Actor-oriented analysis of innovation systems: exploring micro–meso level linkages in the case of stationary fuel cells

Abstract

The innovation systems approach has been successfully established as a conceptual basis for an encompassing analysis of actors, networks and institutions that influence innovation processes. Here we present an approach that concentrates on the actors in a selected technological innovation system. The intention is to specify current or potential contributions of actors, or actor groups, to innovation system performance and dynamics. This will be achieved by explicitly relating actor level characteristics such as innovation strategies and resource endowments to system level characteristics like actor configurations, system functions and performance. We illustrate the approach with a case study on stationary fuel cells in Germany.

Keywords:

• innovation systems
• actor strategies
• resources
• stationary fuel cells

Notes

1. One of the very few studies we have found in this respect is a comparative analysis of firm level competences in the biomedical field and their impact on the performance of regional/national clusters (see Cetindamar and Laage-Hellman 2002).

2. See also the critique that meso level frameworks for the study of technological transitions tend to downplay the importance of agency in this respect (Smith, Sterling, and Berkhout 2005).

3. Much of the empirical data presented in this paper is based on an earlier study, which dealt primarily with the question of how the innovation activities of electric utility companies contribute to technological change in the energy sector (Markard, Truffer, and Imboden 2004; Markard 2004).

4. For a more detailed discussion of the distinction between radical and incremental innovation processes at the system level, see Markard and Truffer (2008). In Malerba (2004) a similar issue is raised with the differentiation of innovation, production and distribution systems.

5. In the following, we concentrate on actors and, for example, their impact on innovation system functions. However, institutions may also be conceptualised in a way as to influence system functions directly, not just through their impact on actors. Either way, the influence of institutions is left aside in the analysis, which is of course a simplification. In a very recent publication, we propose a framework for an explicit institutional analysis of innovation systems (Rohracher, Truffer, and Markard submitted).

6. Note that this definition of innovation activity at the micro-level is different from typologies of innovation activities that have been proposed at a more aggregated level. See Chaminade and Edquist (2005) for an overview.

7. Apart from innovation activities, institutions also have a major influence on the performance of an innovation system. Here, we largely ignore the role of institutions, cf. note 5. A fully fledged analysis of innovation system dynamics, however, would have to explicitly include this dimension as well as the relationships between the actors.

8. Alternatively, we could just differenciate external and internal mobilisation instead of resource and mobilisation deficits.

9. With regard to stationary power supply, for instance, a 200 kWel phosphor acid fuel cell (PAFC) by UTC Fuel Cells, has been made commercially available in 1991. Until 2004, 270 of these units have been installed worldwide (cf. Baker and Jollie 2004).

10. See Carrette, Friedrich, and Stimming (2001) for a technology overview and for a review on recent market developments Adamson (2005, 2006).

11. Depending on the fuel cell type the primary fuel is either used directly or pre-processed in a reformer to extract the hydrogen.

12. See Carlsson et al. (2002b) or Bergek et al. (2008) with regard to the question of how to delineate technological innovation systems.

13. See Geiger (2003). These figures include all, i.e. stationary, mobile and portable fuel cell innovation activities. The 350 actors include research institutes (28%), manufacturers and suppliers (28%), car manufacturers and utility companies (12%) as well as policy makers and associations (12%).

14. For a more encompassing collection of particular actors, see Geiger (2003).

15. In 2005, for example, Sulzer, the incumbent behind Hexis in Switzerland, decided to end their commitment to stationary fuel cells. Although Hexis was a leading manufacturer in stationary solid oxide fuel cell technology at that time, Sulzer had a hard time to find an investor. Finally, the development activities at Hexis had to be cut back dramatically before a new risk taker took over.

16. This restriction is related to the assumption that utility companies control a more or less comparable set of resources, on which basis they can be defined as an actor group. While this is empirically supportable in most cases with regard to the resource access, actual resource amounts in each firm may differ significantly.

17. Altogether 13 interviews with utility employees in response to the fuel cell development activities were conducted in 2002 and 2003 (follow-up). Our sample included eight companies, six of which were located in Germany, one in Switzerland and one in The Netherlands. The two non-German companies were included for two reasons. First, a minimum sample size was needed, e.g. in order to identify strategic similarities. Second, we wanted to check whether firms under a different institutional framework show a much different strategic behaviour (which was not the case).

18. For a more detailed description of how the strategy types were identified, see Markard, Truffer, and Imboden (2004).

19. Here we assume that knowledge created and applied within one organisation serves the innovation system as a whole. We did not check, however, to what extent the knowledge was externalised.

20. Whether such a scenario will become reality is, of course, not just a question of resource endowments but also of actor strategies (cf. Section 4) and other influences such as technology development, institutional change, external events, etc.

21. See Markard et al. (submitted) with regard to a more explicit analysis of future development options of innovation systems.