Part III - Legitimacy and Effectiveness in the Climate-Energy Nexus

7 Disentangling Legitimacy Comparing Stakeholder Assessments of Five Key Climate and Energy Governance Institutions

Naghmeh Nasiritousi and Soetkin Verhaegen

7.1 Introduction

This chapter builds on the insights of the previous chapters about the institutional complexity of the climate-energy nexus. As was shown in Chapter 3, the global institutional complex on climate and energy governance has in recent years developed into a crowded field with the emergence of several international institutions that seek to address both issues in tandem. Hence, multiple actors work in the same area without overarching coordination (Biermann et al. 2009). With different mandates, forms, functions, and values, these institutions both cooperate and compete with one another to further their mission. Given scarce resources amongst policy makers and other stakeholders, these actors need to prioritize which institutions to engage with.

Central to the question of which international institutions warrant support and are prioritized are considerations of the institutions’ legitimacy. With competition over members and resources, international institutions depend on favourable perceptions of legitimacy by a diverse set of global governance stakeholders, such as policy makers, nongovernmental organizations, and businesses, to achieve their objectives (Andresen and Hey 2005; Biermann et al. 2009). As was discussed in Chapter 2, legitimacy broadly refers to ‘the acceptance and justification of shared rule by a community’ (Bernstein 2005, 142). Legitimacy is important for international institutions in order to be able to operate with authority and to attract constructive participation of political and societal stakeholders in the processes of making and implementing governance. Put differently, to achieve their objectives, international institutions must gain acceptance, trust, and credibility amongst the communities that they seek to govern (Andresen and Hey 2005).

The aim of this chapter is to understand how international institutions operating under institutional complexity are perceived by key stakeholders in terms of legitimacy. We present a novel approach to studying legitimacy perceptions as we capture stakeholders’ assessments of a broad range of dimensions of legitimacy and bring those together in a composite measure of legitimacy assessments. Scholarly work on the concept of legitimacy highlights, and debates, that legitimacy is built on institutional qualities such as how the internal decision-making and accountability structures work and how effective and fair the institution is perceived to be (Scholte and Tallberg 2018). We contribute to this debate by showing that the surveyed stakeholders in climate and energy governance indeed perceive these elements as dimensions of the broader concept of legitimacy.

Concretely, by focusing on those aspects of legitimacy that international institutions themselves can influence, i.e. their institutional qualities, we contribute to understanding how perceptions of these – i.e. what we call legitimacy assessments – differ between stakeholder groups. Previous literature has to our knowledge not mapped stakeholder’s perceptions of a set of institutions that work on similar issues and that thereby have overlapping mandates. In terms of empirical novelty, the chapter offers a systematic and comparative mapping of stakeholders’ legitimacy assessments of five institutions. To this end, it uses a hybrid approach focusing on stakeholders’ assessments of those dimensions of legitimacy that concern institutional qualities. Theoretically, the chapter unpacks the meaning of legitimacy under institutional complexity.

This chapter thereby provides innovative insights to the literatures on both legitimacy and institutional complexity, with implications for ways in which climate and energy governance can be strengthened. Moreover, the findings have implications for how institutions may influence perceived legitimacy deficits through legitimation strategies toward different stakeholder audiences (Bäckstrand and Söderbaum 2018).

We gained insight into stakeholders’ legitimacy assessments by fielding an expert survey among energy and climate stakeholders from different world regions. Respondents were asked about five climate and energy governance institutions that exhibit different but overlapping mandates and membership: the Clean Energy Ministerial (CEM), the International Energy Agency (IEA), the International Renewable Energy Agency (IRENA), the Renewable Energy Policy Network for the 21st Century (REN 21), and the United Nations Framework Convention on Climate Change (UNFCCC). All five institutions belong to the subfield of renewable energy. As shown in Chapter 4, they play a key role for this subfield. Based on that chapter and its analysis of institutional coherence and management on renewable energy, we now expand the analysis of this subfield toward institutional legitimacy. The stakeholders who evaluate the five selected institutions comprise both state and nonstate actors, work with different issues (e.g. mitigation, adaptation, energy security, climate finance, and technology) and come from varying regions of the world. The data allow us to show how legitimacy assessments of these five institutions differ across stakeholder types and across stakeholders working with different issues.

The chapter proceeds as follows. The next section presents our framework for studying stakeholders’ legitimacy assessments. Here, we also further develop the conceptual insights on legitimacy introduced in Chapter 2. Next, the five institutions in climate and energy governance are described, paying specific attention to the institutional qualities that are expected to be relevant in guiding legitimacy assessments. Thereafter, the data and methods section outlines how we measured these assessments. The results section then maps stakeholders’ legitimacy assessments of the five institutions in our study. The final section summarizes the insights gained and highlights avenues for further research.

7.2 Theory and Concepts

As argued in Chapter 2, institutional complexity complicates an evaluation of legitimacy of individual institutions because of the interlinkages and overlapping mandates between institutions. In this section, we link back to the discussion in Chapter 2 on the concept of legitimacy and highlight how the cognitive model of legitimacy provides insights into understanding legitimacy under institutional complexity. Thereafter, we discuss the institutional qualities that have been argued to be central to institutions’ legitimacy, deriving nine dimensions of legitimacy.

7.2.1 Congruence and Cognition: Understanding Perceptions of Legitimacy

The traditional view of legitimacy in IR has held that ‘legitimacy depends on the congruence between an organization’s features – specifically, its procedures, purpose, and performance – on the one hand, and the inter-subjectively shared norms and values held by relevant organizational stakeholders, on the other hand’ (Lenz and Viola 2017, 943). Legitimacy in this view depends on the extent to which an institution lives up to certain legitimacy demands that stakeholders have, which are determined by the norms and values of those stakeholders. Recent research by Lenz and Viola (2017) has, however, outlined several empirical and analytical weaknesses in the traditional approach – or what they call ‘the congruence model of legitimacy’. Central to this argument are limitations to stakeholders’ ability to make a precise and complete evaluation of an institution in order to compare this to their normative beliefs.

Instead, Lenz and Viola (2017) introduce a ‘cognitive model’ for understanding how legitimacy perceptions are formed. This model draws on the literature on cognitive psychology to outline the micro-foundations for understanding the formation of legitimacy perceptions and reflects similar approaches in the public opinion literature (Armingeon and Ceka 2014). The three core insights that inform their model are: ‘(1) judgments rely on cognitive schemata and heuristics that bias judgments; (2) they are comparative; and (3) they are sticky, up to a threshold’ (Lenz and Viola 2017, 947–948).

According to these insights, legitimacy perceptions are not formed in a vacuum, i.e. actors do not judge institutions one by one against their held social values and norms. Rather, perceptions of an institution are based on a reference point that is derived from previous experiences. These heuristics consist of perceptions of institutions that stakeholders are most familiar with or which they most recently engaged with, but it may also consist of an ideational prototype of what the perfect institution would look like. Heuristics are presented as rather stable images in stakeholders’ minds. When we ask stakeholders to assess the legitimacy of an institution, we should therefore expect them to compare the perceived qualities of that institution to those of their ‘heuristic’ institution. Moreover, we can expect variations across stakeholders as they will have different reference points, or heuristics, depending on their background, the institutions they are mostly familiar with, and the norms they hold.

While this turns legitimacy assessments into something much more personal than the congruence model proposes, processes of socialization and shared experiences within specific professional sectors lead us to expect systematic similarities in the used heuristics and normative beliefs about legitimacy across individuals within the same sector, and differences among individuals in different sectors. For instance, nonstate actors such as business or civil society actors may assess institutions in relation to the norms of legitimate governance that are central in their respective peer group. Likewise, climate- and energy-related stakeholders that also work on questions of international development are expected to keep development institutions, and their respective norms, in mind when they assess the legitimacy of the climate and energy governance institutions in our study. This very use of heuristics, as well as its dependence upon stakeholders’ specific experiences, provides an additional motivation for studying individual legitimacy assessments (Scholte and Tallberg 2018).

This conception of legitimacy has two key implications for how we can understand legitimacy perceptions. First, this chapter argues that an awareness of cognitive limitations is central to understanding legitimacy beliefs. Rather than assuming that actors, even if they are experts, are capable of capturing the exact way in which institutions function and the extent to which the institution is in line with those actors’ normative beliefs, one should recognize that legitimacy assessments are based on heuristics and underlying experiences, which come with respective limitations. Especially in a highly complex, and therefore cognitively demanding, institutional environment, one may expect actors to base their legitimacy assessments on such heuristic simplifications. When facing several institutions with overlapping and complex mandates, actors may use mental shortcuts to form opinions about some of these institutions (Alter and Meunier 2009).

Second, the norms, values, and experience of actors can both influence how they assess the qualities of an institution as well as how they value these qualities, i.e. the relative importance that they place on the purpose, process, or performance of institutions. In other words, stakeholders’ legitimacy perceptions may differ either because they assess the institutional qualities of institutions differently, and/or because they value different characteristics of legitimacy differently. This means that an actor’s legitimacy perceptions, i.e. the extent to which an institution is viewed as legitimate by an actor, is a combination of that actor’s legitimacy assessment (i.e. an assessment of the institutional qualities of an institution) and that actor’s legitimacy valuation (i.e. the importance attached to certain institutional qualities). This chapter focuses on legitimacy assessments by climate and energy experts along nine dimensions of legitimacy as explained in the next section.

7.2.2 Legitimacy Criteria Used to Map Perceptions

Legitimacy is the assessment and valuation by an audience as to the appropriateness of an authority. What should be considered a legitimate form of authority has preoccupied normative scholars. What is in practice considered a legitimate form of authority is instead the focus of sociological work (Nasiritousi et al. 2016). In this chapter we opt for a hybrid approach, as we study stakeholders’ perceptions of institutions while referring to normative criteria of legitimate governance (cf. Agné 2018). This take thus differs from a ‘purely’ sociological approach where it is left to selected stakeholders to determine relevant criteria for assessing an institutions’ legitimacy. In this type of study, legitimacy is empirically measured as confidence in, or support for, an institution (Gibson and Caldeira 1998; Dellmuth and Tallberg 2015).

The current study, in contrast, combines normative and sociological aspects. It does so by seeking to understand legitimacy in terms of its different dimensions. This approach provides a uniquely fine-grained perspective on legitimacy perceptions (cf. Scholte and Tallberg 2018). The hybrid approach is in line with the work of Beetham (1991), who argues that legitimacy has both a normative and sociological component, as perceptions of institutions’ legitimacy will be based on institutions meeting normative criteria on the exercise of power.

Concretely we seek to provide a comparative mapping of stakeholders’ views of a set of nine institutional qualities or dimensions derived from the normative literature. This helps us to better understand how legitimacy assessments may vary between different institutions and various stakeholder groups. These assessments are expected to be an important indicator for sociological legitimacy (cf. Scholte and Tallberg 2018).

Our conceptual framework therefore begins with identifying dimensions of legitimacy. We do so by advancing normative criteria, i.e. a set of standards that are ‘grounded in normative theories that reflect prevailing sociological standards in society’ (Karlsson-Vinkhuyzen and McGee 2013, 58). Central to the identification of dimensions of legitimacy is the distinction between input and output legitimacy. While input legitimacy refers to the design of political processes, i.e. governance by the people, output legitimacy concerns problem-solving capacity, i.e. governance for the people (Scharpf 1999). By exploring aspects of input and output legitimacy, it is possible to derive criteria for assessing legitimacy anchored in a normative framework.

The normative framework presented in Table 7.1 builds on the works of Bodansky (1999), Karlsson-Vinkhuyzen and Vihma (2009) and Mena and Palazzo (2012). The framework distinguishes source-based and process-based input legitimacy, as well as substantial and distributive output legitimacy. These unfold into a total of nine dimensions of legitimacy.

Table 7.1 Analytical Framework - Dimensions of Input and Output Legitimacy and their Operationalization.

Input or Output Legitimacy	Dimensions of Legitimacy	Operationalization in Survey
		For those institutions in Question 6 that you are familiar with (where you answered 3–5), please evaluate these institutions in their respective column according to the criteria below. Write a score between 1–5 in each cell, where 1 means that the institution is very weak and 5 means it is very strong on the respective dimension.
Source-based (input) legitimacy	Source of authority	Expertise
Process-based (input) legitimacy	Inclusion	Inclusion of all appropriate actors
	Procedural fairness	Procedural (decision-making) fairness
	Transparency	Transparency
	Accountability	Accountability
Substantial (output) legitimacy	Output	Output (what is produced)
	Outcome	Outcome (the effect the output has on its members)
	Impact	Impact (the effect the output has on problem-solving)
Distributive (output) legitimacy	Distributive fairness	Distributive fairness (distributing benefits to members fairly)

Source: CLIMENGO Expert Survey 2017–2018.

Source-based legitimacy refers to how authority is gained by an institution – not by its operations, but through its essence and standing. Three common forms of source-based legitimacy are expertise, tradition, and discourse (Karlsson-Vinkuyzen and McGee 2013). Process-based legitimacy pertains to the design of procedural rules that affect the decision making of the institution. Inclusion refers to how open the institution is in terms of membership. Procedural fairness in decision making means that stakeholders have opportunities to be heard and be treated fairly so as to have a sense of ownership of the decisions made (Raines 2003). Transparency relates to the degree of access to information that the institution provides to members and other stakeholders. Accountability implies that institutions can be held to account for the decisions that they make and for the ways in which they implement these decisions. Substantial legitimacy is concerned with issues of effectiveness. Output concerns performance in terms of what the international institution produces, for example issuing regulations (these can be binding or non-binding), producing reports, conducting research, organizing meetings, providing funding, providing training, etc. (Szulecki et al. 2011). Outcome relates to whether the institution produces behavioural changes, for example in terms of whether the institution increases the level of cooperation and compliance amongst members for instance by improving learning and modifying incentives (Underdal 2002; Gutner and Thompson 2010). To determine an institution’s impact involves making judgements about the extent to which the institution contributes to alleviating the problem it was tasked to resolve (Underdal 2002). Distributive legitimacy, finally, is a dimension that is concerned with the distribution of benefits to the members of the institution.

7.3 The Five Cases: Similarities and Differences in Institutional Qualities

The five institutions whose legitimacy we put under scrutiny in this chapter are: CEM, IEA, IRENA, REN21, and UNFCCC. These institutions have different forms and functions, yet they also have overlapping mandates. We selected these institutions since they pertain to one major subfield of the climate-energy nexus, namely renewable energy. Chapter 4 analyzed the degree of coherence of the renewable energy subfield and identified these as the key institutions therein (Sanderink, this volume). Their importance was further confirmed by climate and energy experts (both state and nonstate actors) that we interviewed prior to designing our questionnaire. The five institutions have thus all achieved a certain level of authority, which makes them interesting cases for a comparative mapping of how stakeholders’ legitimacy assessments differ amongst these key institutions. In what follows, we briefly introduce the five institutions based on their self-descriptions – by representatives we approached or on their websites¹ – and highlight a number of similarities and differences across them in terms of key properties. The descriptions form the context for our expectations that we thereafter derive about how stakeholders make legitimacy assessments.

The most long-standing institution in our sample is the IEA – an intergovernmental organization that was established in 1974 and is based in Paris. The IEA was established within the framework of the Organisation for Economic Co-operation and Development (OECD) in response to the 1973 oil crisis to strengthen the cooperation of industrialized countries to meet the energy needs of oil-consuming countries. The agency draws its thirty member countries from the OECD group of industrialized countries, and, in addition, features eight association countries: Brazil, China, India, Indonesia, Morocco, Singapore, South Africa, and Thailand (IEA 2018a). Association countries may participate in the analytical work of the IEA, but have no rights and obligations. While its main focus has been to tackle global oil supply disruptions, the IEA’s mandate has broadened to ‘ensure reliable, affordable and clean energy for its thirty member countries and beyond’ (IEA 2018b). It has a global scope and works on energy security, sustainability and clean energy transitions, technology, innovation, and energy access. The main decision-making body of the IEA is the Governing Board, which comprises energy ministers or their senior representatives from each member country. Governing Board decisions are legally binding on all member countries. Majority vote is based on a system of voting weights allocated to each member country. Such a vote is required for decisions on the IEA Programme of Work, procedural questions, and recommendations. Unanimity is required for other decisions. The IEA works closely with partners, including industry partners, and other international institutions to gain insights and advice from outside actors (IEA 2018c). There is no formal role for nonstate actors, but nonstate actors may contribute to and peer-review IEA reports, participate in IEA events and programmes, and serve on IEA advisory boards. In terms of output, the IEA collects data, conducts research, provides analysis, makes policy recommendations, produces reports, organizes meetings/workshops/seminars, and offers training.

Almost two decades after the establishment of the IEA, countries adopted the United Nations Framework Convention on Climate Change in 1992. With near-universal membership, the objective of the UNFCCC is to ‘stabilize greenhouse gas concentrations in the atmosphere at a level that would prevent dangerous anthropogenic interference with the climate system’ (UNFCCC 2018). Tasked with supporting the operation of this international environmental treaty, the UNFCCC Secretariat is based in Bonn. The UNFCCC is an intergovernmental institution that makes decisions based on consensus. The UNFCCC deals with a range of issues related to climate change, including mitigation, adaptation, technology, capacity building, and finance. It is also one of the most open international institutions in terms of involving a range of nonstate actors in the yearly conferences compared to other institutions in, for example, trade or security fields (Nasiritousi and Linnér 2016). Nonstate actors also have a prominent role in the Global Action Agenda, an initiative to spur more ambitious climate action amongst stakeholders, as evidenced by the Yearbook for Global Climate Action (UN Climate Change Secretariat 2018) and the NAZCA database of climate commitments by nonstate actors. The UNFCCC’s key outputs have been the 1997 Kyoto Protocol and the 2015 Paris Agreement, both landmark international agreements aimed at addressing the causes and consequences of climate change.

More recent institutions are REN21, IRENA, and CEM. REN21 was launched in 2004 as a ‘global renewable energy policy multi-stakeholder network’ (REN21 2018a). It is based at the office of the UN Environment Programme (UNEP) in Paris. Its mission is to facilitate knowledge exchange and drive a transition toward renewable energy. The members of REN21 come from five stakeholder groups: governments, industry associations, nongovernmental organizations (NGOs), academia, and other international organizations. REN21 tries to keep membership balanced between the five stakeholder groups. By implication, governments are outnumbered by nonstate actors. Government representatives come from the following thirteen countries: Afghanistan, Brazil, Denmark, Dominican Republic, Germany, India, Mexico, Norway, Republic of Korea, South Africa, Spain, United Arab Emirates, and the United States (REN21 2020). REN21 is thus a collaborative network that seeks to connect the public and private sectors on renewable energy (REN21 2018b). The Steering Committee is elected from REN21’s members, ten from each stakeholder group. From that, the seven people of the Bureau are elected. These elections are held at the annual meeting, the General Assembly, and this is the only time REN21 takes decisions by majority vote. Other decisions are typically consensus based. The Bureau provides month-to-month oversight while the Steering Committee conducts the broader, programmatic oversight. REN21’s key output is the annual Global Status Report, which presents a rich set of data on the status of renewables and is widely disseminated among actors in the field.

Founded in 2009, IRENA is an intergovernmental organization that is headquartered in Abu Dhabi. It currently has 161 member states, with further 22 states currently undergoing accession processes (IRENA 2020a). The agency seeks to promote adoption and sustainable use of all forms of renewable energy, in the pursuit of sustainable development, energy access, energy security and low-carbon economic growth. The main decision-making body of IRENA is the Assembly, which includes one representative from each member country. All matters of substance are decided by consensus among the members present, whereas questions of procedure are decided by simple majority. IRENA works with the broader renewable energy community, including companies, NGOs, and other international organizations, to facilitate knowledge-sharing (IRENA 2020b). Examples include a joint project facility, online info and marketplace platforms, initiatives, and the Coalition for Action (IRENA 2020c). In terms of output, IRENA is involved in many activities, including: research and publication of reports, providing member states and nonstate actors with recommendations, issuing non-binding regulations, and providing training and funding to support implementation.

Established in 2009, CEM is a high-level ministerial forum that seeks to advance clean energy technologies by promoting initiatives based on common interests among its members and other stakeholders. Its Secretariat is seated at the IEA headquarters in Paris. CEM members include twenty-seven country governments, but also the European Commission. It is the only regular meeting of ministers focusing on clean energy. Rather than relying on consensus, CEM employs a ‘distributed leadership’ model whereby any government interested in furthering an idea on clean energy technology is encouraged to identify willing partners and proceed. The initiatives, which countries join based on their interests and capabilities, must include three or more CEM members, be endowed with resources, and offer a tangible work plan. CEM’s work is divided into three general work categories: (1) energy supply systems and integration, (2) energy demand, and (3) cross-cutting support. The latter includes, for example, initiatives such as Women in Clean Energy and the Clean Energy Solutions Centre (which provides policy toolkits). In terms of output, each initiative sets its own deliverables and objectives, depending on their goals. Some produce reports and analysis, others focus on policy solutions, yet others use workshops, seminars, webinars, and other forms of knowledge-sharing. CEM also seeks the input of key private sector partners through, for instance, dedicated actions, commitments, or the hosting of workshops (CEM 2018).

These five institutions thus all operate in the complex of institutions that govern the climate-energy nexus within the subfield of renewable energy, but differ in a number of respects that may impact on how stakeholders assess their legitimacy. The first is in membership, where some are intergovernmental organizations with near-universal membership (UNFCCC, IRENA) while others are minilateral institutions (IEA and CEM) or multi-stakeholder partnerships (REN21). This may have implications for stakeholders’ assessments of their inclusion, procedural fairness, and distributive fairness. Second, they differ in terms of the scope of their mandate, where some have a broad mandate focusing on multiple issues (UNFCCC and IEA) whereas others concentrate on more specific questions (IRENA, REN21, and CEM). Third, they vary in terms of the nature of their mandate, with the UNFCCC having a political mandate requiring negotiations on contentious issues between countries, while the other four institutions in the sample are endowed with a more technical mandate focusing on implementation. Their mandate is likely to have implications for stakeholders’ assessments of the output, outcome, and impact of respective institutions. Fourth, the selected institutions differ with respect to how strongly they work with nonstate actors. The UNFCCC and REN21 have a close relationship with a broad range of nonstate actors in terms of access or cooperation. Other institutions are less engaged with such actors or are more selective, with a narrower set of nonstate collaboration partners (IRENA, CEM, IEA). This, in turn, may well affect stakeholders’ assessments of their levels of inclusion and expertise. Fifth and finally, most institutions take decisions of substance based on consensus, whereas CEM has a more flexible decision-making structure where initiatives only need agreement between at least three members. This may have consequences for how stakeholders view procedural fairness and distributive fairness.

7.4 Theory-Based Expectations of Legitimacy Assessments

Some of the differences mentioned in the previous section have theoretical value, since they imply expectations about legitimacy assessments. In what follows, we turn to the question of how stakeholders’ assessments of the legitimacy of the five key institutions governing the climate-energy nexus may vary.

The literature has shown that different types of stakeholders hold different legitimacy demands based on their social values, norms, and previous experiences (Bernstein 2005; Karlsson-Vinkhuyzen and Vihma 2009; Lenz and Viola 2017). We argue that legitimacy demands can therefore vary depending on (1) the type of stakeholder (i.e. government, business, or NGO representative); (2) the issues that these stakeholders primarily work with (for example energy, development, or climate change); and (3) where in the world the person comes from, as social values, norms, and experiences can be expected to vary across different legitimacy-granting communities (Symons 2011; Nasiritousi et al. 2016). Thus, stakeholder type, focus of work and geographical origin can serve as proxies for differences in norms, values, and experiences that may influence legitimacy assessments.

At the same time, institutional complexity – and the logic of the cognitive model – implies that stakeholders, even if they are experts, may face difficulties in distinguishing their assessments of institutions that are similar in their functions due to bounded rationality (Alter and Meunier 2009). If it is indeed too hard for stakeholders to disentangle certain properties across institutions with overlapping mandates, e.g. dimensions such as outcome and impact (Bäckstrand et al. 2018), the cognitive model of legitimacy would lead us to expect that there will not be great variation in stakeholders’ assessments of the institutions governing the climate-energy nexus. Despite the differences in institutional qualities outlined in the previous section, the five institutions are interrelated and fulfil comparable governance functions within the same subfield, such that stakeholders might draw on similar heuristics to form their legitimacy assessments. In sum, the literature provides reasons to expect both variation and similarity in legitimacy assessments of institutions, across different categories of stakeholders.

Expectations can therefore be drawn up based on the nature of institutions as well as on the background of the stakeholders. The following expectations will guide the exploratory analysis that we present in the remainder of this chapter. First, all five institutions are relatively specialized and rely on expert knowledge as source-based input legitimacy. It is therefore of interest to explore whether stakeholders agree with the institutions’ claims that they are strong on expertise. Given that expertise is an important feature of the institutions studied, we have reasons to believe that the expertise dimension will be positively evaluated by stakeholders. Conversely, because most institutions are more concerned with expertise than the empowerment of marginalized groups, procedural and distributional fairness can be expected to be evaluated more negatively (cf. Nasiritousi et al. 2016).

Second, the selected institutions vary along the nine dimensions of legitimacy. Particularly the UNFCCC fulfils many of the respective normative criteria, with, for example, inclusive membership, relative openness toward nonstate actors, and outputs such as the Paris Agreement and can therefore be expected to rank highly on legitimacy (Karlsson-Vinkhuyzen and McGee 2013). Yet, the cognitive model highlights that legitimacy assessments also depend on the prototype used by actors to form their perceptions (Lenz and Viola 2017). This implies that, while an institution fulfils many normative criteria of legitimacy, legitimacy assessments may still vary depending on the norms, values and experiences of the community of stakeholders that grant legitimacy.

Third, and linking to the background of stakeholders, we may expect different legitimacy assessments among state actors on the one hand, and nonstate actors on the other. State actors play an important role in intergovernmental organizations, and are likely to take these as a point of reference. For nonstate actors, on the other hand, the prototype used to make an evaluation is likely to be an institution that the nonstate actor is familiar with or wishes for, i.e. a relatively open institution with formal access for nonstate actor participation (Tallberg et al. 2014). Institutions that are relatively closed are therefore more likely to be negatively evaluated by nonstate actors than by government representatives.²

Fourth, stakeholders also differ in terms of the issue areas they are predominantly working on. Differences in legitimacy assessments could thus also arise from variations in norms and values that go back to different thematic environments. Stakeholders from a certain community are likely to be more familiar with institutions from their own field than from other issue areas and, subsequently, may well use different heuristics or prototypes. For example, those actors working primarily in the energy sector may be much more familiar with institutions such as the Organization of the Petroleum Exporting Countries (OPEC) and the International Energy Forum than those actors that primarily work in the development sector – who, in turn, may be more familiar with, for example, the Global Environment Facility and the Green Climate Fund. In other words, the frame of reference that actors in global climate and energy governance use for their legitimacy assessments can be assumed to reach far beyond the institutions included in this study.

Finally, we expect to observe differences in legitimacy assessments based on where respondents come from. Both legitimacy norms and heuristics are likely to vary depending on the geographical background of the respondents. For instance, governance norms and expectations, political culture, and level of involvement in international organizations may differ considerably across countries. This said, our sample consists of experts largely active in international circles. This might weaken the differentiating effect of geographical origin as these experts may have experienced a certain socialization into more general and international norms of global governance (Flockhart 2006; Greenhill 2010).

In what follows, we use our expectations as an explorative guidance to provide a first empirical mapping of legitimacy assessments for the five selected key institutions governing the climate-energy nexus. This mapping will offer novel insights into how these assessments differ between institutions and stakeholders.

7.5 Data and Methods

This chapter uses unique questionnaire data to capture the assessments of key stakeholders on the different dimensions of input and output legitimacy that we introduced previously (CLIMENGO Expert Survey 2017–2018). Climate and energy experts were surveyed, including representatives from national, regional, and local governments as well as businesses, NGOs, academia, and intergovernmental organizations. The survey was distributed to participants at three venues: the UNFCCC COP23 in Bonn, Germany, November 2017; the UNFCCC intersessional in Bonn, Germany, May 2018; and the Nordic Clean Energy Week that comprised both Mission Innovation and CEM meetings in Malmö, Sweden, and Copenhagen, Denmark, May 2018. At the UNFCCC meetings we handed out questionnaires in side-events with an energy-related focus. We thereby obtained responses from a broad range of public and private stakeholders that work with climate and energy questions.

In addition, we created an online version of the questionnaire to target specific categories of respondents that were not sufficiently covered by the paper version of the survey. As probability sampling was not possible – given that it is not possible to define the population of climate and energy experts in global governance – we aimed at covering a broad variety of stakeholders. This means that, while we can show differences in legitimacy assessments between stakeholder categories, the results cannot be extrapolated to the entire population of climate and energy experts in global governance.

The survey first asked respondents to indicate which type of stakeholder they are, and which issue areas are central to their work. They were also asked to indicate their nationality. Next, respondents were asked how familiar they are with the five institutions of our study. When respondents indicated to be at least somewhat familiar with an institution, they were asked follow-up questions on nine criteria that reflect the different dimensions of legitimacy as identified in the conceptual framework. Respondents were instructed to use a scale that ranges between 1 (very weak) and 5 (very strong) to evaluate each organizations’ expertise, transparency, accountability, inclusion of all appropriate actors, procedural (decision-making) fairness, output (what is produced), outcome (the effect the output has on its members), impact (the effect the outcome has on problem-solving), and distributive fairness (distributing benefits to members fairly).

The survey was completed by 262 respondents in total. Of these, 28 per cent were government representatives, 26 per cent represented an NGO, 23 per cent identified themselves as academics, 17 per cent represented a business organization, and 8 per cent an intergovernmental organization. The largest share of respondents worked with multiple issue areas; most of them with climate mitigation (36 per cent), followed by technology (30 per cent), energy or energy security (30 per cent), development (19 per cent), adaptation (19 per cent), and climate finance and carbon pricing, e.g. carbon markets (17 per cent). Geographically, most respondents hold a European or Western nationality³ (64 per cent); 13 per cent of respondents came from Africa, 15 per cent came from the Asia-Pacific region, and 6 per cent from a Latin American or Caribbean country.

These groupings were used to examine differences in perceptions of stakeholders from different geographical origins. An additional categorization of nationalities was conducted based on the World Bank’s income categories of countries, i.e. low, lower-middle, upper-middle, and high. This two-pronged approach allows us to test whether differences in legitimacy assessments stem from differences in norms, values, or experiences held across world regions as determined by geography or by income.

As respondents could indicate multiple actor types and issue areas they were active in, t-tests (rather than analysis of variance, i.e. ANOVA) were performed in order to explore statistical differences in their legitimacy assessments. For each instance, all respondents who indicated to be active within a certain actor type or to be working with a certain issue area were compared to all respondents who were active in a specific other organization, or active in another issue area. Among the surveyed stakeholders, the most well-known organization is the UNFCCC, with 90 per cent of respondents being rather to highly familiar with this organization. Next are the IEA (87 per cent), IRENA (84 per cent), and REN21 (47 per cent), while only 35 per cent of the respondents are at least rather familiar with CEM.⁴

7.6 Results

7.6.1 Exploring Nine Dimensions of Legitimacy

Figure 1 presents assessments by the respondents for each institution and each legitimacy dimension, as well as the average score for all dimensions taken together (‘total average’ in the figure). On the whole, we see that expertise is positively assessed across the institutions. For the IEA, IRENA, REN21, and UNFCCC it ranges between 4 and 4.5 points on the 5-point scale. Specifically, the data show that the level of expertise of the IEA and IRENA is significantly more positively evaluated by the respondents than any other legitimacy dimension of those institutions (confirmed by a t-test). Similarly, for REN21 and the UNFCCC the level of expertise is more positively evaluated than most other dimensions. This is in line with our expectations, as this dimension represents a key feature of the institutions we study and the survey shows that it is recognized accordingly by the respondents.

Figure 7.1 Mean levels of legitimacy assessments per institution.

Additionally, we observe that for the UNFCCC, evaluations of its input legitimacy are on average more positive than those of its output legitimacy. For the other institutions, however, no such division is visible. Furthermore, the evaluation of the input legitimacy of the UNFCCC is higher than that of CEM, IEA, and IRENA, while the perceived output legitimacy of the UNFCCC is similar to that of the IEA, IRENA, and REN21. One possible explanation for the UNFCCC’s strong performance on input dimensions could be that the highly political negotiations have forced the institution to put increased emphasis on strengthening inclusion and transparency to maintain legitimacy, at least in comparison to the other institutions in our study. This has particularly been highlighted in the aftermath of the Copenhagen conference in 2009 (Karlsson-Vinkhuyzen and McGee 2013). Moreover, transparency and inclusion were key to the French Presidency that was successful in concluding the Paris Agreement (Brun 2016).

7.6.2 Exploratory Factor Analysis of Legitimacy Dimensions

The next step in the analysis uses exploratory factor analysis in order to examine the underlying structure in the data. Exploratory factor analysis is a statistical method that is used to determine how many distinct constructs are captured by a set of measures, in a case where the researcher does not have definite expectations about the underlying structure of correlations between the observed measures. In other words, it shows whether the measures capture different aspects of one broader construct or whether they capture multiple constructs (Fabrigar and Wegener 2012).

In this particular study, we test whether the nine dimensions of legitimacy together measure one underlying construct, ‘legitimacy assessments’, or whether they load on two separate factors, as they might as well capture ‘input legitimacy assessments’ on the one hand, and ‘output legitimacy assessments’ on the other. The factor loadings and Eigenvalues of the exploratory factor analyses for each institution indicate that the items indeed load on one underlying factor which we label ‘legitimacy assessments’. Hence, the individual assessments of each legitimacy dimension can be treated as part of a broader measure for legitimacy assessments, in a multi-faceted manner.

For this reason, the composite indicator (a sum-scale ranging from 1 to 5) for perceived legitimacy of each institution was used in the remainder of the study as a measure for respondents’ legitimacy assessments. On average, respondents’ assessment of CEM across the different dimensions of legitimacy (mean=2.845) is less positive than that of any of the other institutions, while the legitimacy assessment of the UNFCCC (mean = 3.707) is the most positive for all institutions in our study (confirmed by t-tests). This suggests that, in comparative terms, the UNFCCC is perceived to meet best the normative expectations of respondents – which corresponds to the institution’s relatively good formal record on some of the criteria. The average overall legitimacy assessments of the IEA, IRENA, and REN21 do not significantly differ from one another. Thus, this first observation indicates that the extent to which institutions formally meet normative legitimacy criteria has an influence on individual legitimacy assessments (cf. Karlsson-Vinkhuyzen and McGee 2013).

7.6.3 Legitimacy Assessments among Subsets of Stakeholders

We also sought to understand how these legitimacy assessments differ across actors with different backgrounds. As we show in the following, the variation in the data across stakeholder groups qualifies the previous observation: it shows that the formal compliance of an institution with normative legitimacy criteria does not directly translate into stakeholders’ legitimacy assessments. By looking for systematic patterns in the legitimacy assessments of different categories of stakeholders, we seek to better understand what shapes such assessments. Further t-tests were therefore conducted in order to explore how different institutions are perceived by different categories of stakeholders and how the issue areas and geographical backgrounds of respondents might affect their assessments of the different institutions.

Table 7.2 shows how state and nonstate actors ranked the different institutions in terms of legitimacy assessments. The means and reported t-tests in Table 7.2 confirm, for both types of actors, the generally observed pattern of a more negative legitimacy assessment of CEM, and a more positive assessment of the UNFCCC, compared to the other institutions. Yet, in addition to this similarity, we also observe a slightly different rank-order in legitimacy assessments within both categories of respondents. The IEA (mean 3.678) is ranked significantly higher than IRENA (3.479) amongst state actors, while not being ranked significantly lower than the UNFCCC. Among nonstate actors, REN21 is not assessed as significantly less legitimate than the UNFCCC.

Table 7.2 Legitimacy Assessments of Institutions, structured by actor type.

	Mean	SE	95% CI	N		Results t-tests
State actors
CEM	3.026	0.206	[2.590–3.462]	17*		Significantly lower than all other
REN21	3.411	0.259	[2.825–3.998]	10*
IRENA	3.479	0.149	[3.175–3.783]	29
IEA	3.678	0.116	[3.440–3.916]	29		Significantly higher than IRENA and CEM
UNFCCC	3.714	0.130	[3.448–3.979]	31		Significantly higher than all but IEA
Nonstate actors
CEM	2.739	0.149	[2.434–3.045]	29		Significantly lower than all other
IRENA	3.540	0.084	[3.372–3.708]	63
IEA	3.577	0.083	[3.411–3.743]	72
REN21	3.655	0.122	[3.407–3.903]	37
UNFCCC	3.705	0.071	[3.564–3.845]	87		Significantly higher than all but REN21

Source: CLIMENGO Expert Survey 2017–2018.

Notes: * Few respondents within this category were sufficiently familiar with the institution in order to evaluate it on all legitimacy dimensions. Results should be interpreted with this caution in mind. Given the modest sample size, a 90 per cent confidence level is used as the cut-off point for significance testing.

Table 7.3 pairs these figures according to state and nonstate actors’ legitimacy assessments for each institution. The means and t-tests further indicate significant differences in the legitimacy assessments among these two actor groups. Assessments of CEM are significantly lower among the surveyed nonstate actors than among the state actors. By contrast, legitimacy assessments of REN21 are significantly more positive among nonstate actors than among state actors. This observation suggests that the inclusion of nonstate actors in an institution plays a role in shaping legitimacy assessments: CEM is the organization with the least access to nonstate actors in our study, while REN21 is the most open one, being a multi-stakeholder network that reaches out to a broad range of nonstate actors in the public and private sectors. Hence, nonstate actors may be more familiar with REN21 so that this institution might be incorporated in their heuristics of what a legitimate climate and energy governance institution could look like. (Tallberg et al. 2014; Lenz and Viola 2017).

Table 7.3 Legitimacy Assessments of Institutions, structured by institution and actor type.

	Mean	SE	95% CI	N	T-test (t)
CEM
State actors	3.026	0.206	[2.590–3.462]	17*
Nonstate actors	2.739	0.149	[2.434–3.045]	29
					−1.925 (p<0.05)
IEA
State actors	3.678	0.116	[3.440–3.916]	29
Nonstate actors	3.577	0.083	[3.411–3.743]	72
					−1.215 (ns)
IRENA
State actors	3.479	0.149	[3.175–3.783]	29
Nonstate actors	3.540	0.084	[3.372–3.708]	63
					0.727 (ns)
REN21
State actors	3.411	0.259	[2.825–3.998]	10*
Nonstate actors	3.655	0.122	[3.407–3.903]	37
					1.992 (p<0.05)
UNFCCC
State actors	3.714	0.130	[3.448–3.979]	31
Nonstate actors	3.705	0.071	[3.564–3.845]	87
					–0.131 (ns)

Source: CLIMENGO Expert Survey 2017–2018.

Notes: * Few respondents within this category were sufficiently familiar with the institution in order to evaluate it on all legitimacy dimensions. As for CEM and REN21, few state actors are included, and the variances in legitimacy assessments were compared between the largest and smallest group following de Winter (2013). As the variances are relatively equal, the likelihood of Type I error (i.e. observing a false positive result) is low.

Table 7.4 shows how respondents rank institutions differently depending on whether they work in: energy security and technology; climate finance, carbon pricing and mitigation; or adaptation and development. While these categories are partially overlapping, they help distinguish actors according to their main domain (energy, climate, or development).⁵ A simple ranking of the mean legitimacy assessment of the five institutions for each category of respondents again shows that CEM is ranked the lowest and the UNFCCC the highest. Yet, no statistically significant differences are detected between the institutions among the adaptation and development respondents. At least for CEM and REN21, this is most likely due to the limited number of respondents. For the climate finance, carbon pricing, and mitigation group, we do observe that the UNFCCC is ranked significantly higher than all other institutions, the IEA significantly higher than IRENA and CEM, and CEM significantly lower than all other institutions. Furthermore, CEM is ranked significantly lower than all other institutions for the energy security and technology respondents. This last finding is counterintuitive, given that CEM has a clear focus on energy and technology questions.

Table 7.4 Legitimacy Assessments of Institutions, structured by stakeholders’ work focus.

	Mean	SE	95% CI	N	Results t-tests
Energy security and technology
CEM	3.044	0.147	[2.743–3.345]	28	Significantly lower than all other institutions
REN21	3.545	0.183	[3.165–3.926]	22*	‡
IRENA	3.632	0.101	[3.428–3.835]	48
IEA	3.667	0.088	[3.489–3.844]	54
UNFCCC	3.795	0.085	[3.625–3.964]	56	Significantly higher than all but REN21 (given ‡)
Climate finance, carbon pricing and mitigation
CEM	2.667	0.171	[2.306–3.027]	18*	Significantly lower than all other institutions
IRENA	3.452	0.093	[3.264–3.640]	43
REN21	3.545	0.164	[3.202–3.889]	21*
IEA	3.618	0.102	[3.413–3.824]	46	Significantly higher than IRENA and CEM
UNFCCC	3.761	0.088	[3.585–3.937]	57	Significantly higher than all other institutions
Adaptation and development					(no significant differences between the evaluations of the institutions)
CEM	2.570	0.281	[1.967–3.174]	15*	‡
REN21	3.299	0.200	[2.864–3.735]	13*	No significant differences
IRENA	3.379	0.160	[3.050–3.707]	27
IEA	3.449	0.149	[3.143–3.755]	26
UNFCCC	3.511	0.141	[3.227–3.796]	39

Source: CLIMENGO Expert Survey 2017–2018.

Notes: * Few respondents within this category were sufficiently familiar with the institution in order to evaluate it on all legitimacy dimensions. ^‡Following the method of de Winter (2013), no comparison could be made between the mean perceived legitimacy among these respondents. The variance is too high for this small group of observations, compared to variances of the other means in the analysis. Given the modest sample size, a 90 per cent confidence level is used as the cut-off point for significance testing.

Table 7.5, which rearranges these issue-area-based figures along the five institutions, sheds more light on this observation about CEM. Overall, respondents working with energy security and technology tend to have the most positive legitimacy assessments; those working with adaptation and development have the least positive ones. Moreover, for both CEM and IRENA, the difference between respondents working with energy security and technology and respondents mainly working with other issues is most pronounced. In other words, respondents for whom energy security and technology is most central to their work tend to assess the legitimacy of those institutions that focus most strongly on these issues as particularly more positive than other respondents. For institutions that, next to energy security and technology, also focus on mitigation, climate finance and carbon pricing, development, and adaptation (IEA and UNFCCC), we observe that the legitimacy assessments are not significantly different among respondents working with energy security and technology and those working with climate finance, carbon markets and pricing, and mitigation. This again suggests the importance of the thematic foci of the institutions for the legitimacy assessments of stakeholders. Respondents who work with adaptation and development make the most negative legitimacy assessments for all institutions, even for the UNFCCC, although adaptation and low-carbon development feature prominently on that institution’s agenda.

Table 7.5 Legitimacy Assessments of Institutions, structured by institution and stakeholders’ work focus.

	Mean	SE	95% CI	N	T-test
CEM
1. Energy security and technology	3.044	0.147	[2.743–3.345]	28	1 vs. 2: t=2.569, p<0.01 1 vs. 3: ‡
2. Climate finance, carbon pricing and mitigation	2.667	0.171	[2.306–3.027]	18*	2 vs. 1: t=-2.208, p<0.05 2 vs. 3: ‡
3. Adaptation and development	2.570	0.281	[1.967–3.174]	15*	3 vs. 1: ‡ 3 vs. 2: ‡
IEA
1. Energy security and technology	3.667	0.088	[3.489–3.844]	54	1 vs. 2: t=0.551, ns 1 vs. 3: t=2.462, p<0.01
2. Climate finance, carbon pricing and mitigation	3.618	0.102	[3.413–3.824]	46	2 vs. 1: t=-0.477, ns 2 vs. 3: t=1.661, p<0.1
3. Adaptation and development	3.449	0.149	[3.143–3.755]	26	3 vs. 1: t=-1.469, p<0.1 3 vs. 2: t=-1.140, ns
IRENA
1. Energy security and technology	3.632	0.101	[3.428–3.835]	48	1 vs. 2: t=1.779, p<0.05 1 vs. 3: t=2.500, p<0.001
2. Climate finance, carbon pricing and mitigation	3.452	0.093	[3.264–3.640]	43	2 vs. 1: t=-1.932, p<0.05 2 vs. 3: t=0.787, ns
3. Adaptation and development	3.379	0.160	[3.050–3.707]	27	3 vs. 1: t=-1.586, p<0.1 3 vs. 2: t=-0.459, ns
REN21
1. Energy security and technology	3.545	0.183	[3.165–3.926]	22*	1 vs. 2: t=0.003, ns 1 vs. 3: t=2.569, p<0.01
2. Climate finance, carbon pricing and mitigation	3.545	0.164	[3.202–3.889]	21*	2 vs. 1: t=-0.002, ns 2 vs. 3: t=1.496, p<0.1
3. Adaptation and development	3.299	0.200	[2.864–3.735]	13*	3 vs. 1: t=-1.230, ns 3 vs. 2: t=-1.230, ns
UNFCCC
1. Energy security and technology	3.795	0.085	[3.625–3.964]	56	1 vs. 2: t=0.398, ns 1 vs. 3: t=3.355, p<0.001
2. Climate finance, carbon pricing and mitigation	3.761	0.088	[3.585–3.937]	57	2 vs. 1: t=-0.388, ns 2 vs. 3: t=2.849, p<0.01
3. Adaptation and development	3.511	0.141	[3.227–3.796]	39	3 vs. 1: t=-2.018, p<0.05 3 vs. 2: t=-1.777, p<0.05

Source: CLIMENGO Expert Survey 2017–2018.

Notes: * Few respondents within this category were sufficiently familiar with the institution in order to evaluate it on all legitimacy dimensions. ^‡ Given the low N of one of the categories in the comparison, the variances in legitimacy assessments were compared between the largest and smallest group following de Winter (2013). When the variances are relatively equal or when the variance is smaller in the category with the lowest N, the likelihood of Type I error (i.e. observing a false positive result) is low. In these cases, the result of the t-test is presented. Where this criterion is not met, the result of the t-test is omitted.

These observations are in line with one of our aforementioned expectations, namely that differences in legitimacy assessments could stem from differences in norms and values amongst communities of stakeholders working on similar issues, or from differences in the institutions that they are familiar with and use as heuristics or prototypes. According to the cognitive model, the observed positive assessments of respondents working on issues of energy security and technology would be explained by their higher level of familiarity with institutions that fulfil fewer criteria for normative legitimacy, which in turn would reflect in the heuristics they use to compare the five institutions in our study against. In comparison to those, they assess the five climate-energy institutions in our study more favourably. For example, if respondents working mainly with energy issues take institutions such as OPEC and the International Energy Forum as reference points, the five institutions in this study could be considered more legitimate as they fulfil more normative criteria, particularly in terms of openness and transparency.

By contrast, those respondents working on issues of adaptation and development tend to be familiar with institutions that fulfil rather more criteria for normative legitimacy. Compared to such prototypes, they would assess the five climate-energy institutions less favourably. For example, the Global Environment Facility and the Green Climate Fund that focus more on development issues may constitute such reference institutions (in terms of heuristics) for the respondents who work mainly on adaptation and development. Particularly the Global Environment Facility has been discussed as a potential role model for other international institutions due to its inclusiveness and openness toward a diversity of actors (Streck 2001).

In summary, we can expect the prototype institutions to differ considerably across groups of respondents, which could explain a large part of the difference in legitimacy assessments that we found for our sample. This said, this connection needs to be further corroborated, since our survey data does not include information on the heuristic that respondents had in mind when assessing the five institutions. The focus of our study, thus, remains exploratory and descriptive, yet it suggests avenues for further explanatory research.

Finally, we explored whether significant variations can be observed in the legitimacy assessments by respondents from countries with different economic backgrounds, or from different geographical regions (Africa, Asia and Pacific, Latin America and Caribbean, and European and other Western countries). For the UNFCCC, we indeed observe such a significant difference. Respondents from high-income countries (as categorized by the World Bank) perceive the UNFCCC as significantly more legitimate (mean= 3.773) than respondents from other countries (mean= 3.563; t= 2.912; p=0.002). Moreover, no significantly different views about the legitimacy of the UNFCCC are observed for respondents from middle- and low-income countries. Neither did we observe a significant difference in evaluations of the other institutions when we grouped respondents by national-income category.⁶

This is rather surprising, given that both norms and values and what institutions respondents are familiar with would be expected to vary with the geographic origin of the respondent. It may be that many of the respondents are international elites and have therefore been socialized or self-selected into similar norms, and are hence used to similar international institutions. Thus, we might indeed be looking at dynamics of a transnational elite that is divided by professional focus, rather than by nationality – since we did observe distinctions in legitimacy assessments across respondents from different sectors and types of organizations. In fact, previous research supports this assumption: Verhaegen et al. 2018, for instance, showed that there is more variation in legitimacy perceptions of global governance institutions between elites of different societal sectors than between elites from different countries.

7.7 Conclusions

The aim of this chapter was to provide a first mapping of stakeholders’ assessments of the legitimacy of five key institutions governing the climate-energy nexus. Against the backdrop of considerable institutional complexity, and scarce resources amongst public and private actors to enhance participation in global governance, we wanted to better understand to what extent key institutions in global climate and energy governance are seen as legitimate by key stakeholders.

The analyses showed that, on the one hand, there are many similarities in the legitimacy assessments of the five institutions we put under scrutiny – with the mean legitimacy assessments ranging between 2.845 (CEM) and 3.707 (UNFCCC) on a scale from 1 to 5. On the other hand, we also found systematic differences across stakeholders of different types, working with different issues and – to a limited extent – coming from different countries.

Specifically, we observed that CEM is systematically assessed as the least legitimate, and the UNFCCC as the most legitimate, of the five institutions. Second, our analyses showed that the legitimacy assessments of nonstate actors are more positive toward institutions that are more inclusive toward this type of stakeholders. Third, we observed that stakeholders working with energy security and technology, and those working with climate finance, carbon pricing, and mitigation have more positive legitimacy assessments of institutions that more strongly focus on their issues. By contrast, respondents working with adaptation and development issues assessed the legitimacy of the selected institutions more negatively than the other respondents, even for the UNFCCC, which is the global institution in our sample that most strongly engages with these issues. We can only speculate about the reasons for this. Our study has highlighted the possibility that differences in these communities’ norms, values, and experiences contribute to different heuristics being used to make assessments. Yet, whether such differences ultimately stem from processes of socialization or whether they are rather due to functionalist or rationalistic reasons is a pertinent question for future research.

These limitations notwithstanding, the results of our unique survey allow us to draw a set of novel conclusions. First, the results appear to support the view that stakeholders do not adequately disentangle their legitimacy assessments of individual institutions that have similar functions and overlapping mandates. Perhaps this is a reflection of the relatively high level of coordination in the renewable energy subfield (Sanderink, Chapter 4), which means that institutions in this subfield interact extensively with one another and thereby make it difficult for stakeholders to distinguish their respective performance. Within each category of stakeholders, we found comparable assessments and similar legitimacy rankings for these institutions, albeit with some small significant differences. We have reasons to believe that, in order to navigate in a very complex governance field, the surveyed stakeholders form their assessments based on a comparison with institutions that are familiar to, or valued by them. Faced with incomplete information and due to bounded rationality, stakeholders use mental shortcuts to make such comparisons and base their legitimacy assessments thereupon.

Second, the differences in legitimacy assessments found between governmental versus nonstate actors, and across stakeholders working on different issue areas, suggest that international institutions have to pursue different legitimation strategies for different audiences (Gronau and Schmidtke 2016; Bäckstrand and Söderbaum 2018; Verhaegen et al. 2018). Knowing one’s audiences is particularly important for institutions that seek to establish and maintain legitimacy in an increasingly crowded field.

Third, and more generally, a stakeholder’s level of familiarity with an institution appears to be linked to a more positive assessment of legitimacy. It therefore does not come as a surprise that international institutions increasingly engage in outreach activities, especially on social media – in order to promote their work to a diversity of actors and seek input from them.

While these findings advance the research frontier on legitimacy, the reliance on survey data comes with the usual set of shortcomings, which means that the findings need to be confirmed in future studies. First, there is always the possibility that respondents either think of the institution as a whole or just the secretariat or another institutional body, which makes a straight-off comparison difficult (Zaum 2013). Second, we followed a notion of cognitive legitimacy whereby respondents’ assessments are based on comparisons with heuristic or prototype institutions. Which particular heuristic institutions that are used by respondents lies beyond the scope of this study. The first explorative results offered in this chapter should therefore be examined in further, especially interview-based, studies. These could also delve deeper into questions such as how expertise, which is considered key for international institutions in the climate-energy nexus, is conceptualized by stakeholders. Finally, the links to other explanatory variables, such as resources, staffing, or relations to other institutions, should be pursued to further understand assessments of legitimacy.

This study also opens up empirical avenues for further research. It provided a first mapping of stakeholders’ perceptions of nine legitimacy dimensions across five institutions for one particular subfield. An examination of institutions from other subfields could provide insights into how the level of coherence within institutional complexes affects issues of legitimacy. Next steps could also measure differences in how stakeholders view the relative importance of the nine dimensions, or other dimensions of legitimacy not included in this study, to also learn about the sociological legitimacy of the institutions. An interesting and policy-relevant line of inquiry is how low assessments of certain dimensions of legitimacy can be, and amongst which groups of stakeholders, before the institution faces a legitimacy crisis. An answer to that question would, however, require a much larger survey of stakeholders. One limitation of this study has been that the survey includes too few cases (N) in order to do a multivariate analysis that allows comparing the relationship between actor type, issue area, and geographical origin on the one hand, and assessments of legitimacy of the five global climate and energy institutions on the other. A larger research effort would be needed to address this limitation. Such a research effort would also be useful to provide a more fine-grained analysis of differences in legitimacy assessments between different categories of nonstate actors, such as businesses and civil society organizations.

Finally, considerations of legitimacy will always be of major importance for policy makers when deciding on which institution to work with and invest in. Institutional complexity affects these conditions, as institutions and their legitimacy have become highly entangled. Therefore, further research questions, such as about the role of legitimation and delegitimation strategies under institutional complexity, merit further enquiry, as such strategies are likely to affect institutions differently, depending on the norms, values, and experiences of the legitimacy-granting communities.

8 The Performance of the Climate-Energy Nexus Assessing the Effectiveness of the Institutional Complexes on Renewable Energy, Fossil Fuel Subsidy Reform, and Carbon Pricing

Naghmeh Nasiritousi , Lisa Sanderink , Jakob Skovgaard , Harro van Asselt , Cleo Verkuijl , and Oscar Widerberg

8.1 Introduction

What does institutional complexity mean for performance in the climate-energy nexus? As previous chapters have shown, the nexus is made up of a diverse set of institutions that have overlapping mandates and functions. Chapter 3 showed how the institutional complex varies at the meso level, and Chapters 4–6 explored the interactions between different institutions in three selected subfields: renewable energy, fossil fuel subsidy reform, and carbon pricing. Given the large number of institutions that operate and interact in these fields, several questions arise about their performance and environmental effectiveness: what are the consequences of this intricate web of institutions for the performance of the institutional complex of the climate-energy nexus? Is institutional complexity a requirement for effective problem-solving, or does it hamper effectiveness? What management options exist for making the institutional complex at the climate-energy nexus more effective? Considering the magnitude of the climate- and energy-related challenges, the answers to these questions are of great importance to both scholars and practitioners.

Based on these questions, and building on the previous chapters, the aim of this chapter is to assess the effectiveness of each of the three subfields as well as to discuss the overall performance of the institutional complex of the climate-energy nexus. As outlined in Chapter 2 and elaborated on in the next section, effectiveness here refers to how well institutions perform in terms of achieving goals that they have been tasked to fulfil. By examining the outputs, outcomes, and impacts of the three subfields, the chapter shows both the advantages and the disadvantages of institutional complexity of the climate-energy nexus for achieving effectiveness. It further shows that, despite the difficulties with evaluating effectiveness under institutional complexity, such an assessment is a worthwhile exercise in order to identify management options – i.e. options for formally regulating the linkage between institutions – for the climate-energy nexus.

The chapter proceeds as follows. The next section discusses the concept of effectiveness and the challenges to analyzing the effectiveness of institutions, especially when they have overlapping mandates and are interlinked. Thereafter, our methodology section outlines how, in order to respond to these challenges, our research relies on a two-track approach, integrating assessments by researchers and interviews with key stakeholders. Based on this information, we evaluate the outputs, outcomes, and impacts of institutions within each subfield. Thereafter, we examine what the consequences of institutional complexity are for the subfield in question. The insights gained from this analysis are then used to outline management options for the institutions of the climate-energy nexus. The final section concludes with discussing implications of our findings for the governance of the nexus at large.

8.2 Conceptualizing Effectiveness

As discussed in Chapter 2, effectiveness can be evaluated in different ways. Easton (1965) suggested measuring effectiveness across three dimensions: output, outcome, and impact. Output concerns performance in terms of what the institution produces, for example issuing regulations (binding or non-binding), producing reports, conducting research, organizing meetings, providing funding, offering training, etc. (Szulecki et al. 2011). Outcome relates to whether the institution produces behavioural changes, for example in terms of whether it increases the level of cooperation and compliance amongst members, for instance by improving learning and modifying incentives (Underdal 2002; Gutner and Thompson 2010). To determine an institution’s impact implies assessing the extent to which the institution contributes to alleviating the problem it was tasked to resolve (Underdal 2002). Impacts may include effects that are positive or negative, direct or indirect, intentional or unintentional, and these can be short-, medium-, or long-term (Alcamo 2017). This threefold understanding implies that effectiveness is a stronger term than performance, since institutions can perform well in terms of output but nevertheless not achieve the intended impacts necessary for goal attainment.

Measuring effectiveness becomes increasingly difficult as the number of institutions rises. Even just for one institution, assessing effectiveness is challenging because of the need to establish causality between the output of the institution, the behavioural change among the target actors, and the impact on the problem that the institution was tasked to solve. This challenge is multiplied under institutional complexity because of the question of attribution, namely which institutions are responsible for observed impacts in a web of institutions with overlapping mandates? In short, under institutional complexity, the difficulties involved in assessing effectiveness are compounded by challenges in identifying the division of labour between institutions (Alter and Meunier 2009).

Moreover, when evaluating impact for a field with multiple institutions, the analysis shifts from assessing goal attainment for individual institutions to assessing how the work of multiple institutions affects an overall goal, such as the fulfilment of Sustainable Development Goal (SDG) 7 on sustainable energy in the case of the renewable energy subfield. This approach is different from what can be found in much of the previous literature on effectiveness, where the focus is either on assessing institutional or environmental effectiveness (Underdal 2002; Gutner and Thompson 2010; Tallberg et al. 2016). Analyses of institutional effectiveness look at institutional performance, also including assessments of output legitimacy, such as the one presented in Chapter 7. Studies on environmental effectiveness, on the other hand, look at the extent to which specific institutions have an impact on environmental indicators. In contrast, the analysis in this chapter looks at the extent to which the collective contributions of individual institutions within a subfield are successful in fulfilling common goals in the subfield.

Some studies seek to circumvent the challenge of identifying outcomes and impacts of institutions by simply focusing on outputs (Szulecki et al. 2011; Tallberg et al. 2016). By examining outputs, these scholars assess the performance of institutions and thereby look at potential effectiveness. Alternatively, some studies analyze effectiveness by examining whether institutions are producing the outputs that could be expected, given the functions that they have (Pattberg et al. 2012; Chan et al. 2018). However, these approaches are at best useful as first steps and approximations for assessing the actual effects of institutions on the governance of particular issue areas.

The approach employed in this chapter instead seeks to link outputs to observed outcomes and impacts. For each subfield, our approach identifies specific outputs and discusses possible outcomes and impacts. While imperfect due to knowledge limitations, this approach makes explicit how assessments of effectiveness are made and thereby allows for critical reflection and learning about the cause and effect of institutional consequences. The aim of the analysis is hence not to show whether the institutions are effective or not but to discuss and specify in what ways they could be seen as effective (or not) and how institutional complexity affects effectiveness. The value of this approach lies in its context-specific analysis of outputs, outcomes, and impacts for each subfield and in deriving suitable management options to enhance effectiveness. Our own assessments are complemented by interview data from a range of stakeholders with high familiarity of institutions working within these subfields, as explained in the next section.

Assessing effectiveness across the three subfields (renewable energy, fossil fuel subsidy reform, and carbon pricing) was carried out by analyzing key documents and reports, as well as conducting semi-structured interviews with various experts. For the document analysis, we focused on academic journal articles but also included grey literature such as reports from international organizations and nongovernmental organizations. For the interviews, we approached, for each subfield, representatives from national governments, international organizations, NGOs and academia. The interviewees were selected based on thematic expertise and knowledge of the institutions in each case study, with a view to cover a wide variety of actor types, countries and sectors. In total, thirty-eight interviews were carried out across the three subfields.

The interviews covered the following aspects: (1) the overall degree of effectiveness in the subfield; (2) possible bottlenecks that may hamper effectiveness; (3) influence of institutional complexity on effectiveness; and (4) management options by particular institutions to improve the effectiveness of the subfield. The analyses are based on a careful assessment of the data retrieved from the document analyses and the interviews. More specifically, output effectiveness was mainly determined based on the analyses in Chapters 4–6, whereas estimations of outcome and impact were mostly derived from academic and grey literatures. Findings about the influence of institutional complexity on effectiveness as well as management options are mainly based on the experts’ views.

In what follows, we examine effectiveness, the consequences of institutional complexity and management options for each subfield at the meso level. The concluding section offers a comparison of the three subfields and draws out the implications of our findings for the performance of the institutional complex of the climate-energy nexus.

8.3 Assessments of Effectiveness for the Renewable Energy Subfield

8.3.1 Assessment of Outputs, Outcomes, and Impacts

A sustainable energy future hinges on a worldwide uptake of renewable energy. Many of the institutions that operate in the renewable energy subfield relate their work both to the SDG 7 target on clean energy and to the Paris Agreement’s temperature target (Chapter 4). This section assesses the effectiveness of the densely crowded subfield for renewable energy.

First, with regard to output, Chapter 4 showed that the majority of renewable energy institutions focus on information-sharing through dissemination of research and publishing reports. Consequently, the renewable energy subfield displays a diversity of knowledge and expertise on energy sources and technologies from a wide range of perspectives, which is frequently shared at various meetings, conferences, and platforms.¹ In addition, there are a fair number of renewable energy institutions working toward capacity-building and project implementation (see also Chapter 4). These particularly focus on deploying renewables for the purpose of expanding energy access in the developing world. Financing schemes, regulations, standards, and commitments are produced to a lesser extent.

Second, in terms of outcome, awareness and capacity are growing along a wide spectrum of stakeholders. For example, via renewable energy institutions, national governments are increasingly sharing experiences and taking note of best practices.² Simultaneously, nonstate engagement is spreading, which is illustrated by the growing number of private initiatives and multi-stakeholder partnerships for renewable energy (see Chapter 4). Businesses, trade associations, financial institutions, NGOs, and other civil society organizations show increasing interest in renewables. In short, renewable energy institutions appear to stimulate if not behavioural, then attitudinal changes amongst their members and beyond.

Third, assessing the level of goal attainment or problem-solving capacity suggests a low degree of effectiveness. Despite 2017 being a record-breaking year for the share of renewables in the global energy mix, the growth rate is currently falling short of meeting either the ‘substantial increase’ by 2030, as targeted by SDG 7, or the 2-degrees target set by the Paris Agreement (IRENA 2018; REN21 2018; United Nations 2018). It is difficult to determine the level of effectiveness based on broad perceptions, let alone for an institutional complex that includes such a high number of different institutions. This notwithstanding, the currently inadequate growth rate for renewables suggests that the subfield’s institutional complex has suboptimal performance.³

Why is this the case? We could identify various bottlenecks through our interviews and literature review. First, the renewable energy subfield inherited the dominance of national policy making in global energy governance (Karlsson-Vinkhuyzen et al. 2012; Röhrkasten 2015; Van de Graaf and Zelli 2016). Even though renewable energy may be a less strategic issue for national security compared to traditional sources of energy, national governments have remained hesitant to give up sovereignty.⁴ Second, the subfield is steered by three different challenges, each of which controversial in its own right – energy security, energy access, and environmental sustainability – resulting in trade-offs and potential conflicts across institutions (see Chapter 4; Newell et al. 2011; and Röhrkasten 2015). Furthermore, there is no clear definition of what constitutes a renewable source of energy, leading to further controversies, for example related to nuclear power, bioenergy, and hydropower.⁵ These aspects at least hinder the effectiveness of the subfield as a whole in terms of accelerating a renewables uptake. Several other bottlenecks relate to institutional complexity and are therefore discussed in the next subsection.

8.3.2 Consequences of Institutional Complexity: What Are the Implications for Renewable Energy?

To what extent can the low degree of effectiveness be attributed to the institutional complexity of the renewable energy subfield? Compared to the other two cases in this edited volume, carbon pricing and fossil fuel subsidy reform, the renewable energy subfield can be regarded as highly institutionally complex. The subfield is densely populated by a diverse set of institutions, which do not only differ in terms of their structural characteristics but also with respect to the functions they perform, the sources of energy and technologies they cover, and the challenges they seek to address (see Chapter 4). While this complexity makes it difficult to establish a causal relationship with the level of effectiveness, our interviewees on balance expect more advantages than disadvantages with institutional complexity.⁶

On the one hand, institutional complexity is considered to support effectiveness in two ways. First, the variety of institutions involved renders more comprehensive information available from a wide range of perspectives on renewable energy sources and technologies as well as on developments and innovations in the field.⁷ Second, institutional complexity provides the opportunity to disaggregate an intricate issue such as renewable energy into smaller challenges and to work on them in parallel with different degrees of progress.⁸ Such a compartmentalizing approach has proven to be effective for the climate change realm and may be particularly suitable for a subfield such as renewable energy, characterized by the diversity of energy sources and technologies and differing challenges to tackle.

On the other hand, institutional complexity may turn out problematic for effectiveness in several ways. First, the interviewees express concerns about duplication of work and conflictive impacts among the renewable energy institutions.⁹ With several institutions working on similar issues, it is sometimes unclear whether there are overlaps, or worse, incongruences, trade-offs, and conflicts between institutions (Biermann et al. 2009). As a consequence, it is difficult for national governments to decide which organizations, partnerships, and initiatives to participate in, and for the institutions themselves to identify thematic and functional gaps that need to be filled. Second, there exists competition over resources, visibility, sphere of influence, and media attention. This competition may involve institutions that target different renewable energy sources, but also institutions from related issue areas such as energy efficiency.¹⁰ Third, there is no single institution with universal membership in the renewable energy subfield.¹¹ Such an institutional umbrella may, according to some observers, be ultimately necessary to achieve the common goal to substantially increase the share of renewables in the global energy mix. However, with 160 states as members and 23 in accession by 2019, IRENA is well on its way to positioning itself as one and to continue its unique multilateral success story in global (renewable) energy governance (Röhrkasten and Westphal 2013; Urpelainen and Van de Graaf 2015).

Yet, there is also a different perspective. Various scholars have recently argued that the emerging global transition toward renewable energy is not the result of deliberate and integrated international cooperation, but rather the result of an organic proliferation of bottom-up initiatives (e.g. Aklin and Urpelainen 2018; Meckling 2019). Although it is difficult to assess the overall consequences of institutional complexity on effectiveness in the renewable energy subfield, our findings tend to support the argument that the current institutionally complex structure seems fitting, and perhaps even required, for the renewable energy subfield (see also Young 2002). Furthermore, there may be less of a need for institutional integration today than prior to 2015: with the Paris Agreement and SDG 7 agreed upon, ‘discursively there is in any case a high degree of consensus’.¹²

8.3.3 Management Options for the Renewable Energy Subfield

The general view among climate and energy experts interviewed is that the renewable energy subfield is functioning fairly well, institution-wise, partly guided by the targets and principles presented in the Paris Agreement and SDG 7.¹³ This notwithstanding, there is a need for increased coordination among the renewable energy institutions to resolve the potentially negative implications of institutional complexity and, ultimately, to achieve targets to substantially accelerate the worldwide uptake of renewables.

First, it is necessary to map out existing renewable energy institutions, and their functions and targeted impacts, and to keep track of the progress being made.¹⁴ This will help to prevent and resolve duplication of work and conflictive impacts and to identify docking points and gaps that need to be addressed among existing institutions. To clarify the latter, there is no need for new institutions trying to reinvent the wheel, but rather to find ways for collaboration to strengthen the overall outcome.¹⁵ Second, with a plethora of knowledge and expertise comes a variety of scenarios, statistics, and data, based on a range of different methodologies and definitions that are not always compatible across institutions. In order to prevent and resolve competition among different measurements and related practices, more cognitive alignment and agreement is needed with regard to methodologies to determine the uptake of renewables and to find some common understanding for what constitutes a renewable source of energy.¹⁶ Finally, it is necessary to coordinate interaction and collaboration beyond the renewable energy subfield – with sectors that deploy renewables such as transportation and heating, but also with sectors that are reluctant to deploy renewables, as well as with the issue area of energy efficiency.¹⁷

While these coordination efforts appear feasible, these are merely desirable as long as they do not add another level to the management structures of renewable energy institutions.¹⁸ Furthermore, coordination attempts should neither compromise the autonomy of institutions nor constrain them in their functioning and experimenting.¹⁹

8.4 Assessments of Effectiveness for the Fossil Fuel Subsidy Reform Subfield

8.4.1 Assessment of Outputs, Outcomes, and Impacts

There are persuasive economic, social, and environmental reasons to tackle fossil fuel subsidies. Over the past decade, more than a dozen international institutions have begun to address this issue from various angles, from information provision and agenda setting to capacity-building and financing of reform efforts.

Taken together, these activities have led to a range of outputs. Members of several forums – including the G7 (Group of 7), G20 (Group of 20), APEC (Asia-Pacific Economic Cooperation), Friends of Fossil Fuel Subsidy Reform (Friends), and the 2030 Agenda process – have expressed commitments to phase down fossil fuel subsidies, although the precise nature of the commitment varies (Chapter 5). The G20 and APEC have also put follow-up mechanisms in place, which allow countries to report on their subsidies, and to have them reviewed by their peers (APEC Energy Working Group 2013; G20 Energy Sustainability Working Group 2013). One key requirement for reform is to ensure an adequate understanding of the scale and impacts of fossil fuel subsidies. Resources such as the Organisation for Economic Co-operation and Development (OECD) and International Energy Agency (IEA) Inventory of Support Measures for Fossil Fuels (OECD 2018a) and the International Monetary Fund’s (IMF) post-tax estimates of fossil fuel subsidies (Coady et al. 2017) seek to shed light on this question. Drawing on such work, as well as the experience of their members and external experts, institutions such as the World Bank, IMF, APEC, and Friends have supported and facilitated workshops, events, and webinars to improve governments’ understanding of reform.²⁰

The outputs of these various institutions can, in turn, be linked to a range of observable outcomes. Reporting mechanisms introduced under the G20 and APEC have prompted members to provide information on their domestic subsidies, although overall such estimates have been much lower than expected (Asmelash 2016), with some countries even claiming to have no subsidies at all (Van de Graaf and Blondeel 2018). Since 2015, more than a dozen G20, APEC, and Friends members have also voluntarily undergone more in-depth peer- or self-reviews, or are in the process of doing so. While the results of these exercises have in some cases been considered disappointing (e.g. ODI 2017), they have, in other cases, facilitated concrete reform plans and timelines (e.g. China 2016). Moreover, it is likely that engagement in review itself can play a valuable role in increasing internal awareness about a country’s subsidies and ways to address them.²¹ Interviewees have also highlighted the helpful role of workshops and other capacity-building activities to strengthen countries’ understanding of fossil fuel subsidy reform (FFSR).²²

While international institutions’ activities in the FFSR space can be associated with several outputs and outcomes, it is more difficult to determine to what extent their efforts have led to increased reform on the ground. At first glance, the data on national reform activities is promising. According to the IEA, global fossil fuel subsidies dropped by just over US $300 billion between 2009 (the year the G20 and APEC committed to phase out fossil fuel subsidies) and 2015 as a result of reform (IEA 2018). The Global Subsidies Initiative estimates that around forty countries underwent some sort of FFSR between 2015 and 2017 alone (Merrill et al. 2018).

In practice, however, there is limited knowledge about the role that an institutional complex as a whole, or even an individual institution, may have in driving such reform. Indeed, it is likely that low oil prices over the past years have contributed significantly to governments’ decisions to adjust or remove the subsidies they provide to consumers of fossil fuels (Benes et al. 2015): while consumption subsidies are currently in decline (Merrill et al. 2018), upstream production subsidies appear on the rise (OECD 2018a). Domestic fiscal pressures could be another key driver of reform (Skovgaard and van Asselt 2018). Yet, while in many cases such decisions may be taken independently from the international context, global developments may also help to inform decisions taken in this regard. Steenblik, for instance, notes that a focus on tax reform in China’s 2016–2020 five-year plan may have been partially informed by the country’s heightened awareness of shortfalls in its internal tax expenditure monitoring system, following its FFSR peer review under the G20.²³ This may suggest that capacity-building and information-sharing can play a more significant role than high-level international commitments in driving reform. Finally, it is worth bearing in mind that efforts to promote reform may take years to come to fruition (with possible setbacks along the way). As such, this analysis should be considered an initial indication, rather than a decisive assessment, of international institutions’ contributions.

8.4.2 Consequences of Institutional Complexity: What Are the Implications for Fossil Fuel Subsidy Reform?

While the overall effectiveness of international institutions in FFSR is difficult to establish, the fact that multiple international institutions are active in this area appears to be largely advantageous for promoting reform.

One advantage of institutional complexity is that different international institutions are associated with different approaches and types of expertise, such as agenda setting, capacity-building, and research.²⁴ The efficiency of the FFSR institutional complex is strengthened by the fact that different organizations can contribute to FFSR efforts in their speciality area(s), as opposed to one institution needing to specialize in all these approaches. At the same time, the fact that multiple institutions are engaged in this area has likely helped ensure that more resources are dedicated to international reform efforts.²⁵ Having multiple institutions working in this area also provides FFSR advocates with increased opportunities to keep this topic on the international agenda, with the respective framing tailored according to the institution’s financial, climate change, trade, or broader social mandate.²⁶

Moreover, the involvement of multiple forums allows advocates of FFSR to point to developments in other institutions to enhance ambition in their own forum, lest it falls behind.²⁷ One example of such positive reinforcement is the peer-review mechanisms established under both the G20 and APEC, described in more detail in Chapter 5. Nevertheless, it should be noted that such dynamics also create potential for upholding the lowest common denominator. This may constitute one explanation for why otherwise more ambitious groupings such as the Friends have opted not to set a timeline for FFSR (see Chapter 5). Finally, institutional complexity also offers the opportunity to widen the geographic scope of FFSR efforts. For instance, the fact that peer reviews are conducted under both the G20 and APEC has allowed economies from both groups to become engaged. Likewise, engagement under the High-level Political Forum on Sustainable Development (HLPF) has provided opportunities to increase traction for reform among various African countries.²⁸

Despite such benefits, there are also certain challenges associated with the institutional complexity in this area. First, and like in the case of renewables, it heightens the risk of duplication of work. Although the peer reviews undertaken under the G20 and APEC can largely be seen as synergistic (Chapter 5), both forums have had to identify guidelines and approaches for review, which means that the process may have not been as efficient as it would have been under one single institution. Another, perhaps more consequential, example is the subsidy estimation work conducted by the various international organizations that produce estimates of fossil fuel subsidies, namely the IEA, IMF, OECD, and, in the future (as custodians of SDG Indicator 12.c.1), UNEP. For instance, although the IMF reports estimates for more countries than the IEA does, both organizations take a similar approach to subsidy estimates that take consumer price support as a starting point. Better coordination among the various institutions involved could help minimize the risk that labour-intensive efforts are unnecessarily repeated and, indeed, prevent inconsistencies in definitions and data from being used as an excuse to postpone action (see OECD 2018b, 10).

A second potentially problematic consequence of institutional diversity in this area has been competition among standards for defining a subsidy. At US $5.3 trillion a year, the IMF’s ‘post-tax’ estimates of fossil fuel subsidies put such support in another order of magnitude compared to OECD and IEA estimates (Coady et al. 2017; Chapter 5).

The ambiguity has been a contributing factor for countries to claim they do not have any fossil fuel subsidies at all (e.g. South Africa, Burton et al. 2018). At the same time, such contestation of what constitutes a subsidy may have certain advantages: it may enable countries to engage in reform at a pace suitable to their national circumstances;²⁹ and different definitions and valuation approaches can highlight complementary information about subsidies, such as, in the case of the IMF definition, their broader societal cost. The extent to which definitions diverge should also not be overstated since there are several areas of measurement on which international organizations are in agreement (Koplow 2018).

8.4.3 Management Options

These findings suggest that there are a few areas in which improved management between different institutions could enhance the effectiveness of international FFSR governance. In terms of duplication of work, while there may be an ownership rationale for the G20 and APEC to take their own approaches to VPRs, better coordination between the intergovernmental organizations involved in monitoring fossil fuel subsidies, such as the IEA, IMF, OECD, UNEP, and the World Bank, may allow both institutions to generate their data more efficiently, and indeed, is something that their members ‘could be demanding’.³⁰ One challenge in this regard, however, is that the IEA depends on data sales for much of its revenue.

Whether international FFSR governance would benefit from enhanced coordination around the definition of a fossil fuel subsidy remains an open question. The first internationally agreed-upon methodology for the measurement of fossil fuel subsidies, issued in the context of monitoring progress on SDG indicator 12.c.1. on fossil fuel subsidies, marks an important step in this regard (UNEP et al. 2019). If efforts to elevate FFSR to the World Trade Organization (WTO) space prove successful (Verkuijl et al. 2019), the WTO’s overarching definition for subsidies may also gain more traction in this context. Nevertheless, given that fossil fuel subsidies’ definitional ambiguity may also be associated with certain advantages, it will be important to ensure that such a universal definition neither sets too high a threshold for action – creating a risk that some countries may disengage – nor too low a bar, which would see certain policies be needlessly excluded from FFSR discussions. Going forward, it will be important for advocates of reform to get this balance right.

8.5 Assessments of Effectiveness for the Carbon Pricing Subfield

8.5.1 Assessment of Outputs, Outcomes, and Impacts

Placing a (substantial) price on carbon emissions constitutes a key component in the response to climate change and is, according to some scholars, the most or even the only effective instrument for mitigating climate change (Nordhaus 2008; Rabe 2018). As discussed in Chapter 6, a range of public, private, and hybrid institutions promoting carbon pricing internationally have emerged over the last twenty-five years. Due to the partial overlaps in terms of governance functions and the interlinkages between these institutions, it is difficult to pinpoint the effectiveness of the individual institutions. Yet, it is possible to discuss their combined output, outcome, and impact effectiveness.

First, outputs have been substantial and increasing exponentially over time. Companies that join private institutions – such as the UN Global Compact, Caring for Climate, or the Carbon Neutral Protocol – thereby commit to adopt mitigation strategies, including internal carbon pricing or ensuring carbon neutrality inter alia via offsetting. Joining the Carbon Pricing Leadership Coalition (CPLC), for instance, entails a commitment to support carbon pricing, although not necessarily to adopt it. When it comes to aviation, the Carbon Offset and Reduction Scheme for International Aviation (CORSIA) will in the future constitute a mandatory instrument for pricing emissions placed on the individual airlines. Regarding more technical output, most of the institutions in the subfield have been active in producing reports and other information that raised awareness and increased knowledge of carbon pricing. They also established standards for offsets and for carbon-market units. The subfield is less effective, however, when it comes to operational activities such as research and development as well as financing. Here, only the Partnership for Market Readiness provides rather modest amounts for capacity-building.

Second, concerning outcome, there has been significant behavioural change in terms of carbon-pricing policies. By early 2018, sixty-six such policies had been adopted globally by public actors (states as well as supra- and sub-national polities such as the EU or American states), and an additional fifty-two carbon pricing policies were being actively considered (Skovgaard et al. 2018).³¹ The policies were put into place in all regions of the world, with a notable spike from 2015 and onwards, coinciding with the emergence of many of the institutions. While these figures do not include carbon pricing by companies on aviation, the 2016 adoption of CORSIA as a future mandatory pricing scheme in this sector fits the trend.

Yet, it is difficult to assess to what degree the output of the institutions has affected behaviour change among the actors targeted by that output (e.g. states, companies). The decision to adopt carbon pricing is inherently a polity- or business-level decision. With their standards and commitments, the governance functions of the institutions may incentivize the adoption of carbon pricing, but they may well lack the leverage to drastically alter polity- and business-level decisions (e.g. sanctions or shaming of norm violation). From the mid-1990s to the early 2010s, private institutions, particularly IETA, have been important in promoting carbon markets, particularly in industrialized countries (Meckling 2011; Paterson 2012; Paterson et al. 2014). More recently, the Partnership for Market Readiness, as well as other institutions embedded in the World Bank and the United Nations Framework Convention on Climate Change (UNFCCC) Secretariat, have provided expert assistance to several of the developing countries that had adopted or were considering carbon pricing within the last ten years. They also facilitated learning processes among such countries.³² It is thus plausible that this assistance has increased the likelihood of carbon-pricing proposals being adopted, but this is ultimately a subject for future research.

Third, the assessment of impact proves even more difficult. As an approximation, we briefly focus on the impact of the carbon-pricing policies that have already been implemented, while bracketing the question to which degree they have been driven by certain intergovernmental or transnational institutions. The impact effectiveness of the public carbon pricing policies can be measured in three ways. The first is to appraise the share of global emissions covered. The World Bank and Ecofys (2018) estimated that public carbon-pricing policies implemented by 2018 cover about 15 per cent of global emissions, rising to 20 per cent once the Chinese emission trading system (ETS) (scheduled for 2020) is operational. This is below the CPLC’s goal of 25 per cent of global emissions by 2020, but still a sizeable share and a drastic increase compared to about 8 per cent by 2012 and less than 1 per cent by 2004 (World Bank and Ecofys 2018).

A second indicator, and arguably a more important one than the share of global emissions covered, is the incentive for reducing emissions that carbon pricing actually provides. This incentive can be measured in terms of the price of emitting one metric tonne of CO_2-equivalents. Mainstream climate economists argue that getting the price right is the chief objective of carbon pricing (Tol 2011). The price is right if it corresponds to the costs to society of emitting one tonne of CO₂ equivalents, meaning that the externality has been fully internalized. To which degree emitters actually choose to cut emissions is less important in this thinking – a notion that might be at odds with the goal to keep temperature increases to 1.5 or 2 degrees (UNFCCC 2015). Yet, no matter whether the objective is internalizing the externality of climate change or limiting temperature increases, current carbon prices are too low. According to the World Bank and Ecofys (2018), 46 per cent of the emissions subject to carbon pricing are valued below 10 US $/tonne, and most of the emissions above 10 US $/tonne are priced between 10 and 20 US $/tonne. As a reality check, the report by the High-Level Commission on Carbon Pricing (2017) found that price levels should be between 40 and 80 US $/tonne in 2020 to meet at least the 2 degrees target. Likewise, a meta-study of various economic estimates about the social costs of carbon found that the actual average price should be 80 US $/tonne (Tol 2011).

Third, it is difficult to isolate the impact of carbon-pricing policies, since emissions are influenced by various additional factors – including fuel prices, technology development, economic growth, and other emissions-reducing policies (e.g. fossil fuel subsidy reform or policies supporting renewable energies). These difficulties notwithstanding, existing studies of individual carbon-pricing policies indicate that their impact is limited. Carbon taxes have had at least some success in reducing emissions – as they generally have higher price levels than carbon markets and mainly have been adopted by smaller, European countries (Rabe 2018, 9). The impact of carbon markets, however, is more modest. On the upside, the sectors covered by the EU ETS, by far the world’s largest carbon market, have experienced falling emissions since its inauguration in 2005. However, this drop in emissions is predominantly attributed to the economic crisis and renewable energy and energy-efficiency policies, with the ETS playing a smaller role (Hu et al. 2015; Bosello et al. 2016). Likewise, the Californian ETS, which is also among the largest carbon-pricing schemes in the world, has played a limited role in curbing emissions compared to other policy instruments such as renewable-energy policies (Bang et al. 2017).

8.5.2 Consequences of Institutional Complexity: What Are the Implications for Carbon Pricing?³³

The carbon-pricing subfield is characterized by a number of institutions (13) that fall somewhere between the figures for the renewable energy and fossil fuel subsidy reform subfields. The same way that it was much easier to assess the output than the outcome and especially the impact or environmental effectiveness of carbon pricing, it is also easier to do so for the consequences of institutional complexity. We therefore concentrate on output and outcome in this subsection.

Regarding potential positive effects, institutional complexity has improved output in terms of simply increasing knowledge-sharing in terms of technical reports, workshops, and other dissemination formats. Concerning outcome, our informants highlight that the sheer number of institutions promoting carbon pricing should have a positive impact to keep it on the political agenda. Furthermore, the diversity of institutions, especially in terms of their memberships, helps to reach a wide variety of actors (business, policy-makers, civil society) from different sectors, e.g. with CORSIA and IATA_COP specifically targeting aviation emissions.

As for possible negative effects, the output of the various institutions is not necessarily coherent. As discussed in Chapter 6, the institutions’ interpretation of the core norm of carbon pricing may diverge, with different objectives emphasized (e.g. whether the objective is functioning carbon markets or pricing emissions through carbon markets or taxes), and several informants therefore highlighted the risk of incongruent or even conflicting messages. At the same time, the interviewees also stressed that (especially cognitive) interlinkages between the institutions, as well as deliberate management efforts, help to reduce such incoherence. The risk of conflicting messages also pertains to outcome effectiveness. One example is the various institutions that promote particular standards for linking carbon markets, either under Article 6 of the Paris Agreement or for voluntary offsets. Having several standards may result in them undermining each other and incentivizing forum shopping by members. Incongruent messages also characterize the overall choice between carbon markets and taxes, since some institutions only promote carbon markets, whereas others promote both carbon markets and carbon taxes. Finally, different audiences – be they governmental, business, or civil society actors – may be less convinced if they are presented with differing arguments about the merits of carbon pricing, which may undermine the legitimacy of the instrument.

Altogether, it is difficult to gauge whether the positive consequences (larger volume of output, outreach to a diversity of audiences) outweigh the negative ones (incoherent messages and standards). This notwithstanding, and as we discuss in the following subsection, management efforts should target the negative consequences rather than reinforcing the positive ones. The positive consequences are mainly a direct consequence of the number and composition of the institutions and exist independently of management efforts.

8.5.3 Management Options

Given the current moderate levels of (mainly informal) management of the carbon-pricing subfield, it is perhaps not surprising that informants thought that more management efforts would increase effectiveness without being a game-changer. As an option, they particularly pointed to increased coordination between the existing institutions, rather than orchestration by a particular institution. Such coordination could mainly imply scaling up existing coordination efforts, while promoting a common narrative around a shared understanding of key tenets. Such tenets may include, for instance, that there is no one-size-fits-all approach but that the promotion of carbon pricing needs to be adjusted to local circumstances. With such a form of coordination, the aforementioned contradiction or incongruence of messages could be limited, as could incoherent standards for linking carbon markets or voluntary offsets.³⁴

Beyond incoherence, existing informal divisions of labour, e.g. between the Partnership for Market Readiness (PMR) and the UNFCCC Secretariat (see Chapter 6), could be supplemented by a more overarching division of labour – one that also covers areas where such coordinating arrangements do not currently exist. Such arrangements are relevant for determining which institutions should target which actors, not only among specific business sectors or groups of countries but also among different ministries and agencies within the same government.³⁵ A respective division of labour could also take into account actors that may be currently overlooked, since institutions may easily cluster around the same actors when seeking them out independently of each other.

Our interviewees were not in agreement regarding whether coordination should cover all institutions within the subfield or just some, e.g. the UNFCCC and the World Bank–embedded institutions (CPLC, PMR, and Networked Carbon Markets, see Chapter 6). Altogether, they indicated that more coordination could improve effectiveness, mainly in terms of outcome, i.e. behavioural change, since output was already relatively high and would not be significantly affected by coordination. This said, increased coordination would not necessarily improve outcome effectiveness radically, since incoherence is not a major impediment to such effectiveness and is already addressed through existing coordination to some degree.

8.6 Conclusions

This chapter has evaluated the extent to which, and the ways in which, three subfields of the climate-energy nexus – renewables, fossil fuel subsidy reform, and carbon pricing – can be viewed as effective. The study has advanced a nuanced picture of how institutional complexity affects effectiveness, as well as presenting a set of management options for each subfield to enhance effectiveness. In what follows we compare some of the similarities and differences among the results from each subfield and discuss what our findings mean for the performance of the climate-energy nexus as a whole.

The chapter yields three broad insights. First, all three subfields appear to be successful in producing outputs, especially in terms of information-sharing and capacity-building. While such outputs may seem trivial at first sight, it is important to note that institutional arrangements in other fields of global environmental governance have failed to produce significant results even at this first level of effectiveness – such as the more than 300 multi-stakeholder partnerships for sustainable development launched at the World Summit for Sustainable Development in Johannesburg 2002 (Pattberg et al. 2012).

At the outcome level, performance is more difficult to assess due to practical reasons of data-gathering and establishing causal relationships and, where this is possible, the success levels seem somewhat weaker. Results show that the institutions across the three subfields have likely influenced behavioural changes amongst actors, for example, through promoting learning processes and policy reform. However, these outcomes have not been as far-reaching as to successfully alleviate the collective action problems that are underlying the respective subfields.

On the whole, therefore, we find little concrete evidence of the institutional complexes across the subfields significantly contributing to problem solving in terms of scaling up renewables, stringent carbon pricing, and alleviation of fossil fuel subsidies. Having said that, it is unclear whether lack of outcome and impact level effectiveness is simply due to data-gathering and analytical challenges.

Second, institutional complexity can both help and hinder effectiveness. The existence of multiple institutions within a field may result in duplication of efforts and counterproductive competition, but can also create positive feedback loops and productive competition. As seen in the example of carbon pricing, even the mere fact of having many institutions working on one issue promotes political attention. In the area of fossil fuel subsidy reform, the engagement of multiple institutions has created new opportunities for FFSR advocates to keep the issue on the international agenda by framing the need for reform in different ways. A similar effect is noted in the subfield of renewable energy, which has witnessed a proliferation of actors, in particular nonstate and sub-national ones, that participate in various institutions and voluntarily create and abide to new rules and norms. To summarize these observations, the sheer magnitude of new actors and institutions that constitute the climate-energy nexus enhances political attention.

Moreover, institutional complexity may facilitate experimentation and learning at multiple venues, jurisdictions, and scales and allows for targeting actors that significantly differ in their preferences and opportunities and constraints they face, e.g. in countries with varying levels of economic and social development. This flexibility, on the other hand, risks leading to conflictive norms, forum-shopping, and diluted ambition levels. Altogether, this ambivalence is in line with previous literature on institutional complexity and fragmentation that highlights both advantages and disadvantages of such complexity (Keohane and Victor 2011; Zelli and van Asselt 2015).

Third, to the degree that international institutions can influence actors, a coordinated approach is arguably the most important factor in improving effectiveness. This should particularly imply the promotion of a common narrative and establishment of a division of labour between institutions, rather than adding further orchestration efforts for example. This said, such coordination attempts are currently largely lacking across all three subfields – which points to a major challenge: institutional complexity often arises due to a divergence of priorities amongst powerful actors (Keohane and Victor 2011), and this very conflict of interests may also hamper any coordination efforts.

In sum, institutions working within the climate-energy nexus face difficult challenges to address. On the one hand, these institutions are strengthened by having other institutions to collaborate with to reinforce their work. On the other hand, competition over resources, as well as duplications, contradictions, and incongruence of the work of different institutions, undermine some of these benefits.

The renewable energy subfield illustrates this ambiguity. On the one hand, internationally set goals provide a joint vision across institutions. On the other hand, the coordination between the multitude of institutions is far from sufficient. Further research is thus needed to look into how such coordination could be achieved to improve the effectiveness of the climate-energy nexus. Such research should look beyond the meso level and also examine interactions across subfields.

What this analysis has shown is that institutions within the three selected subfields have laid much of the groundwork for effectively contributing to their respective subfields. What is required now is the closing of governance gaps, crucially in finance and implementation, and greater cooperation between institutions to overcome some of the downsides of institutional complexity.

9 Conclusions Coherence, Management, Legitimacy, and Effectiveness in the Climate-Energy Nexus

Fariborz Zelli , Karin Bäckstrand , Naghmeh Nasiritousi , Jakob Skovgaard , and Oscar Widerberg

Combating climate change and transitioning to fossil-free energy systems are two central planetary challenges humanity faces today. The two challenges strongly overlap in their substance and the political choices we have to address them. As a consequence, a plethora of international institutions – public, private, and hybrid ones – fall right into this overlap. They seek to regulate the many complex linkages between climate change and decarbonized energy systems.

The introduction to this volume stressed the urgency of researching these links at the nexus of global climate change governance and energy governance. The highly complex and fragmented nature of the climate-energy governance nexus confronts us with a series of questions that need timely responses. How coherent is this governance nexus – in terms of the memberships, functions, core norms, and interactions across its numerous institutions? How have states, international organizations, and other actors responded to this complexity when they sought to tackle coherence and governance gaps? How does institutional complexity affect the legitimacy and effectiveness of individual institutions and the governance of the climate-energy nexus as a whole? Or, to concentrate these concerns into one question: to which degree do the many governance efforts on climate change and energy transition today add up to a coherent regulatory and institutional global framework?

It has been the purpose of this book to provide first and crucial answers to these questions and to alert a wider audience to their importance. In Section 9.1, the concluding chapter summarizes some of the novel findings from our research. The summary is by no means exhaustive, and we refer the readers back to the previous chapters for the rich, cross-dimensional, and comparative insights they provide. Our research findings show us in great detail what is at stake and how carefully institutions and stakeholders need to navigate in the complexity of the climate-energy nexus with its many and emergent governance arrangements. Against this backdrop, this chapter also discusses possible explanations of our results (Section 9.2) and derives policy recommendations from them (Section 9.3). It also points to some of the next steps we need to take as researchers and stakeholders in order to live up to the tall order of tackling climate change and energy transition simultaneously (Section 9.4).

9.1 Findings

This section is organized along the main components of the analytical framework introduced in Chapter 2. The major backbone of the framework are four evaluative themes or analytical dimensions to grasp the shape and performance of institutions in complex governance systems: coherence, management, legitimacy, and effectiveness. Each of these evaluative themes was operationalized through particular dimensions.

To examine the four themes, the framework combined both rationalist and constructivist accounts derived from theories of International Relations. Coherence and management were to a large extent scrutinized along organizational dimensions such as institutional membership and governance functions. The dimension of the core norm, however, added an important constructivist dimension to the analysis of coherence. For the two themes that concern performances under conditions of complexity, i.e. legitimacy and effectiveness, the authors mostly relied on a constructivist or sociological perspective. They examined perceptions and expectations by selected audiences and stakeholders involved in international negotiations on energy and climate.

In addition to this dual approach, the framework distinguished three analytical levels to get a more encompassing grasp on the climate-energy nexus: the macro level (the climate-energy nexus as a whole); the meso level (the institutional complexes that govern each of the nexus’s subfields analyzed in this volume, namely renewable energy, fossil fuel subsidy reform [FFSR], and carbon pricing); and the micro level (specific interlinkages between individual institutions within each of these subfields).

9.1.1 Coherence

Coherence was defined in Chapter 2 as the degree of harmony or compatibility of institutional features to one another or to an overarching purpose. This theme was studied along four dimensions: institutional membership, coverage of governance functions, adherence to an overarching core norm, and interaction mechanisms. Two of these dimensions (membership and functions) were analyzed across all three analytical levels. The other two were only scrutinized for the meso level (core norm) and micro level (interaction mechanisms), respectively.

In Chapter 3, Sanderink and colleagues advanced a novel and unique dataset. They mapped the institutions across the entire climate-energy governance nexus at the macro level and examined in great detail the membership distribution of public, civil society, and firm-based members. The database they generated contains 108 institutions, which include more than 12,000 members as of early 2017. Through their network analysis, the authors identified the International Renewable Energy Agency (IRENA) and the United Nations Framework Convention on Climate Change (UNFCCC) as the central institutions within the climate-energy nexus, due to their high degrees of universality and inclusiveness.

By the same token, Sanderink and colleagues found for the climate-energy nexus a clear dominance of public members, i.e. country, state, or municipal governments, secretariats to international agreements, and other public agencies. These are involved in no less than seventy-eight institutions and exclusively constitute forty-eight of these, with cities being by far the most frequent type of public member. While the predominant number of public members come from developed countries, public actors from various countries in the global South, such as China, Mexico, and Indonesia, have become widely represented across the nexus. By contrast, firm or business actors are members of less than half of the sample (fifty-one institutions), and only make up seventeen institutions exclusively. Civil society actors are officially engaged in solely thirty-five institutions, seven thereof being purely civil-society-based.

When disentangling these overall observations for the three selected policy subfields, the contributions to this volume equally identified a predominance of public actors, albeit with certain variations across issue areas. As Sanderink demonstrated in Chapter 4, the renewable energy subfield of the nexus exhibits the most institutions of the three cases, forty-six in total. The membership distribution across these institutions mirrors that of the climate-governance nexus as a whole. Twenty-eight of the forty-six renewable energy institutions are solely constituted by public members. Apart from the UNFCCC and IRENA, these include, for instance, the International Energy Agency (IEA), the Association of Southeast Asian Nations, and the Covenant of Mayors. No more than four institutions are merely firm-based and only three are purely civil-society-based, with the remaining eleven institutions featuring mixed memberships.

For fossil fuel subsidy reform, Verkuijl and van Asselt in Chapter 5 identified an even greater imbalance in favour of public actors. The institutional complex is made up of only fourteen institutions, thirteen of them purely intergovernmental, e.g. the UNFCCC, IEA, the United Nations Environment Programme (UNEP), the Friends of Fossil Fuel Subsidy Reform (Friends), and the World Bank. The only exception is the Global Subsidies Initiative, which was started by a Canadian NGO.

By contrast, Skovgaard and Canavan found a nearly balanced distribution for the subfield of carbon pricing (Chapter 6). Of the thirteen institutions they identified for this complex, five were solely public (e.g. the UNFCCC or the International Carbon Action Partnership [ICAP]), four firm-based (e.g. the International Emissions Trading Association [IETA] and the Verified Carbon Standard), and one, the Gold Standard, constituted by civil society actors, with the three remaining ones being hybrid institutions.

For the second dimension of coherence, governance functions, Chapters 2 and 3 differentiated four major dimensions to guide the empirical analyses in this volume: standards and commitments, operational activities, information and networking, and financing. At the macro level of the climate-energy nexus as a whole, Sanderink and colleagues (Chapter 3) found that all these four governance functions were institutionally covered. This notwithstanding, there was no clear division of labour evolving across the nexus as a whole. Instead, they identified a rather uneven distribution of functions. The climate-energy nexus is dominated by institutions that perform information and networking functions, while, for instance, standard-setting and financing are conducted by a much smaller set of institutions. Institutions with public members are clearly in the majority when it comes to information and networking functions. Interestingly, though, the number of public and firm-based institutions is nearly equal for the setting of standards and commitments. Put differently, a key observation is that soft-governance functions – such as information and networking – dominate at the expense of hard-governance functions that involve authoritative rule and standard-setting. Clearly, states are not willing to cede sovereign control of rule-setting on energy to global institutions.

When zooming in on the meso level and comparing the three selected cases, this picture changes and gets more differentiated. The subfield of renewable energy again comes closest to the research findings from the climate-energy nexus as a whole. In Chapter 4, no clear cross-institutional division of labour of governance functions can be found. Instead, global renewable energy governance is geared toward information and networking functions that are dominated by public institutions.

For the subfield of fossil fuel subsidy reform, by contrast, the distribution of institutions across governance functions is much more balanced. As Verkuijl and van Asselt observe in Chapter 5, this is partly the result of active coordination by international organizations and country governments. Some forums are instrumental in agenda-setting and the formulation of commitments (e.g. the Group of Twenty [G20] and Friends), while others concentrate on the provision of information (e.g. the Organisation for Economic Co-operation and Development [OECD] and the IEA) or on financing and implementation on the ground (e.g. the World Bank and the Global Subsidies Initiative).

For the third case study on carbon pricing, Skovgaard and Canavan established yet a different picture, namely one of considerable overlaps over certain governance functions. For instance, ICAP (intergovernmental), IETA (business), and the Networked Carbon Markets Initiative (NCM; hybrid) engage in very similar information and networking activities to promote the linking of carbon markets. However, in all subfields rule- and standard-setting is a less predominant governance function, which reflects that sovereign states have a solid grip on decision-making power in energy governance.

As a third dimension to analyze the degree of institutional coherence in the nexus, Chapter 2 identified the core norm, i.e. the overarching expectation of appropriate behaviour that characterizes a particular subfield. The concept is closely related to the very definition and delineation of a subfield and its respective institutional complex. It provides the substance or goal that unites the institutions governing the area in question. Although the institutions agree on the importance of the norm, they may interpret its precise content and application in strongly diverging ways (see also Wiener 2004). The interpretation of the core norm therefore served as the major benchmark for identifying normative convergence or divergence at the meso level.

For renewable energy, Sanderink (Chapter 4) identified Sustainable Development Goal 7.2 as the core norm, which aspires to substantially increase the share of renewable energy in the global energy mix, especially with a view to ensuring access to, and availability of, clean energy for all (United Nations 2015). The universal nature of this norm notwithstanding, Sanderink ascertained a considerable normative divergence and contestation across the subfield. Only seventeen of forty-six institutions subscribe to the norm literally, while the others prioritize specific aspects over others, i.e. either climate change mitigation, energy access, or energy security.

For fossil fuel subsidy reform, Verkuijl and van Asselt (Chapter 5) found a similar, if not even more divergent picture. They also referred to a key institutional document as the most suitable expression of the core norm, in this case taken from the statement of the third leaders’ summit of the G20 in Pittsburgh 2009: to ‘[r]ationalize and phase out over the medium term inefficient fossil fuel subsidies that encourage wasteful consumption’ (G20 2009, paragraph 29). Verkuijl and van Asselt caution that the key terms in this formula remain contested across actors and institutions, including what should be counted as fossil fuel subsidies in the first place. Moreover, phase-out dates remain ambiguous across the institutional complex. In addition, the United States under the Trump administration has begun to voice overall reservations on the FFSR norm in general. Against this backdrop, it is not surprising that the authors found no major institution to which all others would normatively adhere to.

Normative divergence and contestation, rather than convergence, also mark the institutional complex on carbon pricing (Chapter 6). As a core norm for this subfield, Skovgaard and Canavan established that climate change is best mitigated by giving a price signal to emitters and by leaving the decision of how to reduce carbon emissions to the market. This said, they identify no less than four aspects along which the interpretation of this norm considerably varies across the complex. These four aspects are: quantity versus price instruments, mandatory versus voluntary schemes, carbon pricing within a given jurisdiction versus off-setting, and whether polluters should pay for all of their emissions or not. Variation in institutional positions concerning these four aspects creates normative clusters of largely private institutions on one side and hybrid or public ones on the other.

The fourth and final dimension of coherence introduced in Chapter 2 regards the interaction mechanisms at the micro level of individual institutions. Three types of mechanisms were distinguished through which individual institutions may have an impact on each other within the climate-energy nexus: cognitively, through a flow of knowledge and information; normatively, through an imitation of, or adaptation toward, norms and rules; or behavioural, through the functional or strategic behaviour of specific members or other actors.

Chapters 4–6 found plenty of evidence for all three mechanisms being at play in the three institutional complexes, with no particular dominance of one over the others. Yet, given the aforementioned overall dominance of information and networking functions in the climate-energy nexus, it was arguably easier to identify cognitive interactions than normative interactions. Behavioural interactions depend in part on the overlaps of memberships across institutions. To give one example: in the institutional complex on FFSR (Chapter 6), their membership in both the G20 and the Asia Pacific Economic Cooperation (APEC) allowed the United States and China to push successfully for reform commitments and peer-review efforts in both intergovernmental institutions. Such reform efforts included events and initiatives linked to the UNFCCC, the High-Level Political Forum on Sustainable Development, and the World Trade Organization (WTO). This form of forum shopping (cf. Raustiala and Victor 2004; Orsini et al. 2013) has played an important role in increasing the saliency of FFSR on the international policy agenda.

9.1.2 Management

The second evaluative theme, management, was specifically targeting the micro level, i.e. relations among individual institutions within a given subfield of the climate-energy nexus. Quoting Stokke (2001, 11), Chapter 2 defined management as any deliberate efforts taken ‘by participants in tributary or recipient regimes to prevent, encourage, or shape the way one regime affects problem solving under another’. The chapter distinguished two main dimensions for assessing such efforts. First, the levels (regional, national, global) and agents (domestic or sub-domestic actors, one or several affected institutions, or an overarching institutions) of management; and second, the actual consequences and potential successes of management efforts. In other words: did the management attempts yield any convergence about the core norm or entail any sensible and efficient distribution of memberships and governance functions across the affected institutions?

Notably, all three case studies identified a variety of effective management attempts for the particular institutional interlinkages they put under scrutiny. This signals a strong awareness by institutional members and third parties of the complexity in which they are operating. In Chapter 4 on renewable energy, Sanderink concentrated her micro-level analysis on three hybrid institutions, i.e. partnerships made up by all three major stakeholder types (public, firm-based, civil society). For all three of them – the Renewable Energy and Energy Efficiency Partnership, The Renewable Energy Policy Network for the 21st Century (REN21), and Sustainable Energy for All – she found management attempts initiated by different drivers at various levels. For example, joint coordination efforts between the three institutions included a shared portal and a joint report for common information. Further convergence was reached through unilateral management approaches, with each partnership reaching out toward important third institutions, e.g. IRENA. Finally, Sanderink identified the UNFCCC and Agenda 2030 as overarching institutions that provide core global goals upon which the three partnerships converge. These diverse management activities notwithstanding, she stressed an important limitation of her micro-level study. In light of a total of forty-six institutions and a dominance of public institutions in the renewable energy subfield, the relations between the three hybrid institutions selected should not be mistaken as representative of the management attempts in this institutional complex. In fact, Sanderink found that, aside from the three partnerships that she examined, only thirteen other institutions explicitly link their activities to the UNFCCC and Sustainable Development Goal (SDG) 7.

The results of the micro-level studies were slightly more representative for the two other case studies, FFSR and carbon pricing, since these two subfields exhibit a much lower number of institutions. For the institutional complex on FFSR (Chapter 5), consisting of but fourteen institutions, Verkuijl and van Asselt studied three key intergovernmental institutions in depth: G20, APEC, and Friends. They largely identified informal management efforts between these three and their members, taking the form of meetings on the side during international events, outreach to Friends members for their expertise, and outreach from Friends members to other countries in advance of G20 summits. Despite this informal character, or likely because of it, the management attempts were very effective, since they facilitated the information flow and kept the level of ambition high across the three institutions. Friends members in particular adopted a broker role and proactively enhanced complementarity among the three institutions. For instance, developments at the G20 gave New Zealand, a Friends member, leverage to ensure that similar efforts were undertaken under APEC, in terms of both reform commitments and the introduction of a peer review process in this area.

For their micro-level examination on carbon pricing in Chapter 6, Skovgaard and Canavan arguably chose the most powerful institutions among the thirteen that constitute the meso-level institutional complex. When analyzing interactions between the UNFCCC and World Bank–embedded institutions, they determined various joint management efforts. Such efforts mainly took place through regular, yet often informal, contacts and meetings between institutional officials that are specialized in carbon pricing questions. These experts helped to defuse situations that could potentially have led to open competition or conflict. On the other hand, and unlike for renewable energy and FFSR, Skovgaard and Canavan did not find any noteworthy unilateral or overarching attempts by particular states or institutions to manage carbon pricing. Moreover, most management attempts for this policy field have originated in an ad hoc or incremental manner rather than going back to deliberate and long-sighted planning. On balance, the institutional sample for carbon pricing thus features less encompassing management efforts than the samples in the other two case studies.

Table 9.1 summarizes the major findings for the two evaluative themes of coherence, and management across the three institutional complexes for renewable energy, FFSR, and carbon pricing. The left column lists the various dimensions, while the three columns on the case studies clarify to which extent these dimensions were addressed in a coherent manner, i.e. whether or not the three subfields exhibit high and balanced scope of memberships, functions, interaction mechanisms, a convergence on core norms, and a high and effective scope of management efforts. The bottom row includes our overall assessments of the shape and direction of the institutional complexes for each subfield. The terminology for this overall assessment (here: coordination for all three cases) is guided by the book’s analytical framework (see Chapter 2, Table 2.1) that distinguishes largely non-managed from largely managed institutional constellations under different degrees of coherence.

Table 9.1 Findings for coherence and management in the climate-energy nexus.

Evaluative Themes	Nexus Subfields Renewable Energy	Fossil Fuel Subsidy Reform	Carbon Pricing
Coherence
Scope and distribution of memberships (meso)	46 institutions Imbalance, dominance of public institutions (28)	14 institutions Imbalance, strong dominance of public institutions (13)	13 institutions Balanced distribution (5 public, 4 firm-based, 1 civil society-based)
Scope and distribution of governance functions (meso)	High Imbalanced distribution in favour of information & networking	High Balanced and complementary distribution	High Balanced, but competitive distribution
Convergence on core norm (meso)	Low to medium Strong variations across goal priorities	Low to medium Strong variations on the core definition of fossil fuel subsidies	Medium Strong variations across goal priorities and approaches to reach these goals
Scope and distribution of interaction mechanisms (micro)	High All types (cognitive, normative, behavioural)	High All types (cognitive, normative, behavioural)	High All types (cognitive, normative, behavioural)
Management
Scope of efforts across agents and levels (micro)	High Both formal and informal Largely joint and unilateral, partly overarching	High Mostly informal Joint, unilateral and overarching	Medium Both formal and informal, but largely ad hoc Largely joint, no major unilateral or overarching efforts
Consequences (micro)	Facilitating convergence	Facilitating convergence	Preventing competition
Overall Assessment	Coordination	Coordination, evolving division of labour	Coordination

Our findings indicate that all three subfields are characterized by coordination, rather than competition or outright harmony. This reflects the aforementioned results for coherence and management. On the one hand, all three policy subfields exhibit a medium degree of normative and functional convergence. The institutions in each of them share a certain core norm, for example phasing out inefficient fossil fuel subsidies, but differ considerably in their interpretations of that core norm. While all institutions in each subfield cover all governance functions, they are frequently skewed toward information and networking at the expense of standard-setting, and public actors are over-represented compared to private stakeholders. On the other hand, all three cases exhibit diverse and relatively successful management attempts across key institutions, with a certain caveat regarding renewable energy where more studies are needed to scope out this relatively large institutional complex.

The combination of these two qualities – medium coherence and advanced management – suggests that the meso-level institutional complexes on renewable energy, FFSR, and carbon pricing fulfil notions of ‘coordination’. The differences across the three subfields are, hence, rather minor. FFSR comes closest to a division of labour, due to the widespread management efforts amongst a small number of institutions and the balanced distribution of governance functions across these institutions. The institutional complex on carbon pricing, on the other hand, depends upon ad-hoc management efforts to prevent outright competition or conflict across institutions.

9.1.3 Legitimacy

Moving from the shape or degree of institutional complexity to its consequences, the remaining two evaluative themes of legitimacy and effectiveness have been conceptualized in sociological terms, i.e. they are measured by perceptions among key audiences and stakeholders. To this end, the analyses relied on interviews and questionnaires to explore how certain actors perceive the legitimacy and effectiveness of selected institutions in the three policy areas that we put under scrutiny in this volume.

To examine consequences of institutional complexity for legitimacy, Chapters 2 and 7 introduced a total of nine dimensions. The dimensions capture criteria from the diverse scholarship on legitimacy concepts (normative and sociological, input and output legitimacy), that were geared to be analyzed in terms of perceptions among key stakeholders. They include five dimensions of input legitimacy (an institution’s expertise, inclusiveness, procedural fairness, transparency, and accountability) and four dimensions of output legitimacy (produced output, behaviour-changing outcome, problem-solving impact, and distributive fairness).

In Chapter 7, Nasiritousi and Verhaegen advanced this novel framework, guided by theoretical assumptions and expectations on the legitimacy perceptions of these dimensions (see also Section 9.2). The framework was employed to empirically examine five selected institutions in the subfield of renewable energy: the UNFCCC, IEA, IRENA, the Clean Energy Ministerial (CEM), and REN21. To get reliable results from a broad set of stakeholders, an expert survey was conducted with 262 respondents, and analysed by performing an exploratory factor analysis for the 9 dimensions. The findings are presented and visualized in detail in Chapter 7, differentiated by different stakeholders and institutions.

We can only highlight some of the general findings on legitimacy perceptions among key stakeholders for the renewable energy subfield. One core finding is that the UNFCCC was regarded as the most legitimate institution in the sample, while the Clean Energy Ministerial (CEM) received the least positive legitimacy assessment across the nine dimensions, with a noteworthy margin to the other four institutions. Moreover, the UNFCCC scored higher compared to any other institution with regard to dimensions of input legitimacy. An explanation can be that the UNFCCC is the only multilateral institution with a political mandate in the institutional complex on renewable energy, and thereby has the most elaborated procedures to promote access, inclusiveness, and transparency in particular. Apart from these observations, however, it is striking how similar the overall legitimacy assessments by key stakeholders were for the five institutions. Importantly, this suggests that stakeholders do not adequately disentangle their views of individual institutions from their overall impression of the renewable energy subfield – at least not for those institutions that have similar functions, feature overlapping mandates, and regularly engage in interactions and management efforts.

When distinguishing the results according to different subsets of stakeholders, though, Nasiritousi and Verhaegen found that an institution’s formal fulfilment of normative legitimacy criteria does not necessarily translate into positive legitimacy assessments by all actors. Rather, the varying specializations and backgrounds of stakeholders led to significant variations in legitimacy assessments – owing to differences in their normative orientations, value sets, and familiarity with certain institutions. In this context, the professional background was much more significant for legitimacy assessments than the geographical origins of respondents, which on balance did not render any significant differences.

9.1.4 Effectiveness

Three dimensions of effectiveness were differentiated in Chapter 2, and applied in Chapter 8: the institutions’ production of information, norms, and policies (output), the effect on institutional members and their behaviour (outcome), and the ultimate effect on solving the climate- or energy-related problem that the respective institutions sought to address (impact) (cf. Underdal 2002; Tallberg et al. 2016).

These were identical with the three-tiered legitimacy dimensions discussed in the previous section on legitimacy. What set the study in Chapter 8 apart from that in Chapter 7 was, first, that a broader sample was analyzed, namely across all three institutional complexes – renewable energy, fossil fuel subsidy reform, and carbon pricing. Second, the empirical examination of effectiveness combined both document analyses and qualitative data on the perceptions and expectations of experts (derived from interviews).

A core finding was that all three institutional complexes are relatively successful in producing outputs, especially in terms of information-sharing and capacity-building. Achievements on the outcome and impact levels were harder to establish, but the overall evidence suggested that none of the institutional complexes has been capable of substantially mitigating the collective action problems they set out to address. Not surprisingly, the interviewees called for more coordinated approaches for all three issues to build common narratives and promote a certain division of labour across institutions.

Disentangling these general insights for each of the three subfields showed that institutional complexity can be a supportive as well as hindering factor for effectiveness in the climate-energy nexus. For renewable energy, the interviewees expected more advantages than disadvantages from the institutional complexity in this subfield. This goes in particular for the output level, with benefits arising from multiple perspectives and knowledge bases and thanks to a pragmatic compartmentalizing approach to master different challenges of the energy transition. The most frequently named disadvantages, on the other hand, included: duplication of work; competition over resources, influence, and visibility; and the lack of an umbrella institution to address these issues.

A similarly positive assessment of effectiveness by core audiences was established for the institutional complex on fossil fuel subsidy reform. Similar to renewable energy, respondents viewed the availability of different types of expertise and experiences across the various institutions as highly beneficial, especially for output effectiveness. With regard to outcome, the different memberships of institutions working on FFSR helped to widen the geographical spread of governance efforts on this issue. In addition, interviewees welcomed a race to the top or positive reinforcement arising from institutional plurality, i.e. the incentive to keep or surpass the level of ambition of other institutions in the field. The disadvantages that were most frequently mentioned coincided with those for renewable energy, namely duplication of work and competition among standards at the output level, implying considerable transaction costs and ambiguity regarding outcome effectiveness.

These arguments about the consequences of institutional complexity were more or less repeated for carbon pricing. On the positive side, respondents stressed the knowledge distribution for output effectiveness, as well as positive reinforcement effects and stronger outreach activities for outcome effectiveness. Disadvantages that the interviewees referred to were equally similar to the other two subfields: the competition among standards and the conflicting messages this sends to stakeholders. Importantly, and in comparison to renewable energy and FFSR, these negative implications for carbon pricing were highlighted quite frequently by respondents. This confirmed the aforementioned findings on coherence and management for this subfield, which suggested a somewhat more competitive nature of the institutional complex. The findings for effectiveness echoed this notion and suggest that the degree and consequences of this competitiveness merit further exploration.

9.2 Explanations

As laid out in Chapter 1, this edited volume first and foremost has sought to provide new conceptual and empirical insights into the shapes and consequences of institutional complexity for the nexus of two major policy domains, climate change and energy. Given the lack of previous comparative studies about the climate-energy nexus, this book combined a series of innovative methods and research steps, which, in turn, generated a large set of important and novel research findings for scholars and practitioners alike. In the following, we briefly focus on possible explanations for these findings on the four evaluative themes of coherence, management, legitimacy, and effectiveness.

9.2.1 Explaining Coherence and Management in the Climate-Energy Nexus

Chapter 1 briefly proposed one explanation for variation across the subfields on coherence and management, namely the position of each subfield within the climate-energy nexus. The variation in this position also served as one of our main criteria for selecting the three case studies for this volume. Carbon pricing is mainly connected to climate change, with mitigation as its core objective. Renewable energy, on the other hand, is more at the heart of energy governance, as a policy subfield primarily seeking to promote a segment of energy sources, while mitigating climate change comes as a co-benefit or secondary goal. Fossil fuel subsidy reform, finally, falls somewhere in the middle of the two other subfields, as it is both an instrument for climate mitigation and for de-carbonization of the energy mix.

In Chapter 1, we expected that these different positions of our three case studies within the climate-energy nexus would matter for the coherence of the respective institutional complexes – especially since climate change governance was, unlike energy governance, marked by a central institution, the UNFCCC, as a hub of multilateral climate diplomacy (Biermann et al. 2009; Van de Graaf and Colgan 2016). This central position, and possibly coordinative influence of the UNFCCC, would make it well-equipped to play a stronger role for predominantly climate-related topics such as carbon pricing.

However, when looking at the findings for coherence and management, as displayed in Table 9.1, these expectations were not supported by the empirical evidence. Instead, in spite of their very different positioning in the climate-energy nexus, all three institutional complexes share a number of institutional features, such as wide coverage of different governance functions, dominance of public institutions (albeit to different degrees), and low or medium convergence on core norms, as well as multiple and effective management attempts. Consequently, they score similarly in the overall assessment of ‘coordination’, i.e. they exhibit a relatively well managed, medium level of coherence.

What is more, the convening, steering or orchestrating role that the UNFCCC plays in global climate governance in general did not have the expected impact for the subfield of carbon pricing. On the contrary, Skovgaard and Canavan (Chapter 6) found no evidence for any management attempts by the UNFCCC for this subfield. Rather, as Sanderink showed in Chapter 4, the UNFCCC provides such an umbrella function for some of the institutions governing renewable energy.

At second glance, however, the positioning in the climate-energy nexus appears to play some role for shaping the institutional complexes, albeit not the only or major role. For instance, the normative guidance that the UNFCCC provides for the renewable energy subfield affects a minority of institutions in the complex. Sanderink’s findings point to roughly one-third of institutions that adhere explicitly to core norms of the UN climate regime. By contrast, output from the UNFCCC, particularly the Kyoto Protocol and the Paris Agreement, have historically often acted as stimuli for scaling up efforts across the carbon pricing subfield.

This said, the potential relevance of the nexus position for the degree of institutional complexity may be considerably qualified by other factors. In the case of carbon pricing, for instance, one reason for the slightly more competitive nature of the complex (e.g. with regard to the distribution of governance functions) is the constellation between the UNFCCC on the one hand and World Bank–embedded institutions on the other. These two sets of international institutions differ, for instance, in their confidence in the ability of the market to yield carbon-emission reductions (Chapter 6). While this constellation confirms the important role that the UNFCCC has for this subfield, it shows that, for a more differentiated explanation, other phenomena have to be taken into account.

Concretely, the climate-energy nexus does not exist in a vacuum, but overlaps with other policy domains with their own institutional settings. Carbon pricing is not only a matter of climate change and energy, but also of financing and international development, which explains the significant role the World Bank plays in balancing the UNFCCC’s dominant position in this subfield. Likewise, renewable energy is linked to a plethora of economic, environmental, and social concerns that are in part regulated by other institutions. It is hence not surprising that renewable energy is governed by the biggest institutional complex in our sample. This high number of institutions, in turn, implies both challenges and opportunities that may impact the institutional features of the complex, from more options to cover governance functions on the one hand to more difficulties to reach overarching coordination on the other (Scott 2008; Biermann et al. 2009).

These considerations about thematic scope and the number of institutions demonstrate the need for more theory development in order to explain and understand the variation of institutional features in a governance nexus. This development can build, inter alia, on different theories on international institutions who continuously incorporated the phenomenon of institutional complexity over the past two decades. Rationalist, sociological and discursive accounts of institutionalism yield different theoretical expectations that could be adapted to the study of a governance nexus such as the one on climate and energy. Among the rationalist approaches, proponents of instrumental multilateralism could, for instance, analyze the role of hegemonic countries in determining which core rules are adopted, which governance functions are covered and which management attempts are taken by which actors or institutions (cf. Ikenberry 2003; Morse and Keohane 2014). Likewise, neoliberal institutionalists and organizational ecologists could scrutinize the role of underlying constellations of interests or situation structures (cf. Zürn 1993; Keohane and Victor 2011), problem structures (cf. Rittberger and Zürn 1990; Underdal 2002; Zelli et al. 2017), or resource dependencies (cf. Abbott et al. 2016) in shaping certain areas of a governance nexus. Sociological and discursive institutionalists, finally, could help us understand to what extent the (lack of) institutional coherence in a nexus represents overarching norms or discourses – and the rivalries and contestations among them – in which the various institutions are embedded (cf. Conca 2006; Schmidt 2008, 2017; Arts and Buizer 2009).

9.2.2 Explaining Legitimacy and Effectiveness in the Climate-Energy Nexus

In Chapter 7, Nasiritousi and Verhaegen examined a variety of potential reasons for the legitimacy assessments by stakeholders. They followed Steven Bernstein and other scholars who have shown that different types of stakeholders may hold different legitimacy demands based on their social values, norms, and previous experiences (Bernstein 2005; Karlsson-Vinkhuyzen and Vihma 2009; Lenz and Viola 2017). Nasiritousi and Verhaegen therefore examined whether varying stakeholder characteristics made a difference for legitimacy assessments under conditions of institutional complexity. For their sample of five renewable energy institutions, they found that some of these characteristics indeed mattered for legitimacy assessments. These assessments particularly varied between governmental and nonstate actors and between stakeholders with different work profiles and orientations. The geographic origin, on the other hand, played no major role for how the legitimacy of an institution was valued. These results indicate that norms, values, and experiences of audiences may be of relevance for the sociological legitimacy of institutions in a nexus.

In addition, Nasiritousi and Verhaegen took into account the degree of institutional complexity as a potential explanation for shaping audiences’ legitimacy beliefs. Following cognitivist assumptions about limitations of visibility and shadows of legitimacy in highly complex governance systems (cf. Alter and Meunier 2009; Bäckstrand et al. 2018; Zelli 2018), they examined whether such systems may make it too difficult for stakeholders to differentiate between the processes and performances of individual institutions. The climate and energy nexus, which brings together a large variety of institutions, provides such a complex environment.

The authors indeed found evidence that stakeholders in their sample did not adequately disentangle their legitimacy assessments of the individual institutions under scrutiny, since these overlapped in mandates and governance functions. Moreover, they found that a stakeholder’s level of familiarity with an institution can be linked to a more positive assessment of legitimacy. Hence, there are reasons to believe that knowledge or valuing of an institution provides important markers or mental shortcuts for stakeholders when they navigate a very complex and densely populated subfield such as renewable energy.

These findings on legitimacy regard a small sample from only one of the three subfields of the climate-energy nexus. Future research has to show whether the institutional complexes on fossil fuel subsidy reform and carbon pricing exhibit similar levels of sociological legitimacy. On the one hand, one could expect such similarities, since, as we summarized earlier in this chapter, all fields exhibit comparable degrees of coherence and management. On the other hand, the governance systems on FFSR and carbon pricing are made up of considerably less institutions, which may qualify the impact of institutional complexity on stakeholder assessments.

Chapter 8 analyzed whether the institutional complexity of the three subfields affected various experts and stakeholders in how they assess the effectiveness of certain institutions. The analysis revealed many similar assessments across the subfields, in spite of the institutional differences among them, e.g. the much higher number of institutions for renewable energy than for FFSR and carbon pricing. The main disadvantages that experts and stakeholders in all three subfields highlighted were duplication of work, conflicting messages, and competition, which were all seen as obstacles in the way toward stronger synergies. On the other hand, institutional complexity offered more opportunities and venues to include a broader spectrum of actors and interests.

Potential explanations for these results may be related to our findings for legitimacy and call for further research. One could expect, for instance, that the social values, norms, and experiences that impact legitimacy assessments also play a role in effectiveness assessments. What speaks for this assumption is that the three dimensions of institutional effectiveness (output, outcome, impact) are also relevant dimensions for the output legitimacy of institutions.

An explanation for the differences found in assessments of effectiveness between the three subfields may be the overall level of complexity in a subfield as a more fundamental reason, and the respective explanations for coherence and management discussed in the previous section (Section 9.2.1). The position in a subfield in the climate-governance network or, for instance, the underlying constellations of interests, may not only shape the degree of complexity of that subfield, but also the (perceived) advantages and disadvantages arising from that complexity. All these potential connections point to the need for a groundbreaking research programme – one that provides scholars and stakeholders with novel, theory-driven analyses to explain and understand the complexity of a nexus and its consequences for legitimate and effective governance.

9.3 Recommendations

9.3.1 Enhancing Coherence and Management in the Climate-Energy Nexus

The ensuing recommendations are based on the empirical results of Chapters 4–6 of this volume. These chapters present three case studies on three major subfields of the climate-energy complex: renewable energy, fossil fuel subsidy reform (FFSR), and carbon pricing. The studies found that all three subfields, and the institutional complexes that govern them, are marked by medium coherence; minimal or low levels of management; and significant challenges, trade-offs, and conflicts. Given this similarity across the three subfields, our most important recommendations apply to all of them, and they are also of relevance for a wider universe of subfields within the climate-energy nexus that show similar levels of coherence.

First of all, improving coordination and building awareness of the activities of institutions is important for avoiding duplication of efforts and conflicting messages. Such conflicting messages may create confusion among relevant actors. They also provide alternatives for those actors that are opposed to certain policies and institutional efforts, helping such actors to ‘shop around’ for a different arena more suited to their preferences. Ultimately, conflicting messages may lead to conflicting impacts of institutions. This happens intentionally when institutions promote specific objectives at the expense of other goals. An illustrative example comes from the renewable energy subfield. There, the energy trilemma poses a challenge to all international institutions working on this area – namely to achieve three objectives simultaneously – energy security, energy access, and sustainable energy – that frequently conflict with each other. Many institutions settle for prioritizing one of these goals, and there is, in fact, currently a clear bias in global renewable governance toward clean energy access (Sanderink 2019). Against this backdrop, and to avoid conflicting impacts, there is a strong need for a more integrated, cross-institutional acknowledgement of, and approach to, the energy trilemma. In particular, the potential of renewables to address energy security must be recognized more widely.

Duplication of work is a serious issue given that the institutions have limited resources in terms of staff, budget, and expertise. Improving coordination and awareness of each other’s activities is key – and it does not necessarily call for a single institution that acts as orchestrator or coordinator. Rather, strengthening and expanding existing inter-organizational coordination mechanisms and possibly establishing new ones could improve oversight and integration. To identify the most urgent governance gaps, it is crucial to keep track of which institutions and actors are performing which tasks in which part of the world. Such a clearinghouse approach or information hub could be modelled on, for instance, the Global Climate Action portal of the UNFCCC (NAZCA 2019). Based on such continuous and cross-cutting information, institutions could then ideally adapt or shift their activities or mandates accordingly.

Second, we need clearer definitions. All three subfields we analyzed suffer from differing and even conflicting interpretations of central concepts and norms, such as what constitutes renewable sources of energy, fossil fuel subsidies, and carbon pricing. These diverging interpretations may lead to inconsistencies and tensions when pursuing core objectives. For instance, the various and competing definitions of fossil fuel subsidies entail considerable differences between estimates of annual global fossil fuel subsidization – ranging from several hundreds of billions by the OECD and IEA (OECD 2018a, 2018b) to several trillions by the IMF (Coady et al. 2017). Such definitional ambiguity among international institutions and their members has allowed some countries to maintain that they have no subsidies at all, even though there are numerous methodological aspects on which these institutions agree.

A possible solution to this ambiguity could be some form of joint ‘minimum’ definition put forward by all intergovernmental institutions on FFSR. Such a minimal but flexible consensus could leave the door open to complementary approaches such as those of the IMF, which highlights the broader societal costs of government support for fossil fuels. Likewise, for the subfield of carbon pricing, diverging interpretations could be addressed by further specifying the norm of carbon pricing. A cross-institutional consensus could clarify that carbon pricing entails the payment of a non-trivial price for a significant share of emissions. Such a clarification could avoid conflicting signals and prevent carbon pricing schemes that only place a very low price on overall emissions. While such a clarification is strongly needed, it should still leave room for promoting both carbon taxes as well as emissions trading systems instead of just focusing on one of the two. Put differently, clarification is a balancing act. Too much norm or goal specification can come at the expense of a wider acceptance or support, particularly when competing definitions are rooted in actors’ diverging preferences and worldviews.

Third, silo mentality should be avoided. Actors and institutions operating in the three subfields of renewable energy, FFSR, and carbon pricing should build stronger connections to other policy areas in the climate-energy nexus and beyond. For instance, the institutions in global renewable energy governance appear to compete with energy efficiency institutions over resources and visibility. Such competition can be in part prevented by separating opportunities for (financial) support and resources for renewable energy and energy efficiency, while simultaneously supporting collaborations between the two areas.

Another example of the need for stronger cross-area coordination is the importance of integrating fossil fuel subsidy reform within the UNFCCC (Skovgaard and van Asselt 2019). Moreover, in a broader sense of sharing experiences and spreading knowledge, institutions within the three subfields need to reach out to sectors and institutions beyond the climate-energy nexus. Many actors and institutions operating in other policy domains still need convincing of the urgency of a global energy transition toward renewables.

Fourth, certain institutions could act as orchestrators in order to facilitate these and other measures. To play such an orchestrator role, an institution would ideally have a broad membership and the convening power that comes with it, an extensive mandate, and a high degree of acceptance – qualities associated with major UN institutions – along with the organizational capacity to play such a role.

Within the renewable energy subfield, one candidate could be IRENA. Being positioned at the centre of the energy trilemma, it is one of the few renewable energy institutions that simultaneously promote energy security, energy access, and environmental sustainability concerns. Furthermore, IRENA’s mandate to gather and disseminate comprehensive information would allow the organization to initiate a database on global renewable energy activities and to facilitate discussions on the benefits and shortcomings of various energy sources. Finally, having almost universal membership and being well-known as a focal point for renewable energy, IRENA can have convening power to invite institutions and actors from areas other than energy governance to join renewable energy discussions. All this said, a major limitation is that IRENA – like the International Energy Agency that could potentially also play such a role – is not a UN organization, which may affect its acceptance as an orchestrator.

Within the subfield of fossil fuel subsidy reform, the UN Environment Programme (UNEP) is well-placed to take over the function of an orchestrator. UNEP coordinates efforts to bring together multiple stakeholders to develop a reporting methodology for SDG indicator 12.c.1 on fossil fuel subsidies. It may also play a valuable role in improving coordination and coherence between the institutions working in this area (UNEP et al. 2019). Furthermore, UNEP benefits from having a universal membership. Nonetheless, efforts to avoid duplication of work could also be undertaken by (a subset of) institutions independently. Given the WTO’s existing definition of a subsidy, parallel efforts to promote FFSR through this forum also help socialize a common definition of a fossil fuel subsidy (Verkuijl et al. 2019).

For the subfield of carbon pricing, the two most obvious orchestrators are the UNFCCC and the World Bank. Yet, both of them suffer from specific shortcomings. The World Bank is an institution in which industrialized countries have disproportionate influence compared to the UN institutions, which limits its legitimacy particularly in the eyes of developing countries. The UNFCCC has much greater legitimacy in this respect, but is constrained by the modest resources of the UNFCCC Secretariat and the often-protracted decision-making procedures.

Some of these concerns are also relevant regarding a more general orchestration role within the larger climate-energy nexus. Despite the UNFCCC’s relatively large secretariat, it cannot be expected to coordinate the institutional complex on climate change. It neither has the mandate nor the organizational capacity to do so. The secretariat has, however, been engaging in light-touch coordination using orchestration as a mode of governance. It collaborates with other institutions and actors to provide platforms and data that could help to mitigate cross-institutional coordination gaps (Hickmann et al. 2019).

9.3.2 Enhancing Legitimacy and Effectiveness in the Climate-Energy Nexus

The feedback from selected experts in our surveys and interviews confirmed that greater cross-institutional coordination is also necessary to address some of the negative impacts of institutional complexity for the effectiveness in the climate-energy nexus. In line with the recommendations we outlined in the previous section, a more overarching and clear division of labour, reducing incongruence of messages, and facilitating greater information and data-sharing were all seen as key steps by interview respondents. They cautioned, however, that diverging interests will keep rendering such efforts difficult. Moreover, coordination attempts would need to be designed in such a way as not to overburden the institutions with bureaucracy and to instead strike a balance between greater harmonization and maintaining the autonomy of institutions. Otherwise, effectiveness could be hampered by ill-designed coordination attempts. Here, key member states that are members of several institutions may have a role to play to push for improved coordination.

Effectiveness ultimately implies that the output of the many governance institutions contributes to actual problem-solving – i.e. successfully scaling up renewables, providing stringent carbon pricing, and phasing out of fossil fuel subsidies. As Chapter 8 showed, it is difficult to isolate and trace back the impact of the three institutional complexes on such problem solving, either because that impact is limited in the first place, or because of limitations in data. Nonetheless, there is considerable scope for improving problem-solving in each of the three subfields we analyzed. This includes enhancing the effectiveness of the measures that have already been brought on their way. In the case of carbon pricing, for instance, this would imply a stronger concentration on the adoption of carbon pricing and on effective carbon price levels, e.g. in line with the recommendations of the High-Level Commission on Carbon Prices (2017) that suggests prices between 40 and 80 US $ per metric tonne of CO₂ equivalents.

Chapter 7 and its study of the legitimacy perceptions of five key institutions showed a significant variation across stakeholder groups. Against this backdrop, international institutions, first, need to apply different legitimation strategies to target different audiences and their varied interests. To be visible and reach various audiences in an increasingly crowded governance nexus, it is particularly important for international institutions to provide a credible picture of their purpose, procedures, and performance through multiple channels. The study differentiated the legitimacy assessments according to nine dimensions of institutional legitimacy, e.g. transparency, accountability, and procedural and distributive fairness. The responses on these dimensions provide a useful indication of which institutional qualities the five institutions need to communicate (and improve on) more extensively, in order to be perceived more positively by certain stakeholders.

Moreover, the study showed that stakeholders’ familiarity with institutions matters for their sociological legitimacy. This points to the importance of outreach activities. Institutions operating in a dense institutional complex not only depend on performing their function well in order to be viewed positively, but also need to communicate and engage with a range of stakeholders, both members and non-members. Put differently, institutional complexity requires institutions to make additional efforts to improve perceptions of their legitimacy in two different turfs: first, in an intricate institutional landscape through cooperation and competition with other institutions that have overlapping mandates; and second, in a more complex communication landscape with outreach activities directed at different audiences.

9.4 Outlook

We conclude this chapter, and this book, with identifying urgent research gaps and future research avenues that we have repeatedly come across in our comparative and in-depth assessment of the climate-energy nexus. As Section 9.2 already includes our suggestions for further theory development on the analysis of a governance nexus, we concentrate our outlook on essential empirical questions of high relevance for researchers, policy practitioners, and stakeholders.

We structure our suggestions along our three case studies – renewable energy, fossil fuel subsidy reform, and carbon pricing – and the themes of legitimacy and effectiveness. There are cross-cutting aspects among them that point in the same direction, namely (1) to learn more about the causes of institutional complexity in the climate-energy nexus, and, more concretely; (2) to substantiate claims on the causal links between the degree of institutional complexity (in terms of coherence and management) and the implications of this complexity (in terms of legitimacy and effectiveness); (3) to identify conditions for successful management efforts and spill-over effects across institutions and levels and, while doing all this; (4) to go beyond renewable energy, fossil fuel subsidy reform, and carbon pricing, and expand studies toward other subfields in and beyond the climate-energy nexus.

For renewable energy, we need to know more about the interplay between renewable energy institutions and institutions from other energy-related areas, especially with regard to energy efficiency. Renewable energy and energy efficiency along with low-carbon technologies are key areas in decarbonizing energy systems and transitioning to fossil-free or carbon-neutral societies. However, as mentioned earlier, they more often than not compete in spite of their common vision. Moreover, and with a view to addressing the energy trilemma more appropriately, further research is required to explore how increased coordination efforts can take shape, and what role UN and non-UN organizations can play for this purpose, such as UNFCCC, UN Energy, IRENA, and IEA.

With respect to fossil fuel subsidy reform, one issue that could be examined in more depth is the interplay between limited-membership coalitions, such as the G20 or the Friends, and multilateral arenas. Such research could look into the extent to and conditions under which progress made in small-n settings can be integrated in multilateral decision making. Respective studies could draw on existing literatures in other areas, for instance on the role of regionalism in international trade governance (Baldwin 2014). A second avenue for further research is to parse out the influence of international institutions on the implementation of domestic FFSR policies. As Chapter 5 discussed, countries have taken on commitments to reform their fossil fuel subsidies through various international forums, including the G20, the UNFCCC, the High-Level Political Forum, and Agenda 2030. The extent to which these international commitments trickle down to domestic action needs empirical assessment, through examining which countries have implemented reforms following the introduction of such commitments.

Future research on carbon pricing could, when exploring causes and consequences of coherence and management, focus more on the role of individual members (see also Andonova and Mitchell 2010). Regarding research on causes, one could analyze whether or not overlapping membership between institutions facilitates inter-institutional coordination – in particular through specific officials or units of a joint member. For instance, it is likely that the same states were to a large degree represented vis-à-vis the UNFCCC by their environment ministries (see Skovgaard and Gallant 2015), vis-à-vis the World Bank institutions by their finance and development ministries, and vis-à-vis CORSIA by their transportation ministries. Exploring the impact of identical or different negotiators in different institutions on overall coherence could provide both empirical and theoretical contributions important to academics and policy makers alike. In terms of consequences, future research should focus on whether and how the presence of interaction mechanisms between two institutions influenced the decisions of actors to adopt carbon pricing. If the relationship between two institutions became more coherent, did this also make them more influential toward their members? Comparisons of such constellations across time as well as across different subfields could provide insights of great policy relevance.

Concerning the connection between institutional complexity and perceptions of legitimacy, this book has provided a novel empirical study on five institutions from the subfield of renewable energy. First of all, this type of analysis could be expanded to institutions from other subfields. Such a wider examination can provide comparative insights into how the different levels of coherence across institutional complexes affect perceptions of legitimacy. Second, further research is required to establish why different stakeholders assess the legitimacy of institutions differently in a complex institutional environment. A broader set of potential explanatory factors could be put to the test here to examine whether these differences ultimately stem from processes of socialization or whether they are rather due to functionalist or rationalist reasons. Third, while we asked stakeholders to assess institutions according to nine dimensions of legitimacy, the next step would be to evaluate the relative importance that each stakeholder type assigns to each dimension. This would help to capture the overall legitimacy of an institution in greater detail. Such a research avenue could also explain and understand how different dimensions of legitimacy relate to one another and which factors may lead to a legitimacy crisis. Fourth, and based on these insights, the role of specific legitimation and delegitimation strategies under institutional complexity merits further enquiry. Finally, our study focused on legitimacy at the micro level, i.e. for individual institutions. The next research frontier is to study how institutional complexity affects legitimacy for an entire institutional complex (meso level) or governance nexus (macro level).

With regard to effectiveness in the climate-energy nexus, our empirical findings indicate that output effectiveness, i.e. generating regulations and infrastructure, is high across the three institutional complexes for renewable energy, FFSR, and carbon pricing. However, behaviour-changing outcomes and problem-solving impacts of institutions were less significant in all three cases. Whether this is a result of institutional complexity, or perhaps a reflection of the limited authority of international institutions in these subfields, warrants further investigation.

Moreover, further research is needed to examine interactions across different subfields and how these affect the effectiveness of the climate-energy nexus (Sanderink and Nasiritousi 2019). On the domestic level, the three issues we scrutinized in this volume are often interlinked, e.g. in the sense that fossil fuel subsidies lower the de facto carbon price and make renewable energy less competitive. Likewise, renewable energy may provide a useful instrument for limiting potential negative consequences of FFSR and carbon pricing through energy access and rising energy prices. In turn, FFSR and carbon pricing may finance the expansion of renewable energy installations. Concerns about just transitions to a zero-emissions world (Newell and Mulvaney 2013) are relevant to all three issues and could benefit from addressing all three of them in a joint manner where possible.

Furthermore, studying other governance fields in a similar manner could provide a useful comparison to examine whether institutions in other policy domains have overcome some of the shortcomings of institutional complexity and why. Such research could help identify additional management options for international institutions to close governance gaps and enhance effectiveness in the climate-energy nexus. Finally, it should be noted that effectiveness as defined here does not capture the issue of efficiency. While the three subfields were all considered effective at the output level, future studies should also examine how economic efficiency can be improved under institutional complexity. This is important given strong competition over resources in different policy fields.

The governance of the climate and energy nexus, and the ability of its institutions to address key issues – such as renewable energy, fossil fuel subsidy reform, and carbon pricing – in a timely manner will have great consequences on people and the planet in the years to come. We therefore encourage scholars, stakeholders, and practitioners alike to address the theoretical, empirical, and societal dimensions of the governance of the climate-energy nexus sooner rather than later. We hope that the research presented in this volume provides a first and crucial step to advance this important research frontier and to ultimately help resolve one of the largest planetary challenges ahead of us.