What governs the transition to a sustainable hydrogen economy?

Matthijs Hisschemllera, Ries Bodeb and Marleen van de Kerkhofb a. Institute for Environmental Studies (IVM), Vrije Universiteit, De Boelelaan 1087, 1081 HV Amsterdam, The Netherlands, [corresponding author: matthijs.hisschemoller@ivm.falw.vu.nl] b. Institute for Environmental Studies (IVM), Vrije Universiteit, De Boelelaan 1087, 1081 HV Amsterdam, The Netherlands
Abstract There is a lack of integrated knowledge on the transition to a sustainable energy system. The paper focuses on the relationship between technologies and institutions in the field of hydrogen from the perspective of political theory. The paper unfolds four paradigms of governance: Governance by policy networking, Governance by government, Governance by corporate business, and Governance by challenge, and looks into the major line of argument in support of these paradigms and into their possible bias with respect to hydrogen options. Each of these paradigms reveals an institutional bias in that it articulates specific opportunities for collaboration and competition in order to stimulate the transition to a sustainable hydrogen economy. The paper makes the observation that there is a compelling need to reframe fashionable discourse such as the necessary shift from government to governance or from government to market. Instead, specific questions with respect to the impact of guiding policy frameworks on innovation will highlight that neither neutral nor optimal frameworks for policy making exist, where competing hydrogen options are at stake. The identification of paradigms of governance may be considered a methodological device for (participator) policy analysis.

IVM Working Paper: Keywords: Date:

IVM 06/02 governance; hydrogen economy; energy innovation, transition 23 February 2006

Acknowledgements
This article reports on the findings of the study Paradigms of governing the transition to a sustainable energy system and their implications for technology development: A critical reflection. This study was supported by the NWO/NOVEM Programme to stimulate energy research. This article is also embedded in the current project Governing the transition towards a hydrogen economy: Evaluation of competing hydrogen strategies from a multi-level perspective, supported by the NWO ACTS Sustainable Hydrogen Programme (see: http://www.nwo.nl/ACTS and http://www.h2dialoog.nl).

About IVM
The Institute for Environmental Studies (Instituut voor Milieuvraagstukken, IVM) is the oldest academic environmental research institute in the Netherlands. Since its creation in 1971, IVM has built up considerable experience in dealing with the complexities of environmental problems. Its purpose is to contribute to sustainable development and the rehabilitation and preservation of the environment through academic research and training. The institute has repeatedly been evaluated as the best Dutch research group in this field.
Institute for Environmental Studies Vrije Universiteit De Boelelaan 1087 1081 HV AMSTERDAM The Netherlands Tel. +31 (0)20-5989 555 Fax. +31 (0)20-5989 553 Web: http://www.vu.nl/ivm

1. Introduction The idea that viable technologies conquer the market by virtue of their advantages for individual consumers and society at large, seems nave to all who work in the field of the adoption of innovations. Technologists, economists and policy scientists are only too familiar with the notion that even the best technologies may fail if the necessary socio-economic conditions are not in place (Berkhout et al., 2003). Market conditions for innovations are thereby largely shaped by government policies. One of the basic contributions governments could make is the creation of niche markets, where innovations can fully develop into marketable products. Yet, the governance of technological innovation faces specific difficulties in an era of liberalization of the energy sector. On the one hand, international obligations, e.g. the climate change regime and the emerging issue of security of supply, may urge both the EU and national governments to take action. On the other hand, national and EU policies that go beyond the scope of supporting R&D, may suffer from a lack of legitimacy and credibility. It could be argued that, whereas the market may fail in realizing innovations that trigger transitions towards more sustainable energy efficient systems, government does not have a very good track record either (Luiten, 2001). The current state of affairs in energy policy and science is featured by a lack of integrated knowledge on the transition to a sustainable energy system. Technologists, economists, political scientists and lawyers hardly interact with respect to their disciplinary insights of the issue. This article aims to make a modest contribution to knowledge integration by focusing on the relationship between technologies and institutions. We start with the assumption that energy options may, apart from specific characteristics related to technology, costs and public perception of risk, have institutional requirements. In turn, political institutions show a preference for specific technologies and a dislike of others. By institutional requirements we refer to (a) the guiding policy framework, including laws, regulations, instruments, measures, (b) networks of organizations and actors, which articulate the dominant dimensions of conflict and consensus in a society and (c) the formal and informal rules of the game that shape actors behaviour as well as the actual implementation of policies and laws. 1 As comes out from a Dutch stakeholder dialogue on climate options for the long term, the option CO2 sequestration and storage is believed to benefit from incentives for institutions that provide incentives for competition through market instruments (tradable permit system) (Hisschemller, 2003). However, break-throughs in the field of renewables are believed to benefit from incentives for collaboration, especially the formation of new networks between vested interests and newcomers (Hisschemller and Van de Kerkhof, 2004). Yet, to make the picture more complete,

In political science literature the concepts regime and institution are used interchangeably. In the narrow definition they both refer to the formal and informal rules of the game that shape actors behaviour (Young, 1999). This definition excludes organizations (actors) from being institutions. In contrast, we use a more inclusive definition of institution, referring to rules that shape behaviour of organizations as well as to organizations shaping others behaviour by the creation and application of rules.

however more complex, institutions provide formal and informal rules that articulate both competition and collaboration in a specific way. This contribution reports on a study into the relationship between technologies and institutions, which takes the somewhat unusual perspective of political theory. A critical reading of selected political theory and philosophy has helped us to unfold four paradigms of governance with respect to the transition to a sustainable energy system. These paradigms are: Governance by policy networking, Governance by government, Governance by corporate business, and Governance by challenge. In the sections to come, we will present the major line of argument in support of these four paradigms and look into their possible bias with respect to hydrogen options. The concluding section will address the question how these paradigms can assist policy analysis and participatory integrated assessment.

2. Paradigms of Governance A paradigm is understood as a set of assumptions with respect to a certain policy problem, i.e. a problem which is supposed to require some sort of collective action. Analogue to scientific paradigms in the sense of Kuhn (1962), paradigms of governance are fundamental concepts of viewing the dynamics of society and the way in which governments and/or governance can or cannot give direction to those dynamics. They articulate a consistent line of argument, which includes causes and effects, goals and means, and problems and solutions In order to enable a systematic comparison between the paradigms of governance, we use the same argumentative framework for all (based on Toulmin, 1969). The presentation of each paradigm starts with a distinctive policy claim, which relates to the strategic principles that the paradigm puts forward such as, for instance: leave it to the private sector, which has the power, the knowledge and the ability to make the transition happen! Then, we focus on warrants for this claim. These why questions can be answered by pointing to historic events, (causal) empirical evidence from the present, and normative reasoning. It should be noticed that each paradigm is constructed in such a way that it underlies normative as well as causal argument. Furthermore, for each paradigm, it is tentatively explored what kind of technologies could be advanced or blocked under the respective political paradigm. Finally, each paradigm also includes a rebuttal that is to identify (a) condition(s) under which the paradigm is likely to be unworkable. 3. Governance by policy networking This paradigm claims that the traditional instruments of the state to control and shape society do not work anymore. The power of the traditional nation state is weakening and, instead of enforcing its will on society, the state may help private actors to jointly realize the public interest. Therefore, government facilitates the formation and maintenance of policy networks consisting of socio-economic actors. These networks are supposed to have the knowledge, the power and the inclination to foster some

public interest, such as hydrogen applications. In the Netherlands, with its traditions of enlightened corporatism (polder model) this paradigm appears to be dominant, and is defended by a great deal of policy scientists (De Bruin and Ten Heuvelhof, 1999; Driessen & Glasbergen, 2000). The paradigm is an international trend as well though. Generally speaking, this paradigm is based on a number of assumptions. First of all, there is a shift from government to governance. Whereas government refers to a monocentric interventionist perspective, governance refers to a polycentric model, which envisions concerted actions by decentralized actors, each of which has only limited coercive capacity. The model includes not only governmental actors, but also stakeholders from civil society, especially the business communities and the environmental and consumer NGOs. Secondly, in policy networking, participation becomes associated with co-production rather than with opposition, and with workshops rather than with public hearings. This is observable in national policy making as well as in international environmental negotiations. Thirdly, governance is multi layered, which refers to the way policy making at different levels (local, national, European, global) affect each other. Also, from a normative perspective, this concept refers to the need for better coordination and collaboration between different levels of governance, given its bad track record, which comes out from many studies on implementation. Fourthly, one of the main factors responsible for the participatory trend in environmental policy is the changing role of science and expertise. The status and privilege of scientific knowledge has declined and is likely to decline further, as (environmental) problems are conceived of as increasingly complex (Ezrahi, 1990). Some striking developments during the last decades in energy policy worldwide can illustrate the rise of this paradigm in theory and practice. Traditionally, energy policy was in the realm of national government, especially where issues of environmental risk are concerned. Until recently, many national governments owned companies with the goal to produce and allocate collective goods and services, such as energy. Especially the building and exploitation of large infrastructures became a nation state responsibility, including the funding and the legal and coercive framework needed to implement and maintain the quality of infrastructures and the handling of collective resistance against unwanted facilities. In the Netherlands, national government has, in the case of natural gas, shared the ownership and exploitation of natural resources, but the import and distribution of transport fuels have always been a private activity. In case of the allocation and protection of collective goods and services that are associated with a national interest in 20th centurys industrial states energy being a typical example , political systems (states), even the most pluralist ones, have acted as monolithic actors vis--vis citizens and business. This is also true for The Netherlands, where major decisions related to energy policy, such as closing down the coal mines in South-Limburg (Moharir, 1979), the transition to natural gas (Correlj and Verbong, 2002) or the decision not to build new nuclear power stations (Hisschemller and Midden, 1989), were based on a political consensus. Open conflict was avoided and where this turned out impossible as in the case of nuclear power, it resulted in non-decisions c.q. inaction. Governance by policy networking may be considered a corollary of accommodation politics (see Lijphart, 1968), especially because of its focus on consensus formation between political actors and stakeholders. The Dutch model of environmental agreements between government and business have become famous and are

considered to be an example by many international scholars. Major Dutch corporations have agreed to maintain their leading international position in the field of energy efficiency, while in exchange Dutch government has promised not to take reduction measures that could be harmful for the position of Dutch industry in international competition. This agreement does not affect the level of emissions, as in a period of economic prosperity emissions can go up in spite of improved energy efficiency. Transition management may also be considered to fit in nicely with the Networking paradigm. Consensus on transition goals can be considered a means to stimulate the private sector to take its social responsibility without specific government regulations. However, behind the scene, the possibility of regulation is always there, but the likelihood thereof is not that big, given the shared recognition that parties are dependant upon one another in order to move forward. Another and most remarkable feature of the policy networking paradigm in practice, is the focus on collaboration and, which is the other side of the medal, the lack of competition. To the extent that there is competition between transition views at the political level, the controversy is rather symbolic and does not seem to have much bearing on practice, an example being debates on (long-term) environmental targets. And although at the level of specific solutions creative competition (Teisman, 1997) may be part of the game, Governance by networking suggests that sharing expertise can lead to the best results both for separate companies as for national economy. In the field of energy this seems to be confirmed by the very high correlation found between the overall energy-efficiency of the economies of 36 democracies and the degree of consensual characteristics of these democracies, a correlation that is unaffected by the introduction of the level of economic development as a control variable (Lijphart, 1999). However, Governance by policy networking may also be vulnerable to the critique that it yields conservatism rather than innovation. The networks to be governed become part and parcel of the governance setting itself, in other words; they become institutionalized and resistant to innovative views and actors. In the absence of rival views networks may extend and reproduce specific discourses. Apart from their important symbolic function, discourse may also perform as an important mechanism for inclusion and exclusion. So, it maybe hypothesized that Governance by policy networking, ironically and sadly, may have the unintended effect to prevent what it aims for, i.e. the stimulation of technological innovations. In a recent study on the social and political implications of the project The Greenhouse as a Source of Energy initiated by the Innovation Network and the sector, it was concluded that perhaps the most significant threat for this technological break-through concept may not be lack of support in the sector, but a government based discourse which seems to a priori ignore the feasibility of this concept for the decade to come (Hisschemller et al., 2003). 4. Governance by government This paradigm claims that government should do what citizens or private actors cannot, which, is to safeguard the public interest. The warrant for this is twofold. Firstly, private actors will not voluntarily adjust their behaviour as to realize a public good. Secondly, it is at the hart of (representative) democracy that public affairs are

dealt with by an accountable public agent. The first warrant will be elaborated and justified by using Olsons theory of the logic of collective action (Olson, 1971); for the second warrant we use the argument developed by Ezrahi (1990). Olson rejected the view that groups will organize themselves to promote a collective interest and argued that people will not voluntarily join in defence of a collective good (i.e. a common interest), of which a sustainable energy system is a good example. This is not because people are selfish. Also an altruist person would not act in favour of a collective good, if he makes a rational judgment, since his own contribution would not be perceptible (1971: 64). After all, you never know as to whether others do the same for the public good. The social isolation of individuals, especially when the group whose interest is at stake is large, is the primary cause of free-rider behaviour. So, what Olson shows is the existence of what has been labelled social dilemmas, i.e. discrepancies between collective and individual interest. 2 Olsons theory can be situated in a long tradition of thinking about the legitimacy of a coercive state vis--vis its citizens, of which Hobbes is considered one of the founding fathers. There is evidence to endorse that citizens are aware of the existence of social dilemmas and therefore tend to support government policies that makes them behave in a way they would not voluntary do by themselves. In exchange, citizens must be able to trust some basic qualities of the policy-making system: in particular, the benefits of government interventions should become visible for them. This is a necessary prerequisite for citizens to hold government accountable for its interventions. Ezrahi (1990) stresses the role of the citizen as a capable witness in democratic politics (American politics being his main example), giving acte de presence to actively observe real life experiments of social engineering. In this view, government acts, takes the initiative, experiments and, through trial and error and with the help of science and technology, works to realize an improvement of society. The citizen becomes an active lay-witness. Science may develop into what Lasswell (1951) has labelled the policy sciences of democracy, which was designed to assist policy makers in addressing the major social problems of their time, poverty, unemployment, racism and education. In this vision of democratic governance, what citizens evaluate is not political rhetoric. What counts in the end is as to whether policies have worked in practice. There are two conditions for making the paradigm Governance by government work, which makes it - in our view - both salient and problematic. The first is that there should be a visible act, and some kind of visible / measurable effect of government intervention; the second is that there must be consensus with respect to the public goods (values) to be protected or realized, and that there is an objective knowledge available in support of specific actions. In order to realize a public good, the number of means, interventions and instruments that government has at its disposal is basically unlimited. In case of innovation policy, it may establish generic policies but also support specific technologies; it may even perform as an entrepreneur by making large scale investments in a hydrogen
2

Similar metaphors are the tragedy of the commons (Hardin, 1968) or the prisoners dilemma (Luce and Raiffa, 1957).

economy. Ezrahi (see also Price, 1965; Castells, 1996) point to the experience with Governance by government as an innovation strategy, the governing of innovation of the American national aerospace and defence sectors during the 1940s 1960. Huge funds were spent on research and the production of weapons. In fact, these sectors actually functioned as niches for the development and testing of new technologies. This strategy, which Ezrahi labels as Governance by contract, served several functions at the same time: Firstly, the public-private collaboration served to develop, test and implement specific technical innovations, whereby government acted as what may be referred to as an early mover. This innovation strategy was justified by the size of the costs and risks of the innovations, a risk no private actor could bear on its own. The strategy contributed to national prosperity and employment. But there was also another, more symbolic function: Through the involvement of private parties, they became integrated as agents of public actions (Ezrahi, 1990: 43). The private firms that contributed to the public cause in the US became to symbolize the collective action of the American people. Hence, Governance by Government became legitimized by an appeal to a strategic national interest and -simultaneously- created the conditions for people to feel involved. As this example shows, Governance by government may, in spite of the rhetoric of a free market system, focus on policies aimed at specific technology development and support. By that time there were two conditions that made the strategy work. The first was a successful appeal to an overriding collective good: security. The second was the general perception of science and technology being impartial and objective, which helped to maintain the integrity of the system. The existence of a compelling national interest made it possible to bring to bear a seemingly unlimited amount of resources. The distribution of these resources among the private sector was based on no other considerations but one, i.e. to obtain given objectives. Policy was neither evaluated in terms of efficiency nor in terms of the quality of its underlying procedures. For an understanding of the limitations of Governance by government it is essential to note that a large consensus is a prerogative for any strategy based on this paradigm. Consensus relates, on the one hand, to the public goods (values) to be protected and, on the other, to the objective knowledge base available in support of specific actions. In a not that often quoted paragraph, Olson is the first to acknowledge that his theory on collective goods underlies this unrealistic assumption of full consensus: But the results obtained under this assumption are, for that reason, all the stronger, for if voluntary, rational action cannot enable a large, latent group to organize for action to achieve its collective goals, even with perfect consensus, then a fortiori this conclusion should hold in the real world, where consensus is usually incomplete and often altogether absent (Olson, 1971: 60). And he adds a warning for those who too eagerly may want to apply his theory with respect to state interventions in practice: It is thus very important to distinguish between the obstacles to group-oriented action that are due to a lack of group consensus and those that are due to a lack of individual consensus. In other words, coercion by government policies may only work if the citizens agree on the value and relevance of the public good under consideration. Many have acknowledged the changing role of science and expertise in democratic societies. Rather than a means to impartially solve social problems, scientific knowledge has become contested and debated, especially where new technologies and their potential risks are at stake. The world wide controversy on the risks of nuclear

power belongs to the most cited examples. The nuclear power debate had implications that go far beyond the area of technology policy. It has strongly contributed to a retreat of government in its capacity to support specific technologies. It has led to reconsider the relevance of public participation. And it has shaped a more critical attitude with respect to the often privileged position of expert networks in decisionmaking, giving more attention to practical or lay knowledge. Notwithstanding its built-in limitations, it may be too early to state that for Governance by government the book is closed. In case of a shared awareness and a high sense of urgency, governments may deviate from their course of liberalization and privatization and take the transition of the energy system more in their own hands. The paradigm Governance by Government might still guide a collaborative attempt to implement hydrogen options that require huge infrastructure investments, i.e. investments far beyond those to support R&D. As in the case of nuclear power, such a strategy may be confronted with the need to address potentially large-scale environmental risks. Also today, its appeal to both public and private actors is the promise that collective action, a joint effort to invest as early mover in a precompetitive stage, may be effective in realizing a transition path to a sustainable energy system.

5. Governance by corporate business This paradigm claims that the private sector, especially corporate business has the power, the knowledge and the ability to make the transition happen. In particular the work of the economist and political scientist Schumpeter is interesting to support this claim 3 . In his monumental work Business Cycles (1939) Schumpeter discerns three interrelated cycles of the capitalist economies: Kitchin cycles, that lasted from top to top 3 to 4 years, Juglars, that took 8 or 9 years and Kondratieff waves of about 50 years. Much has been written about his suppositions and the statistical flaws with long-term statistical time series. But for us it is more interesting to learn about the driving forces behind these movements. Much more than other economists Schumpeter focused on entrepreneurs and innovation. An entrepreneur is a person who creates a new production function, a new combination of production factors. It all evolves around leadership, an insight neglected by Marx and the other classical economists according to Schumpeter. Nothing keeps us from considering energy as one of the production factors that play a vital role in new combinations.

Competition between private companies is central in Schumpeters view on technological innovation. Even companies that have reached monopoly-like positions will and have to keep on innovating to prevent themselves from loosing their position in the long run. In what he calls trustified capitalism they are permanently making
3

Joseph Schumpeter is to be considered the latest of the representatives of the Austrian school; the group of economists that ended with their theory of marginality the efforts of the classical economists to build a theory of value of all economic goods and services, based on production costs and in the end on a labour theory of value.

10

R&D efforts to maintain their leading positions. An additional aspect of Schumpeters perspective is that simple cost-benefit calculations are considered inadequate. His concept of creative destruction denotes that what may seem destroying production capacity at a too early stage from a cost-benefit perspective, is in fact an inevitable consequence of the capitalist dynamic. Instead, it is a sign of insufficient dynamics if governments and vested interests would enable the existing production functions to live their physical lives.

For our purpose it is interesting to see that according to Schumpeter, assessing the economic developments in the US at the end of the thirties the electro-technical industry, the automobile sector and related sectors like rubber industry and chemical industry behaved in conformity with his theory on business cycles. So, there was a reasonable expansion in many important sectors, together with many small innovations. But the electricity sector lagged behind in a relative strong manner and the only substantial expansion was via public works, and disappointingly little via privately owned utilities. In Schumpeters view, this was caused by the anticapitalistic character of the public investments during the thirties, which gave rise to fear on the side of private initiative of public company competition.

It is not so much the existence of public investments in the power sector that worries Schumpeter. He states that, if there is a strong state that is not subject to control by specific group interests but has clear goals, a well-directed use of monetary means that prevents waste and inflation, and if the use of means is additive to what the entrepreneurs would have done anyhow, a state can play an entrepreneurial role. It is the political context and the modalities of state intervention in the power sector that caused the problems. Schumpeter points to the enormous investments done by Nazi Germany in energy production technology (via Braunkohle AG), including hydrogenation, that were motivated by an autarkic vision. The problem he saw was not the investments themselves, but accompanying policies that prohibited investments in other energy technologies, ruling out competition between alternative technologies (Schumpeter, 1939: 975). Schumpeters argument for a leading role of corporate business especially in capitalist societies is to some extent reflected in recent publications from various sides. Jeremy Rifkins Creation of the Hydrogen Economy (2002) tends to bear Schumpeterian traces. His long-term view on the economics of energy production and consumption systems, his neglect of cost aspects, his focus on returns, investment possibilities, and on the initiatives of (big) businesses all witness that inspiration. In some significant respect, also Shell Internationals scenario study (2001) echoes the Schumpeterian view. The scenarios recognize the importance of climate reduction targets, the Kyoto protocol and other government policies and interventions. But what counts in the end is how the energy companies address international developments and public demand. This claim is warranted by the observation that the driving force behind corporate strategy is consumer demand. To take this line of argument a little further, the bigger the corporation, the more it tends to focus on the preferences and concerns of the

11

general public. From this perspective, climate policies are by far not the only relevant drivers to a sustainable energy future. Indeed, one Shell scenario, called Business as usual, is based on considerations such as cleanness, security and ultimately sustainability. The development of renewables, like solar, is important in this perspective. The other scenario, called The spirit of the coming age, focuses on the development of (natural gas based) hydrogen, not so much on climate or energy considerations, but because consumers are supposed to prefer a superior technology. The fuel cell will become popular because of the huge increase in consumer comfort. Cost considerations do not play a remarkable role in any technological trajectory for the long term, in spite of what is so often assumed in public policy statements. In conclusion, it might be appealing for politics to follow (corporate) business. This would not mean that government has nothing left to do. To the contrary, companies know exactly when and how to find government if they need support for some activity, which embodies both a private and a public interest. Some might argue that this is already a long-standing practice. For others, the idea that corporate business works to a sustainable energy future might sound very unrealistic as long as the world controls and uses relatively cheap oil and other fossil fuels. In this respect one of the basic Schumpeterian assumptions may be questioned: are not the many failures the big privatization projects of public utilities are due to the neglect of the relevance of entrepreneurship, since the leaders of energy companies are in many instances former civil servants? Are international energy companies still capable of innovation, since they have turned into large quasi-political bureaucracies themselves? From a government perspective, the idea that corporate business might lead the transition to a sustainable energy system might also be frightening. Increased practice of accountability vis--vis the public and openness to some form of control may mitigate public suspicion and distrust. Especially where government lacks the knowledge and where corporate businesses has a firm stake in public trust, like in food production, pharmaceuticals, private healthcare, and quality labelling, government may loose relevance. In this connection, energy companies might initiate actions with respect to security of supply or the quality and safety of energy in a crisis situation as e.g. Shell did during the oil crisis in the seventies on behalf of the Dutch government. So, the risk (or blessing?) inherently linked to the paradigm Governance by Corporate Business is, that, in the end, in our type of societies, politics is rendered irrelevant as far as technological innovation is concerned.

6. Governance by Challenge The main policy claim within this paradigm is that government should address rules, regulations and privileges that may stand in the way of innovation. Government interventions in the market are justified if they actually improve competition between existing and new options and strengthen consumer sovereignty. This paradigm is supported by theories on technological innovation (lock-ins and lock-outs); by critical social and political theories that focus on the knowledge implications of power relationships in society, and by theories that discuss the risks of a too powerful state. Literature on technological innovation has pointed to the importance of technological regimes. A technological regime links scientific and technological knowledge and

12

expertise to a set of applications and actors or sectors of economy. Innovation can occur within and across technological regimes. The inclination to draw upon knowledge and knowledge applications available in a regime may lead to incremental growth within a given technology path, which is referred to as lock-in. When knowledge from the outside is permitted access to a regime, we may witness nonincremental innovations, leading to new technological concepts and new applications. It is stressed that in processes of technological development, innovation and diffusion, vested interests specialized or locked-in in the dominant regime, resist the adoption of new technologies. (Mulder: 2003: 31). Also from a political science and sociology perspective, non-incremental change seldom goes without struggle. As Castells (1996) and e.g. Florida (2002) recognized: the breakthrough of new pervasive technologies does not take place within the dominant networks but in networks that are peripheral at first and even based on deviant subcultures. Castells and others thus point to the observation that technological innovation is accompanied by a new articulation of power. In an analogy with Galbraiths (1952) concept of countervailing power the concept of countervailing networks can be introduced. Such networks may challenge and undermine the dominant technological regimes and the institutions through which they (inter)act. 4 Basically, dominant institutions serve to assist stakeholders with vested interests in sunk investments to avoid competition and, thereby, postpone pervasive innovation. Studies in political economy and international relations from a Marxist perspective, have focused on the relationship between economic, political and technological (f)actors. They have found that specific socio-economic and political conditions, including labour relations, dominant modes of production, styles of policy making and international alliances, are being legitimized in quite coherent concepts of control. Interestingly, such concepts cannot be directly related to specific sectors, political parties, or business interests. A concept of control becomes hegemonic as it succeeds to legitimize itself as an expression of the general interest (Bode, 1978;van der Pijl, 1984). Hence, different social science approaches provide similar answers to the question how power relations interfere with innovation processes. First and most visible, as innovation processes tend to undermine vested interest positions, the powers that be are likely to resist them. For that reason, they may use their resources, including their privileged position within the networks of business, policy and expertise, to keep (potential) challengers in a position of disadvantage. However, this is not the entire story. The other and perhaps more important part of it is that, given the institutionalization of technological regimes, or, given the paradigmatic function of concepts of control, the actors involved may not be able to see or recognize specific innovations. For this type of cases, Lindblom (1997) has introduced the concept cognitive impairment. This concept does not at all refer to a
In political science literature the concepts regime and institution might be used interchangeably, as they both refer to the formal and informal rules of the game that shape actors behaviour. In our vocabulary and in contrast to some other definitions (Young, 1999), the concept of institution may also be used to point to the behavioural impacts of organizations.
4

13

persons intelligence but to the way in which institutional conditions shape peoples capacity to take notice of what happens in the world around them. Cognitive impairment is therefore considered a huge impediment for initiating change. In practice, resistance to change on the side of powerful stakeholders may have a dual background, on the one hand they do not want it, on the other they do not see its opportunities. That cognitive impairment is an important institutional factor in shaping the viability of sustainable energy options is illustrated by the observation that, in spite of the fact that international epistemic communities have come into existence, national states, even within the EU, show quite some variation in this respect. So, historical and political factors can explain for Frances preference for nuclear, Germany as forerunner with respect to biomass, and the US preference for fossil. The same is likely to be true with respect to hydrogen: while stakeholders in the Netherlands express confidence in fuel cell technology, stakeholders in other countries (United Kingdom) maybe more reluctant in this respect. In addressing cognitive impairments, a transition process maybe considered successful to the extent it succeeds in identifying and exploring hypotheses and options that are marginal from the perspective of dominant networks. The ability to bring initially marginal options and hypotheses to bear is for one part dependant on the approaches and methods used in scientific analysis, which can be referred to as problem structuring (Dunn, 2001), for another part it depends on politics. The political task in this respect is to address vested interests and institutions. The guiding principle for Governance by Challenge is to remove, as much as possible, the privileges that vested interests may use to resist innovations. These privileges maybe found for most part in the rules, regulations and institutions maintained by government itself. This notion, which is critical in Governance by Challenge is supported by political theories (liberal, anarchist and conservative) that reject a too powerful government (e.g. Hayek, 1944; Rawls, 1971; and Taylor, 1982). We may especially refer to the work of J.S. Mill (1957) as he combines a view on restricted government vis--vis individual freedom with a plea for specific interventions in the interest of future generations. Interestingly, Mill argues for resisting the power of monopolies and other vested interests for the sake of consumer sovereignty and control. Depending on the case, this value may lead to an extension of specific government tasks (e.g. taking over the London water supply system from private companies) or limiting government regulations where they serve to inhibit competition by newcomers on the energy market. In the sphere of energy generation technological concepts are evolving that tend to decentralize energy production and distribution. Based on either the Stirling motor, the gas turbine technology or on hydrogen and Fuel Cell technology micro combined heat and power production in small buildings is on its way to commercialisation. They might enable a substantial reduction of the use of natural gas and of CO2 emissions compared to central power production. An innovation as distributed generation (DG), e.g. based on renewable hydrogen and fuel cells technologies for mobile as well as stationary purposes (Lovins, 1999) can be introduced by newcomers in the energy markets. New technological artefacts like home fuelling devices are being developed, enabling the dispensing into the FC- or ICE vehicle of home-made hydrogen. A further concept that is being developed is the so-called reversible fuel cell power train,

14

a vehicle that provides for energy of the residence if idle and can take in hydrogen e.g. produced by electrolysing from solar PV roofs. Although the energy efficiency of these new decentralising artefacts that integrate many functions is higher than what we are used to nowadays, they will only be accepted if consumers appreciate them as improvements as far as reliability, comfort and price are concerned. In some countries the breakthrough of DG might be the consequence of consumers lack of trust in the dominant energy companies.

7. Conclusions: a methodological device for policy analysis The previous sections have presented four paradigms of governance, Governance by policy networking, Governance by government, Governance by corporate business and Governance by challenge. None of the paradigms reject government intervention, but each paradigm reveals an institutional bias in that it articulates opportunities for collaboration and competition in a particular way, thereby creating a context for policies, regulations and instruments, which may at first glimpse look identical but are given a specific meaning by their institutional context. Within the paradigm Governance by Corporate Business an instrument such as tradable permits is possible, as long as it does not interfere with corporate strategies. Therefore, it might be based on principles such as grandfathering, which protect vested interests against newcomers. Tradable permits put in the context of Governance by Challenge might be based on principles that allow for newcomers to enter. We have also speculated on the relationship between paradigms of governance and hydrogen options. For the Netherlands, stakeholders found the question as to whether there should be an incremental introduction of hydrogen (mixing 10 to 70 percent into the grid) or making an immediate shift to a full hydrogen infrastructure most critical. A number of governance related issues was identified for transport, including the distribution and storage of hydrogen. Major public acceptance issues given the risks associated with the hydrogen infrastructure might be anticipated (Hisschemller et al., 2001). We hypothesize that the more complex infrastructure requirements, the greater the likelihood that major government interventions will be needed. But also in case of a breakthrough of decentralized electricity production in combination with hydrogen, society may be faced with critical governance issues. This option is likely to be resisted by vested interests in the energy system, which in itself shows that it is in a position of disadvantage vis--vis other options and not only or primarily because of its technical or economic disabilities. So, we find strong reasons to expect that low profile governance (e.g. Networking or Corporate Business) might fail to make the transition to a hydrogen based economy, either because of the collective good character of heavy infrastructures or because advanced options are being excluded from competition. Yet, specific measures to support break-through options (e.g. in the context of Government or Challenge) are vulnerable to the critique that they undermine the level playing field, even as or because- they are meant to provide an equal level playing field for newcomers. Without pretending such findings to be conclusive after all, the relationship between technologies and institutions is the focus of ongoing research the exploration is still relevant for energy policy discourse in an era of worldwide liberalization and priority issues such as climate change and security of supply. The exercise confirms that there

15

is a compelling need to reframe fashionable discourse such as the necessary shift from government to governance or from government to market. Instead, specific questions with respect to the impact of guiding policy frameworks on innovation will highlight that neither neutral nor optimal frameworks for policy making exist, where competing hydrogen options are at stake. A related insight provided by our exercise is that the search for a policy framework for hydrogen must go beyond the narrow perspective of policy instrumentation and the effectiveness of policy instruments. The identification of paradigms of governance may help in broadening and deepening the perspective. However, what we have is more than a set of hypothetical arguments with respect to possible directions for energy policy that could stimulate political debate. We also have a methodological device or, to put it less pretentiously, a tool to assist integrated policy analysis. The method of unfolding and exploring different paradigms of governance fits in with the standing tradition in social science to develop and use ideal type presentations, which draws upon notions developed by Max Weber. Policy analysts can use them for the purpose of ex ante evaluation of competing intervention options and strategies. Different paradigms might help to classify the (often hidden) assumptions that underlie policy makers views with respect to the feasibility of options. Another way to use theory-based paradigms is to explore the relationship between hydrogen technologies and political institutions in a participatory context, i.e. in a dialogue among stakeholders from business, NGOs, policy making, and scientific experts. Here, they could set the institutional stage for backcasting, a way of developing scenarios where one starts with defining the end-stage situation or goal and then reasons backwards to the current situation (Van de Kerkhof, 2004). In conclusion, we suggest that the relationship between technologies and institutions is well worth further researching, thereby using a multitude of approaches from the various social science disciplines. This article has given an illustration of how political theory and philosophy, may deliver a contribution in this respect.

References
Berkhout, F., Smith, A., Stirling, A., 2003. Socio-technological regimes and transition contexts. SPRU Electronic Working Papers, no. 106. Sussex, UK. Bode, R., 1978. De Nederlandse bourgeoisie tussen de twee wereldoorlogen, Cahiers voor de Politieke en Sociale Wetenschappen 4, pp. 9-50. Castells, M., 1996. The Rise of the Network Society, Blackwell Publishers, Cambridge MA. Correlj, A., Verbong, G., 2002. The transition to natural gas in the Netherlands. Paper submitted to the RMNO-workshop July 4-6, Twente. De Bruin, J.A., Ten Heuvelhof, E.F.,1999. Netwerkmanagement. Strategien, instrumenten en normen, Lemma, Utrecht. Driessen, P.P.J., Glasbergen, P. (red.), 2000. Milieu, samenleving en beleid, Elsevier, Den Haag. Dunn, W., 2001. Using the method of context validation to mitigate Type III errors in environmental policy analysis, in: Hisschemller, M., Hoppe, R., W. Dunn et al. (eds.), Knowledge, Power and Participation in Environmental Policy Analysis. Transaction Publishers, New Jersey, 417-436.

16
Ezrahi, Y. 1990. The descent of Icarus, Science and the Transformation of modern democracy, H.U.P, Cambridge MA. Florida, R., 2002. The Rise of the Creative Class, Basic Books, New York. Galbraith, J.K., 1952. American Capitalism-The Concept of Countervailing Power, Boston. Hardin, G., 1968. The tragedy of the commons. Science no.162. Hayek, F.A., 1944. The Road to Serfdom, Chicago Univ. Press, Chicago. Hisschemller, M., 2003. The Dutch energy transition and its institutional problems. In: Faure, M., J. Gupta and A. Nentjes (eds.). Institutions and instruments to control global climate change. Edward Elgar Publishers. London, UK. pp. 292-311. Hisschemller, M., Bode, R., and van de Kerkhof, M., 2004. Paradigms of governing the transition to a sustainable energy system and their implications for technology development: A critical reflection. IVM report. Amsterdam, The Netherlands. Hisschemller, M., van de Kerkhof, M., 2004. Europa, Nederland: Hoe verder na Kyoto? Aanbevelingen van Stakeholders voor het Nederlands EU voorzitterschap, Bilthoven. Hisschemller, M., Ravensbergen, P., Minnesma, M., 2003. Maatschappelijke reacties op het concept De Kas als energiebron, IVM, Amsterdam. Hisschemller, M., van de Kerkhof, M., Kok, M., Folkert, R., 2002. Climate OptiOns for the Long term (COOL): Stakeholders views on 80% emission reduction. In: Kok, et al. (eds.). Global warming and social innovation. The challenge of a climate-neutral society. Earthscan Publications Ltd., London, 196-217. Hisschemller, M., van de Kerkhof, M., Kuik, O., Folkert, R., Kok, M., Spakman, J., Treffers, D.J., Spanjersberg, M., Faaij, A., Oude Lohuis, A., Seebregts, A., Kroon, P., Annema, H. Jeeninga, Tuinstra, W., 2001. Backcasting exercities in COOL De Nationale Dialoog. W01/03. Institute for Environmental Studies. Vrije Universiteit. Amsterdam. Hisschemller, M., Midden, C. 1989. De bestuurlijke vormgeving van de besluitvorming over de bouw van nieuwe kerncentrales in Nederland in het licht van de maatschappelijke acceptatie. Den Haag. Kuhn, T., 1962. The structure of scientific revolutions, Chicago. Lasswell, D., Lerner, H.D., 1951. The Policy Sciences, Stanford University Press, Stanford. Luiten, E., 2001. Beyond Energy Efficiency, Actors, networks and government intervention in the development of industrial process technologies, Utrecht. Lijphart, A., 1968. The politics of accommodation. Pluralism and Democracy in the Netherlands, Univ. of California Press, Berkeley. Lijphart, A. 1999. Patterns of Democracy. Government Forms and performance in Thirty-Six Countries, Yale University Press, New Haven and London. Lindblom, C.E., 1997. Initiating Change; modes of social inquiry, in: L.K. Comfort (ed): Initiating Change: Theory and Practice, special issue of American Behavioral Scientist, Vol. 40/3, 264277. Lovins, A.B., 1999. A strategy for the Hydrogen Transition, Vienna (Virg.), Rocky Mountain Institute, US National Hydrogen Association. Luce, D.R., Raiffa, H, 1957. Games and decisions: introduction and critical survey, New York. Mill, J.S., 1975. Three Essays, London, Oxford University Press. Mill, J.S., 1957. Collected Woks. Routledge and Keegan, Toronto. Mill, J.S., 1876. Principles of Political Economy, Longman, Green, Reader and Dyer, London. Moharir, V.V., 1979. Process of public policy making in the Netherlands: A case study of the Dutch governments policy for closing down the coal mines in South-Limburg, University of Amsterdam.

17
Mulder, P., 2003. On the economics of technology diffusion and energy efficiency, Vrije Universiteit Amsterdam. Olson, M., 1971. The Logic of collective Action. Public Goods and the theory of groups. Harvard U.P., Cambridge (Mass.). Pijl, K. van der, 1984. The making of an Atlantic Ruling Class, Verso, London. Price, D.K., 1967. The Scientific Estate, Harvard U P, Cambridge (Mass.). Rawls, J. 1971. A theory of justice. Harvard U P. Cambridge (Mass.). Rifkin, J., 2002. The hydrogen economy, the creation of the world wide energy web and the redistribution of power on earth. Penguin Putnam Inc, New York. Shell International, 2001. Energy Needs, Choices and Possibilities, Scenarios to 2050. Royal Dutch Shell Group, London. Schumpeter, J.A., 1939. Business Cycles, a theoretical, historical and statistical analysis of the capitalist process, MacGraw-Hill Books Co., New York. Taylor, M., 1982. Community, anarchy and liberty. Cambridge. Teisman, G., 1997. Sturen via creatieve concurrentie; een innovatie-planologisch perspectief op ruimtelijke investeringsprojecten, Katholieke Universiteit Nijmegen Toulmin, S., 1969. The uses of argument, Harvard UP, Cambridge (Mass.). Van de Kerkhof, M., Hisschemller, M., Spanjersberg, M., 2002. Shaping diversity in participatory foresight studies. Experiences with interactive backcasting in a stakeholder dialogue on long-term climate policy in the Netherlands. Greener Management International, 37 (1), 95-99. Van de Kerkhof, M., 2004. Debating Climate Change. A study of stakeholder participation in an integrated assessment of long-term climate policy in the Netherlands. Lemma Publishing, The Netherlands. Young, O., 1999. Institutional dimensions of Global Environmental Change. Science Plan. IHDP Office, Bonn.

18

19

IVM Working Papers Series

IVM Working Papers present some cutting edge results of IVMs scientific research in a concise format to the international research community, policy makers and other interested people. The papers are available as PDF files from IVMs website www.vu.nl/ivm. Up to now, the following papers have been published in this series: 06/01 Buitenhuis, C.J.K., Cenijn P.C., van Velzen, M., Lilienthal, H., Malmberg, T., Bergman, ., Gutleb, A.C., Legler, J., Brouwer, A. Prenatal exposure of rats to hydroxylated PCBs and some brominated flame retardants. 06/02 Hisschemller, M., Bode, R., van de Kerkhof, M. What governs the transition to a sustainable hydrogen economy? 06/03 Dellink, R.B., Hofkes, M.W. Towards more efficient national climate policy: a dynamic CGE analysis. 06/04 Vermaat, J.E., Thampanya, U. Mangroves reduce coastal erosion.