Networking, Innovation, and Firms' Performance

Networking, Innovation, and Firms’ Performance: Portugal as Illustration

Sérgio Nunes, Raul Lopes & Nerys Fuller-Love

Journal of the Knowledge Economy ISSN 1868-7865 J Knowl Econ DOI 10.1007/s13132-017-0508-7

1 23

Your article is protected by copyright and all rights are held exclusively by Springer Science+Business Media, LLC. This e-offprint is for personal use only and shall not be selfarchived in electronic repositories. If you wish to self-archive your article, please use the accepted manuscript version for posting on your own website. You may further deposit the accepted manuscript version in any repository, provided it is only made publicly available 12 months after official publication or later and provided acknowledgement is given to the original source of publication and a link is inserted to the published article on Springer's website. The link must be accompanied by the following text: "The final publication is available at link.springer.com”.

1 23

Author's personal copy J Knowl Econ https://doi.org/10.1007/s13132-017-0508-7

Networking, Innovation, and Firms’ Performance: Portugal as Illustration Sérgio Nunes 1 & Raul Lopes 2 & Nerys Fuller-Love 3

Received: 23 May 2017 / Accepted: 24 October 2017 # Springer Science+Business Media, LLC 2017

Abstract The innovation process takes place not only as a result of research and development but also because of specific combinations of knowledge that firms obtain by accessing both internal and external environments. In this paper, the relationship between networking intensity and the innovation process is investigated in order to analyze the effect of these networks on firm performance. The results show that firms that are engaged more intensively in knowledge networks increase the likelihood of obtaining higher levels of innovation, which can lead to better economic performance. Furthermore, informal mechanisms of interaction have proved to be a fundamental dimension of the innovation process, especially in conjunction with formal networks. These conclusions have strong implications for government innovation policies designed to improve firms’ performance and that of the local economy. Keywords Innovation . Knowledge networks . Economic performance . Innovation policy . Portugal JEL Classifications L14 . L15 . D83

* Sérgio Nunes [email protected] Raul Lopes [email protected] Nerys Fuller-Love [email protected]

1

CIAEGT-IPT, Polytechnic Institute of Tomar, Tomar, Portugal

2

Instituto Universitário de Lisboa (ISCTE-IUL), DINÂMIA CET-IUL, Lisbon, Portugal

3

School of Management and Business, Aberystwyth University, Aberystwyth, UK

Author's personal copy J Knowl Econ

Introduction Innovation is a fundamental element in the economic performance and competitiveness of firms. This was noted by Porter (1990) and is abundantly documented by the literature on the importance of the innovation process in providing a competitive advantage for firms (Damanpour et al. 2009; Carayannis and Gonzalez 2003; Lopes 2001). There is less agreement, however, in the literature about the explanation for the fact that some firms reveal greater capacity to innovate than others. It is precisely this aspect that contributes the main contribution to knowledge of this paper. The objectives are as follows: first, the relationship between innovation performance and firm economic performance was analyzed. Secondly, the relationship between networking intensity and the innovation performance was studied. Underlying these objectives is a set of theoretical elements that emerge from the literature and is presented in the next section. Our approach is based on the paradigmatic understanding that innovation stems from a cumulative and interactive process of collective learning, a process that provides support to the interaction of the different agents of innovation in the knowledge networks, both formal and informal. It adds two main novelties to the extant literature: first, it provides evidence on the influence of innovation on firm’s economic performance; and second, the importance of knowledge networks in firm’s innovation performance. Moreover, it extends the analysis beyond these aspects and highlights informal interaction mechanisms in the firm’s innovation process. The more general result that follows from our analysis is that firm performance (innovation and economic) and knowledge networks are multidimensional phenomena and the keyword is interaction. Interaction in this context is associated with complementarities, synergies, externalities, and learning conditions. In knowledge networks, we looked at formal and informal interaction mechanisms; in innovation, we looked at product, process, and organizational innovation; and in the economic performance, we looked at sales, orders, and exports. The results suggest that when interactions are considered and firms increase the level of interaction, they obtain better results. Theoretically, this article starts from the assumption that the performance of firms depends on innovation and shows that firms’ innovative capacity is conditioned by their degree of integration in knowledge networks. At the empirical level, we focus on a semi-peripheral economy characterized by a moderate innovative intensity, as is the case of Portugal (EC 2016), to discuss the extent to which these theoretical relations are observed. Therefore, the text is structured as follows: in the BInnovation, Firms’ Performance, and Networking^ section, the theoretical assumptions on which our hypotheses are based are set out. In the BMethodology^ section, we present the sample and data, the variables and measures, and the statistical model of analysis. In the BEmpirical Results^ section, we present our empirical results, namely testing economic and innovation performance and in the second model, the relationship between the innovation performance and knowledge networks. Finally, the BDiscussion and Main Conclusions^ section discusses the results and highlights the implications of the conclusions of this research for the development of regional innovation policy.


Innovation, Firms’ Performance, and Networking Innovation and Firm Competitiveness The innovation process has become a central theme in the economic literature in recent years. Firms innovate to improve their economic performance, thereby improving their competitive advantages. The contribution of innovation to economic growth has been well documented since Schumpeter (1934). Despite the importance of innovation, it is the implementation that often leads to success. Although innovation allows firms to gain a competitive advantage, it is likely that it will be eroded over time; therefore, it needs to be a continuous and cumulative process. These innovations can take the form of new and improved products or services to attract new customers and retain existing ones as well as innovative approaches to the organization’s systems. We also believe on the importance of the combination between the different types of innovation for the performance of firms. Firms that are consistently more innovative are those that do not limit their innovation effort to a one domain only. The most innovative ones are those that, at the same time, can articulate multiple typologies of innovation. If a firm innovates in a specific typology (product, for example), the maintenance and exploitation of this innovation forces the firm to innovate in other fields (processes, marketing, organizational, financial, etc.). The most innovative companies are those that can innovate consistently in their different fields. Damanpour et al. (2009) found that innovation led to higher levels of performance, especially when there were different approaches to innovation and that a diverse approach to networks was more likely to create a competitive advantage. Innovation can enhance the dynamic capability of an organization and enable it to adapt to changes in the competitive environment. There are several previous studies on the establishment and the rationale for the relationship between innovation and business performance (see, for example, Kleinknecht and Mohnen 2002; Kemp et al. 2003; Cefis and Ciccarelli 2005; Cainelli et al. 2006; Koellinger 2008; Morone and Testa 2008; Fagerberg et al. 2009; Cappellin and Wink 2009; Hall 2011). In general, these studies have found a positive effect of innovation on a firm’s performance, expressed in both productivity and growth in sales. The first hypothesis is based precisely on this assumption and will be tested in the BEmpirical Results^ section. In this study the relationship between the increase sales, orders and exports of firms and their innovation performance was tested. Hypothesis 1—innovation performance has a positive impact on firm’s economic performance This hypothesis seeks to test the relationship by integrating variables associated with the economic and innovative performance of firms. Both innovation and economic performance are multidimensional phenomena and, as such, this hypothesis will be tested using variables that were designed to incorporate this conceptual and analytical diversity. Additionally, this hypothesis is also supported by observations in the literature that innovations should have a direct benefit on sales and profitability.


Innovation as a Collective and Interactive Learning Process Although the dominant orientation of innovation policy will continue to favor the incentive to R&D, this is not the only mechanism that firms use in the production and transformation of knowledge necessary for their innovation process. Multiple studies have shown that firms do not innovate in the same way (see, for example, Holtskog 2017; Apanasovich et al. 2017; Apanasovich 2016; Parrilli and Alcalde Heras 2016; Parrilli et al. 2016; Nunes and Lopes 2015; Fitjar and Rodriguez-Pose 2013; Gokhberg et al. 2012; Parrilli et al. 2012; Marlon and Lambert 2009; Jensen et al. 2007; Lundvall 2007; Tödtling et al. 2006; Lorenz and Lundvall 2006). Particularly, Lundvall (1985) drew attention to the fact that innovation results from an interactive and collective learning process. In turn, Antonelli and Ferrão (2001) show that the scope of this learning process is not confined to the internal environment of the firm, before it develops into a complex variety of relationships between the firm and its external environment. Through these dynamic relationships, firms can access both tacit and codified knowledge that, together, combine to develop innovation. In other words, the innovation process is complex and relies on a combination of factors including specific combinations of knowledge that firms obtain by accessing both the internal and the external environments (Karlsson and Olsson 1998; Antonelli and Ferrão 2001; Antonelli 2005; Rutten and Boekema 2004). It follows that one way for businesses to develop innovation is through knowledge networks (Powell and Grodal 2005; Fischer 2006). Networks are the organizational support that enables economic actors to pursue innovation and explore new business opportunities through joint efforts, resources, and competences. By doing so, they not only manage the scarcity of resources for the innovation process but also reduce the risks associated with uncertainty. The dynamics of interaction between actors in the networks also contribute to the strengthening of trust between partners, a critical dimension of the innovation process. These interactions occur through mechanisms not only of formal but also of an informal nature (Birley 1986; Fuller-Love 2009). Formal networks are structured and governed by legal-formal mechanisms while informal networks result from interaction processes without legal support, such as the interactions associated with the noncontracted labor market, social and professional relationships, personal relationships, or social networks. Both formal and informal networks are key factors in the innovation process, complementing and mutually reinforcing each other. The formal mechanisms promote stability within a relationship, promoting cognitive proximity and the efficiency to transfer specialized codified knowledge. In turn, informal mechanisms provide greater network flexibility in the transfer of tacit knowledge. The Role of Networks in the Innovation Process Acquiring external knowledge is recognized as essential in the innovation process as it has been identified as the additional factor which makes the difference to progress (Lopes 2001; Acs et al. 2012; Nunes and Lopes 2012; Farinha and Ferreira 2016). According to Antonelli (2005:10), tacit knowledge acquired through the learning process is, Barticulated both internally and externally by means of network relations.^ In this way, networks allow to firms to innovate more quickly and develop new useful


knowledge that’s beyond their individual competences. They need external knowledge, i.e., knowledge that is beyond their hierarchical control, in the innovation process (Rutten and Boekema 2004; Ferreira et al. 2014a, b). The different relevant sources of knowledge required for the innovation process are increasingly specialized and fragmented, and there is a tendency for this knowledge to be increasingly dispersed, either spatially or in respect to the actors possessing it. In order to overcome this fragmentation of knowledge, universities, firms, individuals, research laboratories, and institutional actors are increasingly engaging in cooperative activities and networking. Because learning processes develop both internally and externally in organizations, the dynamics of interaction are inevitable and networks are formed to support and coordinate the whole process of innovation (Oerlemans et al. 1998, 2001; Karlsson and Olsson 1998; Powell and Grodal 2005; Fischer 2006). In general, therefore, a network is formed from the interaction between different actors with a view to obtaining further knowledge for the innovation process. Participation in networks can have various complementary benefits for the innovation process. Five dimensions can be identified which are associated with these benefits: (i) the creation of networking skills necessary for the acquisition of innovative knowledge, (ii) access to strategic information, (iii) coordination between different types of knowledge and different players, (iv) reduction of transaction costs, and (v) a reduction in the uncertainty and the risk of the innovation process. i. Networks can contribute to the development of the relationship skills necessary for the long-term sustainability of the dynamics of innovation. For example, Karlsson et al. (2005) found that members of a network may develop joint knowledge and product specific language over time which helps to reinforce these relationships. These linkages therefore provide a structure to the organization of these firms. The dynamics of these network links therefore have an important impact on the innovation processes. Several studies point to the positive correlation between research efforts and technological sophistication and the number and intensity of strategic alliances (Freeman 1991; Hagedoorn 1995; Powell and Grodal 2005). Faems et al. (2005) also found that the more firms engage in networks, the more likely they are to create new or improved products that are commercially successful. ii. Networks can act as a vehicle for access to information and innovative ideas. Aalbers et al. (2006) and Parker et al. (2002) found that networks played an important role in the transfer of knowledge within a firm. Information may also be transferred more quickly in an informal network, and they may be more flexible. Johannisson et al. (2002) underline that external linkages provide opportunities to access new ideas. Therefore, the innovation becomes more likely with greater number of linkages. In the same way, Hervas-Oliver and Albors-Garrigos (2009) conclude that firms operating in clusters often perform better because they can develop and maintain relationships. One of the key issues is that firms exploit relationships in different ways, enabling some of them to access better information and become more innovative. Fuller-Love and Thomas (2004) found that formal business networks provided economies of scale and a cost-effective way to improve performance, sharing information and resources, and undertaking joint projects. Sammarra and Biggiero (2008) looked at the exchange of technical, market, and managerial knowledge in the aerospace industrial cluster in Rome.


They found that network participants exchange technological, marketing, and managerial information because of the complex nature of the innovation process which requires diverse information. Fritsch and Kauffeld-Monz (2010) looked at the transfer of knowledge and information in regional innovation networks in Germany. They found that embeddedness within an innovation network was positively correlated with the inter-organizational exchange of information and knowledge. iii. Networks create conditions that facilitate the connection and coordination between the different sources of knowledge, enabling the coordination and participation of different actors (see, e.g., Oerlemans et al. 1998, 2000, 2001; Tödtling et al. 2004; Rutten and Boekema 2004; Lambooy 2005; Caravaca et al. 2005; Fischer 2006). They facilitate exchange and sharing of information and specialized resources, collective and inter-organizational learning, joint development of skills and knowledge and make it possible to develop new opportunities and experiences (Powell and Grodal 2005; Caravaca et al. 2005). According to Witt (2004), entrepreneurs with larger and more diverse networks are likely to be more successful than those with smaller networks. Witt (2004) found that networking in some industries, e.g., biotechnology, was more important than in others because of the levels of implicit and tacit knowledge required to maintain competitive advantage. Firms that pursue innovative strategies may need more networking as they rely more on cooperative strategies with other firms. Research into the territorial approach also points out the crucial importance of networks (both formal and informal) to the innovation process (Antonelli and Ferrão 2001; Storper and Venables 2004; Gellynck and Vermeire 2009; Huggins et al. 2012; Nunes 2012; Martin 2013). In general, as Martin (2013: 1431) said, these studies conclude that BEmbeddedness into networks can have positive effects on innovation outcomes as they facilitate the flow of information and knowledge and provide access to tacit forms of knowledge which are not available elsewhere.^ iv. Another recognized benefit of networking is in terms of lower transaction costs (Williamson 1975). By developing relationships with network members, firms can reduce transaction costs and this will, ultimately, have an impact on the economic performance of the firm. According to Coase (1937, 1992), if a contract with a firm is made for a longer period, the costs of making new contracts will be avoided. The game theory approach has also been used to explain the benefit to participants (Cowan et al. 2007). In the Nash equilibrium, there is a payoff for cooperation, i.e., each participant benefits. v. Networks also enable small and large firms to overcome many problems arising from uncertainty and limited resources connected to the innovation process, and the risks associated with the complexity of innovation activities (Tödtling and Kauffman 2001). In summary, irrespective of the innovation in research and development laboratories, networks, in their different forms, have a fundamental importance in the innovation process. Involvement in knowledge networks is relevant not only to enhance the innovative performance of the firm but also to improve its economic performance.


The positive relationship between embeddedness in networks and firms’ innovative performance is tested empirically in the next section, according to our second hypothesis. The analysis, both conceptual and analytical, of the formal and informal nature of knowledge networks is usually performed by separating these two components. Our perspective is that this separation is artificial and, if it contributes to conceptual and analytical clarity, its empirical analysis should be performed in an integrated way. Consequently, the second hypothesis also tests the interaction between the formal and informal nature of knowledge networks and their relationship to a firm’s innovative performance. Hypothesis 2—network intensity has a positive effect on innovative performance The next section looks at the methodology used to test this hypothesis.

Methodology Sample and Data The database used for our empirical analysis is made up of a representative sample drawn from a database of 981 Portuguese firms that simultaneously satisfied the following criteria: had a turnover of over €1 million in 2008 and an increase in turnover of at least 5% between 2007 and 2008.1 The intention was to identify a group of more dynamic firms, from the point of view of their economic performance. The data was stratified according to the following variables: levels of technological intensity and knowledge services—high technology (HT), medium-high technology (MHT), medium-low technology (MLT), and low technology (LT). Knowledge services (KS) firms were also included. This categorization was chosen because it is the most commonly used in the international literature, mainly by reference to entities such as the OECD (2005); firms’ size—classified into Micro (0–9), SMEs (10–250), and large firms (> 250) by number of employees (2008); and NUTSIII (Greater Lisbon and Setubal Peninsula, Pinhal Litoral, and Greater Porto). These NUTS correspond to the Portuguese metropolitan areas and to an intermediate urban region (Pinhal Litoral). As it was not financially possible to carry out an investigation of the entire population, a representative sample was subsequently chosen. This was obtained by a random stratified sample, from telephone interviews conducted by an independent specialized company in late 2010 and early 2011. This produced a database containing 397 observations, representative of the population on which the statistical and econometric work of this paper is based. The sample used for the empirical analysis is reported in Table 1. Variables and Measures Tables 2 and 3 show the key survey questions for the variables used in this paper and the descriptive statistics of the variables. 1

The database was obtained from COFACE SERVICES PORTUGAL, SA. See, please, www.coface.pt.

Author's personal copy J Knowl Econ Table 1 Sample used on empirical analysis

Greater Lisbon and the Setúbal Península

Pinhal Litoral

Total

SME

N

N

%

%

Large

Total

N

N

%

%

LT

1

5.6

34

19.8

11

28.9

46

20.2

MLT

5

27.8

37

21.5

6

15.8

48

21.1

MT

2

11.1

33

19.2

5

13.2

40

17.5

HT

2

11.1

19

11.0

7

18.4

28

12.3

KS

8

44.4

49

28.5

9

23.7

66

28.9

18

100.0

172

100.0

38

100.0

228

100.0

2

40.0

33

39.8

3

23.1

38

37.6

Total Greater Porto

Micro

LT MLT

1

20.0

22

26.5

2

15.4

25

24.8

MT

1

20.0

17

20.5

4

30.8

22

21.8

HT

0

0.0

2

2.4

1

7.7

3

3.0

KS

1

20.0

9

10.8

3

23.1

13

12.9

Total

5

100.0

83

100.0

13

100.0

101

100.0

LT

3

60.0

15

24.2

1

100.0

19

27.9

MLT

2

40.0

38

61.3

0

0.0

40

58.8

MT

0

0.0

8

12.9

0

0.0

8

11.8

HT

0

0.0

0

0.0

0

0.0

0

0.0

KS

0

0.0

1

1.6

0

0.0

1

1.5

Total

5

100.0

62

100.0

1

100.0

68

100.0

LT

6

21.4

82

25.9

15

28.8

103

25.9

MLT

8

28.6

97

30.6

8

15.4

113

28.5

MT

3

10.7

58

18.3

9

17.3

70

17.6

HT

2

7.1

21

6.6

8

15.4

31

7.8

KS

9

32.1

59

18.6

12

23.1

80

20.2

28

100.0

317

100.0

52

100.0

397

100.0

Total

Dependent Variables In model A (hypothesis 1), we used Economic Performance as our dependent variable. Firms’ economic performance is a multidimensional phenomenon and, based on the business survey data, an aggregate measure of economic performance was used to incorporate some of that diversity. We asked firms if, in the last 5 years, they had increased sales, orders, or exports. This variable ranges from zero type (not increasing any of them) to three types (if the firms increased all three). In model B (hypothesis 2), we used Innovation Performance as our dependent variable. As in the previous explanation, innovative performance is a multidimensional phenomenon. Accordingly, three types of innovation output were used: product, process, and organizational innovation. Firms were asked what type of innovation they had introduced in the markets in the last 5 years. This was then assigned an ordinal variable that ranged from Bzero innovation^—if the firm did not introduce any type of innovation—to Bthree types of innovation^—if firms simultaneously introduced all

Author's personal copy J Knowl Econ Table 2 Variables and business survey questions Variables used

Business survey questions

Network Channels

We listed five market (suppliers, customers, consulting services, labor market, and competitors), six institutional (universities, polytechnics, research laboratories, regional innovation centers, professional associations, and public institutions), and one personal (personal relations) channels of interaction and ask the importance given to each one (Likert 1–5)

Learning and interaction mechanisms

Formal Informal

We listed the 12 channels and ask the firms the importance given to formal and informal mechanisms of interaction with each one (Likert 1–5)

Product Innovation

Binary (0-no; 1-yes) If firms have brought to market product innovations in the last 5 years

Process Innovation

Binary (0-no; 1-yes) If firms have brought to market process innovations in the last 5 years

Organizational Innovation Binary (0-no; 1-yes) If firms have brought to market organizational innovations in the last 5 years

types of innovation. As we have already mentioned, we believe that the most innovative firms are those that can innovate consistently in their different fields, and this was the reason why cumulative logic was adopted in the construction of the innovation performance variable. Additionally, innovation is a complex, cumulative, systemic, and dynamic process. The shape of the variable we adopt refers precisely to these dimensions, seeking to move away from a more static view of innovation prevalent in the literature.

Table 3 Descriptive statistics Variables

Number

Min.

Max.

SD

Mean

Network Intensity

397

0.200

0,640

0.071

0.419

Formal Network Intensity

397

0.200

0.650

0.075

0.412

Informal Network Intensity

397

0.200

0.730

0.080

0.427

Interaction

397

0.400

0.403

0.063

0.184

Innovation Performance

397

%

Cum.

• Zero type of innovation

3

0.8

0.8

85

21.4

22.2

• Product and process innovation

151

38.0

60.2

• Product, process, and organizational

158

39.8

100.0

397

%

Cum. 18.1

• Product innovation

Economic Performance • Zero type of economic output

72

18.1

• Sales

102

25.7

43.8

• Sales and orders

110

27.7

71.5

• Sales, orders, and exports

113

28.5

100.0


Independent Variables The independent variable of model A is Innovative Performance. The possible interaction channels that firms could use to access external knowledge are identified in Table 2. In the survey, the firms were asked to identify first the channels they normally used and, secondly, the importance they attached to each channel using a 5-point Likert scale. These channels were classified as market, institutional, or personal channels. For the purposes of this study, for each channel, firms were asked to identify and classify (in the same way as previously) the nature of the interaction mechanism, whether they were formal or informal mechanisms. With this information, we developed the independent variables in model B, as explained below. Formal Network Intensity and Informal Network Intensity are both variables combined from the following transformation: in the numerator, the sum of the classification attributed by firms to the importance of each channel and the denominator is the maximum that can be assigned. This variable ranges from 0.2—if firms acknowledge the importance of each channel to a minimum (Likert = 1)—to 1—if such recognition is the maximum (Likert = 5). The greater the intensity of networking will be the more the variable approaches unity. Network Intensity is the mean of relationships established through the various channels used by firms, which can be formal or informal interactions. Network Intensity is the mean of formal and informal network intensity. Finally, in order to capture the complementarity of formal and informal networks, we developed the Interaction variable that results from the multiplication of formal networks through informal networks. Although the main concern of this article is the general relationship between innovation and knowledge networks, three control variables have also been introduced: firm size, regions, and technological intensity. The Statistical Model of Analysis Given the ordinal character of the dependent variables, the model for the ordered dependent variables can be motivated through an underlying continuous latent variable y∗, where 0

y* ¼ x β þ u with x′ a vector of k regressors, β an unknown vector of k parameters, and u an unobservable error term. We do not observe the latent variable y∗ but some discrete values y or categories such as y ¼ j if and only if μ j−1 < y* < μ j ; j ¼ 1; 2; …; J : If the distribution of u is known with distribution function F(u), we can compute the probabilities of observing y = j,


Pðy ¼ j jxÞ ¼ P μ j−1 < y* < μ j 0 0 ¼ P μ j−1 −x β < u < μ j −x β 0 0 ¼ F μ j −x β − F μ j−1 −x β : The maximum likelihood is the standard procedure to estimate the unknown parameter β. For a sample of n independent pairs of observations, (xi, yi), the likelihood function is given by n

n

Lðθ j y; xÞ ¼ ∏ ∏ Pðy ¼ j jx

δij

;

i¼1 j¼1

where δij = 1(yij = j) is a binary indicator variable and P(y = j |x) is defined above. The ordered logit is used in the following formula, i.e., F(u) is the cumulative logistic function defined by F ð uÞ ¼

expðuÞ : 1 þ expðuÞ

The parameters can be interpreted in several ways. For example, exponentiated coefficients are interpreted as partial odds ratios for being in the higher categories of y rather than in the lower half of the categories of y. To simplify, assume that x is a single variable. For a variation of x in Δx, the partial odds ratio is given by Pðy≤ j jx þ Δx =Pðy > j jx þ Δx ¼ expð−Δx β Þ OR j ¼ Pðy ≤ j jx =Pðy > j jx

Empirical Results Table 4 presents the top five most important network channels, in accordance with the answers given by firms to the survey. Table 4 Most important network channels used by firms (top five above mean) Total

Formal

Informal

Competitors

Customers

Labor market

Customers

Suppliers

Competitors

Labor market

Competitors

Consulting services

Suppliers

Public institutions

Research laboratories

Consulting services

Labor market

Polytechnics


Table 4 shows the most important channels of interaction that firms recognize in their dynamics of networking. Overall, the market channels, or interaction between clients, suppliers and competitors, or access to specific knowledge associated with the labor market are the most valued. From the perspective of the mechanisms utilized, the three most important channels associated with formal mechanisms are the channels of the market—customers, suppliers, and competitors—followed by the institutional and labor markets. In informal terms, the main difference is the substitution of market channels (customers and suppliers) for institutional channels normally associated with specific external knowledge. Economic Performance and Innovation Performance To analyze our hypotheses, two models were estimated (see Table 5). To analyze the first hypothesis, model A was estimated by using ordered logistic regression, according to the nature of the dependent variable. The dependent variable was Economic Performance, and Innovation Performance was the independent variable as previously defined. The Binnovation performance = 1^ was used as a benchmark for interpreting the results of innovation performance. Model A examines the impact of the innovations achieved by firms in their economic performance. The results are shown in Table 6. The reported coefficients show how much the probability of a firm’s economic performance grows with an increase in the independent variable, Innovation Performance. These results show firms that introduced more than one type of innovation in the last 5 years are likely to have better economic performance. The magnitude of the marginal effects also supports our argument. Product innovation increases the likelihood that the firm will be at the highest level of economic performance (sales, orders, and exports) by 30.8%. When the firm has introduced product, process, and organizational innovations, the probability of being in the highest category of economic performance increases by 31.4%. The combination of the different typologies of innovation thus appears to be a critical factor in the economic performance of firms, allowing them to obtain better results in several dimensions relevant to their Table 5 Dependent and independent variables (models A and B) Dependent variable Model A

Model B

Independent variable(s)

Economic Performance

Innovation performance

1 – Zero type of economic output

1 – Zero type of innovation

2 – Sales

2 – Product

3 – Sales and orders

3 – Product and process

4 – Sales, orders, and exports

4 – Product, process, and organizational

Innovation Performance

(1) Network Intensity

1 – Zero type of innovation

(2.1) Formal Network Intensity

2 – Product

(2.2) Informal network Intensity

3 – Product and process

(3) Interaction Network Intensity

4 – Product, process, and organizational

Author's personal copy J Knowl Econ Table 6 Model A: estimation results Dependent variable

Coefficients Marginal effects (Df/dx)

Economic performance

Economic performance

Innovation performance P

PP

Zero

− 0.15 − 0.20 − 0.20

Sales

− 0.14 − 0.14 − 0.14

(0.0270)

Sales + orders

–

1.501**

Sales + orders + exports 30.8

Innovation performance = 1, reference Product (P)

1.367**

Product + process (PP)

(0.0105)

Marginal effects: minimum of * p < 0.1

Product + process + organizational (PPO)

PPO

–

–

31.7

31.4

1.498** (0.0107)

Control variables Region

NO

Firm size

NO

Technological intensity

NO

Observations

397

Wald chi2

6.85

Prob > chi2

0.07

Log likelihood

− 543.34

Pseudo-R2

0.002

Robust p value in parentheses *p < 0.1; **p < 0.05; ***p < 0.01

competitiveness. In this model, the control variables were not statistically significant. These results confirm the first hypothesis, i.e., that innovation performance has a positive impact on firm’s economic performance. Innovation Performance and Network Intensity To analyze the second hypothesis, the following model B was estimated by using ordered logistic regression, according to the nature of the dependent variable. Innovation Performance was used as the dependent variable and Network Intensity (global, formal, informal, and interaction), as previously defined in the former section, was the independent variable. Firstly, Innovation Performance was calculated against Network Intensity (global), the second regression used Innovation Performance against formal and informal network intensity, and finally, the third regression used Innovation Performance against interaction network intensity. These three regressions capture the impact of knowledge networks in different conceptual and analytical perspectives, for the achievement of innovation output. As firms use knowledge networks with higher levels of intensity, their innovation results are also likely to be higher. This model also looks at the potential impact of the combination of formal and informal networks vis-àvis their individual impact. The results show that formal and informal networks have a


positive effect in the innovation output, and the combination of interaction between them should lead to improved performance. The results can be seen in Table 7. The reported coefficients of the first regression show that increasing the network intensity raises the probability of having increased levels of innovation performance. The marginal effects show these results in a more detail. In other words, firms that engage more intensively in knowledge networks increase the likelihood of obtaining higher levels of innovation output. The results shown in column 2 confirmed the same results consistently, with the relevance of the informal networks. Formal and informal networks are found to be extremely important when it comes to explain higher categories of innovation output, especially when firms introduced product, process, and organizational innovation as we can see by the magnitude of the marginal effects (columns 2.1 and 2.2). However, the estimation uncovers an interesting finding that can be shown in columns 3 and 3.1 (marginal effects). The interaction of formal networks and informal networks allows the best results to be obtained in terms of the probability

Table 7 Model B: estimation results Dependent variable Coefficients I n n o v a t i o n (1) performance

(2)

Marginal effects (Df/dx) (3)

Dependent (1)

(2.1)

(2.2)

(3)

Innovation Network Formal Informal Interaction Network Intensity

10.38***

Zero

–

–

(0.000)

One (P)

− 1.55

− 0,66 − 0.89

− 1.89

4.462**

Two (PP)

− 0.85

− 0.36 − 0.49

− 1.09

(0.021)

Three (PPO)

2.45

1.05

3.05

Formal Network (FN)

Informal Network (IN)

–

1.42

6.009*** (0.001)

Interaction (FN × IN)

12.87*** (0.000)

Control variables Region

No

No

Yes

Firm size

Yes

Yes

Yes

Tech. intensity

Yes

Yes

Yes

Observations

397

397

397

Wald chi2s

53.47

55.82

60.47

Prob > chi2

0.00

0.00

0.00

Log likelihood

− 397.67 − 397.02 − 395.54

Pseudo-R2

0.09

Minimum of * p < 0.1

0.09

0.09

Robust p value in parentheses *p < 0.1; **p < 0.05; ***p < 0.01

–

Author's personal copy

0

.2

.4

.6

.8

1

J Knowl Econ

.2

.3

.4 .5 Formal Network Intensity Pr(INOVA4==2) Pr(INOVA4==4)

.6

.7

Pr(INOVA4==3)

0

.2

.4

.6

.8

1

Fig. 1 Innovation performance and Formal Network Intensity

.2

.3

.4 .5 Informal Network Intensity Pr(INOVA4==2) Pr(INOVA4==4)

.6

Pr(INOVA4==3)

Fig. 2 Innovation performance and Informal Network Intensity

.7

Author's personal copy

0

.2

.4

.6

.8

1

J Knowl Econ

0

.1

.2 Interaction Pr(INOVA4==2) Pr(INOVA4==4)

.3

.4

Pr(INOVA4==3)

Fig. 3 Innovation performance and Interaction (formal × informal)

that the firm has introduced higher levels of innovation. Figures 1, 2, and 3 help us to visualize the results obtained in model B. Each figure shows us the predicted probabilities of the Innovation Performance (vertical axis) by the different typologies of networks (horizontal axis) performed in model B. Although the pattern is already marked in Figs. 1 and 2, 3 shows clearly the scope of the obtained results. The interaction dynamics of formal and informal networks seems to have little relevance to firms that have only introduced product innovations. However, higher levels of interaction allow firms to simultaneously introduce product and process innovations, at least up to a certain threshold of interaction. After this threshold, increased interaction does not increase the likelihood of introducing both types of innovation. Finally, as firms increase the intensity of interaction, they can introduce all types of innovation (product, process, and organizational). Without interactions, the probability of introducing all types of innovation is greatly reduced. As the level of interaction increases the likelihood of introducing all kinds of innovation increases dramatically until it reaches its maximum precisely at the highest level of interactions. In this model, the control variables were statistically significant. Given the set of results achieved in this section, hypothesis 2 is confirmed, i.e., that participation in knowledge networks has a positive impact on the firms’ innovation performance.

Discussion and Main Conclusions This investigation revealed three important results: first, the importance of innovation in firm’s economic performance; second, the importance of knowledge networks in firm’s innovation


performance; and finally, the importance of the combination of formal and informal interaction mechanisms in the firm’s innovation process. The importance of innovation for the competitiveness of firms is now a consolidated theoretical acquisition by literature, and the empirical results of this research reinforce it. In addition, our findings confirm some of the previous studies on this topic. Faems et al. (2005), for example, found that increased network linkages led to more successful innovation. Huggins et al. (2012) conclude that the innovation performance of firms is significantly related to network capital investment. Nunes and Lopes (2012) showed that firms need external knowledge to develop their innovation activities and establish cooperation with other organizations to obtain that extra knowledge. The sample includes mainly small- and medium-sized firms with a profile of predominantly incremental innovation, which may affect the scope of the conclusions. It is possible that informal mechanisms have greater importance for SMEs. However, this observation does not challenge the main conclusion from this study, i.e., that knowledge networks play a key role in the innovation process of the firms. Firms need diversified forms of knowledge for their innovation process, and networks are precisely the support mechanism that allows them to access such knowledge. For most SMEs, this is possibly the only way by which they can innovate, and thus ensure their competitiveness. In summary, the development of innovation activities in the context of cooperation substantially increases the innovation performance of firms and hence their economic performance. In turn, Rogers (2004) points out that innovation is not an internal process. He concluded that in the innovation process, the suppliers, customers, and the external environment all have an impact. This study found that customers, competitors, suppliers, and labor market are the most important linkages in the innovation process. In the formal networks, the three most important networks were customers, suppliers, and competitors. This confirms earlier findings (Johannisson et al. 2002; Gellynck and Vermeire 2009, e.g.) that these linkages play an important role in the innovation process as they provide access to new ideas and information. These links with suppliers and customers can also lead to a reduction in transaction costs (Coase 1992) as relationships develop over to time and become more established. This can lead to an increase in trust which is necessary for the transfer of knowledge. The transfer of both implicit and tacit knowledge is especially important in the innovative process (Witt 2004; Antonelli and Ferrão 2001) and is required for firms to develop and maintain competitive advantage. Finally, this paper has highlighted the importance of informal mechanisms of interaction in the innovation process. We found that informal mechanisms are a very important way of managing the relationship between innovative firms and consulting services or R&D centers. In addition, this study confirms that the labor market is of crucial importance in the innovation process. The labor market appears not only as an important mechanism for formal relationships but also as an informal mechanism for access to information and knowledge outside the firm. The regional environment is an important influence in, as well as skilled labor innovation (Karlsson and Olsson 1998). The availability of skilled manpower in an area increases the opportunity for the exchange of knowledge which may have been created either in the region or elsewhere. Because of their size, SMEs will employ fewer qualified staff than larger companies and are more likely to be dependent on the regional network for innovation and product


development. Therefore, access to a qualified labor market is an important factor in the innovation process. This paper shows the importance of knowledge networks on the innovative performance. The main added value of our analysis is that we managed to operationalize the relationship between networks and innovation, which appears in the literature essentially as a theoretical assumption. Moreover, we have also developed a way of measuring the importance of participation in knowledge networks for the firms’ innovative performance, a test that has been shown to be consistent with theoretical assumptions. However, the more general result that follows from our analysis is that firm performance (innovation and economic) and knowledge networks are multidimensional phenomena and the keyword is interaction. Interaction in this context is associated with complementarities, synergies, externalities, and learning conditions. In knowledge networks, we looked at formal and informal interaction mechanisms; in innovation, we looked at product, process, and organizational innovation; and in the economic performance, we looked at sales, orders, and exports. The results suggest that when interactions are considered and firms increase the level of interaction, they obtain better results. These results help to consolidate similar ones, such as Lopes and Farinha (2017). This result helps to support the argument that innovation is a systemic process that structurally depends on interactive learning dynamics. The learning dynamics and interactions are the concepts that give coherence to the innovation process and its relationship to firm economic performance. The multidimensionality of the firms’ innovation and their dynamics of interaction are key factors of their innovative performance and, consequently, of their economic performance. There are also some lessons that can be drawn from the results already obtained for the guidance of innovation policy, particularly in the context of EU countries. In line with the European 2020 Agenda, the EU (2010) approved the document Regional Policy Contributing to Smart Growth in Europe, which assumes the goal of adopting a regional policy that promotes ‘smart specialization’ of the economies of the EU. Knowledge and innovation are recognized as strategic factors of smart specialization. On the other side, in ESPON (2012), three typologies of regions are empirically identified: technologically advanced regions, scientific regions, and knowledge networking regions. This means, among other things, that the knowledge economy is expressed in a territorially differentiated way with multiple complex interactions and possibilities (see e.g., Carayannis and Campbell 2011 for a deep understanding of these complex interdependencies). A knowledge economy region can be identified as a region specialized either in high-tech sectors, or in scientific functions or is capable to obtain knowledge from other economies through cooperation and networking. Promoting networking should be a central dimension of innovation policy, especially in countries such as Portugal, where the business community is overwhelmingly composed of SMEs and where the potential for innovation is mostly incremental. Innovation policy should therefore be designed specifically for SMEs with incremental innovation as one of its main concerns. BThe challenge for regional policymakers is to develop a more targeted approach to particular subgroups with respect to their behaviour in networking and their innovation capacity…^ (Gellynck and Vermeire 2009: 732). This should be a primary concern not for reasons of size but by the specificity of the relationship of this type of firms with the knowledge networks and their role in the


combination of the medium and low-tech sectors with medium and high-tech sector in the EU. It is well known that promoting the firm’s competitiveness requires a vigorous innovation policy. Our conclusions have important implications for the understanding of this policy. Indeed, in economies with a business structure such as in Portugal, the country’s competitiveness will depend more and more on the institutional capacity to transfer the focus of innovation policy from technology and the firms themselves to focus on how to promote learning and the relational dynamics of the regional and the national innovation systems. Finally, as we mentioned earlier, we questioned firms about their innovation activities over the last 5 years (see Table 2). However, our survey was applied at a specific point in the time. Therefore, we do not have longitudinal data to test the tendencies over a longer time span, which is a limitation of our work. However, we would recommend that future studies take a longitudinal approach to the question of how firms utilize formal and informal networks to increase their competitive capabilities through innovation.

References Aalbers, R., Dolfsma, W., & Koppius, O. (2006). On and off the beaten path: How individuals broker knowledge through formal and informal networks. Paper No. 2006/08, Centre for Innovation, Research and Competence in the Learning Economy (CIRCLE), Lund University, Sweden. Acs, Z., Audretsch, D., Braunerhjelm, P., & Carlsson, B. (2012). Growth and entrepreneurship. Small Business Economics, 39(2), 289–300. Antonelli, C. (2005). Models of knowledge and systems of governance. Journal of Institutional Economics, 1(01), 51–73. Antonelli, C., & Ferrão, J. (2001). Comunicação, Conhecimento Colectivo e Inovação. Lisbon: ICS. Apanasovich, N. (2016). Modes of innovation: A grounded meta-analysis. Journal of the Knowledge Economy, 7(3), 720–737. Apanasovich, N., Alcalde-Heras, H., & Parrilli, M. (2017). A new approach to business innovation modes: The ‘Research, Technology and Human Resource Management (RTH) model’ in the ICT sector in Belarus. European Planning Studies, 1–25. Birley, S. (1986). The, role of networks in the entrepreneurial process. Journal of Business Venturing, 1(1), 107–117. Cainelli, G., Evangelista, R., & Savona, M. (2006). Innovation and economic performance in services: A firmlevel analysis. Cambridge Journal of Economics, 30(3), 435–458. Cappellin, R., & Wink, R. (2009). International Knowledge and Innovation Networks, New Horizons in Regional Science. UK: Edward Elgar ISBN: 978 1 84844 441 6. Caravaca, I, González, G., & Silva, R. (2005). Innovación, redes, recursos patrimoniales y desarrollo territorial^. In Revista EURE, Dezembro (Vol. XXXI, N.° 94, pp. 5–24) Santiago do Chile. Carayannis, E., & Campbell, D. (2011). Open innovation diplomacy and a 21st century fractal research, education and innovation (FREIE) ecosystem: Building on the quadruple and quintuple helix innovation concepts and the Bmode 3^ knowledge production system. Journal of the Knowledge Economy, 2, 327– 372. Carayannis, E., & Gonzalez, E. (2003). Creativity and innovation = competitiveness? When, how, and why, vol. 1, chapter 8. In L. Shavinina (Ed.), The international handbook on innovation (pp. 587–606). Amsterdam: Pergamon. Cefis, E., & Ciccarelli, M. (2005). Profit differentials and innovation. Economics of Innovation and New Technology, 14(1), 43–61. Coase, R. (1937). The nature of the firm. Economica, 4(16), 386–405. Coase, R. (1992). The institutional structure of production. The American Economic Review, 82(4), 713–719.

Author's personal copy J Knowl Econ Cowan, R., Jonard, N., & Zimmermann, J. B. (2007). Bilateral collaboration and the emergence of innovation networks. Management Science, 53(7), 1051–1067. Damanpour, F., Richard, M., & Avellaneda, C. (2009). Combinative effects of innovation types and organizational performance: A longitudinal study of service organizations. Journal of Management Studies, 46(4), 650–675. ESPON (2012). KIT-knowledge, innovation, territory. European Union. EU (2010). Regional policy contributing to smart growth in Europe. COM (2010) 553, Brussels. European Commission (2016). European Innovation Scoreboards, project for the European Commission, Directorate-General for Internal Market, Industry, Entrepreneurship and SMEs. Faems, D., Van Looy, B., & Debackere, K. (2005). Interorganizational collaboration and innovation: Toward a portfolio approach. Journal of Product Innovation Management, 22(3), 238–250. Fagerberg, J. et al. (2009). Innovation and economic development. UNU-MERIT Working Papers 2009–032. ISSN: 1871–9872. Farinha, L., & Ferreira, J. (2016). Measuring innovative and entrepreneurial networks performance. World Academy of Science, Engineering and Technology, International Journal of Social, Behavioral, Educational, Economic, Business and Industrial Engineering, 10(8), 2765–2774. Ferreira, J., Fernandes, C., & Raposo, M. (2014a). Cross-industry cooperation as a key factor for innovation. In P. Gupta & B. E. Trusko (Eds.), Global innovation science handbook (1st ed.). New York: McGrawHill ISBN 978–0071792707. Ferreira, J., Raposo, M., & Fernandes, C. (2014b). Inter-firm cross border co-opetition: Evidence from a twocountry comparison. In M. Peris-Ortiz & J.-M. Sahut (Eds.), New challenges in entrepreneurship and finance—Examining the prospects for sustainable business development, performance, innovation and economic growth. New York: Springer. Fischer, M. M. (2006). Innovation, networks, and knowledge spillovers: Selected essays. Springer. Fitjar, R., & Rodriguez-Pose, A. (2013). Firm collaboration and modes of innovation in Norway. Research Policy, 42, 128–138. Freeman, C. (1991). Networks of innovators: A synthesis of research issues. Research Policy, 20, 499–514. Fritsch, M., & Kauffeld-Monz, M. (2010). The impact of network structure on knowledge transfer: An application of social network analysis in the context of regional innovation networks. The Annals of Regional Science, 44(1), 21–38. Fuller-Love, N. (2009). Formal and informal networks in small businesses in the media industry. International Entrepreneurship and Management Journal, 5(3), 271–284. Fuller-Love, N., & Thomas, E. (2004). Networks in small manufacturing firms. Journal of Small Business and Enterprise Development, 11(2), 244–253. Gellynck, X., & Vermeire, B. (2009). The contribution of regional networks to innovation and challenges for regional policy. International Journal of Urban and Regional Research, 33-3, 719–737. Gokhberg, L., Kuznetsova, T., & Roud, V. (2012). Exploring innovation modes of Russian companies: what does the diversity of actors mean for policymaking. Basic Research Program, Working Papers. WP BRP 01/STI/2012. Hagedoorn, J. (1995). Strategic technology partnering during the 1980s: Trends networks, and corporate patterns in non-Core technologies. Research Policy, 24, 207–231. Hall, B. (2011). Innovation and productivity. UNU-MERIT Working Papers. ISSN 1871–9872. Hervas-Oliver, J. L., & Albors-Garrigos, J. (2009). The role of the firm's internal and relational capabilities in clusters: When distance and embededdness are not enough to explain innovation. Journal of Economic Geography, 9(2), 263–283. Holtskog, H. (2017). Forms of innovation—Insights from product development. Journal of the Knowledge Economy, 8(1), 63–76. Huggins, R., et al. (2012). Network capital, social capital and knowledge flow. Industry and Innovation, 19-3, 203–232. Jensen, et al. (2007). Forms of knowledge and modes of innovation. Research Policy, 36, 680–693. Johannisson, B., Ramírez-Pasillas, M., & Karlsson, G. (2002). The institutional embeddedness of local interfirm networks: A leverage for business creation. Entrepreneurship & Regional. Development, 14(4), 297– 315. Karlsson, C., & Olsson, O. (1998). Product innovation in small and large enterprises. Small Business Economics, 10(1), 31–46. Karlsson, C., Johansson, B., & Stough, R. R. (Eds.). (2005). Industrial clusters and inter-firm networks. In Edward Elgar publishing. Cheltenham. Kemp, R. G., et al. (2003). Innovation and firm performance. SCALES. ISBN: 90–371-0875-X.

Author's personal copy J Knowl Econ Kleinknecht, A., & Mohnen, P. (Eds.). (2002). Innovation and firm performance. Econometric explorations of survey data. Basingstoke: Palgrave. Koellinger, P. (2008). The relationship between technology, innovation, and firm performance: empirical evidence on E_Business in Europe. ERIM Reports Series Research Management. Lambooy, J. (2005). Innovation and knowledge: Theory and regional policy. European Planning Studies, 13(8), 1137–1152. Lopes, R. (2001). Competitividade, Inovação e Territórios. Lisboa: Celta editora. Lopes, J., & Farinha, L. (2017). Measuring the performance of innovation and entrepreneurship networks. Journal of Knowledge Economy, 1–22. Lorenz, E., & Lundvall, B.-Ä. (Eds.). (2006). How Europe’s economies learn: Coordinating competing models. New York: Oxford University Press ISBN: 978-0-19-920319-2. Lundvall, B.-Ä. (1985). Product innovation and user-producer interaction. Aalborg: Aalborg University Press. Lundvall, B.-Ä. (2007). National innovation systems—Analytical concept and development tool. Industry and Innovation, 14(1), 95–119. Marlon, F., & Lambert, R. (2009). Exploring non-technological and mixed modes of innovation across countries. OECD—innovation in firms: A microeconomic perspective. OECD Publishing. Martin, R. (2013). Differentiated knowledge bases and the nature of innovation networks. European Planning Studies, 21(9), 1418–1436. Morone, P., & Testa, G. (2008). Firms growth, size and innovation an investigation into Italian manufacturing sector. Economics of Innovation and New Technology, 17(4), 311–329. Nunes, S. (2012). O papel do território no processo de inovação empresarial, PHD Thesis. IUL-ISCTE, Lisbon, July. Nunes, S., & Lopes, R. (2012). The importance of external knowledge in the firm innovation process. In Bernhard, I. (Ed.), 15th Uddevalla Symposium 2012, Entrepreneurship and Innovation Networks, Algarve University, Faro 14–16 June. Research reports 2012:02, University West, Sweden. Nunes, S., & Lopes, R. (2015). Firm performance, innovation modes and territorial embeddedness. European Planning Studies, 23(9), 1796–1826. https://doi.org/10.1080/09654313.2015.1021666. OECD. (2005). Oslo manual: Guidelines for collecting and interpreting innovation data (3rd ed.). Paris: OECD. Oerlemans, L., Meeus, M., & Boekema, F. (1998). Do networks matter for innovation? The usefulness of the economic network approach in analyzing innovation. Journal of Economic and Social Geography, 89, 298–309. Oerlemans, Leon; Meeus, Marius e Boekema, F. (2001). Firm clustering and innovation: Determinants and effects, Papers in Regional Science, 80,337–356 Parker, A., Borgatti, S. P., & Cross, R. (2002). Making invisible work visible: Using social network analysis to support strategic collaboration. California Management Review, 44(2), 25–46. Parrilli, D., & Alcalde Heras, H. (2016). STI and DUI innovation modes: Scientific-technological and contextspecific nuances. Research Policy, 45(4), 747–756. Parrilli, D., González, J., & Peña, I. (2012). Disentangling the relationship between learning modes, innovation types and firm performance: evidence from Spain. The Challenge of Regional development in a World of changing hegemonies: knowledge, competitiveness and austerity. International Conference on Regional Science, XXXVIII Reunión de Estudos Regionales – AECR. 22–23 November. Bilbao, Spain. Parrilli, D., Fitjar, D., & Rodriguez-Pose, A. (2016). Business innovation modes: A review from a country perspective. In M. D. Parrilli, R. D. Fitjar, & A. Rodriguez-Pose (Eds.), Innovation drivers and regional innovation strategies (pp. 197–218). New York: Routledge. Porter, M. (1990). The competitive advantage of nations. London: MacMillan Press. Powell, W., & Grodal, S. (2005). Networks of innovators. In J. Fagerberg, D. E. Mowery, & R. Nelson (Eds.), The Oxford handbook of innovation (pp. 56–85). New York: Oxford University Press. Rogers, M. (2004). Networks, firm size and innovation. Small Business Economics, 22(2), 141–153. Rutten, R., & Boekema, F. (2004). The spatial dimension of learning: Case study and conceptualization. In P. Cooke & A. Piccaluga (Eds.), Regional economies as knowledge laboratories (pp. 181–196). USA: Edward Elgar ISBN: 1-84376-821-6. Sammarra, A., & Biggiero, L. (2008). Heterogeneity and specificity of inter-firm knowledge flows in innovation networks. Journal of Management Studies, 45(4), 800–829. Schumpeter, J. (1934). The theory of economic development. Cambridge: Harvard University Press. Storper, M., & Venables, A. J. (2004). Buzz: Face-to-face contact and the urban economy. Journal of Economic Geography, 4(4), 351–370.

Author's personal copy J Knowl Econ Tödtling, F., & Kauffman, A. (2001). The role of the region for innovation activities of SME’s. European Urban and Regional Studies, 8(3), 203–215. Tödtling F., Lehner, P., & Trippl, M. (2004). Knowledge intensive industries, networks, and collective learning. 44th European Congress of the European Regional Science Association, 25–29 August, Porto University, Portugal. Tödtling, F., Lehner, P., & Kaufmann, A. (2006). Do different types of innovation require different kinds of knowledge interactions? 46th European Congress of the European Regional Science Association, Agosto, Voglos, Grécia. Williamson, O. E. (1975). Markets and hierarchies. New York: Free Press, pp. 26–30. Witt, P. (2004). Entrepreneurs’ networks and the success of start-ups. Entrepreneurship & Regional Development, 16(5), 391–412.