«Heidi Wiig and Michelle Wood R-01 • What Comprises a Regional Innovation System? An 1995 Empirical Study Heidi Wiig and Michelle Wood STEP ...»
Asheim B. (1992) ’Industrial Districts, Interfirm Cooperation and Endogenous Technological Development: The Experienc of Developing Countries’ UNCTAD Symposium on industrial districts and technology, Geneva 16-47th Nov.1992; Storper, M. (1991) ’Technology Districts and International Trade: The Limits to Globalization in an Age of Flexible Production’ mimeo Grad School of Urban Planning and Lewis Centre for Regional Policy Studies, Univ. Of Calif. LA, Sept. 1991 p36);
Porter, M. (1990) The competitive Advantages of nations. Macmillan, London, p. 19-21 and 60-61.
Cooke and Morgan (1994) op cit; Castells M. and Hall P. (1994) Technopoles of the world: the making of twenty-first century industrial complexes, Routledge, London Héraud J-A. (1994) 'Is there a Local System of Innovation in Alsace? An Analysis of the Firms Networks based on an Empirical Study' Paper presented at EUNETIC Conference, Evolutionary Economics of Technological Change : Assessment of results and new frontiers, European Parliament, Strasbourg, Oct. 6-8, 1994 Storper (1991) op cit.
Cooke, P. and Morgan, K. (1994) 'The regional innovation system in Baden-Wurttemberg' International Journal of Technology Management, vol.9, no.s 3/4, pp394-429.
Aydalot, P. and Keeble, D. (1988) High Technology Industry and Innovative Environments. The European Experience, Routledge, London; Castells and Hall (1994) op cit.
4 STEP rapport / report R-01/1995
in other, less technologically-advanced regions or on ’low-technology’ sectors. Thus, studies on regional innovation often cite the lessons which may be learned from successful, usually geographically core regions, without fully concentrating on endogenous capabilities of less-developed regions. This has important implications for regional policy, as Koshatzky13 notes: "the activation and more intensive utilisation of endogenous innovation resources for regional development constitutes an important challenge for a technology-oriented regional policy".
In addressing such issues of innovation capability, another major approach rests on the application of concepts which place less emphasis on geography and use ideas from evolutionary economics, systems theory and innovation theory, giving rise to the idea of systems or network models for mapping innovation 14. These recognise that technology does not exist alone but functions as an integrated part of a socioeconomic system; for example as a national innovation system 15. Thus, the context within which firms conduct innovation may be highly important and may be modelled by analysing the interrelationships between social, economic and technological systems at various scales. The various components and linkages
within and beyond such systems or networks form the basis for analysis, and include:
other firms, such as customers and suppliers; education institutions and research laboratories as sources of skilled labour and knowledge; government agencies as sources of finance, regulatory constraints and support for innovation; financial agencies such as banks or venture capitalists; and providers of business services.
However, in placing less emphasis on geography this can create difficulties, since the role of factors arising from the particular locality or region within which the system operates is ignored or at best explained by the ’embeddedness’ of firms within particular cultural environments. This criticism has been raised particularly by Krugman and it has been suggested that "recently, however, there have been certain developments within economics which may mark the beginning of a closer relationship with economic geography in general and regional development theory more particularly"16. There remain, therefore, key questions concerning the role and importance of geographic factors in the operation of the social, economic and technological systems within a specific region.
In turn, the marrying of theoretical ideas with empirical work has also been problematic17, particularly with the need to apply new and more sophisticated empirical indicators which has emerged with the recognition of the complexity of innovation18. The main existing science and technology indicators, namely R&D data, patents data and bibliometrics, are often irrelevant to regions characterised by Koschatzky K. (1994) "Utilization of innovation resources for regional development - Empirical evidence and political conclusions", Paper prepared for NISTEP Conference,in February 1995.
see especially Lundvall B-Å (1992) op cit; Todtling F. (1994) 'The uneven landscape of innovation poles. Local embeddedness and global networks' in Amin and Thrift (eds) Globalization, Institutions and Regional Development in Europe,OUP, Oxford Lundvall (1992) op cit; Nelson R. (1993)op cit Martin, R. and Sunley, P. (1995) 'Paul Krugman's Geographical Economics and its Implications for Regional Development Theory: A Critical Assessment' Paper presented at IBG Conference, Newcastle-upon-Tyne, Jan.1995, 2 Higgins, (1995) op cit.
OECD (1992) Oslo Manual. OECD Proposed Guidelines for Collecting and Interpreting Technological Innovation Data, OECD, Paris What Comprises a Regional Innovation System? An Empirical Study 5 ’traditional’ industrial structures, large numbers of small firms and an absence of science-based industries and formal scientific institutions. The indicator problems following from this have also been discussed in Norway, and considerable effort has gone in to developing a wider range of official and unofficial statistics on innovation.19 In addition to this, it seems clear that the study of innovation systems in particular localities or regions should be based on an integration of suitable innovation and regional indicators, using methodologies to allow comparisons across different regions20.
2.2 The statistical approach During the 1980s a number of independent research teams attempted to develop survey approaches to innovation which would widen the scope of statistical methods in innovation analysis (for an overview of such work, see Smith, 1992). These surveys mainly attempted to collect data on new product development, and on the firm-level activities which supported such development. In the early 1990s, these approaches were synthesised by the OECD into a statistical manual which recommended a future ‘standard practice’ for the collection of such data. This approach was taken up by the European Commission, in a collaborative action involving DG-XIII (European Innovation Monitoring Initiative) and Eurostat, who implemented a ‘Community Innovation Survey’ in all Member States in 1993/4; this survey collected harmonised data on approximately 40,000 firms.
Simultaneously with this action, the European Commission sponsored a study exploring the possibilities of extending this approach to a regional level21. This project, known as ERIS (European Regional Innovation Surveys), was important background for the study reported here.
The Community Innovation Survey collected three broad types of data. Firstly, economic data on new product introduction and sales, R&D and non R&D inputs to innovation, sales and employment. Secondly it collected binary data on, for example, patterns of technological collaboration. Finally it collected ordinal data, asking firms to rank the importance of various information sources, obstacles to innovation, support measures, and so on.
In this study we use identical definitions and questions on innovation inputs and outputs to those of the Community Innovation Survey. However we also adapted the questionnaire to reflect a range of locational issues, such as location of main suppliers and customers, roles of specific regional agencies, importance of specific regional infrastructural institutions and so on. The questionnaire was applied in two stages in mid-1994 to the gross population of manufacturing firms in Møre and Romsdal; it is, in effect, a census rather than a sample survey. In the first stage a notably K.Smith (1992) ’Technological innovation indicators: experience and prospects’ Science and Public Policy no.19, vol.6, Dec.1992, 383-392; K.Smith and T.Vidvei (1992) ’Innovation activity and innovation outputs in Norwegian industry’ STI Review, OECD, no.11 December 1992 pp11S.O.Nås, T.Sandven and K.Smith (1994) 'The community innovation survey. Status and perspectives' CEC, DGXIII,.Luxembourg see Alderman and Wood (1994) op cit; Nam, Ch.W., Nerb G. and Russ, H. (1990) 'An empirical assessment of factors shaping regional competitiveness in problem regions' IFO Main Report, CEC, Luxembourg Alderman and Wood (1994) op cit 6 STEP rapport / report R-01/1995 postal survey was sent to all firms. In the second stage, all non-respondents were contacted by telephone, and asked to complete a closely similar ’core’ postal questionnaire. Only 110 firms declined to respond. However a large number of firms (approximately 570) were either not relevant (that is, they had been misclassified as being involved in manufacturing production), or were impossible to contact.
Approximately 300 firms failed to respond to letters and phone calls, and there must be a strong supposition that they were out of business. We received a total of 399 responses, which represents a response rate of 78.4% of the firms who we succeeded in contacting, and 48% of the population including the 399 non-contactable firms. A subsequent non-response analysis was carried out with the 110 non-respondents, which suggested that there were no significant differences between respondents and non-respondents.
3. The Møre and Romsdal region The focus of this study is Møre and Romsdal, which was selected because it is a recognised region for innovation activities in traditional industries22 and has a higher share of total industry employment when compared with the Norwegian average (Table 1), and had one of the countries highest numbers of patents in both 1982 and
199223. As such, it may be termed a 'core' region in Norway. However, there are characteristics which distinguish it from other regions in Norway and core regions in other countries. First, in terms of gross value added, it is only the 9th largest in Norway (Figure 1). Second, there are the particular structural differences of the region, where the main industrial base is not founded on high technologies such as electronics, computers and so on, but is largely comprised of three main 'traditional' industries i.e. the manufacture of furniture, fabricated metal products (including shipbuilding) and fish products (see Figure 2 below). It is therefore useful to analyse the particular characteristics of the 'innovation system', and the innovation activities of firms, in such 'traditional' industries to see how these differ from other industrial sectors which have often received more attention.
Table 1. Industry employment as share of total employment and average unemployment rate in Norway and in Møre and Romsdal.
Source: Statistics Norway.
Wicken, O. (1994). 'Entrepenørskap i Møre and Romsdal. Et historisk perspektiv.' (Entrepreneurship in More og Romsdal. A historical perspective) STEP-report, 21/94 Haug R. and Skorge O. (1994) 'Patenter i Norge. Økonomisk utvikling, bedriftsstørrelse og lokalisering som forklaring på variasjon i antall patentsøknader i Norge', Siviløkonomoppgave Bødo Graduate School of Business 31.05.94.
Figure 1. Regional gross value added at market prices24.
Million Norwegian Kroner.
(Total before deduction of imputed output of bank services, by county.) 1986-1990.
25.0 % 20.0 % 15.0 % 10.0 % 1990 5.0 % 0.0 %
Source: Statistics Norway.
Although the raw material bases for the traditional manufacturing industries are available often in abundance in other Norwegian regions, industry in Møre and Romsdal is perceived as particularly innovative in its use of these materials25.
Related to this the history of the region indicates that there are diversities which exist within the region, where innovation activities and industries differ across the three fogderi26 or sub-regions of Sunnmøre, Nordmøre and Romsdal which make up Møre and Romsdal. Further understanding of differences in innovation activities within the region, as discussed here, may be used to support and direct policy objectives in this area.
There is historical evidence to support the idea that Møre and Romsdal as an area has entrepreneural skills. Historically, collective entrepreneurship through cooperation in both productive and commercial phases of economic activity gave rise to an economic vitality in the rural districts and a positive attitude towards entrepreneurship. It seems that the tradition of collective entrepreneurship paved the way for individual entrepreneurs in these regions. The local community supported new enterprises by way of family, community or municipal support in terms of technical, financial and commercial support to initiatives taken by individuals 27. We shall suggest below that these historical dimensions of the region may still be visible in the contemporary data.
The methodological approach has been to allocate national accounts figures of gross value added (GVA) to regions by using distributional keys corresponding to each industry in the national accounts system.
see evidence from Haug and Skorge (1994) op cit. and Wicken O. (1994) op cit.
An archaic jurisdiction akin to a bailiwich.
Wicken O. (1994) op cit.
What Comprises a Regional Innovation System? An Empirical Study 9