Essays on the role of Internationalization on the R&D Activities

Transcript

1 University of Patras Doctoral Thesis Essays on the role of Internationalization on the R&D Activities Patras 2013

2 Acknowledgements This PhD Thesis encompasses the candidate s traits and competencies as they were formed by interaction with others and own self-struggle. I would like to thank Professors K. Tsekouras, D. Skuras and M. Vivarelli for their unequivocal support during my studies. I would also like to extent my gratitude to all those who supported me along the way, each one with their unique manner. -Pg. i-

3 Table of Contents Chapter 1: Introduction... 1 References... 7 Chapter 2: Endogeneity between Internationalization and Knowledge creation of global R&D leader firms: An econometric approach using Scoreboard data Introduction Theoretical and empirical context The drivers of innovation The knowledge base formation Endogeneity and findings from the literature Methodological Underpinnings Data and Variables Definitions Results and Discussion Econometric Issues The EXPINT and RDINT relationship The determinants of EXPINT and RDINT Conclusions References Chapter 3: The profile of R&D activities of Greek Manufacturing Firms and the role of Internationalization. Primary Results of a survey Introduction The population of interest and the sample of the survey The research tool and field research The knowledge capital The construction of knowledge capital The knowledge intensity The GRD firms Absorptive Capacity Mapping R&D Effectiveness The internationalization activities of GRD firms The relationship between GRD firms Export Orientation and Knowledge Base GRD firms Knowledge Stock and Export Intensity Organizational characteristics of R&D activities and Export Orientation 91 -Pg. ii-

4 Innovation performance and appropriation of R&D outcomes GRD firms R&D Collaborations Conclusions References Appendix Chapter 4: Efficiency and Competitive Advantage based on R&D activities: an antecedent of the decision to export? Introduction Theoretical Framework and Hypotheses Formulation Firms R&D based technical efficiency, competencies, capabilities and (in)effectiveness Firms R&D stock, technical efficiency and competitive advantage The relationship between export decision, firm s R&D based competitive advantage, and the role of patterns of innovation Methodological Strategy The efficiency of GRD firms: A Bootstrapped DEA approach A mover-stayer model for GRD firms Conditional Expectations, Treatment, and Heterogeneity Effects Data Structure Variables definition and data specificities Discussion of the empirical results Mover-Stayer model estimation results and hypotheses validation The two regimes of firms The Export decision Treatment and Heterogeneity Effects Conclusions References Chapter 5: Knowledge Base, Exporting Activities, Innovation Openness and Innovation Performance: A SEM approach towards a unifying framework Introduction Literature Review and Hypotheses Formulation Framing the relationship between Innovation inputs, outputs, and R&D collaboration The role of internationalization in shaping the firm s knowledge base, innovation strategy and performance Pg. iii-

5 Controlling for heterogeneity Methodology The SEM approach The WLSMV estimator The MLR estimator The measurement model and variables employed Control Variables Results and Discussion The measurement model The structural model Conclusions References Appendix I. Unstandardized and standardized estimated coefficients II. Model fit indices Chapter 6: Conclusions General remarks The international context The national context Limitations of Research Policy Implications and Future Research Directions 235 -Pg. iv-

6 Εκτεταμένη Περίληψη: Δοκίμια για το ρόλο της διεθνοποίησης στις διαδικασίες Έρευνας και Τεχνολογικής Ανάπτυξης (ΕΤΑ) (Extended abstract in Greek) Εισαγωγή Η ενδογένεια ανάμεσα στις διαδικασίες Διεθνοποίησης και Παραγωγής Γνώσης των ηγέτιδων επιχειρήσεων σε δραστηριότητες ΕΤΑ: Μια οικονομετρική προσέγγιση Το προφίλ των δραστηριοτήτων ΕΤΑ των ελληνικών μεταποιητικών επιχειρήσεων και ο ρόλος της διεθνοποίησης: Βασικά αποτελέσματα της έρευνας πεδίου Αποτελεσματικότητα και ανταγωνιστικό πλεονέκτημα που βασίζεται σε δραστηριότητες ΕΤΑ: πρόδρομοι της απόφασης για εξαγωγές? Γνωσιακή Βάση, Εξαγωγικές Δραστηριότητες, Ανοικτή Καινοτομία και Καινοτομική Αποδοτικότητα: Προς ένα ενιαίο θεωρητικό πλαίσιο μέσω της προσέγγισης διαρθρωτικών εξισώσεων (SEM) Συμπεράσματα Το διεθνές πλαίσιο Το εθνικό πλαίσιο Προτάσεις Πολιτικής, Ερευνητικοί Περιορισμοί και Μελλοντικές ερευνητικές κατευθύνσεις Pg. v-

7 Table of Tables Table 2. 1 Dataset composition and dependent variables distribution Table Dependent and Explanatory Variables definition Table Descriptive Statistics of the employed variables Table Estimates of EXPINT and RDINT system of equations with alternative approaches Table Econometric Tests Table Variance Inflation Factors for the employed variables Table 3. 1 Basic characteristics of the identified population and the obtained sample Table New Industrial distribution in the population and the sample Table 3.3. OECD Industrial Classification based on technological intensity Table 3.4. Basic Descriptive statistics of the KNEMPLand KNASS in relation to GRD firm size Table RDAbscap with respect to technological intensity and firm size Table R&D effectiveness of GRD firms Table Distribution of exporting activities depending on technological sector Table Routes of exporting per export intensity class Table Exporting Barriers of the GRD firms Table KNASS distribution with respect to firm size and export intensity Table Distribution (%) of additional elements of GRD firms knowledge base with respect to exporting status and technological intensity * Table Percentage of Innovative Sales and Innovative Products to Total Sales and Products Spectrum. Distribution with respect to Technological Opportunities and Export Status* Table GRD firms R&D Collaborations intensity with respect to exporting status, firm size, and sectoral technological intensity Table 3. 1 Basic characteristics of the identified population and the obtained sample Table New Industrial distribution in the population and the sample Table 3.3. OECD Industrial Classification based on technological intensity Table 3.4. Basic Descriptive statistics of the KNEMPLand KNASS in relation to GRD firm size Pg. vi-

8 Table RDAbscap with respect to technological intensity and firm size Table R&D effectiveness of GRD firms Table Distribution of exporting activities depending on technological sector Table 3.9. Routes of exporting per export intensity class Table Exporting Barriers of the GRD firms Table KNASS distribution with respect to firm size and export intensity Table Distribution (%) of additional elements of GRD firms knowledge base with respect to exporting status and technological intensity * Table Percentage of Innovative Sales and Innovative Products to Total Sales and Products Spectrum. Distribution with respect to Technological Opportunities and Export Status* Table GRD firms R&D Collaborations intensity with respect to exporting status, firm size, and sectoral technological intensity Table Conditional Expectations, Treatment, and Heterogeneity Effects Table 4.2. Variables Definition and Basic Descriptive Statistics Table 4.3. Estimation results of the Mover-Stayer model Table 4.4. Testing the impact of innovation patterns and the validity of the two regimes hypothesis Table 4.5. Results on Conditional Expectations, Treatment, and Heterogeneity Effects Table Operationalization of the measurement model Table 5.2. Definition and Descriptive Statistics of variables determining Export Performance Table 5.3. Definition and Descriptive Statistics of variables determining Knowledge Base Table 5.4. Definition and Descriptive Statistics of variables determining Innovation Openness and Innovation Performance Table 5.5. Results on the measurement model with WLSMV and MLR estimator Table 5.6. WLSMV results on intercorrelations between latent variables and convergent and divergent validity criteria Table 5.7. Recapitulation of the hypothesized structural model Table 5.8. Estimation results of the structural model Table 5.9. Estimation results of the Export Performance determinants Pg. vii-

9 Table Estimation results of the Knowledge base determinants Table Estimation results of the Innovation Openness and Performance determinants viii- -Pg.

10 Table of Figures Figure 2.1. Kernel density distributions of EXPINT and RDINTvariables Figure 2.2. Mode of operation of Technology Push and Demand Pull mechanisms Figure 2.3. The decreasing marginal learning benefits with respect to knowledge flows Figure 2.4. Illustration of the interrelationship between R&D and export intensity of the RDC firms Figure 3.1. Distribution of GRD firms in sample according to their technological intensity Figure 3.2. Size distribution of GRD firms in the population and the sample Figure 3.3. Knowledge Capital per employee: Distribution based on technological opportunities Figure 3.4. Knowledge Capital as a percentage of GRD Total Assets Figure 3.5. Kernel density estimates of KNEMPL variable Figure 3.6. Kernel density estimates of KNASS index Figure 3.7. Abscap with respect to Technological opportunities and Size Figure 3.8. Exporting activities of GRD firms depending on size class Figure 3.9. Geographical Distribution of GRD firms Figure Routes of exporting depending on the export intensity class Figure Participation in the Global markets for the Exporting GRD firms Figure Participation in the Global markets for the non-exporting GRD firms Figure GRD firms export intensity and Knowledge capital per employee Figure In-house and external R&D with respect to technological intensity and export status Figure Engagement in external R&D activities with respect to technological intensity and exporting regime Figure Innovation Outcomes, Exporting Status and Technological Intensity of GRD firms Figure Appropriability Conditions and Innovation Output with respect to Technological Intensity for the exporting GRD firms Figure Appropriability Conditions and Innovation Output with respect to Technological Intensity for the non exporting GRD firms Pg. ix-

11 Figure Radical and incremental innovation with respect to exporting status and technological intensity Figure 4. 1 The relationship between competencies and capabilities, technical efficiency and competitive advantage Figure 4.2. The relationship between firms export decision and competitive advantage Figure 4.3. Kernel densities between Exporters and Non-Exporters Figure 5.1. Proposed structural model of export performance, knowledge base, R&D collaborations intensity and innovation performance Figure 5.2. Path diagram of a the system of equations presented in (5.6) Figure 5.3. The structural and measurement model of export performance, knowledge base, R&D collaborations intensity and innovation performance Figure 5.4. The full model representing the measurement and structural model along with independent covariates Pg. x-

12 Chapter 1: Introduction Firms of European small open peripheral economies face an increasing globalization of markets, a strengthening of global value chains, a well documented knowledge and technological gap and these in conjunction to the current crisis at least in the southern part of Europe. These conditions compose a demanding and complex environment within which firms must cope and survive. In this direction, analyzing first and improving in turn competitiveness and productivity of European firms has become a primary policy objective of the EU at the national, regional, sectoral and firm level in an attempt to close the growth gap with the United States (Aghion et al., 2008). In this mission, boosting exporting activities and investments in Innovation, R&D and knowledge intensity is of the outmost importance since they are seen as drivers of productivity, growth and competitiveness (EU, 2012). Especially with respect to Greece s economic outlook and as it has been documented in several European policy documents and analyses, the country s innovation performance has been consistently characterized as moderately following (IUS, 2013) the EU s innovation leaders. The same picture is sketched with Greek firms export performance as a crucial component of its overall competitiveness (European Competitiveness Report, 2012). Examining more closely the relationship between firms exporting activities and innovation dynamism, the theoretical and empirical evidence suggests that firms which are presenting innovation activities are more likely to export, more likely to export successfully, and more likely to generate growth from exporting than noninnovating firms (Golovko and Valentini, 2011; Love and Roper, 2013). In other words, innovation and export performance are directly linked with the creation of a sustainable competitive advantage and are considered as a primary precondition for economic growth (Piercy et al., 1998; EU, 2012). More specifically, exporting activities are considered as the primary internationalization mode (Johanson and Valhne, 1977; 2009) and firms knowledge and learning processes are expected to play a pivotal part in the internationalization process; firms need to be in a position to -Pg. 1-

13 apprehend and assimilate new knowledge in order to compete and grow in markets in which they have little or no previous experience (Autio et al. 2000). In this direction, the relationship between the degree of internationalization and the intensity of the production of technological knowledge remains under examination, since the significant heterogeneity in terms of country, industrial distribution, firm size, and other factors has lead to contradictory results (Harris and Li, 2009). In addition, the differential effects of the firms environment which in turn can be further specified in various dimensions such as business culture, organizational characteristics, strategic orientation, national and regional systems of innovationintroduce a significant amount of unobserved heterogeneity in the employed methodological frameworks. At the same time, the causality direction, too closely related to endogeneity issues, between exporting and R&D activities has not been yet addressed adequately. The relevant literature has documented two theoretical strands, the Product Life Cycle and Endogenous Growth theory, which hypothesize on causality direction between exports and R&D activities. More specifically, the Product life cycle theory argues that innovation eventually leads to exporting (Posner, 1961; Vernon, 1966; Krugman, 1979; Dollar, 1986) and this theoretical strand is strongly interrelated with the Market Selection Hypothesis (MSH; Wagner 2007) which favours the argument that exporters have superior performance characteristics than non-exporters. On the other hand, the Endogenous growth models (Grossman and Helpman, 1989, 1990, 1991a; Segerstrom et al., 1990; Young, 1991; Aghion and Howitt, 1998, ch. 11) argue on the reverse direction of causality. The notion behind this is that exporting firms access to foreign markets provides them with feedback from their suppliers and/or customers, which gives them the opportunity to transform this knowledge into innovation. This theoretical strand has been recorded as opposite to the market selection hypothesis and is named Learning by Exporting Hypothesis (LEH; Clerides et al., 1998; Salomon and Shaver, 2005) Both the above hypotheses seem plausible and have been empirically but the relevant literature has provided contradictory results. However, it would only make sense to assume that this causality direction may be not so straightforward since causality may run in both directions that is a two-way linkage between a firm s exporting and -Pg. 2-

14 innovating activities may exist (Filipescu et al. 2013). The starting point of this PhD thesis lies on the idea that both these activities may influence each other and therefore, is focused on the investigation of the endogeneity between established knowledge creation processes (R&D activities) and internationalization activities as they are depicted in exporting activities. It is worth mentioning that towards the direction of seeking proof for the existence of an endogenous relationship between R&D activities and exports different methodological approaches have been employed. All of them however, examine the existence of endogeneity between the two main firm activities as well as identifying the appropriate set of determinants for each one of the firms activities as the relevant literature dictates. In order to (i) sufficiently address the abovementioned multiple heterogeneity, and (ii) be able to compare them, the present research investigates the interrelationship between R&D and exporting activities on two distinctively different groups of firms. More specifically, two different contextual frameworks are employed, one International and one National. The first group of firms focuses on those firms that are considered to be leaders with respect to R&D investments at a global level. The second group of firms under investigation, concerns the Greek firms which are in turn considered leaders within the national system of innovation they operate but have been consistently characterized as moderately followers within the European context (IUS, 2013). Information for the investigation of the international context was provided by the UK Department of Business, Innovation and Skills (BIS, 2007; 2008). Yet, regarding the national context, a profound lack of information exists both with respect to exporting activities but also with respect to R&D activities at the firm level, which inevitably led to the conducting of a field research at the National Level targeting the Manufacturing Sector. In this line, and in order for the gathered information to be comparable with other European surveys on Innovation and in particular with Community Innovation Survey (CIS), the design of the questionnaire was primarily based on the CIS standards. In addition to the data provided by the National survey, all the financial and other information, including annual expenditures on R&D, for the period was provided by the electronic database i-mentor. Based on this information, a better approximation of R&D performance has become feasible through the construction of Greek R&D active manufacturing firms R&D stock -Pg. 3-

15 (Kumbhakar et al., 2012). The main argument supporting this transformation is that fluctuations in R&D investment flows are more volatile than the knowledge stock acquired from such investments (Dierickx and Cool, 1989). The third chapter of this PhD thesis is devoted in presenting the specificities of the field research, the adopted methodology for the construction of firms knowledge stock, along with primary descriptive results sketching the outlook of Greek R&D manufacturing firms. The rest of this PhD thesis involves three essays each one of them examining research questions arising from the endogenous relationship between R&D and export activities. Chapter two is concerned with identifying whether R&D and export performance are endogenously related within the international context. More specifically, it is argued that the Global R&D leaders knowledge base, is augmented by knowledge flows deriving from both R&D and export activities. In this direction, R&D activities are considered as a proxy of the technology push knowledge flows and exporting activities act as demand pull knowledge flows. Employing a pooled cross section dataset of firms characterized as Global R&D leaders; an appropriate econometric methodology (Smith and Blundell, 1984) for the estimation of a simultaneous equation system which incorporates a weak exogeneity test for detection of endogeneity empirical evidence is provided on the endogenous relationship between R&D and export performance of R&D leader firms. Based on the empirical results a theoretical framework for the operation mechanism of Demand Pull and Technology Push knowledge flows is devised. It becomes evident that Technology Push and Demand Pull knowledge flows are by no means equally important. Essentially, knowledge flows from foreign markets are employed in the service of improving the efficiency of Technology Push knowledge flows generation mostly through the inducement of Localized Technical Change (Atkinson and Stiglitz, 1969; Antonelli, 1998). This fact in turn, provides the opportunity to the R&D leader firms, which by their very nature are oriented in generating knowledge via the notable investments in R&D activities, to reinforce and sustain their competitive advantage and to further strengthen their penetration in global markets. Chapters four and five of this PhD thesis are situated within the Greek Framework. In chapter four, research efforts are focused in the existence and subsequent manifestation of differential effects from the Greek R&D active manufacturing firms decision to export on the formation and sustainability of their competitive advantage, -Pg. 4-

16 as the latter is proxied by their technical efficiency. Theorizing in this direction, a framework is composed showcasing that firms technical efficiency, when driven by innovation competences and capabilities, reflects their competitive advantage. In doing so, a modus operandi is sketched of the mechanism which transforms innovation competencies and capabilities into competitive advantage and is dependent upon the firms exploration and exploitation processes (March, 1991) as well as their (in)effectiveness (Mouzas, 2006). In this line, it is argued that firms R&D based competitive advantage, endogenously determines their decision to participate in foreign markets. The existence of endogeneity is grounded theoretically in the fact that firms decision to export has been discounted based on the expected benefits and costs from such an activity. In other words, the existing business model developed inside the firm that transforms competencies in capabilities essentially dictates firms strategic decision to go abroad. In this essay, employing an informed cross section dataset of low-tech Greek R&D active manufacturing firms and applying a two-stage analysis which consists of a bootstrapped DEA approach and a Mover-Stayer model, the following issues are investigated: (i) the impact of innovation patterns on R&D active firms decision to export and competitive advantage, (ii) the existence of endogeneity between the decision to export and firms competitive advantage as the latter is captured by their R&D driven technical efficiency, (iii) the emergence of an exporting and nonexporting regime due to differential R&D patterns in determining both the decision to export and firms competitive advantage. Based on the estimation results, it is argued that innovative firms would self select to an exporting status not because they are more productive than some other peers but because the anticipated net benefits would reinforce their competitive advantage. This dichotomization into exporters and non exporters brings into the surface differential patterns in innovation processes between the emerged regimes. Juxtaposing the World s leaders in R&D and the Greek R&D active firms belonging to the low tech industries of the manufacturing sector two different transformation mechanisms are identified. However, the endogenous relationship between R&D and export performance is confirmed in both examined cases. Having established that R&D and exporting activities are endogenously related both with respect to the International but also the National context, in the fifth chapter of -Pg. 5-

17 this PhD thesis the attention is directed in an effort to depict in a unifying framework the underlying complex interrelationships between Greek R&D active manufacturing firms internationalization and innovation performance and their knowledge base and R&D collaborations intensity respectively. It is argued that these relationships are all part of the firms complex strategy for living up to the challenges of the regional, national and global business interface, and as such they are interrelated. More specifically, the knowledge creation processes as they have been captured by firms knowledge base components and their external search strategy for R&D collaborations have been paired with the firms internationalization performance and innovation performance respectively. Another level of complexity is added when the dual role of R&D collaborations intensity as both a means of knowledge creation and internationalization is considered. In other words, besides the apparent two way relationships examined, innovation openness may also be linked with internationalization performance, while knowledge base may very well be linked with innovation performance. In order to identify the relationships between these four key variables, the sample of the Greek R&D manufacturing firms that came as a result of the field research, is employed. Because of the overwhelming complexity of the above theoretical illustration a Structural Equation Modeling (SEM) approach has been adopted for the simultaneous estimation of a non recursive system of equations containing latent variables. Estimation results indicate that Greek R&D manufacturing firms knowledge base plays a crucial role in shaping their internationalization performance and innovation performance, while export performance is does not exert a statistically significant influence on knowledge base but R&D collaborations intensity in fact does. In addition, a two way relationship between innovation openness as it is captured by R&D collaborations intensity and innovation performance is found. Finally, the last chapter of this PhD thesis provides concluding remarks and policy implications drawn from the empirical results and the theoretical considerations of the four main chapters along with research limitations and further research directions. -Pg. 6-

18 References Aghion, P. and P. Howitt (1998), Endogenous Growth Theory, MIT Press, Cambridge MA. Aghion, P., E. Bartelsman, E. Perotti and Scarpetta, S. (2008), Barriers to exit, experimentation and comparative advantage, LSE Ricafe2 Working Paper No. 56, London School of Economics. Antonelli, C. (1998) The Dynamics of Localized Technological Changes. The Interaction between Factor Costs Inducement, Demand Pull and Schumpeterian Rivalry, Economics of Innovation and New Technology, 6, Atkinson and Stiglitz (1969), A new view of Technological Change, Economic Journal, 79, Autio E, Sapienza H.,, and Almeida J., (2000). Effects of age at entry, knowledge intensity, and imitability on international growth. Academy of Management Journal, 43, BIS, (2007) The 2007 R&D Scoreboard: Report for company data, Department of Business Universities and Skills, UK BIS, (2008) The 2008 R&D Scoreboard: Report for company data, Department of Business Universities and Skills, UK Clerides, S., S. Lach and J., Tybout (1998), Is Learning by Exporting Important? Micro-dynamic Evidence from Colombia, Mexico, and Morocco, Quarterly Journal of Economics, 113, Dierickx I., and Cool K., (1989). Asset Stock Accumulation Sustainability of Competitive Advantage, Management Science, 35, Dollar, D. (1986), Technological Innovation, Capital Mobility, and the Product Cycle in North-South Trade, American Economic Review, 76, European Competitiveness Report, (2012). Reaping the benefits of Globalisation, Filipescu, D., Prashantham, S., Rialp, A., and Rialp J, (2013). Technological Innovation and Exports: Unpacking Their Reciprocal Causality, Journal of International Marketing, 21, Pg. 7-

19 Golovko, E. and G. Valentini (2011). Exploring the complementarity between innovation and export for SMEs' growth, Journal of International Business Studies, 42, 3, p Grossman, G. and E. Helpman (1989), Product Development and International Trade, Journal of Political Economy, 97, Grossman, G. and E. Helpman (1990), Comparative Advantage and Long-run Growth, American Economic Review, 80, Grossman, G. and E. Helpman (1991), Endogenous Product Cycles, Economic Journal, 101, Harris R., and Li Q., (2009). Exporting, R&D, and absorptive capacity in UK establishments, Oxford Economic Papers, 61, Hollanders H. and Es-Sadki N. (2013), Innovation Union Scoreboard Annual Report 2013, European Union _en.pdf Johanson, J., and Vahlne, J., (1977). The internationalization process of the firm: A model of knowledge development and increasing foreign market commitments, Journal of International Business Studies, 8, Johanson, J. and Vahlne, J., (2009), The Uppsala Internationalization Model Revisited: From Liability of Foreignness to Liability of Outsidership, Journal of International Business Studies, 40, Krugman, P. (1979), A Model of Innovation, Technology Transfer, and the World Distribution of Income, Journal of Political Economy, 87, Kumbhakar, S., Ortega-Argilés, R., Potters, L., Vivarelli, M., and Voigt, P., (2012). Corporate R&D firm efficiency: evidence from Europe s top R&D investors, Journal of Productivity Analysis, 37, Love J. and Roper S., (2013). SME Innovation, Exporting and Growth, Enterprise Research Centre, White paper No.5 March, J., (1991). Exploration exploitation in organizational learning, Organization Science, 2, Mouzas, S., (2006). Efficiency versus Effectiveness in Business Networks, Journal of Business Research, 59, Piercy, N. Kaleka, A., and Katsikeas, C., (1998). Sources of Competitive Advantage in High Performing Exporting Companies, Journal of World Business, 33, Pg. 8-

20 Posner, M. (1961), International Trade and Technical Change, Oxford Economic Papers, 13, Salomon R., and Shaver J., (2005). Learning by exporting: new insights from examining firm innovation, Journal of Economics and Management Strategy, 14, SMEs Performance Review, (2012). EU SMEs in 2012: at the crossroads. Annual report on small and medium-sized enterprises in the EU, 2011/12, European Commission, Smith, R.and R. Blundell (1986), An Endogeneity Test for a Simultaneous Equation Tobit Model with an Application to Labour Supply, Econometrica, 54, Segerstrom, P., T. Anant and E. Dinopoulos (1990), A Schumpeterian Model of the Product Life Cycle, American Economic Review, 80, Vernon, R. (1966), International Investment and International Trade in the Product Cycle, Quarterly Journal of Economics, 80, Wagner, J. (2007) Exports and Productivity: A Survey of the Evidence from Firmlevel Data, The World Economy, 30, p Young, A. (1991), Learning by Doing and the Dynamic Effects of International Trade, Quarterly Journal of Economics, 106, Pg. 9-

21 Chapter 2: Endogeneity between Internationalization and Knowledge creation of global R&D leader firms: An econometric approach using Scoreboard data 2.1. Introduction Among the very large number of stimuli a firm has for expanding its knowledge base, conducting R&D and engaging in exporting activities are identified as two especially prominent activities that offer the firm opportunities towards this direction (Johanson and Valhne, 1977; Nonaka, 1994). In this chapter these two firm activities are perceived as potential knowledge flows to augment their knowledge stock. Although their relationship has been investigated thoroughly by scholars (Lall and Kumar, 1981; Schlegelmilch and Crook, 1988; Wilmore, 1992; Lefebvre et al., 1998; Wakelin, 1998; Bleaney and Wakelin, 2002; van Dijk, 2002; Lachenmaier and Woessmann, 2006; Pla-Barber and Alegre, 2007), the existence of endogeneity between these two activities tends to be omitted in both empirical findings and theoretical predictions. Thi chapter, builds on the potential existence of endogeneity between technological change and foreign markets forces, as they are represented by R&D and Export Intensity of the Global R&D leaders. Unravelling such an endogenous relationship, is in line with the theoretical framework arguing that technology push and market pull forces act simultaneously (von Tunzelmann, 1995; Ruttan, 1997; Ito and Lechevalier, 2010) and are essential in evolving technological paradigms and shaping new technological trajectories (Dosi, 1988). The best equipped candidates to shape or redefine existing and new paradigms and trajectories are those firms that distinguish themselves from their peers, in that they spend a great deal of their resources in augmenting their knowledge base. More specifically, it could be argued that the knowledge bases of these firms are fueled, in a particular mode, by both directions, i.e. from the R&D activities, and also -Pg. 10-

22 from knowledge flows generated due to interactions with foreign markets. In this context however, the R&D process is characterized by great uncertainty and risk, and can be perceived as a search for more efficient methods of value creation. (Nelson, 1982; Castellacci, 2007). I demonstrate how Technology Push and Demand Pull knowledge flows are linked and how they affect firms decisions to invest in the creation of new technological knowledge. Specifically, for the RDL firms, Technology Push and Demand Pull knowledge flows are by no means equally important. Essentially, knowledge flows from foreign markets are employed in the service of improving the efficiency of Technology Push knowledge flows generation mostly through the inducement of Localized Technical Change (Atkinson and Stiglitz, 1969; Antonelli, 1998). This fact in turn, provides the opportunity to the RDL firms, which by their very nature are oriented in generating knowledge via the notable investments in R&D activities, to reinforce and sustain their competitive advantage. Drawing information from the UK Government Department of Business Innovation and Skills (BIS), regarding R&D global leaders (RDL firms), I devised a pooled cross-section dataset for period and test for the presence of endogeneity with respect to R&D and export intensity. The econometric approach employed takes advantage of the intensity measures, instead of models of binary variables, exploiting the full information conveyed by the outcome of the RDL firms decision-making processes with respect to their R&D and exporting activities. The rest of the chapter is organized as follows. In the next section a brief review of the literature regarding the empirical and theoretical findings of the relationship between R&D and exporting activities will be conducted. Section 3 presents some modelling issues and the econometric strategy to be followed, while section 4 provides the data and definition of variables. Section 5 is devoted to the discussion of the econometric results; section 6 concludes the chapter and poses some further research directions. 2.2 Theoretical and empirical context The drivers of innovation -Pg. 11-

23 The relevant literature has been quite extensively preoccupied with the identification of the drivers of technical change, as they are reflected in the decision to invest in R&D activities. In the form of hypotheses, different but as it shall be argued complementary aspects have been expressed, and ever since, scholars, have been arguing in favour or against them. In the heart of this discussion, the Technology Push and Market Pull (Schmookler, 1966; Rosenberg, 1990), along with the Schumpeterian hypotheses regarding bigness and fewness (Acs and Audretsch, 1987) have attracted the focus of the efforts trying to answer what drives and determines innovation 1. Additionally, cumulativeness, appropriablity conditions and technological opportunities have been nominated to be among the cornerstones of industrial evolution and dynamics (Malerba and Orsenigo, 1993; Malerba and Orsenigo, 1996; Scherer, 1965). Innovation has been detected using a number of alternative measures. Firm s expenditures on R&D, patents and Intellectual Property Rights, along with innovation indicators -depicting whether a firm has introduced a process or product innovation or both (Peters, 2009), are the most commonly employed proxies of innovation. In this chapter, the focus is shifted on firm s expenditures on Research and Development as an innovation indicator, having one distinctiveness: the R&D expenditures concern firms that have been globally announced to be those that spend the most in this area of investment. This constitutes the backbone of the analysis in the sense that these firms most prominent characteristic is the fact that they have all that it takes to be considered as carriers of technical change and new technological trajectories (Dosi, 1988). At the same time, this extraordinary investment on R&D, arms these firms with the necessary capabilities to cope with global competition (Chandler, 1986). To elaborate further, the intense and sometimes harsh global competition, in which these firms are destined to function and compete, imposes an one-way solution, that of continuous improvement on both products and processes, allowing, and determining to some extent, the scientific evolution and at the same time sustaining a firm s competitiveness (Ernst and Kim, 2002) On the other hand, these firms need to meet a 1 For a comprehensive review of the literature regarding the characteristics of markets and firms that influence industrial innovation the interested reader can draw from Cohen (1995), chapter for Empirical studies of Innovative activity -Pg. 12-

24 highly diversified global demand for their products and they need to do so, timely and effectively exploiting scale and scope economies as well as dynamic returns to learning (Chandler, 1994). Setting off to identify what drives RDL firms intense orientation towards R&D activities, it is essential to sketch first the contextual foundations of the analytical framework. As it was previously mentioned, being among those who invest the most in R&D sets out a signal that competition has been taken one step ahead of the current technological frontier. Investing on R&D boosts scientific research and constitutes a strategic choice (Dierickx and Cool, 1989) towards a sustainable competitive advantage (Rumelt, 1984). From that point of view, the development and expansion of internal firm specific capabilities and routines (Nelson and Winter, 1982; Teece and Pisano, 1994) associated with the generation of new technological knowledge as well as the identification and exploitation of technological opportunities, has been named in the relevant literature as the Technology Push hypothesis (Mowery and Rosenberg, 1979). On the other hand, much attention has been given to the influence of the demand conditions, in directing firms. The relevant literature has documented that demand conditions exert quite a significant impact on the decision and extent of a firm s innovative strategy. In other words, and according to the Demand Pull hypothesis (Schmookler, 1966; Scherer, 1982; Ruttan, 1997; Piva and Vivarelli, 2007), market signals drive the innovation possibility frontier of each and every firm (Dosi, 1988). As Scherer (1982) puts it, the demand-pull theory is inextricably linked with the ability to make profits. Therefore, market opportunities (Kamien and Schwartz, 1982, p. 35) are what matters in directing innovative activities. Nonetheless, Mowery and Rosenberg, (1989, p.8) noted that the acquisition of knowledge for innovation is not an once-and-for-all matter. Rather than a unidirectional, one-time occurrence of transfer of basic scientific knowledge to application, the processes of innovation and knowledge transfer are complex and interactive ones, in which a sustained two-way flow of information is critical. The ability to adopt a new technology, to evaluate a new technique, or even to pose a feasible re- -Pg. 13-

25 search problem to an external research group may require substantial technical expertise within the firm. Specifically, regarding the Global R&D firms, the Technology Push and the Market Pull frameworks might be considered as complementary within which drivers to innovation can be traced. Adopting the view that innovation is a problem solving process (Nelson, 1982), firms R&D activities are responsible for providing scientifically based solutions to problems posed either by the firms internal operation or factors related to their external environment or a combination of the two. As von Tunzelmann (1995) notes, this interaction not only directs the innovative search for new technological knowledge but also, in this way, it improves the efficiency of that search, having as a result the augmentation of a firm s knowledge base The knowledge base formation Dosi (1988) defined a firm s knowledge base as the set of information inputs, knowledge and capabilities that inventors draw on when looking for innovative solutions. With this definition as a starting point, later on, Kogut and Zander (1992) argued that creating new knowledge does not occur in abstraction from current abilities, but rather new learning, such as innovation, is a product of a firm s combinative capabilities to generate new applications from existing knowledge. As a result, the knowledge-based view of the firm (Grant, 1996) conceptualizes the firm as an entity that mainly creates knowledge in a unique way, which in turn constitutes its competitive advantage, and this is why the core competences of a firm are hard to imitate by competitors (Spender, 1996). The generation of technological knowledge can be viewed as a complex system of interactions between actual production experience and formal Research and development efforts (Kline and Rosenberg, 1986). Therefore, the development of new technologies can rely on the knowledge acquired by means of learning by doing and learning by using in the spectrum of techniques currently being used (von Tunzelmann, 2000). Deepening the above considerations, the process of building a firm s knowledge base is by nature a highly recursive and cumulative process associated with path-dependent as well as past-dependent characteristics (David, -Pg. 14-

26 1985; Antonelli, 1998) and dynamic increasing returns of learning (Geroski et al. 1997). More specifically, R&D activities can be directed towards the solution of any problem, but once it has been carried out, the resulting knowledge is specific to the problem addressed (Atkinson and Stiglitz, 1969). In other words, path dependence of R&D investments refers to irreversibility as well as indivisibility characteristics which determine, to a great extent, the direction and the potential of the firms knowledge base augmentation (David 1975). Past-dependent characteristics affect the generation of new technological knowledge due to differential initial conditions existing among industrial structures. These differentials translate in uneven rates of change with respect to factor prices, demand preferences and other industry specific characteristics which in turn affect the firm s technological evolution (Caves and Porter, 1977; Antonelli, 1998). Dynamic increasing returns of learning accrue to the fact that the generation of new technological knowledge can be viewed as a quasi-economic good in the sense that it is acquired and recombined through a joint process of internal and external learning, production and communication (Geroski et al., 1997). As a concomitant of the above, the generation of new technological knowledge can be viewed as a continuous technological search seeking to improve and diversify existing techniques and applications (Nelson 1982). The outcome of this knowledge generating process induces Localised Technological Change (LTC, Atkinson and Stiglitz, 1969; Antonelli, 1998), in the sense that the technological change induced concerns a specific set of techniques and/or problem solutions, and therefore, possible spillovers to the entire production possibility set are limited. Localised technological knowledge should be perceived as the product of systemic bottom-up process of induction from actual experience, sharply contrasting with the top-down process of deduction from general scientific principles on which the received theory of knowledge as a public good is rested (Dosi, 1988). It is evident that such a specific targeting may evoke improvements in a firm s efficiency and productivity Endogeneity and findings from the literature -Pg. 15-

27 The existence of an endogenous relationship between R&D and exporting activities, even though recognised very early in the literature (e.g. Keesing, 1967), it has been neglected for almost twenty years and then only scarce and rather peripheral research has been undertaken especially when it comes to empirical analyses. On the contrary, from a theoretical point of view, and specifically in the context of Product Life Cycle and Endogenous Growth theory, the endogenous relationship between innovation and exporting activities constitutes a pillar of the analysis. More specifically, the Product life cycle theory argues that innovation will eventually lead to exporting (Posner, 1961; Vernon, 1966; Krugman, 1979; Dollar, 1986) and that exports act as a channel for diffusing as well as transferring knowledge and technology (Saggi, 2002). This theoretical strand is strongly interrelated with the Market Selection Hypothesis (MSH; Wagner 2007) which favours the argument that exporters have superior performance characteristics than non-exporters. On the other hand, the Endogenous growth models of International Trade (Grossman and Helpman, 1989, 1990, 1991a; Segerstrom et al., 1990; Young, 1991; Aghion and Howitt, 1998, ch. 11) consider that innovative activity is endogenously determined and predict some dynamic effects from its relationship with international trade. More specifically, it is argued that exporting firms access to foreign markets provides them with feedback from their suppliers and/or customers, which gives them the opportunity to transform this knowledge into innovation. This theoretical strand has been recorded as opposite to the market selection hypothesis and is named Learning by Exporting Hypothesis (LEH; Evenson and Westphal, 1995). At an empirical level, firms exporting activities have occupied the relevant literature from the early 70 s and in particular, the last two decades the research on this topic has been intensified due to the globalization process. Although there is a substantial amount of literature investigating the determinants of exporting activities, this field of research has been guided, implicitly or explicitly, by the two abovementioned hypotheses i.e. MSH and LEH, regarding the identification of determinants and their relationship with export intensity. In particular, with respect to the extensive research on whether R&D activities have any influence on exporting propensity, results have been mixed. Indicatively, Schlegelmilch and Crook (1988) and Lefebvre et al. (1998), found that R&D intensity -Pg. 16-

28 has no influence on export intensity. Lall and Kumar (1981), investigating a sample of Indian firms, revealed a negative relationship between export and R&D intensities. On the other hand, many empirical studies have revealed a positive and statistically significant impact of R&D on export propensity (Willmore, 1992; Wakelin, 1998; Bleaney and Wakelin, 2002; Lachenmaier and Woessmann, 2006; Pla-Barber and Alegre, 2007). These studies however, do not explicitly address the existence of endogeneity nor are they concerned with the causality of this relationship. In addition, exporting activities stemming from developing countries are treated differently from those deriving from developed countries in terms of what determines them respectively (van Dijk, 2002). Towards dealing with the endogeneity issue, between these two firm decisions Hughes (1985), using cross section country level data, applied a Hausman test and found that the relationship between R&D intensity, and export intensity, was simultaneously determined. Towards the same direction, Clerides et al. (1998) apply causality tests in order to define the pattern of causality between R&D and export intensity and find that more productive firms choose to export. Smith et al. (2002), using a sample of Danish firms, tackle the issue of endogeneity and report that R&D increases the probability that a firm will become an exporting firm. Even more interesting is the empirical work by Harris and Li (2008): using a UK sample of firms, they investigate the endogeneity between R&D and export propensity and argue that a crucial factor for the lack of evidence on this endogenous relationship may be due to the lack of appropriate data and problems with econometric methods that allow testing for such an endogenous relationship. In a very recent paper, Ito and Lechevalier (2010), apply a system GMM estimation in a rich 10-year panel of Japanese firms, treating the endogeneity problem between R&D and Export intensity with lagged variables of R&D intensity as Instruments Methodological Underpinnings In this section the adopted methodological route is presented in order to ground the relationship between the export and R&D propensity of RDL firms. The focus lies -Pg. 17-

29 primarily on the formulation and implementation of an efficient test of weak exogeneity between the outcomes of the two aforementioned decision making processes of firms. Consider the i-th RDL firm which is involved in two decision-making processes, namely the determination of its R&D activities RDINT and its global orientation EXPINT in period t. Considering the case where the knowledge intensity of the RDL firm is crucial for the extent of its participation to foreign markets, the following structural equation describes the second decision-making process: + ' EXPINT + i, t RDINT i, t1 zexp δexp u EXP (2.1) i,t i,t where z EXP is the vector of control variables capturing the variation of EXPINT due to exogenous factors that affect the underlying decision-making process, corresponding vector of parameters to be estimated and δ EXP the uexp the corresponding error term. Parameter 1 depicts the influence of the RDL firm s knowledge creation intensity on its global orientation and is also going to be estimated. As regards the observability of the outcome of the decision-making process related to the RDL firm s level of internationalization the following rule applies: 0 if EXPINTit, 0 * EXPINTi, t EXPINTi, t if 0 EXPINTi, t 1 1 otherwise (2.2) where period * EXPINT i,t is the observable value of export intensity of the i-th RDL firm in t. That is, structural equation (2.1) is, by definition, a tobit equation. On the other hand, and taking into account that the examined firms are R&D global leaders, a simple linear regression of the following form may describe the corresponding decision-making process regarding the level of their R&D intensity: RDINT ' it, RD + i,t RD RDi,t z δ u (2.3) where z RD is the vector of the control variables capturing the factors engaged in the RDL firm s decision-making process with respect to the intensity of its R&D activities, δ RD is the corresponding vector of parameters, and u RD the error term with -Pg. 18-

30 usual properties. Hereafter the subscripts i and t are suppressed for simplicity. It should be mentioned that the RDINTvariable is not censored at unity, as it may was expected. In the next section, this mainly data-driven handling of the RDINTvariable I briefly discussed. To summarize, (2.1) and (2.3) are the structural forms of equations that reflect the examined firm s decision-making processes regarding the determination of its level of exporting intensity, under the condition that the firm is a global R&D leader. In the case where the knowledge intensity of the firm is interrelated with its decision-making regarding its export intensity the following condition holds: Cov u,u 0 (2.4) EXP The validity of (2.4) involves the rejection of the exogeneity of the RDINTvariable with respect to the EXPINT equation, or in other words the assumption that E RDINT, u 0is violated. Thus, equations (2.1) and (2.3) form a system of EXP simultaneous estimated equations where two characteristics are dominant: first, the recursive character of the system, and second the censored character of the EXPINT variable. In order to cope with these special features, I follow the approach introduced by Smith and Blundell (1986) closely. In particular, it is assumed that the correlation between the two error terms is of the following linear form: with RD ' u u ε (2.5) EXP RD EXP u 2 EXP EXP RD, EXP ~ NI 0, 2 RD, EXP urd RD (2.6) Regarding the distribution of the EXP 2 ε term it is assumed that EXP ~ N, EXP ε 0 σ and it is 2 independent of RDINTand u RD. At this point it should be pointed out that σ computed taking into account that from the relation (2.5) the following holds: RD is -Pg. 19-

31 uexp -εexp Var urd Var = 1 Var 2 + Var -2Cov uexp εexp uexp, εexp (2.7) It is apparent that the variance of the error term of the RDINTequation depends on the parameter α and the variances of u, ε. The parameter α as reflected in equation EXP EXP (2.5) is the slope of the linear equation or in other words reflects the degree of correlation between the two error terms. Even though the structural form for R&D intensity RDINT is allowed to depend * directly on the latent variable for export intensity EXPINT it does not directly depend on the observable variable EXPINT (Heckman, 1978). Finally, in order to obtain a consistent estimator (with a known asymptotic normal distribution) for and derive an estimator for α where the final estimation is: ' * 1 EXP exp RD 1 EXP δ RD EXPINT RDINT ˆ z δ uˆ (2.8) RDINT ˆ is the estimated dependent variable of the OLS regression and û RD are the residuals taken from the second estimated equation. Following Greene (2007) the EXP, RD parameter is calculated and tested for weak exogeneity using a simple t- 2 RD test of the hypothesis ψ=0, i.e. Corr uexp, urd 0, or α= Data and Variables Definitions The constructed dataset is drawn from the R&D Scoreboard provided by the British Department for Business, Innovation and Skills (BIS) -former Department of 2 For further elaboration on the joint log-likelihood function along with the test for exogeneity, see the paper of Smith and Blundell (1986). -Pg. 20-

32 Innovation, Universities and Skills (DIUS) and covers two time periods. More specifically, the data report financial and other basic economic characteristics of firms for the years 2006 and Attention should be drawn to the fact that the reporting firms were selected on the basis of their R&D expenditures that are funded by themselves. R&D undertaken under contract for other agents such as governments or other companies, as well as the firm s share of any associated firm or joint venture R&D investment, are excluded. Furthermore, the financial statements used in this case, are the consolidated group financial statements of the ultimate parent company. Firms which are subsidiaries of any other company are not ranked separately. The specific data handling procedure incorporates the view that the crucial strategic decisions are taken by the central management of the firm and the degree of freedom which remain for the peripheral management is rather limited to operational aspects (Penrose, 1959). From the 2007 R&D Scoreboard (DIUS, 2007) which provides information regarding the financial year 2006, 595 RDL firms are reported their exporting activities. Among these, 164 firms lacked crucial information and for that reason they were excluded from that year sample. Thus, for the year 2006 the sample consists of 431 firms. From the corresponding 2008 R&D Scoreboard (BIS, 2008), which in turn provides information regarding the financial year 2007, export sales information was available for a total of 667 RDL firms, from which 270 firms were short of other important information and thus, were excluded, leaving 397 firms in the specific year sample. In sum, the two-year dataset, , consists of 828 observations, out of which 42 firms are considered to be entrants in 2007, and 65 firms, even though they appeared in the 2006 sample, they exit in The structure of the dataset is determined by 360 firms present in both years and 54 firms appearing either in 2006 or The final composition of the dataset used along with the basic distribution characteristics of the two crucial variables, that is EXPINT and, is presented in Table 2.1. Table 2. 1 Dataset composition and dependent variables distribution Period No of Firms EXPINT Distribution (%) RDINT Distribution (%) ( ] 330 -Pg. 21-

33 Exits 42 Entries 65 Total 828 (7.18) (76.39) ( ] 78 (18.06) ( ] 65 (15.05) ( ] 112 (25.93) ( ] 16 (3.70) ( ] 152 (35.19) ( ] 7 (1.62) ( ] 55 (12.73) ( ] 5 (1.16) ( ] 4 (0.93) ( ] 7 (1.62) >1 2 (0.46) 0 17 ( ] 265 (4.89) (76.15) ( ] 65 ( ] 54 (18.68) (15.52) ( ] 90 ( ] 17 (25.86) (4.89) ( ] 128 ( ] 4 (36.78) (1.15) ( ] 45 ( ] 4 (12.93) (1.15) ( ] 3 ( ] 3 (0.86) (0.86) >1 1 (0.29) Additionally, in Fig 2.1 kernel density plots of RDINTand EXPINT are presented. Specifically, RDINTkernel density function depicts the well known regularity of a left skewed distribution (Cohen and Klepper, 1992). -Pg. 22-

34 0.5 Density 1 Density University of Patras Figure 2.1. Kernel density distributions of EXPINT and RDINT variables EXPINT kernel = epanechnikov, bandwidth = RDINT kernel = epanechnikov, bandwidth = Pg. 23-

35 Of course, in the case where the RDL firms are the unit of analysis, it is not surprising that the entire distribution is shifted rightwards. It is also not worthless to point out that contrary to what might be expected, RDINTvariable is not censored at 1. This fact may be interpreted by the particularities of a small number of firms that exist in the sample and they invest more in R&D expenditures than they can presently afford, apparently using intensively external financial resources (Bougheas, 2004). Essentially, this admittedly small subgroup of firms discounts that their competitive advantage will arise from the inflation of their knowledge base through the exploitation of a technological opportunity. Thus, combining the non-censored nature of the RDINTvariable I have been led not to use a tobit model for describing the RDINTfunction. For the total of the excluded firms, either because they fail to report export sales or because they had missing values in some variables, an issue of selection bias may arise. For that reason, the non parametric Mann Whitney U statistic has been performed, in order to test whether statistically significant differences exist in the distribution characteristics of the dataset, and in those of the excluded firms. The tests were performed upon a series of firm specific variables (R&D intensity, firm size, industrial and location distribution, profitability, profit margin and labour productivity). The performed non-parametric tests did not reveal any statistically significant differences between the two samples 3. The selected explanatory variables of the EXPINT equation are drawn from what the relevant literature of Market Selection Hypothesis and Learning by Exporting Hypotheses dictate. The corresponding selection of independent variables for the RDINT equation is guided by the relevant literature on Schumpeterian hypotheses of bigness and fewness along with the conditions for firms dynamic evolution, i.e. appropriability, cumulativeness and technological opportunities. The full set of the employed variables, their corresponding definitions and basic descriptive statistics are presented in Tables 2 and 3 respectively. 3 The tests results are not presented here due to space limitations but are available upon request. -Pg. 24-

36 Table Dependent and Explanatory Variables definition RDINT Equation EXPINT Equation EXPINT RDINT CDCGD EXPN1 EXPN12 EURD HRFND ICTD LBPRD MNHTD MNLTD MSEUR Dependent Variables The ratio of Revenues generated by Exports to the firm s total revenues The ratio of Expenditures on R&D to firm s total revenues Explanatory Variables A dummy variable that takes the value of 1 if the firm belongs to the Durable and Capital Goods Industry and 0 otherwise The total revenues generated for the firm due to Exporting activity in the year t-1,divided to firm s Sales in year t-1 EXPN1squared A dichotomous variable that takes the value of 1 if the RDL firm s location is within Europe and 0 otherwise The sum of square of the market shares of each firm. A dummy variable that takes the value of 1 if the firm belongs to the ICT Industry and 0 otherwise The ratio of total revenues to number of employees capturing labour productivity A dummy variable that takes the value of 1 if the firm belongs to the Manufacturing, High-Tech Industry and 0 otherwise A dummy variable that takes the value of 1 if the firm belongs to the Manufacturing, Low-Tech Industry and 0 otherwise The percentage of the firm s attained sales (exports) in the Region of Europe. -Pg. 25-

37 MSNAM The percentage of the firm s attained sales (exports) in the Region of Europe. PRFAB Firm s profitability index calculated as Operating Profit divided to Market Capitalisation. NAMD A dichotomous variable that takes the value of 1 if the RDL firm s location is within North America Region and 0 otherwise PRMRG Another profitability ratio calculated as Operating Profit divided to Sales RDIN1 The R&D intensity of the previous year, i.e. t-1 SIZE RDL Firm s size captured by its Sales SIZE2 SIZE squared SERVD A dummy variable that takes the value of 1 if the firm belongs to the Services Industry and 0 otherwise TIMED A dummy variable that takes the value of 1 if the time period is 2007 and 0 otherwise -Pg. 26-

38 Table Descriptive Statistics of the employed variables Variables Mean (st.dev) Min (max) CDCGD (0.327) EXPINT (0.224) EXPN (0.225) EXPN (0.209) EURD (0.500) HRFND (0.066) ICTD (0.302) LBPRD (0.168) MNHTD (0.608) MNLTD (0.258) MSNAM (0.198) Variables Mean (st.dev) MSEUR (1.000) (0.256) (1.000) RDINT (0.660) (1.000) PRFAB (0.226) (1.000) NAMD (0.409) (1.000) PRMRG (0.740) (1.000) RDIN (0.113) (1.000) SIZE (0.132) (1.848) SIZE (0.123) (1.000) SERVD (0.238) (1.000) TIMED (0.498) (0.997) Min (max) (0.991) (1.194) (1.625) (1.000) (0.960) (1.568) (1.429) (2.041) (1.000) (1.000) 2.5. Results and Discussion In this section, the empirical estimates of the econometric model as it was previously formulated are presented and discussed along with the accompanying tests which evaluate the validity of the empirical results. The structure of this section consists of three subsections. In the first subsection, the treatment of a number of issues that arise in the adopted econometric context is presented. The second subsection is concerned with the discussion about the empirical findings regarding the endogeneity between R&D and Export intensity. FIML estimates of the export and R&D intensity determinants are discussed in the final subsection. -Pg. 27-

39 Econometric Issues 4 On the basis of previous empirical findings and theoretical argumentations regarding the determinants of R&D and export intensity, a meaningful and informed set of explanatory variables is included among the available economic and financial variables. Estimation results of the above two-equation model are presented in Table 2.4. In the context of the variables selection, for each equation, an important issue needs to be addressed. Despite the fact that the main focus for the potential existence of endogeneity lies in the first equation, where R&D intensity is a suspected endogenous variable, the relevant economic theory dictates that export intensity is also a determinant of R&D intensity (Bhattacharya and Bloch, 2002). However, due to the necessary recursiveness of the model it is not possible to introduce export intensity as a control variable in the equation where R&D intensity is the dependent variable 5. Therefore, Export Intensity is used with one time lag EXPN1, as an Instrumental Variable (IV). Later on, this issue is addressed in more detail as it is crucial for disentangling the relationship between these two activities and their underlying economic intuition. In order to select the model with the best econometric properties among alternatives, a forward selection process was followed. This implies that some variables with no statistically significant coefficients have been included in the final model as they are considered to be an important finding and because such an inclusion does not worsen the overall econometric performance of the model. In the econometric framework adopted in this chapter and presented in the previous section, a plethora of econometric issues arise which may result in inconsistent estimates of the parameters, standard errors and diagnostic statistics of the system of equations (2.1) and (2.3). 4 The largest part of the discussion in this subsection is motivated by the comments of two anonymous referees. -Pg. 28-

40 Table Estimates of EXPINT and RDINT system of equations with alternative approaches EXPINT Equation RDINT Equation Variables Smith and Blundell Coefficient Marginal Estimates Effects GMM Tobit Endogenous Covariates Smith and Blundell GMM OLS Cons * (0,109) * (0.124) * (0.143) 0.051* (0.011) 0.074* (0.018) 0.030* (0.007) EURD (0.085) (0.084) (0.055) (0.010) (0.016) 0.034*** (0.025) NAMD (0.095) (0.097) (0.011)** 0.021** (0.012) 0.015** (0.009) (0.088) TIMED (0.050) (0.055) (0.104) (0.006) (0.007) *** (0.018) SIZE 1.829* 1.807* 1.732* * * (0.605) (0.608) (0.555) (0.972) (0.073) (0.101) (0.260)*** SIZE ** ** ** * 0.157* (0.540) (0.537) (0.568) (0.702) (0.067) (0.040) (0.073) MSEUR * (0.021) * (0.019) * (0.018) *** (0.033) MSNAM *** (0.020) *** (0.022) * (0.017) *** (0.053) LBPRD (0.016) (0.020) ** (0.037) Pg. 29-

41 MNHTD MNLTD CDCGD SERVD ICTD RDINT 7.395* (0.823) 7.444* (0.815) 7.423* (1.052) PRMRG 2.214* 2.126* 2.047* (0.262) (0.233) (0.352) HRFND PRFAB (0.014) (0.020) EXPIN EXPN RDIN * (1.233) 1.883* (0.459) (0.028) * (0.001) * (0.002) ** * (0.002) (0.001) ** *** (0.001) (0.001) * * (0.002) (0.003) ** * (0.001) (0.001) * * (0.007) (0.012) 0.033* 0.028* (0.005) (0.009) * 0.148* (0.018) (0.014) * * (0.007) (0.007) 0.023* 0.029* (0.004) (0.009) (0.005) (0.008) (0.007) (0.015) (0.007) (0.669) 0.053* (0.022) 0.183* (0.055) (0.075) 0.032* (0.015) -Pg. 30-

42 Blundell and Smith Statistics GMM Tobit Endogenous Covariates LL=3, Hansen's J=1,642 LL=1, R =0.654 ρ = AIC = ε 1 σ * ε2 ψ=-7.529* One, two and three asterisks denote statistical significance at 1%, 5% and 10% respectively. 2 X OLS F= Pg. 31-

43 A potential and rather serious drawback of the Blundell-Smith procedure may be that the tobit model assumes that the two decisions of whether and how much to export are affected by the same set of factors. If this is not the case, severe heteroscedasticity, due to tobit misspecification, would be present. To cope with such a possibility, Cragg (1971) introduced a general model which allows the testing of whether the initial decision of EXPINT>0 vs. EXPINT=0 should be separate from the decision of how much to export, given that the firm has decided to become an exporter. Cragg s approach requires the estimation of three models, namely the decision whether to export, the decision of how much to export and the restricted version of the latter. In order to avoid this triple estimation procedure, Finn and Schmidt (1984) devised a Lagrange Multiplier (LM) test, equivalent to Cragg s test, which requires the estimation of the tobit model only (Greene, 2008; pp E ). In this case, the estimated value of the LM statistic is equal to which is smaller than the critical value of X 2 12, = Therefore, the null hypothesis that the two decisions are driven by the same factors is not rejected and thus the tobit model may be considered as an acceptable specification (see Table 2.5). Given the result of the Finn and Schmidt (1984) test, one would expect that the possibility of occurrence of the heteroscedastic Smith and Blundell disturbance term is substantially reduced (Wooldridge, 2002; p. 533). Nevertheless, some additional tests have been conducted, regarding the correlation of the major independent variables of the EXPINT equation and the variance of the Smith and Blundell residuals. In all cases, the hypothesis of the homoscedastic error term cannot be rejected, since the influence of each independent variable on the variance of the Smith and Blundell residuals is statistically non significant. -Pg. 32-

44 Table Econometric Tests Ho Hypothesis Criterion-distribution Criterion Value Degrees of freedom Critical Value (1%) Decision with respire to Ho Cragg Specification Heteroscedasticity The δ coefficient of the cons X e 2 uˆ i k, i i function is statistically equal to zero, where k is: SIZE, PRFAB, PRMRG, LBPRD, MSEUR, MSNAM Normality The residuals of the Smith and Blundell model follow a Normal distribution Davidson and MacKinnon LM (Finn and Schmidt) 2 ~ X v ta, SIZE =0.014 LM= Not Reject PRFAB= PRMRG = LBPRD=0.062 MSEUR =1.249 MSNAM = Not Reject Not Reject Not Reject Not Reject Not Reject Not Reject 2 Jarque-Bera ~ X JB= Not Reject v ta, SIZE =0.527 PRFAB= Not Reject Not Reject -Pg. 33-

45 Endogeneity test PRMRG =1.010 Not Reject The δ coefficient of the LBPRD= Not Reject u cons X e ˆi k, i i MSEUR =1.446 Not Reject function is statistically MSNAM =1.255 Not Reject equal to zero, where k is: SIZE, PRFAB, PRMRG, LBPRD, MSEUR, MSNAM GMM estimators Smith and Blundell Not Reject estimators are consistent within the GMM class estimators Tobit Endogenous Covariates estimators 2 Hausman ~ Xv Not Accept are consistent within the GMM class estimators OLS Estimators are consistent within the GMM class estimators Not Accept -Pg. 34-

46 Regarding the normality assumptions of the u EXP and urd it should be mentioned that Wooldridge (2002, p. 531) proves the Smith and Blundell endogeneity test being valid without any distributional assumptions on the reduced form of the endogenous variable, in this case the RDINT. In order to test the normality assumption, which is employed in the case of EXPINT equation, I have performed a simple Jarque-Berra test based on the residuals of the Smith and Blundell specification. The estimated value of the corresponding JB-statistic does not permit to reject the null hypothesis of the normality of the uexp error term (table 2.5). In order to check for the presence of multicollinearity among the regressors, the Variance Inflation Factor (VIF) matrix has also been computed. Results indicated that no serious multicollinearity problems arise (Table 2.6). Table Variance Inflation Factors for the employed variables Variance Inflator Factor (EXPINT Equation) Variance Inflator Factor (RDINT Equation) Cons Cons EURD MNHTD NAMD MNLTD TD CDCGD SIZE SERVD SIZE ICTD LBPRD EURD RDINT NAMD MSEUR TD MSNAM SIZE PRFAB SIZE PRMRG HRFND EXPIN EXPN RDIN PRMRG Although the adopted econometric strategy tests for possible endogeneity of RDINT with respect to EXPINT, one could point out that additional endogeneity issues may be present regarding other crucial variables of the estimated model. In order to deal -Pg. 35-

47 with such a possibility, the results of a number of simple Davidson and MacKinnon (1993) endogeneity tests are presented in the fifth part of Table 2.5. I have tested for endogeneity regarding the variables of SIZE, PRFAB, PRMRG, LBPRD, MSEUR and MSNAM. In all cases, there is not any statistically significant influence of each one of the above variables on the Smith and Blundell residuals and the thus, the null hypothesis of exogeneity for all the above variables, is not rejected. The most problematic case is the one of the PRFABvariable where the corresponding p-value is just over the 10% level. Finally, some issues concerning the potential use of alternative econometric strategies need to be addressed. One could argue that a dynamic panel GMM-IV estimator should be employed as a better solution in order to cope with the possible endogeneity between RDINTand EXPINT variables. In the context of GMM -IV, the presence of endogeneity is only indirectly detected by the corresponding Hansen s statistic (1982), which is concerned with testing whether the employed instruments are valid. In addition, several issues, related to the available data arise, and in particular, panel data GMM, in the case of small number of time periods and large cross-sections, requires the Arellano Bond (1991) difference GMM estimator (Holtz-Eakin, et al., 1988). This manipulation leads to the inclusion of lagged explanatory variables into the model, which due to the nature of the dataset is not feasible to do since only two time periods are available for all of the used variables except RDINT. Furthermore, to the best of my knowledge, a GMM grounded system estimation of a Tobit and a linear model does not exist, and thus the nature of the EXPINT variable is ignored by the available GMM techniques. In any case, and keeping in mind the restrictions imposed by the employed dataset, i.e., the absence of a dynamic panel and the censored nature of EXPINT variable, a GMM system estimation for EXPINT and RDINT has been implemented, in order to exploit the advantage of being a endogeneity-free problems method with which to juxtapose the results from the Smith and Blundell econometric approach. Furthermore, I have separately estimated the EXPINT equation using a tobit model with endogenous covariates (TEC) and the RDINTequation using OLS. In the final part of Table 2.5 a Hausman (1978) specification test is presented which does not indicate inconsistency in the Smith and Blundell estimators used in this chapter -Pg. 36-

48 compared to the corresponding system GMM estimators. On the contrary, the analogous Hausman tests between GMM and Tobit model, with endogenous covariates for the EXPINT equation and GMM and OLS estimators for the RDINT equation, do not permit to accept the hypothesis that the estimators of TEC and OLS are consistent with the GMM class estimators. The econometric results, especially with respect to the relationship of the crucial variables i.e., EXPINT and RDINT, proved to be robust in several dimensions. More specifically, signs and statistical significance of the corresponding estimated parameters remain unaltered when the used dataset is reduced to only one cross section and when outliers and influential observations are excluded. Overall, it could be argue that the adopted Smith-Blundell modified with IV econometric procedure presents in an explicit manner the endogeneity between RDINTand EXPINT on the one hand, and produces consistent estimators of the crucial variables of the model on the other. Of course, it should be stressed here that this is the case for the specific dataset and one should be very cautious, econometrically speaking, when using the Smith and Blundell estimation procedure since a plethora of econometric tests, regarding the assumptions, should be undertaken and the corresponding necessary corrections may not be always feasible The EXPINT and RDINT relationship In this section the focus of attention is drawn on the relationship between the two endogenously determined RDL firms decisions based on the estimation results presented in Table 2.4. At the bottom of the same Table the estimated value and the corresponding asymptotic standard error of the parameter ψ, designed to test for the presence of endogeneity, are also displayed. According to the performed test, the hypothesis of weak exogeneity of RDINTwith respect to EXPINT is not accepted. In the context of the theoretical framework developed previously, Technology Push and Demand Pull mechanisms as they are represented by RDINTand EXPINT respectively are interrelated constituting a particular operational mode of knowledge generation. -Pg. 37-

49 Further below, it is argued that in this mode of operation there exists a crucial element, that of the latent in econometric terms, knowledge base of the firm. In order to further explore how this mode operates, I have devised an informative sketch as presented in Fig 2.2. Some clarifications are in order here. Arrow lines represent Knowledge flows, which are either due to Technology Push (Top down direction) or Demand Pull (Bottom up direction). RDINTand EXPINT represent observed firm decisions. The grey oblong represents the RDL firm stock of knowledge. Ellipses represent effects related to the functional operation of the firm s knowledge base. Figure 2.2. Mode of operation of Technology Push and Demand Pull mechanisms R&D Efficiency Demand Pull Knowledge Flows RDINT Filtering & Feedback Competitive Advantage KNOWLEDGE BASE EXPINT Technology Push Knowledge Flows Performance Observed Latent Technology Push Flows Demand Pull Flows -Pg. 38-

50 Econometric results indicate that the R&D intensity positively affects a firm s exporting propensity. In terms of Fig. 2.2, which illustrates the theoretical framework already discussed in a previous section of this chapter, the Technology Push mechanism as it is illustrated in R&D intensity, generates knowledge flows, which inflate the knowledge base of the RDL firm. This enlarged knowledge base grounds the RDL firms competitive advantage, which in turn results in superior performance in the global market. Following the bottom up direction of Fig. 2.2, export intensity positively affects R&D intensity, however, not in a linear form. The knowledge flows arising from the Demand Pull source implicitly augment the firm s knowledge base, following a quite different approach. In particular, information flows from foreign markets permit filtering and feedback mechanisms to occur, and thus to induce Localised Technological Change (Atkinson and Stiglitz, 1969). LTC targets at a specific technology set, as it is embedded in the exported goods, and forces the RDL firm to make some targeted technological improvements as the latter are dictated by foreign market forces. This fact alone has two immediate implications. On one hand, the RDL firm is obliged to invest more in R&D resources in order to capitalize the information flows from the foreign markets. On the other hand, the knowledge outcomes of such an investment are prolific, in the sense that the firm has minimized the required search costs (Nelson, 1982) for tracking down the optimum technique, among all the available in the known technological space. According to von Tunzelmann (1995), the optimization of this search process, with respect to technical change, can be perceived as an increase of the R&D efficiency of the firm 5. As a direct result, an ensuing question arises concerning the extent of the returns from learning by exporting (Evenson and Westphal, 1995), with respect to R&D intensity. In order to examine such an issue more thoroughly, these two variables have been plotted against each other in Fig. 2.3, holding all other statistically significant variables constant at their sample means according to the following equation: 5 At this point it should be mentioned that the term efficiency is not based on the strict definition of input-output ratio (Farell, 1957), but we rather follow the theoretical argumentation introduced by Nelson (1982) that R&D is a problem solving activity that entails a continuous search process in order to acquire the optimum technique among many alternatives. -Pg. 39-

51 RDINT University of Patras RDINT= * EXPN * EXPN1 2 (2.10) It is profoundly evident that R&D and export intensity exhibit mixed patterns of interrelationship. In more detail, it could be argued that decreasing marginal learning benefits exist with respect to knowledge flows arising from exporting performance. To put it more explicitly, even though the increase of export propensity yields positive learning benefits, which in turn fuel the corresponding increase of RDINT, the latter are continuously decreasing, in terms of the RDL firms knowledge base augmentation. Figure 2.3. The decreasing marginal learning benefits with respect to knowledge flows Reality Virtual Reality Zone I Zone II EXPINT1 Two main areas are identified, namely the Reality and the Virtual Reality area. The area of Reality essentially is where firms operate since it is not possible for any firm to have an export intensity index greater than one. In this area equation (2.10) is characterized by strict monotonicity and concavity. Economically speaking, R&D intensity, as a measure of RDC firms knowledge creation, is ruled by decreasing returns to their international orientation. On the other hand, considering R&D intensity as a measure of the inefficiency of the resources devoted to R&D activities, it seems that higher export intensity leads to a deterioration in the efficiency of the R&D activities but at a decreasing rate. -Pg. 40-

52 Moving to the Virtual Reality area where the export intensity index exceeds the value of one, two zones can be identified. Zone I may be regarded as an extrapolation of the reality area. Positive monotonicity and concavity continue to be present. In this zone the RDC firm would continue to increase its R&D intensity due to its degree of internationalization though at a decreasing rate. Inefficiency of the R&D activities continues to be exacerbated but the rate is decreasing even more than the corresponding trend of the reality area. Zone II depicts the imaginary situation of a new era arising from a drastic structural change, where the RDC firm has reached the point of maximum inefficiency of the resources devoted to augmenting its stock of knowledge. For values of export intensity greater than the corresponding one at this point, the RDC firm achieves such an exploitation of the knowledge flows derived from exporting activities that it permits the reduction of the intensity of its own R&D activities. In terms of orthodox economic theory one could argue that Zone II of the Virtual Reality area represents a technology where the knowledge inflows from exporting, compared to the corresponding level of R&D activities, are close substitutes. Some kinds of filtering mechanisms of the R&D activities through knowledge acquired by the RDC firm s internationalization are in operation here. A more thorough representation of the theoretical argumentation is presented in Fig. 2.4 in which most of the quantitative and technical detail has been eliminated. It could be argued that the early stages of export engagement, in which an RDL firm engages in exporting activities, have a greater impact, via the channels of filtering and feedback information, on expanding the firm s knowledge base. However, these channels are limited in their contribution and the more engaged one becomes the less are the learning benefits from such a particular activity. Based on the anticipated effects of LTC on RDL firms knowledge base, it can rather safely be assumed that the technical change induced upon a specific set of techniques is not sufficient to evoke generalized technical change given the specific characteristics of RDL firms knowledge base. For the RDL firm to pursue a more generalized technical change and draw new trajectories (Dosi, 1988;1997), i.e., in order to sustain its role of leadership in R&D investment, it needs to rely on additional forms of internationalization such as FDI, forms of alliances, etc. -Pg. 41-

53 Figure 2.4. Illustration of the interrelationship between R&D and export intensity of the RDC firms Knowledge Creation Reality Area Positive Monotonicity. DRS of knowledge creation w.r.t. the internationalization process Domestic vs. Global Markets Positive Monotonicity Decreasing Returns to knowledge creation w.r.t. internationalization become more acute Full Internationalization Negative Monotonicity Knowledge originating from R&D is substituted for knowledge from internationalization Virtual Reality or the New Era Area 1 Negative Monotonicity Positive Monotonicity EXPINT Inefficiency continues to increase but the rate is extremely low Inefficiency starts declining. Internationalization as a filtering process permits an ultra-efficient usage of the resources devoted to R&D Negative Monotonicity. Inefficiency increases at a decreasing rate Inefficiency of R&D activity The determinants of EXPINT and RDINT -Pg. 42-

54 Besides the main relationship between R&D and export intensity, which was analytically presented and discussed above, each set of control variables for each equation of the system will be the interest of this section. Considering the EXPINT equation, the explanatory variables capturing firm size, labour productivity, profit marging, profitability and location which are dictated by the Market selection hypothesis are incorporated. Accordingly, the destination of exporting activities as an explanatory variable in order to test for the Learning by Exporting hypothesis has been included. The role of SIZE is investigated since the relevant literature has identified it as being considerably important in explaining export intensity variation. In light of the above, significant quadratic effects of parabolic type are observed. This finding is in accordance with the empirical evidence provided by several authors (Schlegelmilch and Crook, 1988; Kumar and Siddhartan, 1994; Wagner, 1995; Wakelin, 1998). Wagner (1995) argued that firm size advantages are present up to a certain threshold due to coordination costs and bureaucratic issues, while Schlegelmilch and Crook (1988) argued that this nonlinearity is due to the fact that above a certain size large firms find it more efficient to proceed to FDI rather than exporting. These explanations can be complementary rather than contradictory. More specifically, it can be argued that these firms are in a position to absorb and therefore, exploit export oriented demand-pull knowledge flows until a certain size threshold. After they reach a certain size alternative sources of demand-pull knowledge flows may become more attractive, i.e. FDI, intra-firm trade. Such being the case the interrelated decisions to invest on R&D and penetrate foreign markets become more complicated. The PRMRG variable which constitutes an indicator of the market power a firm may possess and exert, when included in the EXPINT equation, is positive and statistically significant. According to this empirical finding the argument that international trade contributes to the increase of the welfare due to international competitive forces does not seem to apply in this case. Market power having a positive influence on export intensity suggests that the competition in foreign markets does not resemble the perfect competition model but on the contrary the oligopolistic model is more suitable to explain export behavior of firms (Krugman, 1979). As far as the LBPRD, PRFAB, -Pg. 43-

55 EURD, NAMD, and the time TIMED are concerned, they were found to exert no statistically significant influence on the extent of Demand Pull Knowledge flows. Moving on to the Demand pull knowledge flows originated from Learning by Exporting Hypothesis it is not worthless to mention that they are not homogenous but instead may be dependent on their geographical origination (UNCTAD, 2005). The results indicate that both MSEUR and MSNAM are negative and statistically significant. The negative influence of the two variables has to be interpreted in relation to the excluded variable, MSROW. The RDL firms that mainly orient their exporting activities towards Europe and N. America, exhibit overall, a lower degree of export commitment in relation to those that mainly participate in RoW markets. Therefore, it could be argued that any demand pull knowledge flows deriving from European and N. American markets are expected to be relatively small and consequently any filtering and feedback mechanisms inducing LTC will be rather limited in scope and extent. The discussion about the determinants of the R&D intensity equation, i.e., the technology push knowledge flows, concerns the Schumpeterian hypotheses regarding firm size and industrial market structure (Schumpeter 1945), and also the three conditions that define the dynamic evolution of the industries, namely appropriability, technological opportunities, and cumulativeness (Malerba and Orsenigo, 1993; 1996). Regarding the role of size in determining the extent of R&D intensity, the econometric results reveal statistically significant quadratic effects of the hyperbolic type. Bearing in mind that these firms have already spent such a considerable amount of resources as to be characterized as R&D leaders, the interpretation of such a non linear relationship draws, on one hand, from the fact that the production and distribution of new innovative products or processes is characterized by sharp economies of scale (Shy, 2004, p.53) and on the other hand, from the seminal work of Cohen and Klepper (1996) and their notion of cost spreading advantage. More specifically, and with respect to the declining part of the U-shaped relationship between firm size and R&D intensity, it could be argued that these firms have invested a significant amount of resources in the development of a new technology as the latter is embedded in the new improved products or processes. The fruits of such -Pg. 44-

56 an investment are capitalized via the disproportionately short-term increase of sales and therefore, increasing turnover spreads the cost of invention. However, this shortterm supranormal profits cannot be sustained diachronically. Eventually, for the RDL firms to sustain their competitive advantage and thus their technological leadership, they have to evolve the existing technological paradigm by investing proportionally more resources into the development of a new process or product than their size changes. This process in turn, creates additional technology push knowledge flows. The Schumpeterian hypothesis concerned with the effect of industry concentration, with respect to innovation, was proxied with the Herfindahl index HRFND. According to the econometric results, increased concentration of an industry strengthens the R&D efforts of individual firms. Following Levin et al. (1985) industrial concentration as a determinant of R&D intensity may lose its interpretive power when one accounts for inter-industry systematic differences, such as appropriability conditions, technological opportunities, and cumulativeness of knowledge. Furthermore, and in order to control the presence of effects due to appropriability conditions, the RDL firms profit margin PRMRG is included as an indicator (Kamien and Schwartz, 1982; p. 28). Empirical results indicate that this index exerts negative and statistically significant influence. Appropriability conditions, as they are captured by the RDL firms profit margin, seem to provide the RDL firms with the necessary safety, in order to crop the fruits of their investments on one hand, and not to decide hastily when and how much to invest in their next project, at least temporarily, on the other (Winter, 2006). At this point one could reasonably argue that the firm s profit margin can only partially depict appropriability conditions especially with respect to RDL firms. Unfortunately, relevant information regarding detailed appropriability policies is not present in this research. In general, however, such information is quite difficult to be depicted in innovation surveys and firm s financial statements. The next influential component of technology push knowledge flows is related to technological opportunities. These effects are tested through industry specific -Pg. 45-

57 variables 6 (Scherer, 1967) as they are defined in Table 2.2. Results of the estimated coefficients show that all industry dummies are statistically significant and negatively affect R&D intensity. This is not a surprising result if one considers that the omitted industry dummy variable, which is used as a reference point, is the ICTHW variable (i.e., the hardware branch of ICTs). The third and final element, cumulativeness, which determines the propensity of R&D activities, is approximated with the RDIN1 variable, which according to the empirical estimations is positive and statistically significant. This finding comes to support the role of path dependency (David, 1985) on R&D activities as it has been recorded in Malerba and Orsenigo (1993). Finally, it should be noted that the variable NAMD is positive and statistically significant, confirming the superiority in terms of knowledge creation flows of the North American firms Conclusions Although RDL firms may be perceived as important carriers of technological change, with respect to the evolution of technological paradigms and trajectories, theoretical considerations are lacking, in terms of a unifying framework, about the interaction of sources and mechanisms with which knowledge is generated. In the context of this chapter a data driven research process is followed, which allows to construct a theoretical framework of the RDL firms knowledge base creation. Emphasis is given on the conspicuous relationship between Technology Push and Demand Pull as mechanisms of knowledge creation. 6 The R&D Scoreboard classifies R&D leaders into 39 industries. While, such a classification reduces problems related to RDL firms technological heterogeneity, various issues related to the subsequent econometric handling arise. In line with the above, we have put our efforts towards reclassifying the existing industrial distribution in wider sectors. For this purpose Pavitt s taxonomy (1984) would be an ideal alternative. Unfortunately, we are not in a position to apply such taxonomy to our sample due to the fact that all firms in this sample operate on the cutting edge of technological frontier. For this reason, we followed a classification method of six wide sectors, allowing to retain RDL firms most important technological characteristics (Bos et al., 2009) and at the same time minimizing the potential econometric problems that may have arisen otherwise. The matching of the original with the finally employed industrial classifications, are provided upon request and are not presented here due to space limitations. -Pg. 46-

58 In this context, Technology Push knowledge flows are captured by R&D intensity and Demand Pull knowledge flows by degree of penetration to foreign markets respectively. In the process of building the theoretical framework, central role has been assigned to the endogenous relationship between export intensity and R&D intensity of the RDL firms. Drawing from what has been recorded in the relevant literature and employing an appropriate econometric methodology it could be argued that in the process of augmenting the RDL firms knowledge base, Technology Push and Demand Pull knowledge flows are operating in conjunction. However, I find that they are not of equal importance. More specifically, the very nature of RDL firms indicates that Technology Push knowledge flows are crucial for them in order to maintain their leadership position through the exploitation of technological opportunities. In terms of access to foreign markets, the RDL firms are provided with knowledge inputs, which have mainly feedback and filtering character, and are thus, rather limited in scope, capable of inducing Localised Technological Change. More specifically, decreasing marginal learning benefits, with respect to knowledge flows arising from exporting performance, are present. In other words, for the firm to pursue a more generalized technical change and draw new trajectories, i.e., in order to sustain its role of leadership in R&D investment, it needs to rely mainly on its own ability to produce new technological knowledge and/or in additional forms of internationalization. Exploring deeper each knowledge generation mechanism, I conclude that the Technology Push side is ruled by the conditions of industries technological evolution, i.e., cumulativeness, appropriability and technological opportunities as they have been recorded in the relevant literature. No exception is made regarding the RDL firms. As far as the validity of the Schumpeterian conditions is concerned, the picture is not so clear. However, the reader should keep in mind the obscurity, which goes hand in hand with the approximation of the Schumpeterian hypotheses, in terms of measurements issues. Regarding the Demand Pull knowledge generation mechanism, a mixed pattern of influence is outlined, sourcing from the market selection and learning by exporting paradigms. RDL firms individual performance characteristics, together with the geographical distribution of targeted foreign markets are the main channels through which the knowledge flows from exporting activities. -Pg. 47-

59 The theoretical framework identified in this data driven research process produces some further research directions. More specifically, it would be quite interesting for one to explore the role of exports geographical distribution on the RDL firms knowledge base observed heterogeneity. A consequent question is related to the unobserved RDL firms heterogeneous knowledge bases, and is directly linked with the causal ambiguity principle. Finally, it should not be ignored that other forms of internationalization may behave in the same pattern when it comes to Demand Pull Knowledge flows creation. -Pg. 48-

60 References Acs, Z. and Audretsch, D. (1987) Innovation Market Structure and Firm Size, Review of Economics and Statistics, 69, Aghion, P. and P. Howitt (1998), Endogenous Growth Theory, MIT Press, Cambridge MA. Antonelli, C. (1998) The Dynamics of Localized Technological Changes. The Interaction between Factor Costs Inducement, Demand Pull and Schumpeterian Rivalry, Economics of Innovation and New Technology, 6, Arellano, M. and Bond, S. (1991), Some tests of specification for paneldata: Monte Carlo evidence and an application to employment equations, Review of Economic Studies, 58, Atkinson and Stiglitz (1969), A new view of Technological Change, Economic Journal, 79, Bhattacharya, M. and H. Bloch (2004), Determinants of Innovation, Small Business Economics, 22, BIS, (2007) The 2007 R&D Scoreboard: Report for company data, Department of Business Universities and Skills, UK BIS, (2008) The 2008 R&D Scoreboard: Report for company data, Department of Business Universities and Skills, UK Bleaney, M. and K. Wakelin (2002), Efficiency Innovation and Exports, Oxford Bulletin of Economics and Statistics, 64, Bos, J.W.B., C. Oikonomidou and M. Koetter (2010), Technology clubs, R&D and growth patterns: Evidence from EU manufacturing, European Economic Review, 54, Bougheas, S. (2004), External vs. Internal financing of R&D, Small Business Economics, 22, Pg. 49-

61 Castellacci, F. (2007), Technological regimes and sectoral differences in productivity growth, Industrial and Corporate Change, 16, Caves, R. and Porter, M. (1977), From Entry Barriers to Mobility Barriers: Conjectural Decisions and Contrived Deterrence to New Competition Quarterly Journal of Economics, 91, Chandler, A. (1986) The evolution of Global Modern Competition, in Porter M. (ed) Competition in Global Industries, Harvard University Press: Boston Massachussets. Chandler, A. (1994) Scale and Scope: the dynamics of Industrial Capitalism, Harvard University Press: Boston Massachussets. Clerides, S., S. Lach and J., Tybout (1998), Is Learning by Exporting Important? Micro-dynamic Evidence from Colombia, Mexico, and Morocco, Quarterly Journal of Economics, 113, Cohen, W. (1995) Empirical Studies of Innovative Activity, in Stoneman P. (ed.) Handbook of the Economics of Innovation and Technological Change, Blackwell: Oxford. Cohen, W. and Klepper, S. (1992) The Anatomy of Industry R&D Intensity Distributions, American Economic Review, 82, p , Cohen, W. M. and Klepper, S. (1996) A Reprise of Size and R&D, Economic Journal, 106, Cohen, W. and Levinthal, D. (1989), Innovation and Learning: the Two Faces of R&D, The Economic Journal, 99, Cragg, J. (1971) Some statistical models for limited dependent variables with application to the demand for durable goods, Econometrica, 39, David, P. (1975) Technical Choice, Innovation and Economic Growth. Cambridge University Press: Cambridge. -Pg. 50-

62 David, P. (1985) Clio and the Economics of QWERTY, American Economic Review Proceedings, 75, Davidson, R. and MacKinnon J. (1993) Estimation and Inference in Econometrics,, Oxford University Press: New York. Dierickx, I. and Cool, K. (1989) Asset stock accumulation and sustainability of competitive advantage, Management Science, 35, Dollar, D. (1986), Technological Innovation, Capital Mobility, and the Product Cycle in North-South Trade, American Economic Review, 76, Dosi, G. (1988) Source, Procedure and Microeconomic Effect of Innovation, Journal of Economic Literature, 26, Dosi, G., (1997) Opportunities, Incentives and the Collective Patterns of Technological Change, Economic Journal, 107, Ernst, D. and Kim, L. (2002) Global production networks, knowledge diffusion, and local capability formation, Research Policy, 31, Evenson, R. and L. Westphal, (1995), Technological Change and Technology Strategy, In: Behrman, J. ans T.N. Srinivasan (Eds), Handbook of Development Economics, Vol. 3, North-Holland, Amsterdam, pp Farrell, M. J. (1957), The Measurement of Productive Efficiency, Journal of the Royal Statistical Society, Series A (General) 120, Fin, T. and Schmidt, P. (1984) A test of the tobit specification against an alternative suggested by Cragg, Review of Economics and Statistics 66, Fryges, H. (2009) The Export Growth Relationship: Estimating a Dose Response Function, Applied Economics Letters, 16, Geroski, P. Van Reenen, J. and Walters C. (1997), How persistently do firms innovate?, Research Policy, 28, Pg. 51-

63 Grant, R.M. (1996), Toward a Knowledge-based Theory of the Firm, Strategic Management Journal, 17 (Winter Special Issue), Greene, W.H. (2002), Econometric Analysis, Prentice Hall: New Jersey. Greene, W.H. (2007) Limdep v.9. Econometric Modeling Guide, vol. 2, E New York: Econometric Software. Grossman, G. and E. Helpman (1989), Product Development and International Trade, Journal of Political Economy, 97, Grossman, G. and E. Helpman (1990), Comparative Advantage and Long-run Growth, American Economic Review, 80, Grossman, G. and E. Helpman (1991), Endogenous Product Cycles, Economic Journal, 101, Hansen, L. (1982) Large sample properties of generalised method of moment estimators, Econometrica, 50, Harris, R. and Q. Li, (2008), Exporting, R&D, and Absorptive Capacity in UK Establishments, Oxford Economic Papers, 69, Hausman, J. (1978) Specification tests in Econometrics, Econometrica, 46, Heckman, J. (1978), Dummy Endogenous Variables in a Simultaneous Equation System, Econometrica, 46, Hughes, K. (1986), Exports and Innovation, European Economic Review, 30, Holtz-Eakin, D. Newey, W. and Rosen H. (1988) Estimating Vector Autoregressions with Panel Data, Econometrica, 56, p Ito K. and Lechevalier S. (2010) Why some firms persistently out-perform others: investigating the interactions between innovation and exporting strategies, Industrial and Corporate Change, 19, Pg. 52-

64 Johanson J. and E. Vahlne, (1977), The Internationalization Process of the Firm: A Model of Knowledge Development and Increasing Foreign Market Commitments, Journal of International Business Studies, 8, Kamien, M. I. and Schwartz, N. L. (1982) Market Structure and Innovation, Cambridge: Cambridge University Press. Keesing, D. (1967), The Impact of Research and Development on United States Trade, Journal of Political Economy, 75, Kline, S. and Rosenberg, N. (1986), An Overview of Innovation in: R. Landau and N. Rosenberg (Eds), The Positive Sum Strategy: Harnessing Technology for Economic Growth, National Academy Press:Washington, DC Kogut, B. and U. Zander, (1992), Knowledge of the Firm, Combinative Capabilities, and the Replication of Technology, Organization Science, 3, Krugman, P. (1979), A Model of Innovation, Technology Transfer, and the World Distribution of Income, Journal of Political Economy, 87, Kumar, N. and N. Siddharthan (1994), Technology, Firm Size and Export Behaviour in Developing Countries: The Case of Indian Enterprises, Journal of Development Studies, 31, Lachenmaier, S. and L. Woessmann, (2006), Does Innovation Cause Exports? Evidence from Exogenous Innovation Impulses and Obstacles using German Micro Data, Oxford Economic Papers, 58, Lall, S. and R. Kumar (1981), Firm-level Export Performance in an Inward-looking Economy: The Indian Engineering Industry, World Development, 9, Lefebvre, E., M. Bourgault and L. Lefebvre (1998), R&D Related Capabilities as Determinants of Export Performance, Small Business Economics, 10, Levin, R., Cohen, W. and Mowery D. (1985) R & D Appropriability, Opportunity, and Market Structure: New Evidence on Some Schumpeterian Hypotheses, American Economic Review, 75, Pg. 53-

65 Nelson, R. (1982), The Role of Knowledge in R&D Efficiency, Quarterly Journal of Economics, 97, Malerba, F. and Orsenigo, L. (1993) Technological Regimes and Firm Behaviour, Industrial and Corporate Change, 2, Malerba, F. and Orsenigo, L. (1996) The Dynamics and Evolution of Industries, Industrial and Corporate Change, 5, Mowery, D. and Rosenberg, N. (1979) The influence of market demand upon innovation: a critical review of some recent empirical studies, Research Policy, 8, Mowery, D. and Rosenberg, N. (1989) Technology and the pursuit of economic growth, Cambridge University Press: Cambridge Nonaka, I. (1994), A Dynamic Theory of Organizational Knowledge Creation, Organization Science, 5, Pavitt, K. (1984) Sectoral Patterns of Technical Change: Towards a Taxonomy and a Theory, Research Policy, 3, Penrose, E. (1959), The Theory of the Growth of the Firm, Wiley: New York. Peters, B. (2009), Persistence of innovation: stylised facts and panel data evidence, Journal of Technology Transfer, 34, Piva, M. - Vivarelli, M. (2007), Is Demand-Pulled Innovation Equally Important in Different Groups of Firms?, Cambridge Journal of Economics, 31, Pla-Barber, J. and J. Alegre (2007), Analysing the Link between Export Intensity, Innovation and Firm Size in a Science-based Industry, International Business Review, 16, Posner, M. (1961), International Trade and Technical Change, Oxford Economic Papers, 13, Roberts, M. and Tybout, J. (1997) The decision to export in Colombia: an empirical model of entry with sunk costs, American Economic Review, 87, Pg. 54-

66 Rosenberg, N. (1990) Why do firms do basic research (with their own money)?, Research Policy, 19, Rumelt, R. (1984). 'Towards a strategic theory of the firm'. In R. Lamb (ed.) Competitive Strategic Managemen,. Prentice-Hall: Englewood Cliffs, NJ, Ruttan, V. W. (1997) Induced Innovation, Evolutionary Theory and Path Dependence: Sources of Technical Change, Economic Journal, 107, Saggi, K. (2002), Trade, Foreign Direct Investment and International Technology Transfer: a Survey, World Bank Observer, 17, Scherer, F. (1965), Firm Size, Market Structure, Opportunity, and the Output of Patented Inventions, American Economic Review, 55, Scherer, F. M. (1982) Demand-pull and Technological Invention: Schmookler Revisited, Journal of Industrial Economics, 30, Schlegelmilch, B. and J. Crook (1988), Firm-Level Determinants of Export Intensity, Managerial and Decision Economics, 9, Schmookler, J. (1966) Invention and Economic Growth, Cambridge (Mass.): Harvard University Press. Schumpeter, J. (1942), Capitalism, Socialism and Democracy, Harper and Row: New York. Segerstrom, P., T. Anant and E. Dinopoulos (1990), A Schumpeterian Model of the Product Life Cycle, American Economic Review, 80, Shy, O. (2004) The Economics of Network Industries, Cambridge University Press: Cambridge Smith, R.and R. Blundell (1986), An Endogeneity Test for a Simultaneous Equation Tobit Model with an Application to Labour Supply, Econometrica, 54, Pg. 55-

67 Smith, V., E. Madsen, and Dilling-Hansen, M. (2002), Do R&D Investments Affect Export Performance? WP 2002/4, The Danish Institute for Studies in Research and Research Policy, Aarhus. Spender, J. (1996), Making Knowledge the Basis of a Dynamic Theory of the Firm, Strategic Management Journal, 17 (Winter Special Issue), Teece, D. and Pisano, G. (1994) The Dynamic Capabilities of Firms: an Introduction, Industrial and Corporate Change, 3, UNCTAD, (2005), Determinants of Export Performance, Developing Countries in International Trade, Trade and Development Index. van Dijk, M. (2002), The Determinants of Export Performance in Developing Countries: The Case of Indonesian Manufacturing, Working paper for Eindhoven Centre for Innovation Studies (Ecis). Vernon, R. (1966), International Investment and International Trade in the Product Cycle, Quarterly Journal of Economics, 80, von Tunzelmann, N. (1995) Technology and Industrial Progress: the foundations of economic growth, Edward Elgar: Cheltenham von Tunzelmann, N. (2000) Technology generation, technology use and economic growth, European Review of Economic History, 4, Wagner, J. (1995), Exports, Firm Size, and Firm Dynamics, Small Business Economics, 7, Wagner, J. (2007) Exports and Productivity: A Survey of the Evidence from Firmlevel Data, The World Economy, 30, p Wakelin, K. (1998), Innovation and Export Behaviour at the Firm Level, Research Policy, 26, Wilmore, L. (1992), Transnationals and Foreign Trade: Evidence from Brazil, Journal of Development Studies, 28, Pg. 56-

68 Winter, S. (2006), The logic of appropriability: From Schumpeter to Arrow to Teece, Research Policy, 35, Wooldridge, J. (2002), Econometric Analysis of Cross Section and Panel Data, MIT Press: Cambridge MA. Young, A. (1991), Learning by Doing and the Dynamic Effects of International Trade, Quarterly Journal of Economics, 106, Pg. 57-

69 Chapter 3: The profile of R&D activities of Greek Manufacturing Firms and the role of Internationalization. Primary Results of a survey 3.1. Introduction This chapter presents the basic findings of the field research that was conducted among the Greek manufacturing firms that have been engaged in R&D activities during The field research was carried out from members of the, University of Patras with Prof. Kostas Tsekouras being the scientific supervisor. In addition, significant contribution in the implementation of the field research came from the intra-university research network Economics and Management of Knowledge and Innovation (EMaKI) 7. It should be mentioned that recent firm level information regarding innovation and R&D activities of Greek firms are not available either by the European Statistical authority (EUROSTAT) or by the General Secretary for Research and Technology (GSRT). Hence, in order for the gathered information to be comparable with other European surveys on Innovation and in particular with Community Innovation Survey (CIS), the design of the questionnaire was primarily based on the CIS standards. It should be mentioned that regarding the data (i) on R&D expenditures and other financial indicators from the electronic database for the period and (iii) from the field research, provide comprehensive and up-to-date information about both R&D and exporting activities at the firm level for the entire Greek Manufacturing sector. As far as Greece is 7 I would like to thank for the important involvement in carrying out this project Manolis Tzagarakis and Konstantinos Kounetas, both Lecturers at the Department of Economics of the University of Patras. Important help in the realization of the survey have also offered, by M.Sc students of the Departent of Economics, C. Ioannidou, A. Kougias and V. Vorissi, of University of Patras. -Pg. 58-

70 concerned there is not any information available about neither innovation and export performance of Greek firms. This also applies for the European level since firm level information regarding both R&D and exporting activities is not available for the same set of firms The population of interest and the sample of the survey The identification of the reference population was made based on published accounts of the Greek Manufacturing firms for the period More specifically, and for the ten-year reference period, the electronic database i-mentor has been employed in order to locate nationwide Greek manufacturing firms that have included in their published financial accounts expenditures on R&D either as part of their assets and/or as part of their income statements. A number of issues have been raised due to the firms transition from National Accounting Standards (NAS) to the International Accounting Standards (IAS) which has been treated by the research team according to accounting standards. Details on how these issues have been handled are provided further below. The identified population included 764 manufacturing firms. With respect to the reliability of the analysis of the population and specifically with respect to ensure the time-persistence (Peters, 2009) of firms R&D activities, two participation criteria have been established i. t 2 ii. where: T T t 2 N t : The number of years that the firm is active in R&D activities T : The extend of the time period which is defined as the difference between the end of the period under investigation that is 2010, and the year that the firm first appeared R&D expenditures in its published financial accounts. N : The number of intermissions that the firm may present in its R&D activities that are depicted in its financial accounts -Pg. 59-

71 Twenty four firms did not satisfy the above participation criteria (3.35% in the total population) thus, leaving a population of 740 firms. It is not worthless to mention that the above criteria were not applied for 48 business entities (5.13% in the total population) which first appeared in the population as R&D active in 2007 or 2008, since it was not feasible to control for time persistence. Eventually, these 48 firms have been included in the population under investigation. The field research was carried out during the second half of Members of the research team have come in contact with all the firms included in the population. Eventually, 316 firms replied reaching a response rate of 45%. Table 3.1 summarizes the basic information presented above. Table 3. 1 Basic characteristics of the identified population and the obtained sample Population of Greek Manufacturing firms presenting R&D expenditures during the period Firms with time window that does not satisfy the first criterion of R&D persistence t 2 Firms with time window that does not satisfy the second criterion of R&D persistence T T t 2 N Number of Firms Percentage (%) ,0 9 (1,22) 15 (2,02) Contacted firms ,76 Firms participated in the survey-response rate ,71 8 Firms with usable information ,67 The sample of 316 firms was weighted based on two characteristics: The industrial distribution following NACE. Considering the specificities of the industrial distribution of the Greek R&D manufacturing (GRD) firms and the particularly small absolute number of firms active in R&D that belonged to 8 It is based on the number of firms that have been come to contact (740) for the purposes of the survey. The corresponding percentage based on the overall population is 41.83%. -Pg. 60-

72 some two-digit industrial classifications, small modifications were made in forming a modified industrial distribution. Hence, two criteria were taken into account for industrial inclusion; the technological relevance between two or more industries and the very small number of firms (<3) in some two digit NACE sectors. In Table 3.2. the merged industries are presented along with the distribution of firms in the identified population and the sample. The firm s size as it has been approximated by the number of employees of each firm and following the definition of EU (2003). More specifically Large firms with more than 250 employees Medium firms with more than 51 and less than 250 employees Small firms with less than 50 employees. Table New Industrial distribution in the population and the sample. Two-digit Number of Number of NACE firms in the Percentage firms in the Percentage population (%) sample (%) % % % % % % % % % % % % % % % % % % % % Total ,0% ,0% Basic statistical tests demonstrated that the sample collected presents the same characteristics as the identified population with respect to firm size and industrial distribution. Industries such as Food and Beverage, non-metallic minerals and -Pg. 61-

73 Chemicals (including Pharmaceuricals) present the most significant engagement in R&D activities within the population of GRD firms. It is interesting to note that the industrial distribution from the perspective of technological intensity (OECD, 1997; 2003) where firms that are characterized as high-tech are expected to account for a great portion of a nation s R&D active manufacturing firms. The case of the Greek manufacturing sector does not sketch the same picture since the R&D active firms belonging to the high-tech sectors account approximately only for the 6% of the total population of the GRD firms. It is not worthless to mention that according to Kumbhakar et al. (2012) among the leader R&D European firms, the corresponding high-tech sectors account for 64% of the total sample of firms. This fact alone reveals on one hand, the technological idiosyncrasies, or technology lag in terms of technology structure of the Greek Manufacturing Sector and on the other hand, the need to carefully apply the sectoral distributions (Kirner et al., 2009) when crosscountry studies are conducted. It is also worth mentioning that with the Greek manufacturing spectrum declining sectors or sectors that present low technological opportunities (NACE 17, 18, 19, 25, 34, 35) have a small participation in the identified population and therefore in the obtained sample. Table 3.3 presents the industrial distribution which is based on the international literature (OECD, 2003) on industrial classification based on technological intensity, where industrial sectors are classified as High-tech (HT), Medium-high tech (MHT), Medium-low tech (MLT) and Low-tech (LT). Table 3.3. OECD Industrial Classification based on technological intensity NACE Industry Technological twodigit Intensity code 15 Manufacture of food products and beverages LT 16 Manufacture of tobacco products LT 17 Manufacture of textiles and textile products LT 18 Manufacture of wearing apparel; dressing and dyeing of fur LT 19 Tanning and dressing of leather; manufacture of luggage, handbags, LT saddlery, harness and footwear 20 Manufacture of wood and of products of wood and cork, except LT -Pg. 62-

74 furniture; manufacture of articles of straw and plaiting materials 21 Manufacture of pulp, paper and paper products; publishing and printing LT 22 Publishing, printing and reproduction of recorded media LT 23 Manufacture of coke, refined petroleum products and nuclear fuel MLT 24 Manufacture of chemicals, chemical products and man-made fibres MHT 25 Manufacture of rubber and plastic products MLT 26 Manufacture of other non-metallic mineral products MLT 27 Manufacture of basic metals MLT 28 Manufacture of fabricated metal products, except machinery and MLT equipment 29 Manufacture of machinery and equipment n.e.c MHT 30 Manufacture of office machinery and computers HT 31 Manufacture of electrical machinery and apparatus n.e.c MHT 32 Manufacture of radio, television and communication equipment and HT apparatus 33 Manufacture of medical, precision and optical instruments, watches and HT clocks 34 Manufacture of motor vehicles, trailers and semi-trailers MHT 35 Manufacture of other transport equipment MLT 36 Manufacture of furniture; manufacturing n.e.c. LT Combining the industrial distribution adopted in the context of the present PhD Thesis with the classification of industries in High-tech (HT), Medium-high tech (MHT), Medium-Low tech (MLT) and Low tech (LT) the picture emerging from the sample of GRD firms is presented in Figure 3.1 below. Figure 3.1. Distribution of GRD firms in sample according to their technological intensity -Pg. 63-

75 Distribution of GRD firms according to their technological intensity 40.7% 10.4% 29.6% 19.2% HT MHT MLT LT With respect to the GRD firms size distribution, it is interesting to note that the proportion of the large firms in the population of Greek Manufacturing firms, irrespectively their engagement in R&D activities, is differentiated from the identified population of R&D active firms. In particular, large firms take over a bigger percentage within the identified R&D active firms population, a finding that is in accordance with the relevant literature where there is a positive relationship between firm size and R&D engagement (Cohen and Klepper, 1996). A picture of the corresponding firm size distributions in the identified population of the GRD firms and the firms included in the sample is presented in Figure 3.2. Figure 3.2. Size distribution of GRD firms in the population and the sample Distribution of firm size in the identified population and the obtained sample 60% 40% 20% 0% Small Medium Large Population Sample 3.3. The research tool and field research -Pg. 64-

76 As it has already been mentioned the survey was carried out during the second half of All firms identified in the sample were called to complete a specially designed questionnaire 9 which is composed of four sections. The first section was interested in depicting the general economic environment within which the GRD firm operated. The second section involved questions regarding the GRD firms exporting activities. More specifically, departing from the key question of export decision, a series of following questions regarding first year of export, export intensity, export volume growth, means of exporting, barriers to exporting as well as other means of internationalization were included. The third section entailed a series of thorough questions surrounding the Greek firms R&D activities. In particular, it involved, among others, questions regarding the internal organization of R&D activities, as well as information about the innovative outcomes of these activities, along with potential barriers encountered in the process of conducting R&D. The fourth and final section of the questionnaire involved information about domestic and international cooperation in the context of the Greek firm s R&D activities The knowledge capital Despite the fact that the degree of knowledge production is commonly approximated with the intensity of R&D activities defined as the ratio of R&D expenditures to firm s annual sales, the most informative way of measuring such a production is to employ the firm s knowledge capital which encompasses a dynamic scope of knowledge production contrary to the static one provided by R&D intensity (Dierickx and Cool, 1989; Goto and Suzuki, 1989, Mairesse and Mohnen, 2010; Kumbhakar et al 2012). However, the availability of R&D expenditures on a yearly basis and for a quite significant period of time is often an insurmountable barrier. Based on firms annual published financial accounts it has been made possible to construct their knowledge stock for the period Further below the handling of the 9 The e-link to the questionnaire is: (in greek) and is also given in the Appendix of this chapter. -Pg. 65-

77 information from the financial accounts along with the adopted methodology for the construction of the GRD firms knowledge stock is presented The construction of knowledge capital In order to get a grasp of a firm s knowledge base one has to approximate it since it is not possible to have an accurate measure containing all the elements which comprise the firm s knowledge base. What one can do, however, is to exploit the firm s investments in the creation and/or acquisition of knowledge assets and construct its knowledge stock for a certain period of time. Towards this direction and following the relevant literature, the most prominent way for the calculation of a firm s knowledge stock (Hall et al. 2010) is the perpetual method: 1, 1 K K R (3.1) it i t it where K t is the knowledge stock of firm i at time t, R t denotes annual investments in the creation and/or acquisition of knowledge assets 10 at time t and is a suitably chosen (private) depreciation rate. Some issues need to be discussed at this point regarding the nature, or in other words, the decomposition of Greek firms R&D expenditures as well as their initial level of knowledge capital stock. First of all and regarding the initial level of the Greek firms knowledge capital stock, the time period under examination is the decade Within the sample of firms, quite many of them report R&D expenditures the year t However, it is not unreasonable to assume that their R&D activities had been taking place before t 0. In order to cope with this issue I follow Hall et al. (2010) and calculate the knowledge capital stock at t 0 as follows: 10 I follow the definition of R&D investments given by International Accounting Standards (IAS 38), where particular emphasis is given to the term intangible assets. Therefore, I am interested in measuring the investments in knowledge creation and/or acquisition, as they are perceived and reported by Greek firms in their annual financial statements. -Pg. 66-

78 K it0 Rit 0 g is is (3.2) where is is the depreciation rate of the i-th firm which depends on the four sectors s established by OECD 11 and g ij is the knowledge investments growth rate at the s-th 12 sector level. Turning to the components of Greek firm s R&D expenditures, the focus is shifted on the process of their calculation. It is commonly known that published data for R&D expenditures of Greek firms do not exist in a systematic and organized manner. Therefore, and in order to gather such information, the annual published financial records of Greek firms for the period have been employed, in order to identify those who have reported expenditures of that sort. In the following paragraphs encountered issues during the management and construction of the annual R&D investments and thus, the Greek firms knowledge stock during the period , will be discussed more elaborately. Essentially, annual R&D investments R are calculated employing two main components from the firms published financial accounts, each one of them having different properties not only from an accountant s point of view but also from an economist s point of view. More specifically, the first source of identified R&D investments lies in the firm s published intangible assets. Within this category, knowledge investments that the firm has capitalized have been isolated, entailing not only expenditures on R&D but also on licenses, patents and other corresponding permissions. Finally, the third source of R&D investments comes from the GRD firms Annual Income Statement where the firms report annual expenses on R&D that have not been yet capitalized. In the following sections, a condensed but informative it 11 Hatzichronoglou (1997) defined four sectors based on their technological dynamism namely, (i) High-tech, (ii) Medium High-tech, (iii) Medium Low-Tech and (iv) Low-tech Sector. It should be noted that the depreciation rate varies from 12% to 8% and the more technologically advanced a sector is the more annual depreciation in its knowledge stock will suffer. 12 It is calculated as the cumulative average growth for the period and data are taken from the OECD ANBERD database. -Pg. 67-

79 description will be presented in order for the reader to understand the procedures that have taken place for the construction of the Knowledge stock database of Greek firms Handling the Greek firms published annual accounts During the period under investigation a major change has occurred involving the Standards that Greek firms were obligated to follow, in order to publish their annual accounts. In particular, in 2004, a legislative norm had been passed obliging those firms who either participate directly in the stock market or are part of a Group to publish their financial accounts following the International Accounting Standards (IAS), whereas the rest of them had the option to continue publishing their annual financial accounts following the National Accounting Standards (NAS). The main differences regarding R&D expenditures measurement and publishing are the following: (i) Firstly, the IAS dictate that Research should be separated from Development which in NAS this is not necessarily the case; (ii) secondly, the method for the valuation of (intangible) assets differs between the two standards in that the IAS dictate that the valuation of assets is done at their current price taken the last day of the financial year, whereas the NAS dictate that the valuation of assets is done by selecting the lowest value between current and purchase price. One can see that the IAS tend to be less strict than NAS and that impacted our R&D expenditures measurement; (iii) thirdly, the IAS from their core philosophy are less detailed than NAS and therefore, crucial information needed in order to calculate annual R&D investments and thus, Greek firms knowledge stock had to be estimated 13. Shifting the attention towards the adopted approach, in order firstly, to calculate the annual R&D investments and based on that apply formula (3.1) and construct the knowledge stock of Greek firms in the sample. As previously stated R&D investments at the firm level were not available readily made by some corresponding institution at National and/or European level. Therefore, after the identification of the firms that report investments in the creation and/or acquisition of 13 These differences were made clear after several discussions with Professional and Academic Accountants and Dr. D. Tzelepis. I owe thanks to all of them. -Pg. 68-

80 new knowledge in their annual financial accounts for the time period under examination, the values reported had to be further processed, since depending on whether they were reported as assets or expenses they had to be treated differently. More specifically, for the expenditures reported as assets the book values were provided for the cumulative (i) R&D investments (iii) net value CRNV t t CR, (ii) depreciation CRd t and, and also cumulative (i) investments in various forms of industrial property rights such as licences and patents CIPR corresponding depreciation t CIPRd and net value t t, along with their CIPRNV for the year t or else and CRNV t CRt CRdt (3.3) CIPRNV t CIPR t CIPRd t (3.4) It becomes obvious from (3.3) and (3.4) that in their current form these values are not suitable for the calculation of annual firm R&D investments. Hence, these two categories that belong to firm intangible assets had to be treated with the same way. In order to calculate the annual investments in knowledge creation and/or acquisition for the year t from the category of assets the following were applied and RDt CRt CRt 1 (3.5) IPRt CIPR t CIPR t 1 (3.6). The third category of R&D expenditures is reported in the Profit and Loss account and entails annual R&D expenses ARD reported by the firm that are not yet capitalized. Hence, there is no particular handling of these expenditures rather than adding them with the other two components of the annual R&D investments R sum R it is calculated as follows it. In Rit RDit IPRit ARDit (3.7). -Pg. 69-

81 The above formula is applied in the entire sample of Greek firms. In the following paragraphs some issues will be further discussed involving frequently encountered problems during the calculation of R it Handling of financial statements that continuously followed NAS Essentially, the entire sample can be divided into two main categories, namely (i) those firms that publish their annual financial accounts following continuously the National Accounting Standards (NAS) and (ii) those firms that at one point within the time period studied made the transition to IAS. Regarding the first category the main and most frequently encountered problem is when for some reason the firm decides to depreciate the cumulative expenditures on either of the two categories of the intangible assets described above. In more details, from (3.7) one can see that the annual investments in knowledge assets cannot be less than zero. Therefore, the outcome of the difference in (3.5) and (3.6) should be non negative. Such a constraint is posed because in the process of constructing the firm s knowledge stock, the flows which fuel the augmentation of the firm s knowledge base, cannot possibly be negative. From a theoretical perspective, knowledge base formation is fueled by knowledge flows which in turn, may entail augmenting, complementing, substituting, decaying or even destructing elements of knowledge but in any of the above cases and under no circumstances could it be argued that a knowledge flow could be subtracted from the firm s existing knowledge capital. At this point, it should be noted, that the treatment within the two components of intangible assets i.e. RD t and IPR t is somewhat different for reasons that concern the realistic assumptions described above. More specifically, and regarding firms R&D investments the knowledge flows can in no case be negative, in the case of investments in Industrial property rights a negative knowledge flow could imply not a decrease in knowledge but a decrease in the economic exploitation of it. However, this assumption can only hold if the outcome of (3.7) remains non negative. If not, then the handling of such cases is described below. When the cumulative expenditures where depreciated, for instance at time t+1, I replace them with the cumulative expenditures of the previous year t so as to minimize the knowledge flow and equate it with zero or in other words CR * t1 CR t -Pg. 70-

82 * and thus, RD 0 t, i.e. the knowledge flow for the year t+1=0. However, if the 1 cumulative investments of the firm, which is usually the case, increase at a later year, for instance at t+2 then the annual knowledge flows are calculated based on the original values, that is CR CR RD, and then, the difference is added in the * t2 t1 t2 replaced value of the year which had suffered the depreciation or in other words, CR RD CR RD. * * * * t1 t2 t1 t Handling financial statements that switched at some point to IAS As it was previously stated a quite significant number of firms in our sample have made the transition at some point within the time framework under investigation from NAS to IAS. This transition has impacted our measurement for several reasons that have been mentioned above. In order to handle the measurement problems created due to the transition, some rules were applied so as to have a unifying, to the extent possible, adjustment procedure given the structural break occurred. First of all, it should be mentioned that the NAS had distinct codes in which expenditures were reported not only for R&D but also for IPR, whereas in IAS the category of Intangible assets is more aggregated and many other expenditures are also reported. In addition, the legislative norm established in 2004 dictated that when the firms made the transition, were advised to publish their financial statements following both standards in order for their investments to get acquainted with the differences. Furthermore, the period has been characterized as a transitional period in which the IAS and NAS had quite many resemblances. For the object of interest that is investments in knowledge assets, during this period extra information was provided and more specifically, the cumulative (i) expenditures in intangible assets, (ii) depreciation and (iii) net value. It should be mentioned at this point, that normally when publishing financial statements following the IAS and with respect to assets, the firm reports only the net value of its assets. Based on the above facts I was able to come up with a two step-procedure in order firstly, to estimate the cumulative expenditures on knowledge assets for the period , and secondly, to adjust these expenditures to the equivalents when they were published following the NAS. More specifically, and as it was previously -Pg. 71-

83 mentioned, the cumulative Net Value is calculated as CNVt CEt Cdt where CEt is the cumulative expenditure in intangible assets at year t and Cd t is the cumulative depreciation. In order to estimate the cumulative expenditures for the period the average growth rate of the cumulative depreciation for the period was first calculated. Then the following formula was applied: t t1 t0, t1 Cd Cd 1 (3.8) where is the average growth rate of the cumulative depreciation of Intangible 0 Assets investments for the period 2004 t until the last available year t 1. Having estimated the cumulative depreciations for all years the cumulative annual expenditures in intangible assets were calculate as CEt CNVt Cdt. Then it remained to adjust the cumulative investments in intangible assets to the corresponding values of the categories following NAS. For that purpose, the base year 2004 was exploited where firms published their annual financial statements following both standards. More specifically the following was estimated: CE tnas CR CIPR CE CEt (3.9) where CEt NAS are the adjusted cumulative expenditures in intangible assets expressed in NAS values and at year t, CEt are the cumulative expenditures expressed in IAS values, CE 2004 are the are the cumulative expenditures expressed in IAS values for the year 2004 and CR2004 CIPR 2004 is the sum of cumulative expenditures on R&D and IPR for the year 2004 expressed in NAS values. Last but not least, it should be mentioned the case where a firm published R&D expenditures in its Income Statement and after the transition to IAS. Again the reference year was exploited and the following were estimated IARD tnas ARD IARD IARDt (3.10) -Pg. 72-

84 where IARD t, are the reported expenses expressed in IAS values at year t, IARD 2004 are the reported expenses expressed in IAS values at year2004, IARDt NAS are the estimated expenses in R&D expressed in NAS values at year t, and ARD2004 are the annual expenses reported and expressed in NAS values. After having completed all the adjustment procedures (3.7) is applied somewhat differentiated and specifically Rit Eit IARD it (3.11) where Eit is the equivalent knowledge flow from the category of intangible assets expressed in NAS values and values. IARDit are the annual expenses expressed also in NAS 3.5. The knowledge intensity In this section the interest is shifted in presenting detailed characteristics of the constructed knowledge stock as they have emerged from the survey of the Greek R&D manufacturing firms. More specifically, the mapping of the knowledge intensity is attempted, which is defined as the degree of involvement of knowledge in business activities either through its integration directly into GRD firms outputs or indirectly through their inputs. In order to explore the extent of the of knowledge intensity two indices are employed; the ratio of knowledge capital per employee KNEMPL and the ratio of knowledge capital to GRD firms total assets KNASS. Specifically the indices are defined as and Knowledge Capital KNEMPL (3.12) Total Number of Employees Knowledge Capital KNASS (3.13) Total Assets -Pg. 73-

85 KNCAP/EMPL University of Patras In figures 3.3 and 3.4 the basic descriptive statistics of the two indices are presented respectively, both for the total sample of GRD firms as well as for the three 14 technological sectors (High-tech, Medium-tech and Low-tech). Figure 3.3. Knowledge Capital per employee: Distribution based on technological opportunities KNEMPL and Technological Opportunities 1,200,000 Even though the relevant differences among the three technological sectors are expected and follow the findings of the relevant literature, within each technological sector one can observe great discrepancies with the technological sectors. On the other hand, it should be noted that these asymmetries follow the distributional specificities of R&D expenditures identified by Cohen and Klepper (1992; 1996). Given that the distribution of R&D expenditures is the one that essentially underlies the distribution of Knowledge Capital, the abovementioned knowledge capital asymmetries could reasonably interpreted based on the findings of Cohen and Klepper (1996). 800, ,000 0 All LT MT HT Max 986, , , ,105 Min Average 26,320 10,869 27,816 79,399 However, it is worth mentioning that the higher values, in average terms, of both KNEMPLand KNASS indices that correspond in the High-tech sector. More specifically, the average value of the KNEMPLindex in the High-tech industries is 14 For analytical purposes the medium-high tech sector and the medium-low tech sector have been merged into one medium tech sector. -Pg. 74-

86 KNCAP/TOTASS University of Patras approximately eight times higher than the corresponding value in the low-tech industries and approximately three times higher than the Medium-tech sector. Figure 3.4. Knowledge Capital as a percentage of GRD Total Assets KNASS and Technological Opportunities 80% 60% 40% 20% 0% Average Min Max All 14.11% 0.01% 77.50% LT 5.83% 0.01% 68.06% MT 7.55% 0.02% 77.50% HT 19.33% 0.11% 76.98% Fewer are the discrepancies in the KNASS index. In particular, the average value of the index in the case of GRD high-tech firms is about two times higher than the corresponding value of the GRD medium-tech firms and three times higher than the corresponding value of low-tech firms. The differential sectoral values of KNEMPL and KNASS indices could easily be interpreted from the corresponding differential values in terms of employment-capital intensity indices of the three technological sectors examined. Table 3.4 presents the basic descriptive statistics of the joint distribution of KNEMPLand KNASS indices in relation to GRD firms size. A complete picture of the distribution of both these indices as they were approximated by kernel densities estimates is presented in figures 3.5 and 3.6 respectively. KNEMPLand KNASS indices are differentiated in terms of size and technological sector, which provides a strong hint for the existence of a severe underlying heterogeneity with respect to Greek manufacturing firms R&D activities. It could be argued that the existence of heterogeneity is rather expected since the field research is extended in the entire Greek Manufacturing; however, it should be taken into consideration in subsequent analyses employing the particular dataset. According to the Schumpeterian -Pg. 75-

87 hypotheses large firms should exhibit greater values, in average terms, relative to smaller firms. In this particular case though, this expectation does not seem to be confirmed. On the contrary, in average terms, small firms present greater valued of the KNEMPLindex, however, with great variation existing among them. Table 3.4. Basic Descriptive statistics of the KNEMPLand KNASS in relation to GRD firm size KNEMPL Average Min Max Small Medium Large KNASS Small 7,84% 0,02% 77,50% Medium 9,18% 0,01% 83,98% Large 8,07% 0,10% 76,98% Figure 3.5. Kernel density estimates of KNEMPL variable -Pg. 76-

88 Density University of Patras Kernel Density estimate of KNEMPL index Small Medium Large KNEMPL Regarding the KNASS index medium firms exhibit slightly greater average values, again with great variation existing among them, while the existing differences between size classes are evidently smaller than the corresponding KNEMPLindex. It should be noted that in every case, significant heterogeneity exists with respect to KNEMPLand KNASS indices within each size class, which raises questions as to the core of Schumpeterian hypothesis as well as with respect to the rather arbitrary boundaries that differentiate small, medium and large GRD firms. Regarding the higher average values of the KNEMPLindex that refer to large firms, it should be mentioned that according to Cohen and Klepper (1996), and while the absolute values of their R&D expenditures which are the principal input of the constructed Knowledge Capital are clearly greater than the corresponding of medium and small firms, the scale or in other words the size of large firms eventually leads to smaller analogies of any knowledge measure relative to firm size. In other words, it is about an R&D idiosyncrasy that Cohen and Klepper (1996) have identified. Figure 3.6. Kernel density estimates of KNASS index -Pg. 77-

89 Density University of Patras Kernel density estimates of KNASS index 10 Small Medium Large KNASS 3.6 The GRD firms Absorptive Capacity The approach of absorptive capacity follows the findings of the relevant literature. More specifically, following Zahra and George (2002) a firm s absorptive capacity is defined a set of routines and processes by which firms acquire, assimilate, transforms and exploit knowledge to produce a dynamic organizational capability and can be quantitatively approached by the Abscap variable (Harris and Li, 2009): Number of Employees with tertiary Education Abscap Total Number of Employees (3.15) In the same direction, absorptive capacity has been defined as the intensity of R&D activities (Cohen and Levinthal, 1991) as the ability to recognize the value of new information, assimilate it, and apply it to commercial ends and it is quantitatively approached by the RDAbscap: -Pg. 78-

90 Number of R&D Employees RDAbscap (3.16) Total Number of Employees In the context of the RDAbscap variable, absorptive capacity depends greatly on prior related knowledge and diversity of background. For that reason, a firm s R&D investments play a crucial role in approximating absorptive capacity. Absorptive capacity is perceived as a cumulative measure, which is the outcome of a continuous path dependent process where the firm is required to pursue continuously the sustainable growth of its absorptive capacity in order to be able to evaluate, assimilate and exploit new information in a particular field. Following again the Schumpeterian hypotheses, the above two definitions of absorptive capacity are examined in terms of firm size and technological opportunities as the latter are depicted in the High, Medium and Low-tech sectoral distributions. Table RDAbscap with respect to technological intensity and firm size LT MT HT All 7,77% 10,22% 18,94% (0,121)* (0,161) (0,264) Small 13,57% 15,82% 25,81% (0,160) (0,201) (0,329) Medium 3,23% 4,91% 9,60% (0,025) (0,061) (0,102) Large 2,16% 2,62% 19,06% (0,037) (0,021) (0,268) *Numbers in parentheses are standard deviations Regarding the distribution of the RDAbscapvariable, as expected, firms belonging to high-tech industries exhibit higher levels of absorptive capacity followed, by medium and low-tech firms. On the contrary, firm size distributional asymmetries related to R&D expenditures (Cohen and Clepper, 1992; 1996) that have been identified in the case of KNEMPLand KNASS indices are also present in this particular case. More specifically, a negative relationship seems to be formed between firm size and in the medium and low-tech sectors, and a U-shaped relationship with the high-tech industries. Regarding the Abscap variable as it is presented in figure 3.7 basic descriptive statistics indicate that the relationship between firm size and absorptive -Pg. 79-

91 capacity is differentiated to a small extend whereas the distributional specificities with respect to technological sectors remain unchanged. In particular, it may be argued that the more affluent technological opportunities exist, as the latter are approximated by technological sectors, the more increased absorptive capacity is required by GRD firms belonging to of a particular technological sector. The difference of the Abscap variable distribution with the corresponding distribution of RDAbscap concerns firm size, while, the U-shape distribution applies for all firms, independently of the intensity of technological opportunities they face. Nevertheless, approaching GRD firms absorptive capacity should be treated carefully due to the significant underlying heterogeneity that exists in the sample. Figure 3.7. Abscap with respect to Technological opportunities and Size 60% 50% 40% 30% 20% 10% Abscap with respect to size and technological opprtunities 0% LT MT HT Small 27% 30% 53% Medium 15% 19% 41% Large 18% 21% 45% 3.7. Mapping R&D Effectiveness An important issue surrounding innovation analysis has been recorded in the relevant literature and it involves the effectiveness of R&D activities. R&D effectiveness refers to the mechanism of transforming knowledge competencies related to R&D activities into technological capabilities that are in turn reflected in GRD firms innovation performance and overall business performance. In the context of the -Pg. 80-

92 conducted survey the R&D effectiveness was investigated employing questions 3.13 and 3.14 of the survey questionnaire. Table R&D effectiveness of GRD firms Technological Opprtunities Size Motivation Exporters NonExporters Exporters NonExporters All LT MT HT All LT MT HT Sm Md Lr Sm Md Lr Products and Services Profile New Customer Attraction Investment Attraction Foreign Market Penetration Expansion of Market Share Quality Improvement JIT Production Production Capacity Increased Profits Increased Value Added Reduced Production Costs Safety and Hygiene Conditions Environmental Impacts Replacement of existing product/processes Complying with Regulations Ineffectiveness Efefctiveness Super-Effectiveness For each of the questions separately, the responses have been normalized, in order to reduce subjectivity. Then the difference has been computed between each of the available indicators in questions 3.13 and A positive difference reflects a refutation of the expectations with respect to R&D activities and is placed within the R&D ineffectiveness area. A negative difference outcome reflects the fulfillment -Pg. 81-

93 above the original expectations of the purposes of R&D activities and form the super-effectiveness area, while zero difference outcomes denote the exact fulfillment of expectations regarding R&D activities and form the effectiveness area. Numbers in cells depict the relative significance of the corresponding indicator within the corresponding colored area of effectiveness per category of GRD firms. It should be noted that the hierarchy of effectiveness is structured to the vertical and not to the horizontal dimension. However, the horizontal dimension reveals the importance of several (in-) effectiveness issues between GRD firms distinct groups when the latter are established with respect to exporting status, size characteristics and technological opportunities. Non-colored-cells correspond to issues that were not considered as significant incentives for GRD firms to be engaged in R&D activities and thus no expectations were formed regarding the corresponding issues. Despite the complexity of the individual effectiveness mode, the emerging overall picture indicates that the greater R&D ineffectiveness is concentrated in the large and non exporting GRD firms belonging to the low-tech sector. However, in some individual issues related to investment attraction, JIT production and reduction of production costs this type of firms exhibits important super-effectiveness. The greater ineffectiveness is exhibited in expectation regarding increased profits and market share expansion. On the contrary, significantly improved effectiveness and supereffectiveness is presented between GRD firm groups regarding investment attraction, compliance with regulations, safety and hygiene conditions and replacement of existing product/processes. Issues related with R&D effectiveness, the formation of competitive advantage and the decision to export are analyzed in the fourth chapter of this PhD thesis. -Pg. 82-

94 Number of Firms University of Patras 3.8. The internationalization activities of GRD firms One of the most important forms of internationalization activities is that of export orientation 15. Complementary information of internationalization activities of GRD firms are examined in the following two chapters of this PhD thesis, even though the locus of attention remains their internationalization activities. In the investigated sample of GRD firms 78% are engaged in exporting activities while the remaining 22% operates exclusively in the domestic market. The distribution of GRD firms exporting activity per size class is presented in figure 3.8. Figure 3.8. Exporting activities of GRD firms depending on size class Exporting Activities and Size Class Small Medium Large Exports No Exports Medium and large firms share approximately the same percentage of exporting firms, 89% and 86% respectively, while small exporting firms correspond to a significantly lower percentage of 66%. Table Distribution of exporting activities depending on technological sector Technological Exports Intensity Yes No LT 35,53% 56,25% 15 Besides exporting activities, the relevant literature has recorded other means of internationalization that is, Foreign Direct Investment (FDI), Joint Ventures and Strategic Alliances, Licencing and Outsourcing, (Saggi, 2002). -Pg. 83-

95 MT HT (69.22%) (30.81%) 53,07% 37,50% (83.37%) (16.43%) 11,40% 6,25% (86.70%) (13.30%) In table 3.6, percentages outside parentheses denote the fraction of GRD (non)exporting firms that belong to a particular technological sector to the total GRD sample, while, percentages within parentheses denote the fraction of (non)exporting firms that belong to a particular technological sector to the total number of firms that belong to the same technological sector. It becomes evident that regarding HT and MT sectors, the percentage of exporting firms is significantly higher compared to LT sector. In addition, more than half GRD firms in the sample that do not present any exporting activity belong to the LT sector. Turning to the Export Market Destinations, Eurozone countries take over more than 50% of the total exports of GRD firms, followed by European countries that are not members of the Eurozone and Countries belonging to the Rest of the World. Export share destined to North America (including Canada) is the smallest (13.7%). From a first perspective, taxation and tariff related barriers, in conjunction with geographical distance may be the primary determinants that define this particular distribution. With respect to the routes of exporting that GRD employ in order to export their products, primary results on the answers provided depending on the GRD firms export intensity class are presented in table 3.7. Direct exports DIREXP is the most frequent means of exporting, followed by exporting through a representative EXPREPR, exporting through an intermediary through subcontracting SUBCONTR. INTERM, and finally exporting -Pg. 84-

96 Figure 3.9. Geographical Distribution of GRD firms Geography of Exporting Activities Eurozone 60% 40% 20% ROW 0% ROE NAM No significant variations seem to be present among export intensity classes. It should be noted that for the GRD firms that realize the 25%-50% of their annual sales through exports, a preference in using intermediaries and subcontractors exist in contrast to the direct exports and export representatives. Table 3.8. Routes of exporting per export intensity class SUB EXPREPR EXPINT DIREXP Total (0,25] 9 (5,26%) 47 (27,49%) 21 (12,28%) 94 (54,97%) 171 (25,50] 7 (10,77%) 13 (20,00%) 14 (21,54%) 31 (47,69%) 65 (50,75] 3 (6,52%) 11 (23,91%) 8 (17,39%) 24 (52,17%) 46 (75,100] 5 (8,62%) 15 (25,86%) 9 (15,52%) 29 (50,00%) 58 Total 24 (7,06%) 86 (25,29%) 52 (15,29%) 178 (52,35%) The graphical representation of the above distribution is presented in the figure Pg. 85-

97 Figure Routes of exporting depending on the export intensity class 200 Routes to Exporting and EXPINT class SUB EXPREPR EXPINT DIREXP (75%,100%] (50%,75%] (25%,50%] (0%,25%] Besides exports the dominant alternative strategy for participating in the global market is importing (Figure 3.11). In addition, Outsourcing appears to be a frequent means of internationalization. On the contrary, FDIs, Licencing and forming Joint Ventures and Strategic Alliances are the internationalization means less frequently employed 16. Figure Participation in the Global markets for the Exporting GRD firms. Exporters' Additional Sources of Internationalization 80.00% 60.00% 40.00% 20.00% 0.00% IMP FDI JVSA LIC OUT Never 13.01% 66.10% 58.90% 65.41% 50.34% Occasionaly 54.45% 14.38% 20.89% 16.44% 31.85% Regularly 23.63% 3.42% 5.82% 3.08% 4.45% No response 8.90% 16.10% 14.38% 15.07% 13.36% 16 In the context of the survey no distinction is made between inward and outward Licencing, Outsourcing and FDIs. -Pg. 86-

98 Number of Firms University of Patras It should be noted that the above frequencies solely concern the exporting GRD firms. Potentially alternative modes of internationalization have been investigated also for the non-exporting GRD firms. Figure 3.12 presents the corresponding distribution. Figure Participation in the Global markets for the non-exporting GRD firms. Non-Exporters' Internationalization 60% 50% 40% 30% 20% 10% 0% IMP FDI JVSA LIC OUT Never 18.46% 58.46% 47.69% 53.85% 38.46% Occasionally 43.08% 6.15% 21.54% 13.85% 27.69% Regularly 13.85% 4.62% 4.62% 4.62% 7.69% No Response 24.62% 30.77% 26.15% 27.69% 26.15% Based on the above, it is difficult formulate an overall picture of the GRD firms internationalization activities. From the percentages presented above and after performing simple non-parametric tests, it could be argued that only exporting GRD firms with strong exporting orientation (with an export intensity greater than 75%) are often involved, in relation to non-exporting GRD firms) in other forms of internationalization and mostly through imports and outsourcing. At this point, it is worth mentioning that in the sample of the GRD firms, there exists a small though significant number of Multinational Enterprises. In the context of this research, it has not been possible to control for the effects of intra-firm trade. The relevant literature has not investigated in depth the effect of exporting barriers not only on the decision to export (Leonidou, 1995) but also on firms export intensity. In this direction, in the survey questionnaire the evaluation of a series of export barriers have been incorporated as the latter have been identified from other analogous -Pg. 87-

99 surveys in other countries. The series of export barriers under evaluation are depicted in the question 2.9 of the questionnaire 17 and have been evaluated based on a Likert three scale. The answers have been normalized for each observation in order to reduce the subjectivity of the responses. In this line, for the total answers and controlling for size class and technological intensity, the five, hierarchically, most significant barriers have been identified. The corresponding results are presented in table 3.8. It could be argued that independently of the size class and technological sector, the most important export barriers are linked with institutional issues and bureaucracy, followed by the issue of low price competitiveness of GRD firms. Table Exporting Barriers of the GRD firms. Barriers to Exporting Language Exchange Rates Volatility Price of Competitive Products Greek Institutions/Bureucracy Restricted information for market destination Transportation Costs Strict Legislation in European Market Dest. Strict Legislation in non European Market Dest. High Entry Costs Capital Restrictions Participation in International Distrib. Networks Unfavourable National Legislation Politically Unstable Export market Dest. Lack of qualified HR Product Quality/Characteristics Difficulties due to Intermediaries Distrust towards Greek Economy All Technological Size Intensity LT MT HT SM MED LRG Low High 17 See the Appendix Section -Pg. 88-

100 In individual categories the low price competitiveness is highlighted as the most important barrier in the exporting GRD firms that belong either to the LT sectors or they are have a large firm size. The transportation cost is enlisted as the third most important barrier, while as is the case with the low price competitiveness, Low tech or large firms consider it as their number two most important barrier. It seems that both these categories of GRD firms are extremely sensitive, with respect to their exporting decisions, to issues that are directly linked with cost competitiveness. Regarding the rest of the barriers, GRD firms have been evaluated as most important the following: i. National Legislative Framework for Exports ii. Distrust towards Greek Economy iii. High entry costs in International markets It is worth mentioning that as it is presented in Table 3.8 the overall importance of the export barriers exhibits a persistent character across technological sectors and size classes. As Hessels and Terjesen (2010) note, business units that are engaged in internationalization activities, among which is exporting, present high levels of external isomorphism to host country and internal institutional environments. From the responses of the GRD firms it seems that this may also be the case for exporting GRD firms. This issue will be analyzed further in the next two chapters of this PhD thesis The relationship between GRD firms Export Orientation and Knowledge Base In the next few paragraphs I present some features surrounding the relationship between the characteristics of the GRD firms knowledge base and their decision to participate in the International Markets through exporting. More specifically, the bundle of GRD firms knowledge inputs is presented depending on their exporting status. In this line, the basic components of innovation performance are presented as well as the GRD firms attitude towards R&D collaborations. Further investigation of the relationship between GRD firms export orientation and their knowledge stock, -Pg. 89-

101 EXPINT University of Patras according to the Schumpeterian variables of firm size and technological environment, are also presented in several occasions GRD firms Knowledge Stock and Export Intensity Taking into consideration that GRD firms exporting activities may be characterized by significant heterogeneity with respect to its relative export volume, GRD exporting firms have been divided into four classes, based on the fraction of their annual sales that comes from International markets or in other words their export intensity. The relationship between GRD firms export intensity and KNEMPL as well KNASSindices are presented in figure 3.13 and Table 3.9 respectively. Figure GRD firms export intensity and Knowledge capital per employee KNEMPL and EXPINT >75% 51%-75% 25%-50% <25% KNEMPL There may be a rather clear positive relationship between the KNEMPL index and export intensity. An exception appears to exist for the export intensity class of 25%- 50%.. Table KNASS distribution with respect to firm size and export intensity Exporting Status Total Export Intensity No Exp. Size (0-25] (25-50] (50-75] (75-100] Small Pg. 90-

102 (0.453)* (0.319) (0.107) (0.616) (0.451) Medium (0.081) (0.086) (0.479) (0.210) (0.104) Large (0.021) (0.151) (0.024) (0.297) (0.548) Total *Numbers in parentheses are standard deviations Table 3.9 presents the distribution of KNASSindex with respect to GRD firms size and export class. More specifically, numbers outside parentheses correspond to the average values of KNASSdepending on the export status and intensity as well as firm size. Overall, it could be argued that small firms have on average a higher value of KNASSrelative to medium and large firms, while the more export intensive a firm may be the higher the average KNASSvalue. This finding may be indicating that the smaller and more export intensive a GRD firm may be the more knowledge intensive it may be Organizational characteristics of R&D activities and Export Orientation Within the relevant literature the organizational characteristics of R&D activities are considered of particular interest both in relation to firms innovation performance and their formulated strategy (Caloghirou et al., 2004). Firms knowledge stock organizational characteristics have also been reported as important determinants of R&D efficiency. This particular issue will be more extensively analyzed and discussed in the next chapter. The distinction between in-house and external R&D is one of the most employed organizational R&D characteristics (Piga and Vivarelli, 2003). The distribution between in-house and external R&D activities depending on technological sector and exporting status of the sample of GRD firms is presented below in figure It is worth mentioning that 14 GRD firms (4.6% in the total sample) answered that they engage in both types of R&D. From these, 11 belong to the high-tech sector, 2 in low tech sector and 1 medium tech sector. In addition, 12 out of 14 declared that they are engaged in exporting activities. During -Pg. 91-

103 The basic findings indicate that exporting GRD firms to a large extent perform their internal R&D activities internally. More specifically, this high percentage of in-house R&D activities is presented in Medium and High tech firms. From a closer inspection, a negative relationship between technological intensity and external R&D activities appear to be in place. Equally important is the feature of duality between exporting and non exporting firms. More specifically and according to figure 3.15, it could be argued that non exporting GRD firms reveal a strong preference towards external R&D comparing to exporting GRD firms. Figure In-house and external R&D with respect to technological intensity and export status 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% InHRD ExtRD InHRD ExtRD InHRD ExtRD HT MT LT nonexp EXP The percentage of the non-exporting firms that select to acquire external knowledge presents an inverted U-shape form with respect to technological intensity. On the contrary, and regarding the exporting GRD firms, the relationship between engaging in external R&D and technological intensity is negative. These relationships should be interpreted with special caution since the specific organizational characteristic of R&D activities is only related with the dimension of technological intensity of GRD the presentation these 14 GRD firms have been incorporated in the corresponding in-house and external R&D. -Pg. 92-

104 External R&D propensity University of Patras firms while it is evident that the decision of in-house or external R&D (or both) has many dimensions. Apart from in-house or external R&D engagement and in the context of examining the organizational characteristics of R&D activities, question 3.7 of the survey questionnaire entails additional elements which are related to the rest of the inputs employed in R&D activities (Vega-Jurado et al., 2008). Figure Engagement in external R&D activities with respect to technological intensity and exporting regime NonExp Exp Low Medium High Technological Intensity Hierarchically, the importance of these inputs is the following: i. Acquisition of machinery, equipment and software (37.5% 19 ) ii. Purchase or licensing of patents and non-patented inventions, know-how, and other types of knowledge from other enterprises or organizations (14.8%) iii. Internal or external training for business personnel specifically for the development and/or introduction of new or significantly improved products and processes (8.1%) 19 These percentages are calculated over the total sample of GRD firms -Pg. 93-

105 A clear distinction between exporting and non exporting firms becomes apparent with respect to the elements composing GRD firms knowledge base. The relevant results are provided below in table More specifically, the exporting GRD firms employ more inputs in forming their knowledge base than the corresponding non-exporters not only when considering this distinction but also when the dimension of technological intensity is taken into consideration. In any case, the acquisition of R&D related new equipment, tools and software appears to be the primal input employed in R&D activities of GRD firms. The frequency of the use of the abovementioned additional knowledge base inputs is positively related with the technological sectors the GRD firms belong to. Table Distribution (%) of additional elements of GRD firms knowledge base with respect to exporting status and technological intensity * Exporting Status R&D All Exporters Non-Exporters additional inputs LT MT HT LT MT HT New equipment, tools and software R&D (0.45) (0.54) (0.95) (0.45) (0.44) (0.39) (0.41) related Purchase patents, license and IPR (0.36) (0,45) (0.51) (0.78) (0.15) (0.22) (0.38) Personnel Training R&D (0.27) (0.33) (0.41) (0.61) (0.22) (0.00) (0.00) related *Numbers in parentheses are standard deviations Innovation performance and appropriation of R&D outcomes The outcomes of R&D activities are primarily depicted with the mapping of innovation outcomes as they are further categorized in product innovation, process innovation and service innovation (Oslo Manual, 2005). In figure 3.16 the innovation outcomes, or in other words the GRD firms technological capabilities, are sketched with respect to GRD firms exporting status. Based on the figure below, it becomes -Pg. 94-

106 HT MT LT University of Patras evident that in all cases (except the case of product innovation in Medium tech sector) exporting GRD firms present a superior innovation performance. The exposure in international markets and to a lesser extent in domestic markets, raises the issue of internalizing the outcomes of R&D activities (Kafouros et al., 2008), particularly when the innovation outcomes form the basis of GRD firms competitive advantage. Figures 3.16 and 3.17 respectively, provide the distributions of R&D activities internalization type for the exporting and non exporting GRD firms respectively. Figure Innovation Outcomes, Exporting Status and Technological Intensity of GRD firms Innovation type by Technological Intensity and Exports Status Prod Serv Proc Prod Serv Proc Prod Serv Proc 0% 10% 20% 30% 40% 50% 60% 70% 80% nonexp EXP Figure Appropriability Conditions and Innovation Output with respect to Technological Intensity for the exporting GRD firms -Pg. 95-

107 Exporters IPR and TI 70% 60% 50% 40% 30% 20% 10% 0% -10% Patents IndDes Copyright OnGoing None LT MT HT Figures and 3.17 illustrate a failure trap area (ongoing R&D activities and none innovation outcomes) of the transformation process of competencies that arise in the context of investing in GRD firm s knowledge base augmentation into technological and innovation capabilities (Auh and Menguc, 2005). Of course, for the area of failure trap there is not an issue of internalization of innovation outcomes. This particular theoretical context is presented more analytically, in the next chapter of this PhD Thesis. It becomes evident that this area on entrapment is greater for non exporting GRD firms compared to the corresponding area of exporting GRD firms. Figure Appropriability Conditions and Innovation Output with respect to Technological Intensity for the non exporting GRD firms Non-Exporters Appropriability conditions 70% 60% 50% 40% 30% 20% 10% 0% Patents IndDes Copyright OnGoing None LT MT HT -Pg. 96-

108 Towards the direction of extending the appropriability conditions it could be argued that exporting GRD firms require significantly more resources compared to non exporting GRD firms in order to develop processes for the appropriation of the outcomes of R&D activities. It is also worth mentioning that exporting GRD firms belonging to high tech sectors present greater intensity of R&D internalization in relation to medium and low tech GRD firms. Patenting is the most important means of protection over imitation. However, the emerging picture, with respect to R&D internalization should be treated with caution, since as it has been previously mentioned the failure trap is significantly higher in the case of non exporting GRD firms and thus, the incentives for internalization in these firms is clearly weaker. Mapping the R&D and innovation performance has been approximated by the contribution of innovation in (i) the ratio on innovative sales to total firm sales and (ii) in the ration of innovative products to the total product proliferation (Laursen and Salter, 2006). In this line, and with respect to GRD firms exporting status and technological intensity table 3.11 presents the distribution of innovative performance as it is approximated by these two indices. Table Percentage of Innovative Sales and Innovative Products to Total Sales and Products Spectrum. Distribution with respect to Technological Opportunities and Export Status* Non-Εxporters Exporters %sales %products %sales %products LT (29.34) (30.28) (34.06) (33.81) MT (34.37) (35.32) (29.16) (102.94) HT (7.07) (2.50) (32.01) (30.35) *Numbers in parentheses are standard deviations Exporting GRD firms exhibit higher innovation performance in terms of both indices compared to non exporting firms. A noticeable exception can be traced in the Medium tech sector for the case of the innovative sales index. Even though the differences between exporting and non-exporting firms have not been investigated from a more -Pg. 97-

109 nonexp EXP University of Patras thorough statistical perspective and thus, such a difference is yet not known whether is statistically significant, it may not be groundless to assume that at least regarding the High tech sector the differences between exporting and non exporting GRD firms are quite striking. The relationship between exporting orientation and R&D activities innovation outcomes as the latter have been approximated by the percentage of innovative sales over the total GRD firm sales, as well as the percentage of innovative products over the total products produced by the GRD firm, is investigated analytically in the context of the present PhD Thesis and specifically in Chapter 5 where the role of GRD firms R&D collaborations and knowledge base is investigated. Furthermore and in the context of R&D activities orientation, the relationship between exporting activities and the innovation type as it is distinguished in radical and incremental innovation (Antonelli, 1998) determines to a great extent the performance of the innovation system itself. Regarding the global R&D leaders this issue has been investigated in the previous chapter of this PhD thesis. Figure Radical and incremental innovation with respect to exporting status and technological intensity. Innovation type by Techn. Opport and Export Status RadInnov IncrInnov RadInnov HT MT LT IncrInnov 0% 20% 40% 60% 80% Figure 3.18 depicts the relationship between radical and incremental innovation for the exporting and non-exporting GRD firms. In the same figure GRD firms are also distinguished with respect to their sectoral technological intensity. At a first glance, -Pg. 98-