an inverted market stagnation variable and open innovation is consistent with
Allen’s (1983) proposal. Although this explanation of open innovation has not been thoroughly
investigated, research on open source software (e.g., Cowan & Jonard, 2003: 514) implicitly
assumes this condition: “The motivating examples concern groups of firms operating in an era of
rapid technical change and market growth….”
Overall, our results are consistent with the idea that the decision to adopt a broad open
innovation model depends much more on the organization’s growth aspirations and strategy rather
than (directly) on financial shortages, lack of knowledge, or the market situation per se. Open
innovation is stimulated by growth aspirations, but discouraged by stagnating markets. As Rigby
and Zook (2002: 87) note “…the tools of open-market innovation are pointless—indeed,
dangerous—if they’re not used to support a coherent strategic goal.” Thus, our results indicate that
open innovation is foremost a tool for achieving growth goals rather than a defence mechanism
for sustaining general innovation processes, which is in line with the findings in Chesbrough and
Crowther (2006) for example. In other words, open innovation should be regarded predominantly
as an offensive weapon that managers choose not because they perceive it as a better alternative
to innovating alone but because they see open innovation as a means of ensuring growth and
competitive advantage.
Our work provides three main contributions. First, we have proposed a theory-based
explanation for the variations in the degree to which firms implement the broad open innovation
28
model. We have suggested some theoretical mechanisms underlying the idea that managers’
perceived financial, knowledge-related and market-centered resource constraints regarding
innovation, contain information about the aspiration level of the focal firm: A high level of
perceived constraints demonstrates high organizational aspirations in the context of innovation.
We have also proposed some theoretical arguments for the idea that variations in these aspirations
give rise to variations in the application of an innovation-based growth strategy, and that, in turn,
a focus on this strategy should increase the degree to which the focal firm will adopt an open
innovation model. Second, we have extended the behavioral theory of the firm to beyond corporate
performance and growth to account for the adoption of business models such as the broad open
innovation model. Third, we use innovation survey data (CIS data) to model complex constructs
(however, see Foss, Laursen, & Pedersen, 2011). Indeed, this study represents one of a very few
attempts to analyze complex constructs and relationships emerging through the process of open
innovation by applying SEM and the robust WLS method to categorical indicators. We suggest
that the use of SEM allows investigation of a deeper causal structure than is afforded by more
traditional regression analysis—in our case, the structure runs from resource constraints, through
an innovation-based growth strategy to implementation of an open innovation business model
involving both inbound and outbound elements.
Our study has some limitations. First, it draws on a rich and detailed cross-section of Norwegian
firms’ innovative activities; however, these data make it difficult to totally exclude endogeneity
concerns due to unobserved heterogeneity in establishing the relationship between resource
constraints, growth strategy, and open innovation implementation. We do not think that this is a
major concern given that we experimented with several control variables in supplementary
regressions analyses, and given our group analyses. However, we cannot totally rule out
29
unobserved heterogeneity concerns, and future analysis, based on longitudinal data using several
waves of CIS data, would be valuable. In this study, the empirical focus was the Norwegian
context. Although we believe that our theory would hold in other empirical contexts, future
research could examine the generalizability of this work by using data from other geographical
contexts.
Our analysis focused on a range of variables linked to behavioral theory. Future research could
examine these variables as well as a wider range of reasons for adopting open innovation as a
business model. Specifically, we emphasized how the interaction between resource constraints and
the strategic responses to those constraints could lead firms to adopt an open innovation model
that involves both outbound and inbound elements. However, preexisting firm-internal routines
might constrain or enhance near-term possibilities (see, Laursen & Salter, 2006) for successfully
adopting open innovation, and these routines might interact with resource constraints and
innovation-based growth strategies in explaining open innovation implementation. Future research
to account in more detail for these factors and interactions would provide valuable insights into
the antecedents to open innovation and their relation to corporate strategy. We hope that this paper
provides a first step towards exploring this exciting research agenda.
References
Allen, R. C. (1983). Collective invention. Journal of Economic Behavior & Organization, 4(1), 1-
24.
Antonelli, C. (1989). A Failure-Inducement Model of Research and Development Expenditure -
Italian Evidence from the Early 1980s. Journal of Economic Behavior & Organization, 12(2), 159-
180.
Argote, L., & Greve, H. R. (2007). A Behavioral Theory of the Firm—40 Years and Counting:
Introduction and Impact. Organization Science, 18(3), 337–349.
Armstrong, J. S., & Overton, T. S. (1977). Estimating non-response bias in mail surveys. Journal
of Marketing Research, 14(3), 396–402.
30
Arora, A., & Ceccagnoli, M. (2006). Profiting from licensing: The role of patent protection and
commercialization capabilities. Management Science, 52(2), 293-308.
Audia, P. G., & Greve, H. R. (2006). Less Likely to Fail: Low Performance, Firm Size, and Factory
Expansion in the Shipbuilding Industry. Management Science, 52(1), 83-94.
Audia, P. G., Locke, E. A., & Smith, K. G. (2000). The Paradox of Success: An Archival and a
Laboratory Study of Strategic Persistence Following Radical Environmental Change. Academy of
Management Journal, 43(5), 837-853.
Banbury, C. M., & Mitchell, W. (1995). The effects of introducing important incremental
innovation on market share and business survival. Strategic Management Journal, 16(Special
Issue), 161–182.
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social
psychological research: Conceptual, strategic and statistical considerations. Journal of Personality
and Social Psychology(51), 1173-1182.
Bartholomew, D. J., Steele, F., Moustaki , I., & Galbraith, J. (2008). Analysis of Multivariate
Social Science Data (2nd ed.). Boca Raton: CRC Press.
Bianchi, M., Cavaliere, A., Chiaroni, D., Frattini, F., & Chiesa, V. (2011). Organisational modes
for open innovation in the bio-pharmaceutical industry: an exploratory analysis. Technovation,
31(1), 22-33.
Bierly, P., & Chakrabarti, A. (1996). Generic Knowledge Strategies in the US Pharmaceutical
Industry. Strategic Management Journal, 17(Winter Special Issue), 123-135.
Blanchard, P., Huiban, J.-P., Musolesi, A., & Sevestre, P. (2013). Where there is a will, there is a
way? Assessing the impact of obstacles to innovation. Industrial and Corporate Change, 22(3),
679-710.
Blundell, R., Griffiths, R., & Reenen, J. V. (1999). Market Share, Market Value and Innovation in
a Panel of British Manufacturing Firms. The Review of Economic Studies, 66(3), 529-554.
Bolton, M. K. (1993). Organizational Innovation and Substandard Performance - When Is
Necessity the Mother of Innovation. Organization Science, 4(1), 57-75.
Bridge, J., & Dodds, C. (1978). Planning and the growth of the firm. London: Croom Helm.
Brusoni, S., Prencipe, A., & Pavitt, K. L. R. (2001). Knowledge Specialization and the Boundaries
of the Firm: Why Do Firms Know More Than They Make? Administrative Science Quarterly,
46(4), 597-621.
Byrne, B. (1998). Structural Equation Modeling with LISREL, PRELIS and SIMPLIS: Basic
Concepts, Applications, and Programming. Mahwah: Lawrence Erlbaum Associates.
31
Cassiman, B., & Veugelers, R. (2002). Spillovers and R&D Cooperation: some Empirical
Evidence. American Economic Review, 92(4), 1169-1184.
Chen, W. R., & Miller, K. D. (2007). Situational and institutional determinants of firms' R&D
search intensity. Strategic Management Journal, 28(4), 369-381.
Chesbrough, H. (2003). Open Innovation: the New Imperative for Creating and Profiting from
Technology. Boston: Harvard Business School.
Chesbrough, H. (2006). Open Business Models: How to Thrive in the New Innovation Landscape.
Boston: Harvard Business School Press.
Chesbrough, H. (2007). Why Companies Should Have Open Business Models. MIT Sloan
Management Review, 48 (2), 22–28.
Chesbrough, H., & Appleyard, M. (2007). Open Innovation and Strategy. California Management
Review, 50(1), 57–74.
Chesbrough, H., & Crowther, A. K. (2006). Beyond High-Tech: Early Adopters of Open
Innovation in Other Industries. R&D Management, 36(3), 229–236.
Cowan, R., & Jonard, N. (2003). The Dynamics of Collective Invention. Journal of Economic
Behavior & Organization, 52, 513-532.
Cyert, R. M., & March, J. G. (1963). A Behavioral Theory of The Firm. Englewood Cliffs, N.J.:
Prentice-Hall.
D’Este, P., Iammarino, S., Savona, M., & von Tunzelmann, N. (2012). What hampers innovation?
Revealed barriers versus deterring barriers. Research Policy, 41(2), 482–488.
Dahlander, L., & Gann, D. (2010). How open is innovation? Research Policy, 39(6), 699-709.
Desai, V. M. (2008). Constrained Growth: How Experience, Legitimacy, and Age Influence Risk
Taking in Organizations. Organization Science, 19(4), 594-608.
Dodgson, M., Gann, D., & Salter, A. (2006). The role of technology in the shift towards open
innovation: the case of Procter & Gamble. R&D Management, 36(3), 333-346.
Dosi, G., Faillo, M., & Marengo, L. (2008). Organizational Capabilities, Patterns of Knowledge
Accumulation and Governance Structures in Business Firms: An Introduction. Organization
Studies, 29(8-9), 1165-1185.
Dougherty, D. (1992). A practice-centered model of organizational renewal through product
innovation. Strategic Management Journal, 13(S1), 77-92.
Drechsler, W., & Natter, M. (2012). Understanding a firm's openness decisions in innovation.
Journal of Business Research, 65(3), 438-445.
32
Farrell, A. M. (2010). Insufficient discriminant validity: A comment on Bove, Pervan, Beatty, and
Shiu (2009). Journal of Business Research, 63(3), 324-327.
Flora, D. B., & Curran, P. J. (2004). An Empirical Evaluation of Alternative Methods of Estimation
for Confirmatory Factor Analysis with Ordinal Data. Psychological Methods(9), 466-491.
Fosfuri, A. (2006). The Licensing Dilemma: Understanding the Determinants of the Rate of
Technology Licensing. Strategic Management Journal 27, 27(12), 1141-1158.
Foss, N. J., Laursen, K., & Pedersen, T. (2011). Linking Customer Interaction and Innovation: The
Mediating Role of New Organizational Practices. Organization Science, 22(4), 980–999.
Gambardella, A., & Giarratana, M. S. (2013). General technological capabilities, product market
fragmentation, and markets for technology. Research Policy, 42(2), 315-325.
Gambardella, A., & Hall, B. H. (2006). Proprietary versus public domain licensing of software and
research products. Research Policy, 35(6), 875-892.
Garriga, H., von Krogh, G., & Spaeth, S. (2013). How Constraints and Knowledge Impact Open
Innovation. Strategic Management Journal, 34(9), 1134–1144.
Garud, R., Gehman, J., & Kumaraswamy, A. (2011). Complexity Arrangements for Sustained
Innovation: Lessons from 3M Corporation. Organization Studies, 32(6), 737-767.
Gavetti, G., Greve, H. R., Levinthal, D. A., & Ocasio, W. (2012). The Behavioral Theory of the
Firm: Assessment and Prospects. The Academy of Management Annals, 6(1), 1-40.
Greve, H. R. (1998). Performance, Aspirations, and Risky Organizational Change. Administrative
Science Quarterly, 43(1), 58-86.
Greve, H. R. (2003). A behavioral theory of R&D expenditures and innovations: Evidence from
shipbuilding. Academy of Management Journal, 46(6), 685-702.
Greve, H. R. (2008). A Behavioral Theory of Firm Growth: Sequential Attention to Size and
Performance Goals. Academy of Management Journal, 51(3), 476-494.
Greve, H. R., & Taylor, A. (2000). Innovations as catalysts for organizational change: shifts in
organizational cognition and search. Administrative Science Quarterly, 45(1), 54-80.
Grewal, R., Cote, J. A., & Baumgartner, H. (2004). Multicollinearity and Measurement Error in
Structural Equation Models: Implications for Theory Testing. [Article]. Marketing Science, 23(4),
519-529.
Henkel, J. (2006). Selective Revealing in Open Innovation Processes: the Case of Embedded
Linux. Research Policy, 35(7), 953–969.
33
Holmbeck, G. (1997). Toward terminological, conceptual, and statistical clarity in the study of
mediators and moderators: Examples from the child-clinical and pediatric psychology literatures.
Journal of Consulting and Clinical Psychology, 65(4), 599-610.
Hu, L., Bentler, P. M., & Kano, Y. (1992). Can Test Statistics in Covariance Structure Analysis
be Trusted? Psychological Bulletin, 112, 351–362.
Huizingh, E. (2011). Open Innovation: State of the Art and Future Perspectives. Technovation,
31(1), 2-9.
Iacobucci, D., Neela S., & D., X. (2007). A Meditation on Mediation: Evidence that Structural
Equations Models Perform Better than Regressions. Journal of Consumer Psychology, 17(2), 139-
153.
Jeppesen, L. B., & Lakhani, K. R. (2010). Marginality and Problem-Solving Effectiveness in
Broadcast Search. Organization Science, 21(5), 1016-1033.
Jeppesen, L. B., & Laursen, K. (2009). The role of lead users in knowledge sharing. Research
Policy, 38(10), 1582-1589.
Joreskog, K. G. (2002). Structural Equation Modeling with Ordinal Variables using LISREL.
Retrieved from http://www.ssicentral.com/lisrel/techdocs/ordinal.pdf
Joreskog, K. G. (2004). On chi-squares for the independence model and fit measures in LISREL.
Retrieved from http://www.ssicentral.com/lisrel/techdocs/ftb.pdf
Kamien, M. I., & Schwarz, N. L. (1982). Market Structure and Innovation. Cambridge: Cambridge
University Press.
Katila, R., & Ahuja, G. (2002). Something Old, Something New: A Longitudinal Study of Search
Behaviour and New Product Introduction. Academy of Management Journal, 45(8), 1183-1194.
Kline, R. (2011). Principles and Practice of Structural Equation Modeling (3rd ed.). New York:
The Guilford Press.
Kumar, D. (2010). Enterprise Growth Strategy: Vision, Planning and Execution. Farnham: Gower
Applied Business Research.
Kyriakopoulos, K. (2011). Improvisation in Product Innovation: The Contingent Role of Market
Information Sources and Memory Types. Organization Studies, 32(8), 1051-1078.
Laursen, K., & Salter, A. J. (2006). Open for Innovation: The role of openness in explaining
innovative performance among UK manufacturing firms. Strategic Management Journal, 27(2),
131-150.
Laursen, K., & Salter, A. J. (2014). The Paradox of Openness: Appropriability, External Search
and Collaboration. Research Policy, forthcoming.
34
Lavie, D. (2006). The Competitive Advantage of Interconnected Firms: An Extension of the
Resource-Based View. The Academy of Management Review, 31(3), 638-658.
Leiponen, A., & Helfat, C. E. (2010). Innovation Opportunities, Knowledge Sources, and the
Benefits of Breadth. Strategic Management Journal, 31(2), 224-236.
Levin, R., Klevorick, A., Nelson, R. R., & Winter, S. (1987). Appropriating the Returns from
Industrial Research and Development. Brookings Papers on Economic Activity(3), 783-820.
Love, J. H., Roper, S., & Vahter, P. (2013). Learning from openness: The dynamics of breadth in
external innovation linkages. Strategic Management Journal, forthcoming.
Mansfield, E., Schwartz, M., & Wagner, S. (1981). Imitation Costs and Patents: an Empirical
Study. Economic Journal, 91(December), 907-918.
Nelson, R. R., & Winter, S. (1982). An Evolutionary Theory of Economic Change. Cambridge,
Massachusetts: Harvard University Press.
Nickerson, J., & Zenger, T. (2004). A Knowledge-based Theory of the Firm: The Problem-Solving
Perspective. Organization Science, 15(6), 617–632.
Ocasio, W. (1997). Towards an attention-based view of the firm. Strategic Management Journal,
18(Summer Special Issue), 187-206.
Oliver, A. L. (2001). Strategic Alliances and the Learning Life-Cycle of Biotechnology Firms.
Organization Studies, 22(3), 467-489.
Podsakoff, P. M., MacKenzie, S. B., Lee, J.-Y., & Podsakoff, N. P. (2003). Common Method
Biases in Behavioral Research: A Critical Review of the Literature and Recommended Remedies.
Journal of Applied Psychology, 88(5), 879–903.
Podsakoff, P. M., & Organ, D. W. (1986). Self-reports in organizational research: Problems and
prospects. Journal of Management, 12(4), 531-544.
Postrel, S. (2002). Islands of Shared Knowledge Specialization and Mutual Understanding in
Problem-Solving Teams. Organization Science 13(3), 303-320.
Powell, W. W., Koput, K. W., & Smith-Doerr, L. (1996). Interorganizational collaboration and the
local of innovation: Networks of learning in biotechnology. Administrative Science Quarterly, 41,
116-145.
Rigby, D., & Zook, C. (2002). Open-Market Innovation. Harvard Business Review, 80(10), 80–
89.
Roper, S., Vahter, P., & Love, J. H. (2013). Externalities of openness in innovation. Research
Policy, 42(9), 1544-1554.
35
Rosenberg. (1969). The Direction of Technological Change: Inducement Mechanims and
Focusing Devises. Economic Development and Cultural Change, 18, 1-24.
Rosenberg, N. (1963). Technological Change in the Machine Tool Industry, 1840-1910. Journal
of Economic History, 23(4), 414-443.
Rothaermel, F., & Alexandre, M. (2009). Ambidexterity in Technology Sourcing: The Moderating
Role of Absorptive Capacity. Organization Science, 20(4), 759-780.
Rothwell, R., Freeman, C., Jervis, P., Robertson, A., & Townsend, J. (1974). SAPPHO Updated -
Project SAPPHO Phase 2. Research Policy, 3(3), 258-291.
Salancik, G. R., & Pfeffer, J. (1977). An examination of need-satisfaction models of job attitudes.
Administrative Science Quarterly, 22(3), 427-456.
Segarra-Blasco, A., Garcia-Quevedo, J., & Teruel-Carrizosa, M. (2008). Barriers to innovation
and public policy in Catalonia. International Entrepreneur Management Journal(4), 431–451.
Shan, W., Walker, G., & Kogut, B. (1994). Interfirm cooperation and startup innovation in the
biotechnology industry. Strategic Management Journal, 15(5), 387-394.
Stolzenberg, R. (Ed.). (2003). Sociological Methodology. Oxford: Basil Blackwell for the
American Sociological Association.
Teece, D. (1986). Profitting from technological innovation: Implications for integration
collaboration, licencing and public policy. Research Policy, 15(6), 285-305.
Tether, B. S. (2002). Who co-operates for innovation, and why: An empirical analysis. Research
Policy, 31, 947-967.
Tippmann, E., Mangematin, V., & Scott, P. S. (2013). The Two Faces of Knowledge Search: New
Solutions and Capability Development. Organization Studies, 34(12), 1869-1901.
van de Vrande, V., de Jong, J. P. J., Vanhaverbeke, W., & de Rochemont, M. (2009). Open
innovation in SMEs: Trends, motives and management challenges. Technovation, 29(6–7), 423-
437.
Vissa, B., Greve, H. R., & Chen, W.-R. (2010). Business Group Affiliation and Firm Search
Behavior in India: Responsiveness and Focus of Attention. Organization Science, 21(3), 696-712.
von Hippel, E. (1987). Cooperation Between Rivals: Informal Know-How Trading. Research
Policy, 16(6), 291-302.
von Hippel, E. (1988). The Sources of Innovation. New York and Oxford: Oxford University Press.
von Hippel, E. (1994). "Sticky Information" and the Locus of Problem Solving: Implications for
Innovation. Management Science, 40(4), 429-439.
36
West, J., & Bogers, M. (2014). Leveraging External Sources of Innovation: A Review of Research
on Open Innovation. Journal of Product Innovation Management, 31(4), forthcoming.
Wezel, F. C., & Saka-Helmhout, A. (2006). Antecedents and Consequences of Organizational
Change: ‘Institutionalizing’ the Behavioral Theory of the Firm. Organization Studies, 27(2), 265-
286.
Wilhelmsen, L. (2011). Innovasjon i Norsk Næringsliv 2006-2008 Rapporter 32/2011. Oslo:
Statistisk sentralbyrå.
Wooldridge, J. M. (2002). Econometric Analysis of Cross Section and Panel Data. Cambridge
Massachusetts: The MIT Press.
Yayavaram, S., & Ahuja, G. (2008). Decomposability in Knowledge Structures and Its Impact on
the Usefulness of Inventions and Knowledge-base Malleability. Administrative Science Quarterly,
53(2), 333-362.
37
Appendix 1. NACE codes of the sample firms
Code Economic activity
C10-33 Manufacturing
G46 Wholesale and retail trade, except of motor vehicles and motorcycles
H49-53 Trans
p
ortation and stora
g
e
J58 Publishing activities
J59 Motion picture, video and television program production, sound recording and
music
p
ublishin
g
activities
J60 Programming and broadcasting activities
J61 Telecommunications
J62 Services affiliated to information technolo
gy
J63 Information services
K64-66 Financial services and insurance activities
M70 Activities of head offices; mana
g
ement consultanc
y
activities
M71 Architectural and engineering activities
M72 Scientific research and developmen
t
M74.9 Other professional, scientific and technical activities
N82.9 Business support service activities
38
Appendix 2. Constructs and their measures
Construct Indicators Description
Open Innovation 1.SALE Question 9. Has the enterprise sold/delivered R&D services to others during 2008? (recoded
as 1 if
y
es and 0 if not
)
2.BUY Question 11. Provide costs to R&D services bought from others during 2008 (sum of four
variables: other Norwegian enterprises; other foreign enterprises; research institutes,
universities and high schools in Norway; foreign research institutes, universities and high
schools, recoded as 1 if more then 0, and 0 if not)
3.BREADTH Question 24. Tick the type of co-operation partner and where it is geographically situated,
several alternatives are possible (a sum of dichotomous variables reflecting cooperation with
suppliers, clients/customers, competitors, consultants, commercial laboratories and R&D
institutions, universities or university colleges, public/private research institutions, takes
value from 0 to 7)
Innovation-
Based Growth
Strategy
4.ERANG Question 22. Innovation goals. How important were the following objectives for enterprise
for development of new products (goods or services) or processes in 2006-2008: Increase
range of goods and services (Very important/ Quite important/Slightly important/Not
relevant = 3/2/1/0)
5.EMARC Question 22. Innovation goals. How important were the following objectives for enterprise
for development of new products (goods or services) or processes in 2006-2008: Enter new
markets (Very important/ Quite important/Slightly important/Not relevant = 3/2/1/0)
6.EMSHA Question 22. Innovation goals. How important were the following objectives for enterprise
for development of new products (goods or services) or processes in 2006-2008: Increase
market share
(
Ver
y
im
p
ortant/ Quite im
p
ortant/Sli
g
htl
y
im
p
ortant/Not relevant = 3/2/1/0
)
7.ECAP Question 22. Innovation goals. How important were the following objectives for enterprise
for development of new products (goods or services) or processes in 2006-2008: Increased
capacity of production or service provision (Very important/ Quite important/Slightly
important/Not relevant = 3/2/1/0)
Financial
Shortage
8.HCOS Question 25. How hampering were the following factors for innovation activities: Too high
innovation costs (Very important/ Quite important/Slightly important/Not relevant =
3/2/1/0
)
9.HFENT Question 25. How hampering were the following factors for innovation activities: Lack of
financing inside enterprise or corporation (Very important/ Quite important/Slightly
important/Not relevant = 3/2/1/0)
10.HFOUT Question 25. How hampering were the following factors for innovation activities: Lack of
finance from source outside enterprise (Very important/ Quite important/Slightly
im
p
ortant/Not relevant = 3/2/1/0
)
Market
Stagnation
11.HDOM Question 25. How hampering were the following factors for innovation activities: Market
dominated by established enterprises (Very important/ Quite important/Slightly
important/Not relevant = 3/2/1/0)
12.HDAM Question 25. How hampering were the following factors for innovation activities: Uncertain
demand for innovative goods or services (Very important/ Quite important/Slightly
im
p
ortant/Not relevant = 3/2/1/0
)
13.HMAR Question 25. How hampering were the following factors for innovation activities: No need
because of no market demand for innovations (Very important/ Quite important/Slightly
important/Not relevant = 3/2/1/0)
Lack of
Knowledge
14.HPER Question 25. How hampering were the following factors for innovation activities: Problems
with holding or recruiting qualified personnel (Very important/ Quite important/Slightly
im
p
ortant/Not relevant = 3/2/1/0
)
15.HTEC Question 25. How hampering were the following factors for innovation activities: Lack of
technological information (Very important/ Quite important/Slightly important/Not relevant
= 3/2/1/0)
16.HINF Question 25. How hampering were the following factors for innovation activities: Lack of
market information (Very important/ Quite important/Slightly important/Not relevant =
3/2/1/0)
17.HPAR Question 25. How hampering were the following factors for innovation activities: Difficulty
in finding cooperation partner for innovation (Very important/ Quite important/Slightly
im
p
ortant/Not relevant = 3/2/1/0
)
39
Appendix 3. Polychoric correlations for construct indicators
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
1 1.00
2 0.34 1.00
3 0.49 0.49 1.00
4 0.30 0.20 0.42 1.00
5 0.24 0.12 0.31 0.70 1.00
6 0.15 0.23 0.28 0.68 0.74 1.00
7 -0.01 0.18 0.20 0.44 0.45 0.54 1.00
8 0.12 0.14 0.27 0.41 0.34 0.30 0.27 1.00
9 0.20 0.15 0.29 0.37 0.34 0.27 0.24 0.80 1.00
10 0.23 0.10 0.28 0.36 0.33 0.25 0.24 0.75 0.85 1.00
11 0.08 0.05 0.09 0.27 0.29 0.27 0.27 0.57 0.57 0.57 1.00
12 0.16 0.08 0.17 0.35 0.30 0.29 0.23 0.61 0.60 0.57 0.72 1.00
13 0.05 0.13 0.09 0.24 0.23 0.21 0.23 0.50 0.51 0.49 0.62 0.71 1.00
14 0.19 0.19 0.22 0.37 0.36 0.31 0.31 0.60 0.59 0.61 0.58 0.57 0.51 1.00
15 0.08 0.09 0.15 0.32 0.33 0.26 0.32 0.63 0.62 0.63 0.64 0.62 0.59 0.72 1.00
16 0.12 0.00 0.15 0.36 0.37 0.31 0.28 0.60 0.61 0.62 0.67 0.69 0.61 0.69 0.83 1.00
17 0.17 0.06 0.22 0.31 0.31 0.23 0.26 0.62 0.63 0.67 0.68 0.66 0.59 0.63 0.74 0.74 1.00
40
Appendix 4. Factor loadings for the latent variables
Latent Variable Observed Variable Factor Loadings
Open Innovation 1.SALE 0.59
2.BUY 0.54
3.BREADTH 0.92
Innovation-Based
Growth Strategy
4.ERANG 0.86
5.EMARC 0.88
6.EMSHA 0.88
7.ECAP 0.66
Financial Shortage 8.HCOS 0.90
9.HFENT 0.94
10.HFOUT 0.92
Market Stagnation 11.HDOM 0.87
12.HDAM 0.91
13.HMAR 0.81
Lack of Knowledge 14.HPER 0.84
15.HTEC 0.92
16.HINF 0.93
17.HPAR 0.88
