The dismantling of renewable energy policies: The cases of Spain and the Czech Republic

Energy Policy 133 (2019) 110881

Contents lists available at ScienceDirect

Energy Policy journal homepage: www.elsevier.com/locate/enpol

The dismantling of renewable energy policies: The cases of Spain and the Czech Republic

T

Konrad Gürtlera, Rafael Postpischilb, Rainer Quitzowa,∗ a b

Institute for Advanced Sustainability Studies, Berliner Straße 130, 14467, Potsdam, Germany Environmental Policy Research Centre, Freie Universität Berlin Ihnestraße 22, 14195, Berlin, Germany

GRAPHICAL ABSTRACT

ARTICLE INFO

ABSTRACT

Keywords: Renewable energy Solar energy Feed-in tariffs Political economy Policy dismantling Policy feedbacks

Despite their increasing cost competitiveness, the continued expansion of renewable energy remains dependent on policy support. Moreover, the dismantling of renewable energy policies in a number of former pioneer countries indicates that continued policy support is not a foregone conclusion. Yet, in light of the accelerating expansion of renewable energy, the dismantling of renewable energy policies has captured comparatively less attention than the rapid spread of support schemes. This article seeks to fill this important knowledge gap by developing and testing a framework for the analysis of policy dismantling processes in the renewable energy sector. It applies the framework to conduct a comparative analysis of policy dismantling in Spain and the Czech Republic. Both countries represent European pioneers of renewable energy support who subsequently dismantled their policies. The paper finds that the inter-relationship between policy design and the broader configuration of the political economy in the energy sector are key for understanding dismantling processes. It offers a number of conclusions for the design of more robust renewable energy support policies.

1. Introduction The expansion of renewable energy represents a major, accelerating trend in the global energy sector. Spearheaded by the introduction of feed-in tariff (FiT) policies in Germany and a number of other European countries, the diffusion of renewable energy now represents a worldwide phenomenon (Quitzow et al., 2016). Since 2015, global capacity

∗

additions in renewable energy have outstripped those in fossil-based electricity generation (IEA, 2016c, 2017b). This trend is driven by the dramatic cost reductions over the past decade (Quitzow et al., 2016; IEA, 2017b). Nevertheless, the growth of renewable energy markets still remains a policy-driven phenomenon (REN21, 2018). In the medium term, the transition to a low-carbon, renewable energy system is expected to require dedicated policies aimed at facilitating the further

Corresponding author. E-mail addresses: [email protected] (K. Gürtler), [email protected] (R. Postpischil), [email protected] (R. Quitzow).

https://doi.org/10.1016/j.enpol.2019.110881 Received 23 December 2018; Received in revised form 31 May 2019; Accepted 10 July 2019 0301-4215/ © 2019 Elsevier Ltd. All rights reserved.

Energy Policy 133 (2019) 110881

K. Gürtler, et al.

expansion of renewable energy (EEA, 2016). As prices continue to drop, the pressure on governments around the world to adopt such policies is increasing steadily. Indeed, renewable energy policies are now the rule rather than the exception in countries around the globe. At the same time, a number of countries have reversed previous support for renewable energy and have dismantled their renewable energy policies with important implications for market and industry development in these countries (REN21, 2015). These examples show that continued policy support for renewable energy is not a foregone conclusion. Despite the impressive developments of the past decade, the sector's continued dependence on policy support makes it vulnerable to political changes. In light of the accelerating growth of renewable energy, the dismantling of renewable energy policies has captured comparatively less attention than the rapid spread of support schemes. The same is true for the literature on policy change, which has paid considerable attention to understanding what enables or constrain the introduction of new policy measures. Relatively less attention has been paid to the phenomenon of policy dismantling. Given the urgency of a global transition to a low-carbon energy system, the lack of analysis of policy dismantling in the renewable energy sector represents an important knowledge gap. A better understanding of the factors influencing the dismantling of renewable energy policies promises to yield insights on how to develop more robust policy support that is able to resist short-term political pressure. Ultimately, the durability of policy support will decide the speed and scope of renewable energy development. This article seeks to fill this important knowledge gap by developing and testing a framework for the analysis of policy dismantling processes in the renewable energy sector. It applies the framework to conduct a comparative analysis of policy dismantling in Spain and the Czech Republic. Both countries represent European pioneers of renewable energy support who subsequently dismantled their policies. The paper is structured as follows. Section 2 provides an overview of the political science literature on policy dismantling. Section 3 outlines the proposed analytical framework for the analysis of policy dismantling in the renewable energy sector. Section 4 outlines the rationale and limitations of the selected case studies and describes the methods for data collection. Section 5 presents the empirical results of the two case studies. Section 6 provides a discussion of key insights and their implications for furthering a better understanding of policy dismantling in the field of renewable energy and how this might contribute to the broader literature on policy dismantling. Section 7 discusses key implications for policy making in the renewable energy sector.

regarded in the context of austerity pressures (Pierson, 2001), later also with regard to globalization and international regulatory competition (Knill et al., 2009). Overall, studies in the field have focused primarily on single case studies drawn from the American political system and have not attempted to develop a comprehensive framework for understanding dismantling across countries or policy fields (Jordan et al., 2013). In the past decade, efforts have been made to develop comparative approaches for the study of policy dismantling (Bauer et al., 2012; Bauer and Knill, 2014; Jordan et al., 2012). Among other things, this research has demonstrated that the politics of dismantling shows important variation across different policy fields. Taking this as a point of departure, this paper focuses on policy dismantling in the field of renewable energy. It seeks to combine generic factors, derived from existing political science literature, with a nuanced treatment of the policy field. 3. A framework for the analysis of policy dismantling in the field of renewable energy Within the more recent literature on policy dismantling, Bauer and Knill (2012, 2014) provide the most comprehensive analytical framework for studying policy dismantling processes. The authors propose an analytical framework composed of three main groups of factors influencing the policy dismantling process: (1) external factors, (2) institutional constraints and opportunities as well as (3) situational factors (Bauer and Knill, 2012, 2014). In the following, we adapt and further specify this framework for the analysis of policy dismantling in the field of renewable energy policy. In particular, we argue that it is necessary to further elaborate the under conceptualized category of “situational factors”. While the other two categories relate to factors that are external to the chosen policy field, “situational factors” are intended to capture factors that result from the dynamics within the specific field of policy itself, in this case (renewable) energy policy. To facilitate a more in-depth analysis of the latter, we propose to replace the category of “situational factors” with the following two categories: the political economy of the policy field (1) and aspects related to the policy design (2). The former captures the dimension of politics within the policy field, while the latter focuses on the dimension of policy. Continuing this logic, the existing category of “institutional constraints and opportunities” represents the dimension of polity, i.e. the overall configuration of the political system and how it enables or constrain policy dismantling. The category of “external factors” relates to exogenous socio-political and economic events or trends, which may have an influence on the dismantling of a chosen policy but that are not themselves influenced by the policy in question. As stated above, both these exogenous factors and the institutional constraints and opportunities are considered “external” to the policy field, while aspects related to the political economy and the policy design are considered internal to the policy field. For the sake of analytical clarity, the category “external factors” is relabeled as “macro-level factors” in the revised framework. Fig. 1 provides a graphical representation of the new framework, illustrating the relationship between the different categories and the dependent variable (i.e. policy dismantling). As the figure suggests, the final dismantling decision is assumed to be primarily a result of the political economy of the policy field. This may in turn be influenced by the specific policy design as well as macro-level factors. Institutional constraints and opportunities have a mediating role. They represent the institutional setting through which political preferences are translated into actual policy dismantling decisions. Taking these basic assumptions as a starting point, the remainder of the section provides a more detailed description of the framework, discussing each of the four categories and their interrelationships.

2. Policy dismantling: a brief review of the literature In comparison to other phases of the policy cycle, policy dismantling has so far received rather limited attention in political science. Policy dismantling comprises phenomena such as complete termination, weakening, replacement, or a lack of policy updates (Bauer, 2006). It is viewed as a subcategory of the broad field of policy change, with the direction of reducing, decreasing or even terminating existing policy arrangements. This specific direction of policy change is thought of as being subject to particular dynamics of causes, conditions and strategies, which demand a distinct analytical approach (Bauer et al., 2012). Research on the more narrowly defined phenomenon of policy termination started in the 1970s in the USA (Bardach, 1976; Behn, 1978; Brewer, 1978; DeLeon, 1978). With a major focus on US social policy, the early policy termination literature contributed to understanding the effects of policy termination and why complete termination has been rarely successful. The 1980s and 1990s saw research on deregulation in the US. This focused on deregulation strategies and their effects (Jordan et al., 2013). In the 2000s, a new wave of welfare state dismantling or retrenchment literature developed. Dismantling decisions were 2

Energy Policy 133 (2019) 110881

K. Gürtler, et al.

Fig. 1. Overview of analytical framework for the analysis of policy dismantling based on Bauer and Knill (2014).

structural features shaping actor interests have an important influence on whether renewable energy policies are more or less prone to dismantling. In a first step, structural conditions create an environment that is more or less favorable to a continued expansion of renewable energy. This in turn translates into more or less pressure to dismantle existing policies to promote the deployment of renewable energy. For instance, in energy systems that exhibit strong growth in demand – as is the case in many developing countries – the interest of actors in dismantling renewable energy policies is likely to be weaker than in systems characterized by stagnating or even shrinking demand. In a second step, the specific actor configuration will determine whether such interests are translated into effective political pressure to dismantle policies. For example, energy markets characterized by one or a small number of large, state-owned energy companies with large fossil-based power generation capacities are likely to be subject to more targeted political pressure in favor of dismantling policies than markets characterized by a larger number of smaller, private suppliers with a more diverse set of investments. A systematic assessment of these structural features and how they translate into political actor constellations is essential to understanding whether the dynamics of the policy field are likely to favor the dismantling of renewable energy policies.

3.1. Policy design Policy design refers to how the specific design of policies influences the process of policy dismantling. As indicated in Fig. 1, the intricacies of policy design are unlikely to directly shape the preferences of political decision-makers for or against policy dismantling. Instead, the design of a policy may influence the political dynamics within a given policy field. In doing so, the design of a given policy may translate into pressure for or against its continuation. For instance, policy design may have implications for the resulting distribution of costs and benefits across those affected by a policy. Thereby, the design encourages the emergence of networks around a specific policy, which strive to lock-in or overcome previous decisions (Bauer and Knill, 2012; Pierson, 1994, 2001; Steinebach and Knill, 2017). In the case of market support policies for renewable energy, a key question relates to how the costs of financing the scheme are distributed (Jacobs, 2012) and to what extent the economic benefits accrue to local populations or firms (Lewis and Wiser, 2007). More recently, political discussions are increasingly revolving around the role of market support policies in enabling benefits for local populations, while limiting negative impacts on the local environment (Liebe and Dobers, 2019). These and other intricacies of policy design and their implications for policy dismantling are addressed in the corresponding category of the analytical framework.

3.3. Macro-level factors

3.2. Political economy of the policy field

In the subsequent analysis, macro-level factors represent factors relating to the wider exogenous environment. This includes major socio-political or economic changes or events. Macro-level factors are characterized by the fact that they are beyond the direct influence of actors, and therefore difficult to change deliberately. They are unlikely to directly influence a dismantling decision. Rather, they shift the balance of power within the policy field, thus supporting or undermining policy dismantling. They can comprise macro-level changes or events, including but not limited to major economic shocks, countrylevel political developments, or supranational political pressures. Bardach (1976) holds that economic uncertainty or crisis can act as a trigger for the dismantling of policies. A change of government represents another major macro-level factor. In particular, the transfer of power to a different political party or coalition seems to provide a window of opportunity for the termination of policy measures. Newly

The political economy of the policy field refers to the specific political dynamics of the relevant policy field and the underlying institutional and structural features that shape these dynamics. By addressing dynamics within the chosen policy field, we seek to identify features of the policy field itself that may enable or constrain policy dismantling. For the field of energy policy, we define the political economy as the inter-relationship between structural features of the energy system, on the one hand, and the configuration of actors, their interests and political power, on the other. Structural factors refer to the relatively static physical and institutional infrastructure of the energy system. This translates into specific actor constellations with particular interests and resources to pursue their interests in the political realm. We propose that this actor constellation and the 3

Energy Policy 133 (2019) 110881

K. Gürtler, et al.

elected politicians feel less bound to their predecessor's policies or even incentivized to dissociate themselves from the past (Bardach, 1976). It should also be noted that in some cases the political dynamics of the policy field may contribute to electoral outcomes, thus creating a feedback on the balance of power in the policy field. This is depicted by arrow 2 in Fig. 1. We also consider other external events that have an important influence on the legitimacy of certain policy decisions. Bardach argues that a “delegitimation of the ideological matrix in which the policy is embedded” (Bardach, 1976) may work as an enabling factor of policy dismantling. This might include major catastrophes, such as the Fukushima nuclear accident, as well broader societal trends as well as supranational political pressure.

the framework. The empirical data constitute the results of semi-structured interviews with open-ended questions, which are commonly used for expert interviews (Leech, 2002). Fifteen interviews were conducted between July 2015 and October 2015. Interviewees included consultants, NGO representatives, scientists, lawyers, interest group representatives, journalists and government representatives (see Table 1). Since some of the interview partners preferred to answer anonymously, we employ a shortcut for country code and interviewee number in the following (e.g. ES01 or CZ02). Our conclusions are drawn from statements made in the interviews based on both frequency as well as persuasiveness of the arguments expressed. The information gathered from the interviews have been matched by findings from relevant primary sources such as annual reports or policy documents as well as secondary sources, including articles published in specialized media and academic journals.

3.4. Institutional constraints and opportunities

5. Empirical results

Finally, institutional constraints and opportunities refer to checks and balances within a specific political system or polity, which make policy dismantling more or less likely. These institutional factors comprise features of the legislative, executive and judicial branches of government as well as the overall party system and the election cycle (Bauer and Knill, 2012, 2014; Korte et al., 2012; Steinebach and Knill, 2017; Tsebelis, 2002). The sum of these institutional constraints and opportunities – in number as well as intensity – determines to what extent a change in preference of a political actor may lead to a change in policy. The number of independent political institutions with veto powers in a given polity as well as the alignment or heterogeneity of the political actors’ preferences within those institutions represent important elements in this context. These institutional veto points of the political system are strongly intertwined with the struggle of interest within those institutions for and against a dismantling decision. Policy dismantling is marked by politicians and political parties seeking to overcome or circumvent such veto points and thereby making policy dismantling more likely (Bauer, 2006; Behn, 1978). In some cases, macro-level trends, such as the trend towards more populist and authoritarian governments, may also influence the functioning of the political system or culture, which may affect the configuration of the institutional opportunities and constraints that mediate dismantling decisions (see arrow 3 in Fig. 1).

The following section presents the detailed empirical results of the two case studies. Each case study begins with a brief introduction of the case, followed by a systematic assessment of each category of the analytical framework. 5.1. Spain 5.1.1. Overview of renewable energy policy dismantling in Spain Spain has the fourth-largest share of installed wind power capacity and the sixth-largest share of installed PV capacity worldwide (REN21, 2015). Up to the year 2008, Spain was a global leader in the promotion of renewable energy (Pew Report, 2014; REN21, 2015). Renewable energy promotion was ramped up in 1997 when Spain introduced a Special Regime for the promotion of Renewable Energy in its Electricity Sector Law (Law 54, 1997). While wind energy installations grew continuously since the mid-1990s, the number of PV installations remained low until 20042 (del Río and Mir-Artigues, 2014). Royal Decree 661, 2007 kick-started an accelerated deployment of renewable energy facilities in Spain. Main features included priority access to the grid, more favorable conditions for facilities larger than 100 kW, and a revision of the FiT rates only every four years. This triggered a boom in solar PV: installations rose from 103 MW in 2006 to 544 MW in 2007 and to 2708 MW in 2008. These dramatic developments in the PV sector represented a determining factor in the dismantling process that ensued. After the boom, costs of the solar FiT grew to around 50 percent of total renewable support costs, although solar PV only accounted for around 10 percent of electricity generation from renewables (see del Río and Mir-Artigues, 2014). At the end of September 2008, the Spanish government pulled the emergency brake and applied several amendments to the renewable energy support scheme, particularly targeting the solar PV sector. The most severe changes included an annual 400 MW cap on support for new PV installations,3 a reduction in tariff levels for future PV installations, and a maximum period for which PV subsidies were available (Royal Decree 1578, 2008). In 2010, FiT rates for solar PV were further reduced and an annual cap on the hours of production that

4. Case selection and data collection The cases Spain and Czech Republic were selected for in-depth analysis and comparison, as these countries were the first two European countries with substantial policy programs for the deployment of renewable energy that subsequently chose to abandon their FIT schemes. Spain terminated its FiT for new installations in January 2012 (REN21, 2013), the Czech Republic in January 2014 (REN21, 2014). Both countries represent extreme examples of the phenomenon under investigation.1 As suggested by Seawright and Gerring (2008), the selection of extreme cases is a suitable strategy for an exploratory analysis of a phenomenon. We expect the factors outlined above to be relatively easily recognizable, which is useful for an initial application of the analytical framework. While the results will not be immediately generalizable, they should help in generating initial insights and the development of hypotheses for further research. Moreover, we expect them to generate important practical lessons for the design of more robust renewable energy support policies. In a second stage of research, it will be useful to draw on the resulting insights to select cases based on carefully chosen independent variables for a more nuanced testing of

2 First steps to promote renewable electricity had already been introduced in the 1980s, e.g. through the Plan de Energías Renovables in 1986. However, their influence remained limited, especially with regards to solar PV. 3 Royal Decree 1578, 2008 newly introduced a distinction between rooftop (max. 2 MW capacity) and ground-mounted installations (max. 10 MW capacity). Of the 400 MW supported through the new FiT scheme, 267 MW were dedicated to rooftop installations and 133 MW to ground-mounted installations. The introduction of a registry for pre-assignment and the quarterly assignment of variable tariffs constituted new administrative hurdles. Only the tariffs for the first tendering round had been fixed; in subsequent bidding rounds tariffs further decreased.

1 Italy represents a third European frontrunner, which has seen the dismantling of its ambitious support scheme focused mainly on solar PV (see Di Dio et al., 2015). Due to capacity constraints, the decision was taken to limit the analysis to only two countries.-

4

Energy Policy 133 (2019) 110881

K. Gürtler, et al.

minimum transition period of 12 months. For solar PV, this point was reached in September 2007. The announcement of the FiT phase-out resulted in a sudden and exponential increase in investment from developers eager to make use of the remaining window of opportunity. This made Spain the country with the highest annual installed solar PV capacity in 2008. This was underpinned by FiT rates considered by numerous experts to be significantly higher than needed, particularly for solar PV (ES08, see also del Río and Mir-Artigues, 2014). Moreover, the scheme lacked a mechanism to reduce FiT rates in line with the cost of technology. As a result, the 2008 solar boom generated a very significant cost burden to the Spanish electricity system (see Fig. 2). As solar PV was targeted as the main driver for the accelerating costs, the initial reforms hit the solar PV significantly harder than the wind sector (Haas, 2017). In addition, it was not entirely clear who would ultimately cover the costs of the subsidy. In general, the FiT in Spain is refinanced through a surcharge on consumer electricity prices, similar to many other countries with FiT systems (Haas, 2017, p.53). What makes the Spanish case special is that final electricity prices are controlled as a measure to protect consumers. Thus, FITs for solar PV and other renewable energy technologies did not lead to price increases for ratepayers. Instead, they contributed to a so-called “tariff deficit” (TD) 6 (del Rio and MirArtigues, 2014: 8f.). The TD is the result of a difference between fixed consumer electricity prices and higher reported actual costs by the energy companies. In Spain, this tariff deficit was initially borne by the five vertically-integrated utilities (European Commission, 2014). The TD began increasing from 2000 (European Commission, 2014; based on CNE data) and reached €28.5 billion at the end of 2013, which equals nearly 3% of Spain's GDP (European Commission, 2015). Opponents of renewable energy expansion, including the affected utilities, made the contribution of renewable energy subsidies to the TD a central point of political debate (ES02, ES06).

Table 1 Anonymized list of interviews. Source: Authors. Interviewee

Index

Medium

Date

ES01 ES02 ES03 ES04 ES05 ES06 ES07 ES08

In-Person Skype Phone Skype Phone Phone Phone Skype

22/07/2015 03/08/2015 06/08/2015 11/08/2015 12/08/2015 12/08/2015 23/09/2015 09/10/ 2015

CZ01 CZ02 CZ03 CZ04 CZ05 CZ06 CZ07

In-Person Skype Phone Phone Email Email Email

04/08/2015 21/08/2015 02/09/2015 27/08/2015 28/08/2015 04/09/2015 26/10/2015

Spain Consultant NGO representative Researcher Government representative Consultant, Lawyer Representative of Solar PV Federation Consultant, Lawyer Researcher Czech Republic Consultant Consultant Government representative Government representative Journalist working on Renewable Energy Energy Consultant Employee of the Czech Energy Regulatory Office

facilities could deliver electricity4 was introduced (Royal Decree Law 14, 2010). In January 2012, under a new government led by the conservative People's Party, financial incentives for all RES-E sources were suspended (del Río and Mir-Artigues, 2014; see also European Photovoltaic Industry Association (EPIA), 2013; Royal Decree Law 1, 2012). A special electricity tax of 7% for all electricity generators (conventional and RES-E) was introduced in December 2012 (Law 15, 2012). In mid-2013, the remuneration system was completely changed (Royal Decree Law 9, 2013). Renewable energy installations no longer receive a FiT for the energy produced, but a special remuneration in reference to the installed capacity, which shall guarantee a so-called reasonable return5 for the installation (Bird and Bird, 2014). Resulting uncertainty and reduced profitability have led to stagnation in the expansion of renewable energy (Red Eléctrica de España, 2015). After a four-year period of standstill, Spain restarted renewable electricity auctions in 2017 (Clean Technica, 2017). For a chronological overview of key policy changes to Spain's renewable energy policy since 1997, see Table 2.

5.1.3. Political economy of policy field The growing TD is closely intertwined with a number of structural features of the Spanish electricity sector, which pose a particular challenge for the continued expansion of renewable energy in Spain. Firstly, following the 2008 financial crisis, electricity demand decreased considerably between 2008 and 2014 (Red Eléctrica de España, 2015). As a result, Spain has had significant overcapacity in its electricity system, i.e. a relatively high installed capacity in comparison to electricity demand7 (Deloitte, 2015). The growth of renewable energy generation (as well as gas-fired power plants) have added to this. In addition, the so-called merit-order effect,8 caused by renewables priced at zero marginal cost, has reduced the load factors (and thus profitability) of conventional power plants. Adding to this, the Spanish electricity system represents an energy island with limited possibilities to export power (del Río and Mir-Artigues, 2014). Therefore, nearly all the power produced has to be consumed within the country (del Río and Mir-Artigues, 2014; Red Eléctrica de España, 2015). These structural features, coupled with the inability to price electricity at cost, contributed to the increasing TD, which translated into mounting

5.1.2. The role of policy design In the case of Spain, design flaws and inflexibility contributed significantly to the decision to suspend financial incentives for renewables. Firstly, the design of the FiT led to a rapid and relatively uncontrolled growth of renewable energy, in particular in the solar photovoltaics sector. Spain's FiT lacked a mechanism for controlling the amount of new installations as well as the resulting costs. In addition, monitoring was poor due to a “significant lag time in the reporting of investments by regional governments” (del Río and Mir-Artigues, 2014). Instead, it included a poorly designed technology-specific termination mechanism. This led to a dramatic boom and bust cycle in Spain's solar PV market. When 85% of a technology-specific target was achieved, the government could announce a termination of the respective FiT with a

6 In a publication from the European Commission, tariff deficits are defined as “a shortfall of revenues in the electricity system, which emerges when the tariffs for the regulated components of the retail electricity price are set below the corresponding costs borne by the energy companies.” (European Commission, 2014: p.19). 7 In 2014, the maximum peak power demand was 38,666 MWh, while the installed capacity was 102,262 MW (Red Eléctrica de España, 2015, p. 11). Spain had experienced a continuous growth of electricity demand until 2007. Since the onset of the economic crisis, the electricity demand is below the level of 2008 and has not recovered yet. 8 The term “merit order effect” describes the lowering of power prices at the electricity exchange due to an increased supply of renewable energy, which can be priced at zero marginal cost.

4 While the annual cap existed previously for wind and solar thermal installations, it was now introduced for solar PV installations as well. Once the limit had been reached, electricity had to be sold at wholesale market prices. 5 The reasonable return is referenced to the average performance in the secondary market of Spanish government bonds over a 10-year period, plus a margin (for further detail, see Royal Decree Law 9, 2013). In practical terms, this meant a further reduction of approximately 2.7%, according to the European Photovoltaic Inudstry Association (EPIA, 2013: 20).

5

Energy Policy 133 (2019) 110881

K. Gürtler, et al.

Table 2 Key policy changes to Spain's renewable energy policy since 1997. Source: Authors. FiT Dismantling in Spain 1997 June 2007 2007/2008 September 2008 November/December 2010 January 2012 July 2013

Electricity Sector Law (Law 54/1997) introduces FiTs. Royal Decree (RD) 661/2007 revises FiT and introduces increased renumeration for PV installations between 100 kW and 10 MW. Installation rate of solar PV rises to extraordinary levels due to favorable conditions (“solar boom”) – parallel rise in subsidy costs; constant levels of wind energy deployment rates. RD 1578/2008 revises tariff levels and sets quarterly capacity quotas to regain control. RD 1565/2010 reduces duration of support for existing installations (to 25 years); RD Law 14/2010 implements a cap on operating hours for existing solar PV plants. RD Law 1/2012 introduces an indefinite moratorium on new installations. RD Law 9/2013 completely changes renumeration scheme and replaces FiT.

Fig. 2. Cumulative installed renewable energy capacities and related support costs in Spain.

economic pressure on the major utility companies. Two incumbent utilities – Endesa and Iberdrola – and three smaller competitors (Gas Natural, EdP, and Viesgo) dominate the Spanish electricity sector. Together with their umbrella association UNESA (Asociación Española de la Industria Eléctrica), these utilities aimed at maintaining their market dominance (Haas, 2017). Experts agree that lobbying by the incumbent energy utility companies played a major role in driving the policy dismantling process (ES01, ES02, ES03, ES04, ES06, ES08). Although some incumbent utilities had invested in the emerging sector, they perceived renewables as unwanted competition and as a threat to their business models. Experts pointed to the close personal connections between energy utility companies, the Ministry of Industry (Ministerio de Industria, Turismo y Comercio) and the main political parties in Spain. This was particularly true for the conservative party, which took the final step in ending renewables support in 2012 (ES02, ES06, ES08; Haas, 2017). On the opposite side, a coalition of “green” actors formed, including the environment ministry and a number of associations from the renewables sector. However, this coalition remained relatively fragmented and ineffective in influencing the government (ES01, ES06, ES08; Haas, 2017).

Spanish utilities who had securitized part of their debts on international financial markets. To alleviate the increasing debt burden, the government created a special entity to take over securitization from 20109 (del Río and Mir-Artigues, 2014). While alleviating pressure on the utilities, these financial obligations added to the concerns of financial analysts, assessing the credit worthiness of a struggling Spanish government during the Euro crisis (Robinson, 2013)10. This added further pressure to eliminate the additional costs created by the support for renewables. These economic shocks were accompanied by a change in government from the Socialists to the Conservatives after the 2011 elections. The Socialists had already introduced a number of measures, including retroactive reductions in support, to dampen the budget impact of the FiT. The conservative government that took power in 2011 delivered the final blow to the renewables support scheme by suspending financial incentives for new installations in 2012 (del Río and MirArtigues, 2014). According to interviewees (ES02, ES06), the subsidy cuts applied under the Socialist government were a direct result of the austerity imposed on the Spanish state by the financial crisis, while the

5.1.4. Macro-level factors These political dynamics within the energy sector took place in the context of the global economic and financial crisis. The resulting financing constraints further exacerbated the economic challenges facing

9 While the major energy utilities accumulated an increasing debt burden, the Spanish government guaranteed that it would secure the deficits resulting from controlled consumer prices (Del Rio, 2014: 9)). 10 See, for instance, Moody's (2013).

6

Energy Policy 133 (2019) 110881

K. Gürtler, et al.

termination of renewable support was heavily influenced by more fundamental political opposition among conservatives (Haas, 2017).

5.2.2. The role of policy design The Czech FiT was financed via a surcharge included in the consumers’ electricity bill (EEA, 2014). Distinct from other FiT designs, the Czech FiT includes no exceptions for electricity-intensive industries (Council of European Energy Regulators, 2015). In addition, the FiT did not differentiate between rooftop and ground-mounted photovoltaic systems, making the remuneration exceedingly favorable for the latter (CZ01, CZ03). Another important design feature was the mechanism for adjusting the level of remuneration, which allowed the FiT to be adjusted by a maximum of 5% per year (Czech Parliament, 2005). Introduced to provide stability and investment security, it deprived Czech policy makers of their ability to respond appropriately to the changing market environment. The majority of interviewees considered the policy's inflexibility to adapt to a changing market situation as amongst the most important reason for the dismantling (CZ01, CZ02, CZ03, CZ04, CZ06). The price drop of photovoltaic (PV) modules between 2008 and 2011 (Fraunhofer ISE, 2015) led to a rapid expansion of PV installations in the Czech Republic. This in turn increased the costs of the FiT considerably, significantly exceeding the targeted amount (EEA, 2014). The electricity price for household consumers rose steadily from 0,11€/kwh in 2007 to about 0,17€/kwh in 2013, after which it fell back to around 0,14€/kwh (Eurostat, 2018a). The cost of the FiT as a share of the total electricity price rose from around 2% in 2009 to 16% in 2011 (Pavelková; Živělová, 2016). Initially a payback period for investments on PV installations of around 15 years was envisaged by the FiT (CZ1). The time required to recover the cost of PV investments came down to nearly five years after the price drop in 2009. Non-household consumers, such as industry, had to pay about the same as households for electricity in 2007, spiking to 0,13€/kwh in 2011 and falling back to 0,09€/kwh in recent years (Eurostat, 2018b). Another critical factor concerning the policy design was that the FiT was inter alia employed to achieve the renewable energy target introduced by the European Directive 2009/28/EC (EUR-Lex, 2015). This mandated a share of 13% of renewables in total gross domestic energy consumption by 2020, including a 14% share in the power sector (IEA, 2016b). Due to the dynamic development of the sector in the Czech Republic and globally, both targets were almost met as early as 2013 (Czech Republic- Ministry of Industry and Trade, 2018). Interviewees viewed it as politically infeasible to prolong the support of renewable energy along with the related costs to the electricity consumer after the targets had been met (CZ04, CZ05, CZ06, CZ07).

5.1.5. Institutional constraints and opportunities The institutional landscape in Spain was also favorable to the process of policy dismantling and can be considered an enabling factor. In the bicameral parliament, the Cortes Generales, legislative power is divided between the Congress of Deputies and the Senate. The lower chamber, the Congress of Deputies, exerts greater legislative power, while the Senate can veto decisions. These veto decisions can then be outvoted by an absolute majority of the Congress of Deputies. Tsebelis (1995) argues that imperfect bicameral systems in which the lower chamber can ultimately overrule the objections of the upper chamber have to be considered as de facto unicameral systems. The veto power can only delay but not prevent policy decisions. Moreover, in Spain, the two chambers do not exhibit significant difference in their political configuration. As both legislative bodies are elected on the same election day, political power relations are usually similar in both chambers. In addition, in the last decades, the two main political parties – the conservative People's Party (ruling between 2011 and 2018) and the Spanish Socialist Worker's Party – have governed alternately without the formation of coalition governments.11 This further reduces the number of veto players – such as a partner in a coalition government that might prevent a dismantling decision (Tsebelis, 1995). Finally, decision-making in the case of the FiT dismantling took place mainly through Royal Decrees and Royal Decree Laws. Royal Decrees (RD) are issued by the central government and do not need approval through the parliament, so that the majority of dismantling decisions could be taken unilaterally by the government. 5.2. Czech Republic 5.2.1. Overview of renewable energy policy dismantling in the Czech Republic Over the past decades, the Czech Republic has made significant progress in increasing the share of renewable energy in the final consumption, especially in the electricity sector (CZ Energy Regulatory Office, 2017; EEA, 2014). The main support policy for renewable electricity was a FiT, established in 2005 under the Renewable Energy Support Act No. 180/2005 (Czech Parliament, 2005). The largest increase took place between 2010 and 2011, driven primarily by new photovoltaic installations (ERU, 2012). The dismantling of renewable support began in 2010 with the introduction of a levy of 26% on the income of operators of solar PV systems built after 2008 (CZ Constitutional Court, 2012). In 2012 the initial Renewable Energy Support Act (Czech Parliament, 2005) was replaced by Act No. 165/2012 on promoted energy sources (Czech Parliament, 2012). The new Act reduced the maximum capacity for new wind installations to 100 kW and for new solar installations to 30 kW, essentially eliminating support for commercial-scale systems. A further amendment (No. 310/2013 Coll) to the Act led to the de facto12 termination of support for renewables by the end of 2013 (EurObservER, 2018; RES Legal, 2017). In May 2014 an amendment introduced a review mechanism that could retrospectively adjust the investment conditions for each project after 10 years in operation and level the return on investment for all projects to about 3.5% per year (EurObservER, 2015). For a chronological overview of key policy changes to Czech renewable energy policy, see Table 3.

5.2.3. Political economy of policy field Next to renewable energy, overall electricity production in the Czech Republic is dominated by locally produced coal, with a 54% share in the energy mix in 2015, followed by nuclear with a share of approximately 30% (IEA, 2016a). The most dominant actor in the Czech electricity market is the ČEZ Group. It is majority state-owned utility and integrates more than 60% of total generation capacity. While ČEZ also undertook smaller investments in several types of renewable energies, its core business remains the operation of nuclear and coalfired power plants. Partly due to the rising share of renewables, ČEZ share of total generation capacity has been declining over the last decade (IEA, 2016a). One interviewee claimed that ČEZ, with its strong ties to the government, has exerted strong influence on the Czech government against the further development of renewable energy to protect its core business in nuclear and coal-fired power generation (CZ05). Interviewees pointed out that renewable energy interest groups also increased their lobbying activity but were less successful than the incumbent industry (CZ04; PV Magazin, 2013b). The lobbying efforts by ČEZ were supported by a business sector, which was dissatisfied with the fact that they had to pay for the profits of the renewable electricity installations via increased electricity prices (CZ01, CZ02, CZ04, CZ05). In particular, heavy industry, dependent on low electricity prices, strongly opposed the FiT scheme. Even though

11 The development towards a slightly more fragmented party system is a very recent phenomenon. 12 Renewable electricity generation from small plants up to 100 kW (30 kW in case of PV or 10 MW in case of hydro power and onshore wind) is still supported through a feed-in tariff. Onshore wind plays a rather minor role in Czech renewable power generation.

7

Energy Policy 133 (2019) 110881

K. Gürtler, et al.

Table 3 Key policy changes to the Czech renewable energy policy since 2005. Source: Authors. FiT Dismantling in the Czech Republic 2005 2010 May 2012 September 2013 May 2014

Renewable Energy Support Act No. 180/2005 introduces FiTs. Amendment No. 402/2010 Coll. Introduces levy of 26% on operators' incomes. Initial Law is replaced by Act No. 165/2012. Substantially slowing down renewable electricity development. Amendment No. 310/2013 Coll. de facto abolishes FiT (except small hydro). Amendment drafted for retrospectively adjusting investment conditions to yearly ROI of 3.5%, final status unclear.

from 2012 on about half of the costs for the FiT were borne by the state budget and exemptions were introduced for large electricity consumers, the opposition from influential industrial players might have been avoided, had the Czech government opted for exemptions from costs related to the FiT for energy intensive industry earlier.

renewable energy. This was exacerbated by a number of policy design features, leading to the boom in solar energy and its related impacts on the TD. The economic fallout from these developments was aggravated by the onset of the global financial crisis. Finally, the centralization of decision making in the form of Royal Decrees provided a favorable institutional environment for a number of dismantling decisions by the Socialist government. It then took a change of the national government for the full dismantling of the support scheme via a parliamentary decision, underlining the importance of government change in the Spanish case. Examining the Czech case, it becomes clear that it was largely the interplay between the political economy of the policy field and the specific policy design that brought about the dismantling of the Czech renewable energy support scheme. The following three factors stand out in the Czech process of policy dismantling: (1) an inflexible policy design; (2) the lack of industry exemptions, which caused extensive lobbying by powerful incumbents against the scheme; and (3) the early achievement of the Czech renewable energy target for the year 2020. All three factors are located at the intercept of the FiT's design and the political economy of the energy system. Indeed, the interaction between policy design and the political economy are critical for understanding the policy dismantling decisions in both countries. In both cases, a flawed design of the support scheme led to an uncontrolled boom in solar energy installations, leading to rapidly rising policy costs. The resulting political pressure aligned with the vested interests of incumbent utilities. In the absence of strong coalitions in support of renewable energy, the FiT schemes collapsed. This dynamic inter-relationship between policy and politics, known as “policy feedback” in the political science literature, represents an important line of inquiry for better understanding both policy dismantling and policy durability in the renewable energy sector. Pierson (1993) has stressed the importance of better understanding how policy lock-ins are created via positive feedback processes. The related literature has taken a relatively long-term approach to the study of policy feedback processes, considering how large-scale policy reforms, such as the creation of the US social security system, influence interest group formation or voter preferences (Patashnik, 2003; Patashnik and Zelizer, 2013). There is still a lack of research on how specific aspects of policy design are translated into positive or negative feedback effects. In the case of renewable energy policy, it is essential to also consider how feedback processes are mediated via the development of technology. Schmidt and Sewerin (2017) propose a number of avenues through which this may take place in the field of energy. In addition to traditional categories from political science, such as actors, institutions and power, they stress the importance of how physical conditions and the technology characteristics may influence policy feedback. The case studies presented in this article strongly underline the importance of considering the specific configuration of these variables within a local political economy setting. The further development of a comparative perspective on policy dismantling processes can offer important insights for deepening our understanding of how policy design and political economy factors interact within the field of renewable energy policy. Furthermore, as is widely acknowledged in the literature, it is crucial that the analysis is not limited to the design of single policy instruments. Rather a more integrated approach is needed, which takes

5.2.4. Macro-level factors As in Spain, the dismantling process in the Czech Republic was accompanied by the global financial crisis. In relative terms, however, the Czech Republic was modestly affected by the crises (Tvrdoň, 2010), leading to a shrinking of gross domestic product by 4.8% in 2009 (Eurostat, 2016). After 2009 the economy recovered, but negative growth rates returned in 2012 and 2013. The fact that end-consumers had to bear the costs of the support scheme was seen by some experts as an enabling factor for policy dismantling, which was exacerbated by the economic crisis (CZ02, CZ07). Others dismissed this as a minor factor (CZ01, CZ041, CZ06). Politically, the period was marked by several changes in government (Government of Czech Republic, 2017). The major FiT dismantling decisions were taken in May 2012 by the conservative coalition of Nečas’ and in September 2013 by the left-wing coalition of Rusnok. Both coalitions enjoyed majority support in the chamber of deputies but not in the Senate (Czech Senate, 2018). Nevertheless, the bills passed at first attempt in both the Chamber of Deputies and the Senate (PV Magazin, 2013a, b). Hence, it seems fairly clear that macro-political changes did not represent major drivers of policy dismantling. 5.2.5. Institutional constraints & opportunities The Czech Republic is a bicameral parliamentary democracy with a Chamber of Deputies and a Senate (IEA, 2017a). Since the upper and lower house are subject to differing election cycles, political majorities in the two chambers frequently diverge, as was the case at the time of dismantling. However, except for changes to constitutional and procedural laws, the Chamber of Deputies can outvote the Senate (Kysela, 2018), making it a de facto unicameral system, where the upper chamber can only delay, but not veto regular legislative proposals (Tsebelis, 1995). As indicated in the previous section, both houses passed the bills and amendments dismantling the FiT without delay. The overall institutional constraints were thus considerably higher compared to the Spanish case, which indicates that institutional opportunities did not substantially foster the dismantling of the FiT in the Czech Republic. 6. Discussion of key findings and implications for future research Table 4 provides a comparative overview of findings from the two case studies. All analytical categories yielded relevant findings in both cases, although the dynamics of the Spanish case provide a stronger validation of the framework. Both the design of the FiT scheme and the severe economic contraction interacted with the political economy in the energy sector to create an increasingly hostile environment for the further expansion of renewable energy. A combination of structural and institutional features of the Spanish electricity system – manifested in the form of a growing TD – led to mounting political pressure from incumbent electricity firms to dismantle Spain's support scheme for 8

Energy Policy 133 (2019) 110881

K. Gürtler, et al.

Table 4 Policy dismantling in Spain and the Czech Republic – comparative results. Source: Authors. Main Causes of Policy Dismantling in Spain

Main Causes of Policy Dismantling in the Czech Republic

Policy Design Overgenerous FiT rates for certain technologies (esp. utility-scale solar PV) Lack of annual caps, inflexible FiT rate adjustment Technology-specific FiT termination followed by transition phase leads to boom and bust in PV Uncertainties regarding the ultimate responsibility to cover for the increased “tariff deficit”

Lack of industry exemptions in the FiT Lack of discrimination between PV system sizes results in overly generous remuneration for large systems Inflexible policy design, unable to react to substantially changing market circumstances (especially PV modules’ price drop accompanied by accelerated deployment) Early achievement of Czech renewable energy 2020 target, making it politically troublesome to uphold the remuneration scheme

Political Economy of the Policy Field Overcapacities in the electricity sector with limited possibilities for electricity export Predatory competition Lack of phase-out scenarios for fossil-based generation Intensified lobbying efforts by incumbents Fragmented and ineffective renewable energy industry coalition

Lack of phase-out scenarios for fossil-based generation Influence of influential and still dominantly state-owned utility Successful lobbying against FiT by energy intensive industries and conventional energy companies with vested interests in fossil-based system

Macro-level Factors Shrinking power demand in wake of economic crisis

Mixed evidence found for payment problems of electricity consumers during the economic crisis as cause of FiT dismantling Mixed evidence found for changes in government as cause of FiT dismantling

Government securitization for “tariff deficit” put Spanish government under pressure in times of budget constraints/ economic crisis Dismantling process reinforced by election of conservative government Institutional Constraints and Opportunities Centralized decision-making on FiT through Royal Decrees and Royal Decree Laws in de facto unicameral parliament with single-party government facilitates dismantling process

Czech bicameral system does not provide an enabling environment for policy dismantling

reveal the important role that lobbying by powerful incumbents had for the process of policy dismantling. The exemption of certain industrial consumers represents one strategy for mitigating this potential source of opposition. At the same time, such exemptions may represent a source of political resistance in their own right, given the unfair distribution of costs across actors that this implies (see also Patashnik, 2003). A middle way might be exemptions that are phased-out over time. More generally, these examples suggest the need for an explicit strategy for managing the political economy of renewable energy deployment. As pointed out in Quitzow (2015a), the political dimension of policy making remains a blind spot in the broader literature on policies for the promotion of environmental innovation and technological change. The findings derived from this study offer first entry-points for building such politically sensitive policy strategies. Controlling the development and distribution of costs via the smart design of support policies represents a first strategic building block. A second element will have to address the vested interests of powerful incumbents and the politics of phasing-out incumbent industries (see also Kivimaa and Kern, 2016). In a first instance, this requires a sound understanding of the political economy of the policy field (see Stein et al., 2006). Depending on the particular configuration of the energy system, the interests of an incumbent power industry may be more or less affected by the expansion of renewable energy. For instance, the absence of viable export markets for surplus electricity in Spain resulted in strong economic interests from incumbent utilities to limit further renewable energy deployment. In this context, an active policy in support of increased inter-connections with France may have relieved some of this pressure. In most mature electricity markets, i.e. in markets without increasing power demand, similar patterns are likely. More generally, it might be stated that renewable energy

into account the broader policy strategy (Quitzow, 2015a) or policy mix (Edmondson et al., 2018). The key question is how the overall design of policy strategies, composed of policy goals, policy measures, policy processes and institution building, can maximize positive policy feedback and minimize negative policy feedbacks to lock-in desired reform trajectories. Finally, as the Czech case reveals, this research agenda should not be limited to the national level. As outlined above, the early achievement of an EU policy target had an important influence on the political environment to the disadvantage of the remuneration scheme. Meckling (2019) has argued that feedback mechanisms span geographic boundaries, raising important questions regarding the exact nature and role of such transnational feedbacks (see also Imbert et al., 2017; Quitzow, 2015b). 7. Implications for policy While further research on the topic is crucial for drawing robust policy lessons, a number of preliminary lessons can be drawn. As has been pointed out by Oberlander and Weaver (2015), fiscal feedbacks may have a strong influence on policy durability. For the design of durable policies in support of renewable energy, this implies that it is important to retain a degree of control over the costs of the support scheme, both to contain overall costs and to avoid the development of windfall profits (see also Altenburg and Engelmeier, 2013). It is important that costs are not perceived as damaging by stakeholders as well as politicians and the general public. Moreover, the perception that policy costs are providing a small stakeholder group with excessive benefits at the expense of others should be avoided. The shift to auctioning mechanisms witnessed over the past years already partially reflects this insight. The distribution of the costs across actors is another important point, although policy implications may be less clear. Both case studies 9

Energy Policy 133 (2019) 110881

K. Gürtler, et al.

deployment policies will typically require strategies that address the challenge of declining load factors in fossil-based power plants. This comes on top of the multiple techno-economic challenges of integrating increasing amounts of renewable energy in electricity systems. In the absence of strategic interventions to mitigate the economic damage to incumbent industries, resistance to renewable energy deployment is likely to increase in proportion to the economic devaluation of these capacities. Alternatively, renewable energy expansion could be coupled with the increasing electrification of end-use sectors, such as transport or heating. This would shift the initial CO2 reduction derived from the deployment of renewable energy to these end-use sectors, while offering a safeguard to renewable energy support schemes. This might represent a transition strategy to facilitate the development of stakeholder coalitions that are able to act as counterweights to political resistance from incumbent industries (see Lauber and Jacobsson, 2016). Of course, in the medium-term, explicit phase-out policies for fossilbased electricity generation will be inevitable for enabling the continued expansion of renewable energy and the reduction of greenhouse gas emissions. The importance of supranational targets is underlined in the case of the Czech Republic. Both the introduction and the dismantling of the Czech renewable energy support scheme were linked to policy targets negotiated at the EU level. Critically, the early achievement of the renewable energy target, as stipulated in the EU Renewable Energy Directive, disrupted the political momentum of Czech renewable energy support by legitimizing policy dismantling. This highlights the importance not only of the design of national policy but also supranational targets and related frameworks. The static European target setting could not be adapted to the dynamics of innovation and technological change, which led to the early achievement of the Czech target. A more adaptive framework, which rewards early progress and prevents stagnation in ambition, might have maintained needed political pressure on Czech decision-makers. The Paris Agreement's “ambition mechanism” offers an important point of reference for the design of such an adaptive approach to supranational target setting (Falkner, 2016).

Clean Technica, 2017. Spain Awards 4 Gigawatts Solar & 1 Gigawatt Wind in Renewables Auction. https://cleantechnica.com/2017/07/28/spain-awards-4-gw-solar-1-gwwind-renewables-auction/, Accessed date: 2 December 2018. Council of European Energy Regulators, 2015. CEER Status Review of Renewable and Energy Efficiency Support Schemes in Europe in 2012 and 2013. http://www.ceer. eu/portal/page/portal/EER_HOME/EER_PUBLICATIONS/CEER_PAPERS/Electricity/ Tab4/C14-SDE-44-03_Status%20Review%20on%20RES%20Support%20Schemes_ 15-Jan-2015.pdf, Accessed date: 12 February 2015. CZ Constitutional Court, 2012. 2012/05/15 - Pl. ÚS 17/11: Photovoltaic Power Plants. http://www.usoud.cz/en/decisions/20120515-pl-us-1711-photovoltaic-powerplants/, Accessed date: 11 June 2015. CZ Energy Regulatory Office, 2012. The Energy Regulatory Office Intends to Stop Subsidies to New Supported Energy Sources. http://www.eru.cz/documents/10540/ 477987/TiskovkaOZE_ENG_konec.pdf/4ed505d2-5b64-47b7-8735-5cca3ee30043, Accessed date: 4 November 2015. CZ Energy Regulatory Office, 2017. Yearly Report on the Operation of the Czech Electricity Grid. . https://www.eru.cz/documents/10540/462820/Annual_report_ electricity_2016.pdf/bfbc88b7-a814-498f-99bf-98b50148c805, Accessed date: 2 December 2018. Czech Parliament, 2005. Act 180/2005 on the Promotion of Electricity Production from Renewable Energy Sources and Amending Certain Acts. http://www.czrea.org/files/ pdf_en/zakony/RES_act_english.pdf, Accessed date: 16 January 2016. Czech Parliament, 2012. Act No. 165 of 31 January 2012 on Promoted Energy Sources and on Amendment to Some Laws. https://www.eru.cz/documents/10540/473304/ 165_2012_AJ.pdf/5e2856ee-dbb6-4e0f-891a-ef771f5511cf, Accessed date: 18 July 2019. Czech Republic- Ministry of Industry and Trade, 2018. Development of Renewable Energy Shares. http://translate.google.com/translate?hl=de&sl=auto&tl=en&u=https %3A%2F%2Fwww.mpo.cz%2Fcz%2Fenergetika%2Fstatistika%2Fobnovitelnezdroje-energie%2Fvyvoj-podilu-obnovitelne-energie–235054%2F&sandbox=1, Accessed date: 2 December 2018. Czech Senate, 2018. Volby Do Senátu, Poslanecké Sněmovny a Evropského Parlamentu. http://www.senat.cz/senat/volby/vdsapsp.php?ke_dni=17.10.2017&O=6, Accessed date: 18 July 2019. del Río, P., Mir-Artigues, P., 2014. A Cautionary Tale: Spain's Solar PV Investment Bubble. International Institute for Sustainable Development. https://www.iisd.org/gsi/sites/ default/files/rens_ct_spain.pdf, Accessed date: 18 July 2019. DeLeon, P., 1978. A theory of policy termination. In: May, J.V., Wildavsky, A.B. (Eds.), The Policy Cycle. Sage Beverley Hills, pp. 279–300. Deloitte, 2015. European Energy Market Reform. Country Profile: Spain. https://www2. deloitte.com/content/dam/Deloitte/global/Documents/Energy-and-Resources/gxer-market-reform-spain.pdf, Accessed date: 18 July 2019. Di Dio, V., Favuzza, S., La Cascia, D., Massaro, F., Zizzo, G., 2015. Critical assessment of support for the evolution of photovoltaics and feed-in tariff(s) in Italy. Sustain. Energy Technol. Assess. 9, 95–104. Edmondson, D.L., Kern, F., Rogge, K.S., 2018. The co-evolution of policy mixes and sociotechnical systems: towards a conceptual framework of policy mix feedback in sustainability transitions. Res. Policy 1–14. EEA, 2014. Country Profile – Czech Republic. https://www.eea.europa.eu/publications/ energy-support-measures/czech-republic-country-profile/view, Accessed date: 18 July 2019. EEA, 2016. Transforming the EU Power Sector: Avoiding a Carbon Lock-In. EEA Report No 22/2016. . https://www.eea.europa.eu/publications/transforming-the-eupower-sector, Accessed date: 18 July 2019. European Photovoltaic Industry Association (EPIA), 2013. Retrospective Measures at National Level and Their Impact on the Photovoltaic Sector. https://www. qualenergia.it/sites/default/files/articolo-doc/Restrospective_Measures_at_national_ level.pdf, Accessed date: 18 July 2019. EUR-Lex, 2015. Promotion of the Use of Energy from Renewable Sources. http://eur-lex. europa.eu/legal-content/EN/TXT/?uri=celex:32009L0028, Accessed date: 6 November 2015. EurObservER, 2015. Country Profile Czech Republic. https://www.eurobserv-er.org/ pdf/res-policy/EurObservER-RES-Policy-Report-Country-Profile-2015-12-CzechRepublic.pdf, Accessed date: 26 June 2015. EurObservER, 2018. Renewable Energy Policy Factsheet Czech Republic. https://www. eurobserv-er.org/pdf/renewable-energy-policy-factsheet-2018-czech_republic/, Accessed date: 18 July 2019. European Commission, 2014. Electricity tariff deficit: temporary or permanent problem in the EU? In: Linden, Johannesson, Asa, Kalantzis, Fotios, Maincent, Emmanuelle, Pienkowski, Jerzy (Eds.), Directorate General Economic and Monetary Affairs (DG ECFIN). European Commission, 2015. Country Report Spain 2015 - Including an In-Depth Review on the Prevention and Correction of Macroeconomic Imbalances. pp. 85. COM (2015). http://ec.europa.eu/europe2020/pdf/csr2015/cr2015_spain_en.pdf, Accessed date: 4 October 2015. Eurostat, 2016. Real GDP Growth Rate - Volume. http://ec.europa.eu/eurostat/tgm/ table.do?tab=table&plugin=1&language=en&pcode=tec00115, Accessed date: 7 November 2015. Eurostat, 2018a. Electricity Prices for Household Consumers. bi-annual data (from 2007 onwards). http://appsso.eurostat.ec.europa.eu/nui/submitViewTableAction.do, Accessed date: 25 May 2018. Eurostat, 2018b. Electricity Prices for Non-household Consumers. bi-annual data (from 2007 onwards). http://appsso.eurostat.ec.europa.eu/nui/submitViewTableAction. do, Accessed date: 25 May 2018. Falkner, R., 2016. The Paris Agreement and the new logic of international climate politics. Int. Aff. 92, 1107–1125.

Funding sources This work has been supported by funding from the German Federal Ministry of Education and Research, Grant Number 01US1701. Acknowledgements The authors would like to thank Tobias Haas (Freie Universität Berlin), Helge Jörgens (ISCTE-Instituto Universitário de Lisboa), Alejandro Nuñez-Jimenez (ETH Zürich) and Tobias Schmidt (ETH Zürich) for their feedback on previous versions of the article. References Altenburg, T., Engelmeier, T., 2013. Boosting solar investment with limited subsidies: rent management and policy learning in India. Energy Pol. 59, 866–874. Bardach, E., 1976. Policy termination as a political process. Policy Sci. 7, 123–131. Bauer, M.W., 2006. Politikbeendigung als policyanalytisches Konzept. Polit. Vierteljahresschr. 47, 147–168. Bauer, M.W., Knill, C., 2012. Understanding policy dismantling: an analytical framework. In: Bauer, M., Green-Pedersen, C., Héritier, A., Jordan, A. (Eds.), Dismantling Public Policy: Preferences, Strategies, and Effects, pp. 30–56. Bauer, M.W., Knill, C., 2014. A conceptual framework for the comparative analysis of policy change: measurement, explanation and strategies of policy dismantling. J. Comp. Policy Anal.: Res. Pract. 16, 28–44. Bauer, M.W., Green-Pedersen, C., Héritier, A., Jordan, A., 2012. Dismantling Public Policy: Preferences, Strategies, and Effects. Oxford University Press, Oxford. Behn, R.D., 1978. How to terminate a public policy: a dozen hints for the would-be terminator. Pol. Anal. 393–413. Bird & Bird, 2014. Impending Reduction of Spanish Feed-In Tariffs for Renewable Energy. http://www.twobirds.com/en/news/articles/2014/spain/impending-reduction-ofspanish-feed-in-tariffs-for-renewable-energy, Accessed date: 18 July 2019. Brewer, G.D., 1978. Termination: hard choices-harder questions. Public Adm. Rev. 338–344.

10

Energy Policy 133 (2019) 110881

K. Gürtler, et al.

self-undermining policy feedbacks. Forum 13, 37–62. Patashnik, E., 2003. After the public interest prevails: the political sustainability of policy reform. Governance: Int. J. Policy Administr. Inst. 16, 203–234. Patashnik, E.M., Zelizer, J.E., 2013. The struggle to remake politics: liberal reform and the limits of policy feedback in the contemporary american state. Perspect. Politics 11, 1071–1087. Pavelková, E.M., Živělová, I., 2016. Pricing electric power in the Czech republic and in selected countries. Acta Univ. Agric. Silvic. Mendelianae Brunensis 64, 1001–1011. Pew Report, 2014. Who's Winning the Clean Energy Race? Edition 2013. Pew Charitable Trusts, Philadelphia, Washington. Pierson, P., 1993. When effect becomes cause: policy feedback and political change. World Politics 45, 595–628. Pierson, P., 1994. Dismantling the Welfare State?: Reagan, Thatcher and the Politics of Retrenchment. Cambridge University Press, Cambridge. Pierson, P., 2001. Coping with permanent austerity: welfare state restructuring in affluent democracies. Rev. Fr. Sociol. 369–406. Quitzow, R., 2015a. Assessing policy strategies for the promotion of environmental technologies: a review of India's National Solar Mission. Res. Policy 44. Quitzow, R., 2015b. Dynamics of a policy-driven market: the co-evolution of technological innovation systems for solar photovoltaics in China and Germany. Environ. Innov. Soc. Trans. 17, 126–148. Quitzow, R., Röhrkasten, S., Jänicke, M., 2016. The German Energy Transition in International Perspective. Institute for Advanced Sustainability Studies, Potsdam. Red Eléctrica de España, 2015. The Spanish Electricity System 2014. http://www.ree.es/ en/publications/statistical-data-of-spanish-electrical-system/annual-report/spanishelectricity-system-2014, Accessed date: 18 July 2019. REN21, 2013. Renewables 2013 Global Status Report. Renewable Energy Policy Network for the 21st Century, REN21 Secretariat, Paris. REN21, 2014. Renewables 2014 Global Status Report. Renewable Energy Policy Network for the 21st Century, REN21 Secretariat, Paris. REN21, 2015. Renewables 2015 Global Status Report. Renewable Energy Policy Network for the 21st Century, REN21 Secretariat, Paris. REN21, 2018. Renewables 2018 Global Status Report. Renewable Energy Policy Network for the 21st Century, REN21 Secretariat, Paris. RES Legal, 2017. Feed-in Tariff (State-Purchasing Price). http://www.res-legal.eu/ search-by-country/czech-republic/single/s/res-e/t/promotion/aid/feed-in-tariff-acton-the-promotion-of-the-use-of-res/lastp/119/, Accessed date: 18 July 2019. Robinson, D., 2013. Comment: Pulling the Plug on Renewable Power in Spain. Oxford Institute for Energy Studies Publications. Royal Decree 1578, 2008. Real Decreto 1578/2008, de 26 de septiembre, de retribución de la actividad de producción de energía eléctrica mediante tecnología solar fotovoltaica para instalaciones posteriores a la fecha límite de mantenimiento de la retribución del Real Decreto 661/2007, de 25 de mayo, para dicha tecnología. Agencia Estatal Boletín Oficial del Estado (BOE). https://www.boe.es/buscar/doc. php?id=BOE-A-2008-15595. Royal Decree 661, 2007. Real Decreto 661/2007, de 25 de mayo, por el que se regula la actividad de producción de energía eléctrica en régimen especial. Agencia Estatal Boletín Oficial del Estado (BOE). https://www.boe.es/buscar/doc.php?id=BOE-A2007-10556. Royal Decree Law 1, 2012. Real Decreto-ley 1/2012, de 27 de enero, por el que se procede a la suspensión de los procedimientos de preasignación de retribución y a la supresión de los incentivos económicos para nuevas instalaciones de producción de energía eléctrica a partir de cogeneración, fuentes de energía renovables y residuos. Agencia Estatal Boletín Oficial del Estado (BOE). https://www.boe.es/boe/dias/2012/01/ 28/pdfs/BOE-A-2012-1310.pdf. Royal Decree Law 14, 2010. Real Decreto-ley 14/2010, de 23 de diciembre, por el que se establecen medidas urgentes para la corrección del déficit tarifario del sector eléctrico. Agencia Estatal Boletín Oficial del Estado (BOE). https://www.boe.es/boe/ dias/2010/12/24/pdfs/BOE-A-2010-19757.pdf. Royal Decree Law 9, 2013. Real Decreto-ley 9/2013, de 12 de julio, por el que se adoptan medidas urgentes para garantizar la estabilidad financiera del sistema eléctrico. Agencia Estatal Boletín Oficial del Estado (BOE). https://www.boe.es/diario_boe/ txt.php?id=BOE-A-2013-7705. Schmidt, T.S., Sewerin, S., 2017. Technology as a driver of climate and energy politics. Nat. Energy 2, 1–3. Seawright, J., Gerring, J., 2008. Case selection techniques in case study research: a menu of qualitative and quantitative options. Political Res. Q. 61, 294–308. Stein, E., Tommasi, M., Echebarría, K., Lora, E., Payne, M., 2006. The Politics of Policies. Inter-American Development Bank. Steinebach, Y., Knill, C., 2017. Still an entrepreneur? The changing role of the European Commission in EU environmental policy-making. J. Eur. Public Policy 24, 429–446. Tsebelis, G., 1995. Decision making in political systems: veto players in presidentialism, parliamentarism, multicameralism and multipartyism. Br. J. Polit. Sci. 25, 289–325. Tsebelis, G., 2002. Veto Players: How Political Institutions Work. Princeton University Press, Princeton. Tvrdoň, M., 2010. Consequences of the Global Economic Crisis on the Czech Economy, vol. 17 Scientific papers of the University of Pardubice. Series D, Faculty of Economics and Administration (2/2010).

Fraunhofer ISE, 2015. Current and Future Cost of Photovoltaics. https://www.ise. fraunhofer.de/content/dam/ise/de/documents/publications/studies/ AgoraEnergiewende_Current_and_Future_Cost_of_PV_Feb2015_web.pdf, Accessed date: 9 November 2015. Government of Czech Republic, 2017. Vláda České Republiky. https://www.vlada.cz/cz/ clenove-vlady/historie-minulych-vlad/prehled-vlad-cr/1993-2007-cr/, Accessed date: 14 October 2017. Haas, T., 2017. Die politische Ökonomie der Energiewende. Deutschland und Spanien im Kontext multipler Krisendynamiken in Europa. Springer, VS, Wiesbaden. IEA, 2016a. Energy Policies of IEA Countries - Czech Republic. https://www.iea.org/ publications/freepublications/publication/Energy_Policies_of_IEA_Countries_Czech_ Republic_2016_Review.pdf, Accessed date: 2 December 2018. IEA, 2016b. National Renewable Action Plan. https://www.iea.org/ policiesandmeasures/pams/czechrepublic/name-39468-en.php?s= dHlwZT1jYyZzdGF0dXM9T2s,&return=PG5hdiBpZD0iYnJlYWRjcnVtYiIPGEgaHJlZj0iLyISG9tZTwvYT4gJnJhcXVvOyA8YSBocmVmPSIvcG9saWNpZXNhbmRtZWFzdXJl cy8iPlBvbGljaWVzIGFuZCBNZWFzdXJlczwvYT4gJnJhcXVvOyA8YSBocmVmP SIvcG9saWNpZXNhbmRtZWFzdXJlcy9jbGltYXRlY2hhbmdlLyIQ2xpbWF0ZSBDaGFuZ2U8L2E-PC9uYXY-, Accessed date: 2 December 2018. IEA, 2016c. Renewable Energy Medium-Term Market Report 2016. . https://www.iea. org/Textbase/npsum/MTrenew2016sum.pdf, Accessed date: 18 July 2019. IEA, 2017a. Czech Republic - Energy System Overview. https://www.iea.org/media/ countries/CzechRepublic.pdf, Accessed date: 7 February 2018. IEA, 2017b. Renewables 2017. International Energy Agency (IEA), Paris. https://www. iea.org/renewables2017/, Accessed date: 18 July 2019. Imbert, E., Ladu, L., Morone, P., Quitzow, R., 2017. Comparing policy strategies for a transition to a bioeconomy in Europe: the case of Italy and Germany. Energy Res. Soc. Sci. 33, 70–81. Jacobs, D., 2012. Renewable Energy Policy Convergence in the EU: the Evolution of FeedIn Tariffs in Germany, Spain and France. Ashgate Publishing Company, Surrey, UK. Jordan, A., Green-Pedersen, C., Turnpenny, J., 2012. Policy Dismantling: an Introduction. Dismantling Public Policy: Preferences, Strategies, and Effects. Oxford University Press, Oxford, pp. 3–29. Jordan, A., Bauer, M.W., Green-Pedersen, C., 2013. Policy dismantling. J. Eur. Public Policy 20, 795–805. Kivimaa, P., Kern, F., 2016. Creative destruction or mere niche support? Innovation policy mixes for sustainability transitions. Res. Policy 45, 205–217. Knill, C., Tosun, J., Bauer, M.W., 2009. Neglected faces of europeanization: the differential impact of the EU on the dismantling and expansion of domestic policies. Public Adm. 87, 519–537. Korte, S., Jörgens, H., Bauer, M., Jordan, A., Green-Pederson, C., Héritier, A., 2012. Active dismantling under high institutional constraints? Explaining the bush administrations attempts to weaken US air pollution control policy. In: Bauer, M.W., Jordan, A., Green-Pedersen, C., Héritier, A. (Eds.), Dismantling Public Policy: Preferences, Strategies, and Effects. Oxford University Press, Oxford, pp. 81–102. Kysela, J., 2018. Bicameralism in the Czech Republic: Reasons, Functions, Perspectives. http://www.senat.cz/informace/z_historie/studie/sbornik/bikameralismus-eng. php?ke_dni=4.8.2016&O=10, Accessed date: 21 October 2018. Lauber, V., Jacobsson, S., 2016. The politics and economics of constructing, contesting and restricting socio-political space for renewables – the German Renewable Energy Act. Environ. Innov. Soc. Trans. 18, 147–163. Law 15, 2012. Ley 15/2012, de 27 de diciembre, de medidas fiscales para la sostenibilidad energética. Agencia Estatal Boletín Oficial del Estado (BOE). https://www.boe. es/diario_boe/txt.php?id=BOE-A-2012-15649, Accessed date: 3 October 2015. Law 54, 1997. Ley 54/1997, de 27 de noviembre, del Sector Eléctrico. Agencia Estatal Boletín Oficial del Estado (BOE). https://www.boe.es/diario_boe/txt.php?id=BOEA-1997-25340, Accessed date: 3 October 2015. Leech, B.L., 2002. Asking questions: techniques for semistructured interviews. Pol. Sci. Pol. 35, 665–668. Lewis, J., Wiser, R., 2007. Fostering a renewable energy technology industry: an international comparison of wind industry policy support mechanisms. Energy Policy 35, 1844–1857. Liebe, U., Dobers, G., 2019. Decomposing public support for energy policy: what drives acceptance of and intentions to protest against renewable energy expansion in Germany? Energy Res. Soc. Sci. 47, 247–260. Magazin, P.V., 2013a. Czech Lower House Votes through End to FiT Payments. https:// www.pv-magazine.com/2013/09/03/czech-lower-house-votes-through-end-to-fitpayments_100012616/, Accessed date: 25 May 2018. Magazin, P.V., 2013b. Czech Republic Ends FIT Program, Extends Solar Tax. https:// www.pv-magazine.com/2013/09/16/czech-republic-ends-fit-program-extends-solartax_100012748/, Accessed date: 25 May 2018. Meckling, J., 2019. Governing renewables: policy feedback in a global energy transition. Environ. Plan. C: Pol. Space 37, 317–338. Moody's, 2013. Rating Action: Moody's Downgrades Spanish Electricity Tariff Deficit Transactions. https://www.moodys.com/research/Moodys-downgrades-Spanishelectricity-tariff-deficit-transactions–PR_271163, Accessed date: 2 December 2018. Oberlander, J., Weaver, R.K., 2015. Unraveling from within? The affordable care act and

11