تجزیه و تحلیل های سیستم نوآوری و انتقال پایداری: مشارکت و پیشنهاداتی برای تحقیقات

This paper argues (1) that technology-specific policies are necessary if we are to meet the climate challenge and (2) that a main contribution of innovation system (IS) analysis to the study of sustainability transitions is that it allows policy makers to identify the processes and components in a system where intervention is likely to matter most. We demonstrate that an IS framework can identify a diverse set of system weaknesses in the field of environmental innovation and identify five venues for further research that can help strengthen the framework and improve its application to environmental innovations.

Greenhouse gas emissions need to be reduced by 80% if a stabilisation is to occur in terms of concentrations of carbon dioxide (CO2) equivalents at the level of 500 ppm (Stern, 2006). Much of this transformation needs to have taken place by 2050, and will entail an almost complete decarbonisation of the electricity sector. To achieve this within the suggested time frame is a formidable challenge, which requires that adequate climate policies are put into place. The neoclassical guidance for policy is the “market failure” approach. With respect to the introduction of new energy technologies, two failures are usually emphasized: positive knowledge externalities and negative environmental externalities (e.g., Jaffe et al., 2002). With respect to the former, the argument is that firms tend to under-invest in research and development compared with a socially optimal level due to, e.g., knowledge spill-overs that benefit other actors than the investor.1 The policy approach to handling this inability to appropriate the full benefits of R&D is to fund basic R&D and co-fund some industrial R&D. With respect to the latter, it is generally accepted that energy markets fail to internalise the environmental costs of energy supply. Favoured solutions are to either make polluters pay or to make good environmental performance someone's property and create markets where this property can be traded. Policy instruments that do not distinguish between different technologies are usually advocated, for example general CO2 taxes, tradable emission permits and tradable green certificates. Under the influence of such instruments, the main selection mechanism is marginal cost; investments will first occur in the currently most cost-efficient technologies and only after these encounter rising cost levels will more costly technologies be fostered. Market-failure based climate policies, thus, rest on two legs: R&D support and general, often “market-based”, economic incentives. In this paper we will argue, with Azar and Sandén in this volume, that such general policies are not enough to stimulate sustainability transitions; technology-specific policies are also required. However, implementing such policies raises the question of how policy makers can identify the processes that are of critical importance to the dynamics of specific technologies and to which policy intervention should be addressed, i.e. where intervention is required.2 This is where innovation system analyses make their prime contribution to the study of sustainability transitions. “Innovation systems” was developed as a policy concept in the mid-1980s. In the context of debates over industrial policy in Europe and as a reaction to perceived inadequacies of neoclassical economics and the spread of neoliberalism, a range of system approaches emerged (Sharif, 2006). These include national innovation systems (NIS) (Freeman, 1987 and Lundvall, 1992), sectoral innovation systems (SIS) (Breschi and Malerba, 1997 and Malerba, 2004), technological innovation systems (TISs) (Carlsson and Stankiewicz, 1991) and regional innovation systems (RISs) (Cooke, 1996). Whilst differing in system boundaries, these approaches have many shared features, in particular, that the innovation and diffusion process is both a collective and an individual act. The various systems are, furthermore, interdependent. For instance, an innovation system that is specific to a technology is situated in a context of systems at higher levels of aggregation (NIS, RIS, SIS) (Markard and Truffer, 2008). The key contribution of innovation system analyses to the study of sustainability transitions is, we argue, that it provides policy makers with a tool for identifying system weaknesses. It promises, therefore, to inform policy makers of the problems that an intervention needs to solve in order to promote the growth of a particular system or to influence its direction. So far, the bulk of innovation system scholars have not focussed on technologies within the environmental field.3 It is primarily the technological innovation system (TIS) approach that has been used to study the emergence of new energy technologies. This has given rise to a fairly large literature which provides an opportunity to review what we have learnt with respect to its capacity to inform the policy making process by identifying system weaknesses. The purpose of this paper is therefore twofold: (1) to take stock of the achievements of innovation systems research in identifying system weaknesses in the area of environmental innovations and (2) to provide some ideas for further research that can improve its usefulness for policy in this field. The structure of the paper is as follows. Section 2 develops the arguments for technology-specific policies. Section 3 discusses system weaknesses as a core concept in innovation system analyses. Section 4 presents a selective review of the literature on innovation systems and sustainability, demonstrating the diversity of system weaknesses identified. Section 5 states the main conclusions and identifies five venues for further research.

Building on work in the 1990s on system failures and an extension of the IS framework with a stronger process focus, we have demonstrated the ability of the TIS framework to identify a diverse set of system weaknesses in the area of environmental innovations. This ability contributes to the study of sustainability transitions in that it allows policy makers to identify the points in a system where intervention is likely to matter the most, opening up for rich but also challenging climate policies. As Geels et al. (2008, pp. 524–525) put it, with respect to socio-technical approaches in general: “Socio-technical approaches refrain from simple policy recipes, because they highlight co-evolution, multi-dimensionality, complexity and multi-actor processes. They argue that constellations of policy instruments should vary, depending on specific challenges, opportunities and problems in sectors, technologies and social networks. While this message may be unpleasant for policy makers who hope for silver bullet solutions, we argue that a deeper understanding of socio-technical dynamics provides policy makers (and other actors) with a more solid base for policy interventions.” In what follows, we will outline areas for further research that can improve the usefulness of the framework for policy in the field of environmental innovations. 5.1. Measuring functionality A central task in a TIS analysis is to assess the strength of the functions. This can be done in a number of ways which include conventional indicators such as patents but also less conventional ones such as measures of the supply of specialized human capital and of the legitimacy of a new technology. Interview based assessments are common (e.g. Hellsmark, 2010). Negro et al. (2008) and Suurs (2009) combine these with quantitative analyses of events. van Alphen (2011) uses expert assessment to quantify the strength of the functions in five countries. A number of tools have, thus, been tried but as yet, no standard combination of indicators for measuring the strength of the functions has been developed. The avenue for research of methodological nature is, therefore, quite open. A first research venue would be to review and empirically test possible indicators for each function, the aim of which is to recommend a standard set of indicators that allows for comparisons across technologies, time and space in the field of environmental innovations. 5.2. Interaction of TIS with higher system levels As mentioned above, the functional approach was developed to handle an integration of technology-specific and more general influencing factors. Yet, as argued by Borup et al. (2007), there is a risk that a focus on a specific technology leads us to overlook the higher system levels (NIS, RIS, SIS). This is problematic since new TIS are formed through accessing resources from their contexts. In particular, new entrants into the TIS come from related industries (Porter, 1990). Bergek and Jacobsson (2003) pointed to how German entrants into the wind turbine industry came from a range of industries, such as shipbuilding and gearbox manufacturing, and Cooke (2010) explored the concept of relatedness at the regional level. Hellsmark (2010) developed the relationship between “system builders” and their contexts in the case of second generation biofuels and argued that the context in which a new TIS emerges influences not only the resources available but also the direction in which the TIS grows. For instance, in the German case, he situates the system builders in a rich SIS and NIS structure composed of a range of capital goods firms and transport equipment manufacturers as well as a diverse institutional framework, composed of both technology-neutral and technology-specific instruments for stimulating the emergence of renewable energy technologies. The system builders (Hellsmark, 2010, pp. 324–325) “… mobilise resources and attract actors from the existing and primarily national industry structure by aligning the technology to their interests and existing technologies. For example, the German system builder … FZK could draw resources from the incumbent gasification capital goods industries by offering a solution suitable for its existing reactors and downstream processes used for fossil gasification … the system builders have made use of the structure in which they are embedded to create elements of a new structure specific for biomass gasification.” Understanding the formation of a TIS and, indeed, assessing the likelihood of its emergence, requires therefore systematic and deep insights into its larger context. A second research venue would be to complement analyses at the TIS level with studies of SIS and NIS levels with respect to environmental innovations. Another aspect of the relationship between TIS and higher system levels is the former's impact on the latter. The growth of new TIS has been conceptualised as a process of cumulative causation involving both functional and structural dynamics (e.g. Jacobsson and Bergek, 2004, Suurs, 2009 and Suurs et al., 2010). Some of the structural change may go beyond the TIS level and spill over to higher system levels.30 For instance, the German feed-in law came about as a result of combined lobbying from the German industry associations for small scale hydro and wind power (Jacobsson and Lauber, 2006). Advocates of these technologies consequently had an impact on the SIS level institutional context in Germany. As Negro (2007) demonstrated, it led to strategic investments by firms to develop a turnkey capability in the field of biomass digestion. The initial efforts to align institutions to the need of small scale hydro and wind power thus came to benefit, through positive externalities, other emerging TIS, now sharing an institutional relatedness. As a feed-in law is one (of several) candidates for forming a market for second generation biofuels, the dynamics may eventually go beyond the SIS level, impacting on the NIS level. In policy assessments, such wider repercussions have to be included and a third venue for research is to improve our understanding of externalities that cross system boundaries. 31 5.3. Competence, organisation and politics of policy Acquiring the deep understanding that is required to pursue rich and challenging climate policies implies that a key topic for research is the competence, organisation and integrity of public policy bodies. Identifying the relevant system weaknesses (at various system levels) necessitate that policy makers must have a high analytical and deep domain-specific competence.32 Moreover, using system thinking implies that a range of government bodies, at various administrative levels, needs to be involved in the analysis and implementation of policy. No ministry or other government body can be expected to have all the required instruments in its arsenal. In addition to domain-specific competence, competence (and the power to use it) to coordinate policy intervention must, therefore, exist (Teubal, 2001). A fourth research venue is to analyse the competences and organisational set-ups in different countries and see how they affect policy making in the environmental field. This competence should, arguably, be combined with a high integrity. Transformation processes are characterised by a great deal of uncertainty in several dimensions (Rosenberg, 1996 and Meijer et al., 2007). Given these uncertainties, much lobbying work is undertaken by advocacy coalitions for low carbon technologies to influence the perceived desirability of these. Ultimately, the objective is to shape expectations of policy makers. This is done in many ways, some of which are subtle and difficult to unmask. Lobbying can also have the opposite purpose. Firms (and other organisations) are frequently adverse to a technology that may threaten their position or they may be adverse to policy scheme that places a financial burden on industry. For instance, the Austrian feed-in law was weakened after a successful lobbying efforts by the Austrian paper and pulp industry (Hellsmark and Jacobsson, 2009). Policy makers, consequently, have to manoeuvre in a political minefield and must develop an independent competence to critically assess attempts to shape expectations as well as to form an own vision of the future of a specific technology or groups of technologies (Geels et al., 2008 and Watson, 2008). A fifth and final research venue would be to address the “politics of policy” by, for instance, tracing the decision making process for legislative changes of relevance to environmental innovations, identify external influences and relate the impact of those to the competence of the involved policy makers.