De Facto Rule-Making Below the Level of Implementing Acts: Double-Delegated Rule-Making in European Union Electricity Market Regulation

Abstract

Within the area of electricity market regulation, a practice has emerged in which the chain of delegation has gone beyond the European Commission, resulting in double delegation. During 2015–2017, the European Commission adopted implementing regulations requiring detailed European terms, conditions and methodologies (TCMs) for electricity markets and system operation to be jointly adopted by national energy regulators. Should the latter fail to agree within a predefined time limit, rule-making would move to the Agency for the Cooperation of Energy Regulators. This rule-making procedure entails that, depending on the dynamic within the procedure, different actors would adopt the TCMs. This article examines how double-delegated rule-making unfolds in a novel and emerging practice, evolving beneath implementing acts. By analysing the factors behind whether TCMs are adopted jointly by national agencies or not, the study investigates whether this form of delegated rule-making in a network setting delivers decisions or whether rule-making by a European Union agency is needed.

I. Introduction

Legislative delegation of rule-making power to the executive branch occurs when the legislator decides on basic elements and leaves filling out the details to the executive branch. In the European Union (EU), the European Parliament and Council can delegate powers for the European Commission to supplement EU legislation (delegated acts) or set conditions to ensure uniform implementation thereof (implementing acts). ¹ Previous research on delegated rule-making ² in the EU has examined the origin and practice of Commission rule-making ³ while also identifying how other actors, such as EU agencies, can be formally required to provide input. ⁴ However, recent developments in EU electricity market regulation have introduced a new form of delegated rule-making at the EU level. During the years 2015–2017, the Commission adopted eight regulations as implementing acts below the EU Electricity Regulation. ⁵ Four of these included a new delegated rule-making procedure at the EU level (capacity allocation and congestion management (CACM), forward capacity allocation (FCA), electricity balancing (EB) and system operation (SO) guidelines). ⁶ According to this procedure, national regulators from all Member States were jointly to adopt terms, conditions and methodologies (TCMs) based on proposals drafted by electricity transmission network operators (TSOs). ⁷ Included in the procedure was an escalation clause. If the national regulators could not achieve joint approval within a given schedule, they would have to refer the proposal to the EU Agency for the Cooperation of Energy Regulators (ACER) for final approval. This meant that the question of who adopts a TCM would depend on the dynamics of the delegated rule-making process itself.

Delegated rule-making on TCMs is striking in several ways. First, European TCMs are formally adopted as binding individual decisions but are still directed towards the full community of TSOs or electricity exchanges in every EU Member State, with significant implications for a broader set of groups. De facto, then, TCMs constitute general rules. ⁸ Second, the procedure for TCM development and adoption was not foreseen when the EU Electricity Regulation was adopted in 2009. ⁹ It was later introduced by the abovementioned implementing acts, although these referred to the higher-order Electricity Regulation. Hence, in this area, there is not just delegated rule-making, but “double” delegation of rule-making in the sense that the implementing acts themselves introduced additional Union-level rule-making. Moreover, the adoption of rules was not delegated to the Commission, but to all national regulators or, conditionally, to ACER.

TCM adoption based on consensus amongst all national regulators in Europe may be conceptualised as rule-making within a regulatory network. In practice, the national energy regulators processed the TCMs within a loose network structure called the European Regulators’ Forum (ERF). Given the escalation clause, the success or failure of this regulatory network in reaching a decision would be a precondition for whether the EU agency would be authorised to engage in European rule-making. Previous research has examined how a European regulatory network may evolve into an EU agency, or how the presence of such networks could prevent the establishment of an EU agency in the same policy area. ¹⁰ However, such networks and EU agencies can also coexist in the same policy area. ¹¹ Little is known about regulatory networks and EU agencies within Union-level rule-making as they unfold in practice, including when they coexist and, potentially, are both involved in the same rule-making process. This article therefore asks: under what conditions did the European regulatory network adopt decisions on TCMs or not? It provides new knowledge about the impact of joint involvement of a European regulatory network and an EU agency on the rule-making process. By examining European regulatory networks’ ability to adopt decisions, the study sheds light on whether and when hierarchical rule-making (by an EU agency) rather than consensual rule-making might be needed to reach decisions. ¹² Finally, the article shows the practical operation of a potentially unique case of double-delegated rule-making in the EU – that is, a delegated rule-making process provisioned in rules that themselves have been adopted via delegated rule-making. ¹³

To analyse whether a European network of national regulators was (not) able to decide on a TCM, we apply a bureaucratic politics perspective. Prior research has shown the relevance of this perspective in explaining (non-)cooperation amongst national regulators on the European scene. ¹⁴ Moreover, a policy-network perspective is applied to analyse how rule-making by the European regulatory network is shaped by a dependence on expert knowledge of TSOs entitled to draft proposals for decisions. ¹⁵ Empirically, we study the process and adoption of six European TCMs selected to achieve case representation from all four Commission regulations and for variation in the route of adoption (some by all national regulators and some by ACER). Additionally, the cases represent substantially important and/or contentious issues in which the involved actors have a stake in rule-making.

Our research shows that rule-making by the European regulatory network was strongly influenced by whether national regulators considered a given TCM as interfering with their national core mission. Rule-making by the European regulatory network was moreover shaped by the extent to which TSOs, when drafting proposals, had been responsive to the input from national regulators. For these reasons, the European regulatory network was frequently unable to agree on proposals for European TCMs, which were, therefore, escalated to ACER for adoption. This was an important rationale for why later reforms ¹⁶ made ACER the default rule-maker for European TCMs. ¹⁷ Nevertheless, the regulatory network managed to adopt some TCMs, indicating that decentralised rule-making (with centralised decisions by an EU agency as a fallback solution) sometimes offered an alternative to full centralisation by default. Finally, we discuss how procedural design entails trade-offs amongst several considerations. Delegating rule-making based on subsidiarity, such as to the lowest level possible, could improve the legitimacy of the adopted output. However, this might reduce the efficiency of rule-making, including in terms of longer processes or less harmonised output.

II. Emergence of delegated rule-making in the electricity sector

In the 1990s, the EU adopted several directives to achieve free competition and trade in electricity and gas within and across national borders in the energy sector. However, legal provisions to ensure non-discriminatory access to transmission networks within and across borders performed poorly in dismantling barriers to cross-border trade. This was due to differences in national modes of regulating and operating market platforms and networks. In the early 2000s, the Commission therefore established and supported various initiatives for voluntary cooperation at the European level to harmonise the more technical aspects underpinning market transactions and network operation. ¹⁸ Considering the barriers to cross-border trade to be persistent, the Commission proposed to formally set up EU-level cooperation bodies of national regulators and TSOs, as well as procedures for harmonising regulatory rules and operating practices. This was included in the third energy market package adopted by the EU institutions in 2009. ¹⁹

The third package established the European Network of Transmission System Operators for Electricity (ENTSO-E) for EU-level cooperation amongst TSOs, as well as ACER for cooperation amongst national energy regulators. These would support the harmonisation of technical market and SO rules through delegated rule-making on implementing acts called “network codes”. These would be adopted by the Commission, assisted by a committee of Member State representatives. ²⁰ In practice, however, some of the rules adopted through the network code process were not complete. In four thematic areas, important details and clarifications of crucial elements were lacking. Four Commission regulations on “guidelines” (rather than network codes) therefore provided for further rule-making to fill in the missing parts that were essential to increase the volume and efficiency of trade via cross-border interconnectors. Although labelled as guidelines, these Commission regulations are binding rules and not soft legal instruments. Hence, rules that themselves were products of delegated rule-making provided for an additional process of delegated rule-making, implying a new form of double-delegated rule-making (see Figures 1 and 2).

Figure 1. The relationship between European Union (EU) legislation, delegated rule-making and double-delegated rule-making. ACER = Agency for the Cooperation of Energy Regulators; ENTSO-E = European Network of Transmission System Operators for Electricity; ERF = Energy Regulators’ Forum; EU = European Union; NRA = national regulatory authority; TSO = transmission network operator.

Figure 2. The procedures for delegated rule-making on network codes and for double-delegated rule-making on terms, conditions and methodologies (TCMs).

Commission Regulation (EU) 2015/1222 of 24 July 2015 established a guideline on CACM, which aimed at harmonising existing approaches to allocating cross-border capacity and managing congestion, as diverging practices could hinder trade. Commission Regulation (EU) 2017/2195 of 23 November 2017 established a guideline on EB, which aimed at setting up European platforms for and harmonising TSO practices directed at ensuring necessary instant matching of the power taken out of and fed into the electricity network. The rationale for this was to improve competition by expanding access to balancing resources, including across borders. Commission Regulation (EU) 2016/1719 of 26 September 2016 established a guideline on FCA. It was to improve market participants’ access to hedging when engaging in cross-border trade by promoting liquid and competitive forward markets in a coordinated way across Europe. Commission Regulation (EU) 2017/1485 established a guideline on electricity transmission SO. It concerned harmonisation of SO (of how TSOs keep the electricity network secure) to prevent TSO practices and their security standards from randomly or discriminatorily curtailing the allocation of network capacity for trade. Each of the four Commission regulations provided more detailed lists of issues to be regulated via TCMs.

The TCM procedure mandated that the TSOs, with the assistance of ENTSO-E, cooperate closely to develop the common European TCMs and submit them for approval to the national regulators within set deadlines. Should no consensus be reached among TSOs, they would decide by a qualified majority. Each regulatory authority would need to agree to approve a TCM proposal. Where the regulatory authorities could not reach an agreement, or upon their joint request, ACER would adopt the TCM instead. ²¹ In the latter case, ACER adoption required a favourable opinion from two-thirds of all national regulators within ACER’s Board of Regulators.

III. Studying delegated rule-making by a European regulatory network

To study how and why a network of European regulators might be able to adopt decisions on TCMs, we build on theoretical insights regarding regulatory behaviour and regulators’ reliance on cooperation with others for the resources needed to regulate.

A bureaucratic politics approach conceptualises regulators as fundamentally driven by concerns for their own autonomy: “the ‘wriggle room’ to interpret and carry out their mandate as they see fit”. ²² Regulators seek autonomy by performing their core mission well, which may be equated with their core regulatory tasks, ²³ as good performance protects a regulator’s turf. ²⁴ National regulators conceive of their own tasks as helping fulfil public interest objectives, as adopted by policymakers. ²⁵ The core mission of a national regulator is usually situated at the national level, even as national regulators have been drawn into the European scene and engage in EU-level processes. ²⁶ Although autonomy-seeking regulators trying to protect their turf are often associated with non-cooperation, ²⁷ regulators seek cooperation if this helps them accomplish their work. ²⁸

National regulatory arrangements across Europe are embedded in distinct regulatory traditions emerging from national industrial and administrative structures. ²⁹ Such differences can give rise to different understandings of how to address a policy problem or what might constitute an efficient solution. EU-level decisions might be seen by national regulators as aligned with or colliding with their national mission. Hence, a bureaucratic politics perspective expects TCM adoption by the European regulatory network if national regulators cooperate, which they will do if this supports their core mission at the national level. In contrast, if a TCM challenges the existing national regulatory arrangements or collides with the core mission, a national regulator will bargain to defend their national status quo. Given the institutional setup, with consensus required by the European regulatory network for TCM adoption, resistance from a single national regulator is sufficient for adoption to fail. We interpreted statements on points of disagreement in non-papers or letters of referral – as jointly issued by the national regulators when a TCM moved to ACER for rule-making – as evidence of conflict amongst national regulators.

Cooperation may also be affected by national regulators’ expectations about intervention from a higher-level authority (ie ACER). A “shadow of hierarchy” exists in which the threat of intervention from a higher-level authority prompts cooperation to find a common solution amongst organisations to avoid rules being imposed upon them from above. ³⁰ This rests upon an assumption of conflict between lower- and higher-level organisations, with the former seeking to overcome their differences to avoid rule-making by the latter. However, national regulators would also cooperate with the EU-level agency if they considered this helpful for their national tasks, typically for complex issues. ³¹ Hence, cooperation or conflict between national regulators and EU agencies depends on whether national regulators consider the EU agency in their policy area to be an asset or a threat to their core mission and turf. ³² We interpret statements from national regulators in non-papers that seek to prevent escalation to ACER, limit its discretion or stress subsidiarity and the importance of national rule-making on a particular issue as evidence of conflict amongst national regulators and ACER.

A policy-network perspective conceptualises rule-making as shaped by a wider network of public and private actors. ³³ Here, converging or diverging perceptions and preferences affect actors’ interaction towards conflict or cooperation. ³⁴ In the policy network on electricity, we have national regulators as public actors and TSOs as public monopolies variously owned by governments or in private ownership, allowing for a rate of return under limits set by monopoly regulation. ³⁵ For the sake of simplicity, and as the TSOs are not part of the public administration, we refer to them as “private actors”. The main interests of TSOs are linked to their tasks of investing in and securely operating the electricity network to ensure a reliable supply of electricity whilst securing payment of dividends to its owners. ³⁶ National regulators may have wider interests linked to tasks of ensuring a cost-efficient and secure supply of electricity, focused on reducing TSO costs to keep tariffs low and stable for consumers (ie catering to wider interests in the electricity system). Although the interests of TSOs and national regulators often meet, entailing cooperative relations, they are not necessarily always aligned: TSOs tend towards a stronger emphasis on secure operation of the network and seeking leeway for doing so, whereas national regulators tend towards also considering cost-efficiency in terms of how the network is operated, reflecting an emphasis on keeping tariffs low and stable. ³⁷ If these two groups have diverging interests, this would entail conflict.

Public and private actors hold different types of power that make them mutually dependent on each other in rule-making. ³⁸ Public actors, such as national regulators, are formally authorised to adopt binding rules and are also seen as legitimate actors for this role. ³⁹ However, regulators are not as informed about the activities within a regulated industry to the same extent as the regulated industry itself, which gives rise to information asymmetry between regulators and private actors in the industry. ⁴⁰ This provides private actors with informal power and often also access to rule-making, as granted by public actors that may need such information to regulate. The role of private actors in rule-making could be informal or formalised through public consultation or even co-drafting. ⁴¹ Such access allows private actors to seek to influence regulation in a specific policy field to their advantage. ⁴²

The policy-network perspective expects the European regulatory network to adopt a TCM when there is cooperation amongst TSOs and national regulators because the rule-making of national regulators depends on the technical expertise of TSOs. Cooperation is present and sufficient when the proposal by TSOs meets the national regulators’ need for expertise, which will enable national regulators to agree on a common decision. There is conflict when TSOs do not provide national regulators with a sufficiently complete or informed proposal. This will derail the rule-making process, reflecting that national regulators themselves do not have the necessary expertise to tackle issues not resolved by the TSOs. We interpreted requests for amendments by the national regulators, in which they send proposals back to the TSOs, or statements from national regulators regarding deficiencies in a proposal, such as a lack of information or that some pieces are missing, as evidence of conflict amongst TSOs and national regulators.

The two perspectives have partly conflicting and partly complementary expectations. Both anticipate European regulatory network adoption if cooperation exists between TSOs and national regulators and amongst national regulators. However, they prioritise different rule-making aspects. The bureaucratic politics perspective expects adoption regardless of TSO–regulator relations, whereas the policy-network perspective does not foresee major conflicts amongst national regulators if TSOs provide them with a satisfactory proposal.

A case study approach was selected because it is well suited for examining the mechanisms behind an outcome. If rule-making moved to ACER, this was interpreted as non-adoption by the regulatory network, and if they adopted a request for amendments (effectively sending a TCM back to the TSOs for revision), adoption was assessed based on whether the resubmitted proposal was adopted or not. As part of a larger project, ⁴³ descriptive statistics regarding all European TCMs adopted during 2015–2021 were assembled from the legal text of each guideline and webpages of ENTSO-E and ACER. ⁴⁴ This showed which TCMs had been adopted by the European regulatory network and which by ACER, along with process information, such as (1) deadlines for various steps, (2) when proposals were submitted and decisions adopted or escalated to ACER and (3) whether requests for amendments or referrals were adopted. Stratifying the population of European TCMs by guideline, one to two substantially important TCMs were selected for each, in which the involved actors would have a stake in the outcome. While escalation to ACER might be more probable for such TCMs, it would also make the national regulators more wary of such escalation to ACER, making these cases suitable for identifying the different causal mechanisms at work in a technically complex area. Six TCMs were purposely selected to ensure relevant variation on explanatory factors and in the outcome of interest, ⁴⁵ with some adopted by the European regulatory network and others not. Data about the process leading up to adoption were examined via documents and informal background interviews.

IV. Empirical cases of delegated rule-making in practice

By the end of 2021, thirty-four European TCMs had been adopted under the four Commission regulations (see Table 1). ACER had frequently adopted European TCMs following request from or failure of agreement amongst national regulators. However, not all European TCMs were escalated to ACER. In this section, case studies of select European TCMs are presented.

Table 1. Overview of European terms, conditions and methodologies adopted or pending as of 2021.

ACER = Agency for the Cooperation of Energy Regulators; aFRR = automatic frequency restoration reserves; CACM = capacity allocation and congestion management; CCR = capacity calculation region; CSAM = coordinating operation security analysis; EB = electricity balancing; FCA = forward capacity allocation; KORRR = key organisational requirements, roles and responsibilities; mFRR = manual frequency restoration reserves; SAP = single allocation platform; SO = system operation.

1. Capacity calculation regions

For this TCM, a set of regions would be defined as cooperating more closely on capacity calculation. How to define regions was not very contested, but controversy emerged as to whether regions should be based on existing bidding zone borders or also encompass specifying new ones. A key issue was the proposed split of the German–Austrian bidding zone (along the border between these two Member States). Following an early request from the Polish regulator, ACER gave an opinion arguing that such a split was required under the 2009 Electricity Regulation and annexed guidelines. ⁴⁶ When the TSOs later developed the TCM proposal and submitted it to the regulatory network on 17 November 2015, ⁴⁷ they took this on board despite opposition from the Austrian TSO (which was fended off by the majority voting rule).

The deadline for the national regulators to decide was 13 May 2016, but as consensus was not attained (due to opposition from the Austrian national regulator E-Control), the case was referred to ACER a few days later (17 May 2016). ⁴⁸ E-Control submitted a separate request for amendments to the TSOs on 13 May 2016, within the legal deadline. ACER – after receiving confirmation from the Commission that it was competent to decide despite E-Control’s amendment request – adopted this TCM on 17 November 2016. ⁴⁹ Before adopting this TCM, ACER carried out a public consultation in which the Austrian regulator and TSO objected to the inclusion of the German–Austrian bidding zone split. ⁵⁰ Nonetheless, ACER included such a split in its final decision. ⁵¹ The decision was later appealed. ⁵²

Voluntary regional cooperation that had been launched by national regulators and the Commission existed prior to this TCM. With the adopted TCM, regions were formally defined that would engage in mandatory cooperation on capacity calculation (ie a common methodology to calculate capacity). Improving cooperation on capacity calculation could help optimise the use of the existing electricity network. Removing barriers could facilitate more cross-border trade and enhance electricity market integration.

2. Single allocation platform

A single allocation platform (SAP) was to be established as a market platform for the financial settlement of allocated long-term transmission rights to make such allocations more transparent and non-discriminatory. Prior bilateral cooperation structures for allocating cross-border capacity already existed in the European market, initiated by the TSOs and supported by the national regulators. ⁵³ To meet the requirements of cooperation in the 2009 Electricity Regulation, the TSOs had established two joint auction offices tasked with operating the “allocation of available interconnection capacities, through auctions, on behalf of their respective TSOs”. ⁵⁴ The TSOs decided to merge the two offices into the Joint Allocation Office (JAO) in anticipation of the FCA guideline and the future SAP. JAO was established as a joint venture owned solely by TSOs, composed mostly of EU TSOs issuing long-term transmission rights (eighteen EU TSOs out of twenty-eight). Harmonised allocation rules were also developed in anticipation of this. Hence, JAO had already allocated forward capacities in line with harmonised alloZcation rules since 1 January 2015. JAO was the only entity doing this and the only one already in a position to allocate long-term cross-zonal capacities on EU borders. ⁵⁵

The deadline for the first proposal was 18 April 2017. Missing this deadline, the TSOs submitted their first and only proposal on 15 June 2017. Here, the TSOs examined four options based on different types of ownership. They went for a solution in which the SAP would be operated by an entity owned by themselves, opting for JAO as the only feasible solution. The TSOs justified their proposal with the fact that already-existing entities were available to perform such tasks efficiently without incurring the cost of establishing new entities. By using JAO as a SAP operator, its information technology (IT) tools would be compliant with already-existing harmonised allocation rules. The IT tools of JAO and procedures with TSOs had already been tested and considered efficient. The TSOs also found that using JAO as the SAP operator would increase transparency. ⁵⁶

The TSO proposal also mentioned that certain concerns of the national regulators had been considered. Although the national regulators had suggested that the SAP operator be obliged to provide information directly to them, the TSOs found this unnecessary, as the TSOs and ENTSO-E were already obliged to do so. The national regulators had also asked the TSOs to include a time limit for confirming new SAP parties and called for further clarification, implying that there had been close cooperation between the TSOs and the national regulators in the drafting process.

Within the ERF, the national regulators agreed to approve of this TCM on 18 September 2017. ⁵⁷ They found that the TSO proposal fulfilled their objectives and complied with the functional and governance requirements of the FCA guideline. ⁵⁸ The national regulators describe having “assessed, consulted and closely cooperated and coordinated to reach agreement”. ⁵⁹ There is no mention of diverging interests amongst national regulators. It is important to mention, however, that this TCM would not apply to Finland and Sweden because these two countries chose not to issue long-term transmission rights (Finland later modified its position).

3. Key organisational requirements, roles and responsibilities in data exchange

This TCM would require the TSOs to develop harmonised rules for data exchange to improve national grid models and build common grid models at the regional level, to be used for security analysis and security operations at this level. The TSOs’ first proposal for consultations reflected TSO preferences for greater flexibility to unilaterally decide on the formats, scope and frequency of data exchange with network-using actors (notably distribution system operators (DSOs) and significant grid users (SGUs); ie power producers and major consumers, respectively), as well as the organisation of data exchange processes. This also revealed common TSO preferences for minimising their costs associated with the building and maintenance of the necessary data communication system and for these costs to be shared by other actors. ⁶⁰

The proposal was heavily contested by DSOs and SGUs, who wanted a stronger role in deciding on data exchange rules. ⁶¹ However, the TSOs failed to account for these objections when they submitted their first formal proposal on time. The national regulators agreed to request significant amendments, echoing the requests from DSOs and SGUs but also reflecting that informally given input via early bilateral talks with the TSOs had not been heeded. ⁶² The national regulators saw the proposal as going beyond the mandate given by the SO guideline and requested amendments to ensure legal consistency with the subsidiarity and proportionality principles of this regulation. Specifically, they emphasised that the SO guideline had no legal basis for requiring the TSOs to determine the same rules for all Member States. The regulators requested inserting into the text that specific elements be decided at the national level (formats, scope and communication systems for data exchange, quality checks). This implied an upgraded role for DSOs in decisions on data exchange and limitations in responsibilities to fund data communication systems by DSOs, operators of so-called closed distribution systems and SGUs. Moreover, clearer wording was requested to limit the data exchange of confidential information between the TSOs and to include only data strictly needed for the joint operational security analysis. Hence, the text should be aligned with and should not lead to additional data exchange than already decided in the SO guideline. The key organisational requirements, roles and responsibilities (KORRR) should not be used to provide the data needed by the TSOs to perform tasks under other guideline regulations, such as the CACM guideline, where separate data exchange provisions had been established. The national regulators also wanted more reciprocity between TSOs and DSOs as well as SGUs in specific data exchange obligations and to limit data exchange obligations to the most significant SGUs. ⁶³

Following up on these requests by the national regulators, the TSOs came up with an amended KORRR proposal. ⁶⁴ The national regulators demanded further amendments, including more clarifications of provisions to be decided at the national level. After yet another round of amendments, the national regulators concluded that the text had been brought in line with what they had initially requested in bilateral consultations – provision of sufficient room for national specifics concerning data exchange, as provided by the SO guideline. Based on these second-round amendments, the national regulators adopted the KORRR TCM on time in 2018, thus avoiding escalation to ACER. ⁶⁵

4. Coordinating operation security analysis

This TCM was aimed at standardising security operation analyses at least per synchronous area to improve the security of the operation of interconnected networks and to ensure optimal use thereof to benefit free cross-border trade. The TCM was to include some methods for assessing the influence of DSO elements and SGUs located outside of the control area of a TSO to identify its extended observability area, principles for common risk assessment, principles for assessing and dealing with uncertainties of generation and load, requirements on coordination and information exchange between regional security coordinators, what role to be given to ENTSO-E in governing and monitoring tools, data quality rules and the coordination of operational security analysis and operations. ⁶⁶

The TSOs consulted on a draft version of the proposal, but despite many objections by stakeholders, only minor adjustments were made before the proposal was submitted to the national regulators in July 2018. ⁶⁷ Stakeholders representing DSOs and generators regarded the TSOs as introducing excessive requirements for new data provision – beyond those required by the already adopted KORRR – that would impose additional costs. ⁶⁸

This European TCM was developed together with another TCM due to interlinkages. ⁶⁹ Although the TSOs submitted a formal proposal, the national regulators from all Member States were not satisfied; they stated that the proposal was not compliant with the requirements outlined in the guideline itself and that the TSOs had not taken their informal input in early consultations into account. The national regulators therefore decided (ahead of the deadline) to refer this European TCM to ACER (together with the linked TCM) to avoid delays. A delay would affect not only this TCM process but also the linked TCM, as well as other TCMs that depended on the adoption of this coordinating operation security analysis (CSAM) TCM. ⁷⁰ The national regulators further validated their referral by suggesting that the TSOs would not respond to a formal request for an amendment anyway – such a step would not make a difference and would only delay the process. ⁷¹ ACER adopted a decision in 2019 but asked TSOs for further amendments. ⁷² ACER’s decision navigated the diverging positions of TSOs, national regulators and stakeholders: on the one hand, it maintained some discretion and added time to the TSOs to develop amendments. On the other hand, it supported stakeholders and national regulators by adding details and specifications that would reduce the discretion of TSOs in making security assessments and coordinating security operations at the regional level.

5. European balancing platforms for manual and automatic frequency restoration reserves

The electricity system requires an instantaneous balance between the production and consumption of electricity across the network to keep the frequency at 50 Hz. TSOs are responsible for balancing and maintaining this frequency and use different measures to do so. This is increasingly regulated at the EU level, including by the EB Guideline. This guideline prescribed the development and adoption of two European TCMs detailing implementation frameworks for common European balancing platforms – to be used by market parties for the sale of frequency restoration reserves (FRR) that the TSOs can use to bring the frequency back to 50 Hz. Two platforms were to be set up: one with manual activation of FRRs (mFRRs) and one with automatic activation of FRRs (aFRRs).

The TSOs developed the proposal for a mFRR TCM within ENTSO-E. Between May and July 2018, ENTSO-E consulted on a draft proposal. The national regulators discussed the draft informally within the ERF and, in June 2018, submitted an informal shadow opinion to the TSOs. In December, the TSOs submitted their formal proposal. ⁷³ A similar process took place for the aFRR TCM. ⁷⁴

The national regulators, having continued discussion within the ERF, did not agree on the proposal for the mFRR TCM and, in July 2019, referred the case to ACER – after the deadline by which they were to have adopted a decision. They included a non-paper outlining their disagreements, as well as what they had agreed on. ⁷⁵ The same occurred for the aFRR TCM. ⁷⁶ As the national regulators had not reached an agreement on time regarding either the mFRR or the aFRR proposals, both were escalated to ACER. Although ACER organised a public consultation and was in regular dialogue with the national regulators and the TSOs before adopting the decisions in 2020, ⁷⁷ these nonetheless proved controversial and were later legally challenged. ⁷⁸

The introduction of balancing platforms would represent a major change regarding harmonisation, with these having not existed in the past at the European level. Moreover, these European platforms would replace existing national balancing markets and, in some cases, voluntary regional cooperation on balancing that had already existed. However, the two TCMs sought to allow for different approaches to how a TSO would balance its area (eg by allowing for both proactive balancing based on forecasts and reactive balancing based on the occurrence of incidents). Moreover, bids would still be activated by the TSOs (at least for the mFRRs). Finally, although standard products were defined via these European TCMs, it would also be possible to use different national products defined under a national TCM for specific products.

V. Analysing the (non-)adoption by the European regulatory network

This section analyses whether the observed outcome for the six examined TCMs – adoption or non-adoption by the European regulatory network – can be explained by the bureaucratic politics and policy-network perspectives.

For the CCR TCM, the TSOs provided the national regulators with a full proposal indicating cooperative relations between TSOs and national regulators. Due to qualified majority voting amongst the TSOs, disagreements on part of the Austrian TSO did not prevent the TSOs as a group from presenting a joint proposal that was accepted by most national regulators. Hence, opposition from a single TSO was not enough to affect relations between the TSOs and national regulators. However, whilst the policy-network perspective expected cooperative relations across TSOs and national regulators to allow adoption by the national regulators, the TCM was instead escalated to ACER. The Austrian regulator opposed splitting the German–Austrian bidding zone as part of this TCM because this could reduce profits and increase electricity costs in Austria. Thus, the conflict could be traced back to how the proposal would interfere with one of the core national missions of the Austrian regulator: to maximise the welfare gains of the national electricity system. ⁷⁹ The opposition from a single national regulator prevented this TCM from being jointly adopted by the national regulators due to the consensus requirement. ⁸⁰ Hence, non-adoption by the regulatory network could be accounted for by the conflict amongst national regulators resulting from a TCM perceived as interfering with the core national mission of a regulator. Overall, the bureaucratic politics perspective rather than the policy-network perspective accounts for the non-adoption of this TCM by the European regulatory network. This case also illustrates the presence of territorial (country-based) rather than functionally based preferences, stemming from the varying roles of TSOs and national regulators, respectively.

For the SAP TCM, the TSOs provided a proposal that included an assessment of different hedging options, with one recommended for the allocation of long-term transmission rights. The proposal mainly reflected practices already being implemented in Central Europe. Moreover, the overarching Commission regulation had provided for flexibility in the choice of hedging approach with an opt-out clause from issuing long-term transmission rights. The TSO proposal provided the national regulators with the expert knowledge needed to adopt the TCM, indicating cooperative relations across TSOs and national regulators. As expected from the policy-network perspective, such cooperation enabled adoption by the European regulatory network. Relations were also cooperative amongst the national regulators, which may be linked to the TCM’s flexibility and orientation to the status quo, both of which reducing interference with regulators’ national mission. This aligns with expectations from the bureaucratic politics perspective. The two perspectives hence highlight different aspects of what enabled adoption: the policy-network perspective pointing to the role of TSOs cooperating with national regulators by providing them with a basis for adopting a TCM and the bureaucratic politics perspective emphasising how non-interference with the national core mission facilitated cooperation amongst the national regulators.

For the KORRR TCM, the TSOs provided the national regulators with a proposal that was contested amongst other stakeholders (electricity producers and users). The national regulators sided with these actors and opposed the proposal (apparently acting in a united fashion) and asked for amendments in two rounds of exchange with the TSOs. The way in which the national regulators sided with the interests of other stakeholders (to contain the scope and costs of data exchange) points to the wider interests of national regulators compared to those of the TSOs. The KORRR TCM was adopted by the European regulatory network despite these initial disagreements between the national regulators and the TSOs, contrary to expectations from the policy-network perspective. Unity amongst the national regulators gave them the power to steer TSO amendments towards their interests. It also indicates that the national regulators wanted to avoid escalation of the KORRR to ACER, as this would come with a risk of ACER siding with the view of the TSOs on the scope of data exchange needed to perform security analysis at the regional level or even extend this scope. Hence, the cooperation amongst the national regulators explains why the European regulatory network was able to adopt this TCM in support of the bureaucratic politics perspective.

The proposal of the TSOs for the CSAM TCM faced opposition from stakeholders and a unified group of national regulators. The TSOs seemed to aim for a rematch regarding the scope of data exchange (see the discussion of the KORRR TCM above). The national regulators could not decide within the allocated time as they lacked the expertise to revise the proposal. They also did not expect the TSOs to be responsive to requests for amendments. As expected from the policy-network perspective, conflict between the TSOs and the national regulators blocked the adoption of the CSAM TCM by the European regulatory network. However, the national regulators cooperated on actively referring this TCM to ACER – they felt less concerned because another TCM (the KORRR TCM) would constrain ACER from adopting conflicting provisions in the CSAM TCM. Although the bureaucratic politics perspective expected cooperation amongst national regulators to enable adoption by the European regulatory network (which did not happen), it also anticipated that national regulators might be willing to cooperate with an EU agency to the extent that this did not interfere with their national core mission. However, only when combined with the policy-network perspective (drawing attention to national regulators’ inability to adopt a proposal unless TSOs were cooperative) is a satisfactory explanation rendered. Neither perspective had theorised on how connections across processes influenced TCM rule-making. National regulators preferred ACER adoption to ensure coordination with other related TCMs, and they could live with escalation due to the constraints provided by yet another TCM that they had already adopted (ie KORRR).

The aFRR and mFRR TCMs were developed simultaneously, with the proposals of the TSOs seeking to cater to diverse approaches to balancing across Europe, all the while phasing in a common European platform. Despite having given input informally earlier, the national regulators were dissatisfied with the final proposal, revealing a conflict between the TSOs and the national regulators. This conflict, as predicted by the policy-network perspective, prevented adoption by the European regulatory network. The non-papers issued by national regulators clearly illustrate the importance of TSO expertise in TCM adoption, with the national regulators also requesting further details from TSOs to ACER on various issues. However, there was also conflict amongst the national regulators. Although they agreed on the need for changes, they were unable to reach a consensus on requesting specific amendments. This divergence reflected varied national approaches to regulating TSO procurement of balancing energy, making it challenging to find a common solution for Europe. Any one solution for balancing across Europe would interfere with the national mission of several national regulators. The European regulatory network referred the TCM to ACER for adoption. Although this occurred a few weeks before the deadline, it was clear from the distance between the national regulators that they would not resolve their conflicts within the scheduled timeframe. Consequently, conflict amongst multiple national regulators prevented the European regulatory network from adopting the two balancing TCMs, as expected from the bureaucratic politics perspective. Nevertheless, the extensive detail of the non-papers indicates that the national regulators aimed to limit ACER’s discretion to safeguard their core national missions. ⁸¹

Table 2 provides an overview of the studied TCMs. Upon review of the findings for the bureaucratic politics perspective, it is evident that this perspective was relevant for explaining (parts of) five of the studied cases. Conflict amongst the national regulators prevented the European regulatory network from adopting TCMs on CCR, mFRR and aFRR, whereas cooperation enabled its adoption of the TCMs on SAP and KORRR. The bureaucratic politics perspective did not explain why, despite cooperation amongst national regulators, the European regulatory network did not adopt the CSAM TCM. However, it did show how that a European regulatory network could seek to cooperate with an EU agency within the same policy area.

Table 2. Explanatory factors and observed outcomes for the studied terms, conditions and methodologies (TCMs).

aFRR = automatic frequency restoration reserves; BP = bureaucratic politics perspective; CCR = capacity calculation region; CSAM = coordinating operation security analysis; KORRR = key organisational requirements, roles and responsibilities; mFRR = manual frequency restoration reserves; PN = policy network perspective; SAP = single allocation platform; TSO = transmission network operator.

The summarised findings of the policy-network perspective show that it explained (parts of) four of the studied cases. Cooperation across the TSOs and the national regulators facilitated adoption by the European regulatory network of the SAP TCM, whereas conflict prevented its adoption of the TCMs on CSAM, mFRR and aFRR. For the latter two, conflict amongst the national regulators also contributed to their non-adoption. In contrast, the CSAM TCM shows that a lack of cooperation between the TSOs and the national regulators made it difficult for the European regulatory network to adopt the TCM, despite internal cooperative relations. The policy-network perspective could not explain why the European regulatory network did not adopt the CCR TCM (no adoption despite cooperation amongst TSOs and national regulators) and the KORRR TCM (adoption despite an initial lack of cooperation from the TSOs).

Some factors not anticipated by the two perspectives were identified in the case studies. The CSAM TCM became linked with other TCMs, and this affected the decision by the national regulators to refer this TCM to ACER. This shows that interdependence and simultaneity across multiple rule-making tasks may impact the adoption process. Moreover, the scope of a TCM was important for the level of conflict. The narrow delimitation of scope and flexibility provided for the SAP TCM facilitated cooperation, whereas the broadening of the CCR TCM gave rise to conflict amongst the national regulators.

Hence, our research shows that national regulators’ ability to cooperate within a European regulatory network is affected by whether a proposed rule interferes with their national core mission. It also sheds light on the reliance of national regulators on cooperation with TSOs due to the latter’s expert knowledge. Although not studied here, ACER’s ability to adopt TCMs is probably affected by access to the expertise of TSOs, too. Moreover, a lack of cooperation between the TSOs and the national regulators, including due to internal disagreements amongst the TSOs, complicated cooperation amongst national regulators (as shown by the EB TCMs). In turn, conflict amongst the national regulators made it more difficult for them to put pressure on the TSOs to revise the proposals in line with their wishes. Hence, conflict amongst the TSOs and the national regulators could make referrals by the European regulatory network to an EU agency attractive. This was seen in how the national regulators considered referral of the CSAM TCM to ACER more viable than getting the TSOs to amend the proposal in line with their preferences.

VI. Conclusions

Delegated rule-making in the EU has evolved over time, as seen in the changes introduced by the Lisbon Treaty. Most attention has been given to the role of the Commission. However, rule-making may also be delegated to other actors, such as EU agencies and European regulatory networks. In the area of EU electricity market regulation, an additional rule-making process emerged after the adoption of implementing acts by the Commission assisted by a Member State committee. ⁸² Here, the TCM procedure foresees adoption by a European regulatory network or an EU agency depending on case-specific development processes. Should consensus on a TCM not be achieved on time amongst all national regulators from all Member States (conceptualised here as a European regulatory network), the TCM would instead be adopted by ACER.

Considering the inherent lack of clarity in the TCM procedure regarding who would adopt the TCMs, this study examined under what conditions the European regulatory network would or would not adopt European TCMs. Most but not all of the cases saw non-adoption by the European regulatory network (see Table 1). Hence, escalation or referral to ACER was often but not always necessary for a European TCM to be adopted. Delegation to this EU agency was greater in practice than signalled in the legal text of the Commission implementing acts. This could be traced back to how adoption by the European regulatory network depended on cooperation amongst national regulators and on cooperation with private actors such as the TSOs (see Table 2).

By highlighting the factors behind (non-)adoption by a European regulatory network, this article underscores the conditions under which European rule-making based on consensual cooperation amongst national regulators was possible and when more hierarchy was needed via adoption by an EU agency. For cases perceived by national regulators as not interfering with their core national mission, consensus-based rule-making within a European regulatory network appeared to be more attainable than for cases seen as interfering with core missions. The same occurred for cases with little conflict between the groups of involved actors (eg between the TSOs and national regulators). This finding indicates the coexistence of functional and territorial conflict lines within European cooperation. ⁸³

Given the requirement for consensus amongst all national regulators from all Member States, it was not entirely surprising that ACER would step in according to its dispute resolution role to adopt many of these TCMs. Perhaps reflecting this experience, the TCM procedure has recently been revised, making the adoption of European TCMs by ACER the default rule (rather than this occurring only after failure of agreement amongst national regulators). ⁸⁴ While it remains beyond the scope of the current study to assess the implications of this change (such as for effectiveness), the change suggests the establishment of a faster process for adopting European TCMs and clarity in advance as to who will adopt them. ACER’s internal governance structure means that national regulators will remain heavily involved. ACER needs a favourable opinion from its Board of Regulators, requiring a two-thirds majority amongst the national regulators, to adopt a TCM. Thus, cooperation amongst national regulators within ACER on European TCMs can now proceed based on qualified majority voting. Whereas the reliance on TSOs to develop proposals based on their expert knowledge of the electricity system might not change, more effective coordination amongst national regulators within ACER could improve regulators’ leverage vis-à-vis TSOs.

Although the TCM procedure has been changed for European TCMs, the same procedural setup still applies to a series of TCMs to be adopted at the regional level. Here, unclarity remains as to who will adopt TCMs that apply for a region (defined as a group of physically interconnected countries). Should all national regulators in a region not reach an agreement on time, escalation to ACER ensues. Locating delegated rule-making at the lowest level possible, such as the regional level, follows from the principle of subsidiarity. The possibility of escalation to ACER in the case of disagreement amongst national regulators in a region balances the principle of subsidiarity with efficiency in rule-making. This is reflected in how non-adoption of TCMs may be avoided via escalation to an EU agency as a form of dispute resolution. For delegated rule-making by regions composed of many Member States (such as the “Core” region encompassing thirteen Member States), we may expect the adoption of many regional TCMs by ACER due to the high number of national regulators and TSOs involved in the process. ⁸⁵ A balance between subsidiarity and efficiency is also evident in the possibility of a deadline extension from ACER: ACER’s interpretation of its competence, allowing it to grant a deadline extension to the national regulators in a region, suggests an emphasis on subsidiarity. Such requests must, however, come from all the national regulators of that region. ⁸⁶ This prevents multiple individual requests from the many national regulators, which could reduce rule-making efficiency. ⁸⁷

Delegated rule-making through the TCM procedure may also be assessed in terms of the balance between legitimacy and quality of output. In terms of legitimacy, TCMs that are escalated to ACER are subject to a longer process. This extends the time for the involved actors to collect relevant information and negotiate whilst maintaining pressure for reaching an agreement. Moreover, good quality of output depends on ACER having sufficient resources to take on this workload. Lack of clarity on whether a high number of TCMs might be escalated creates uncertainty regarding ACER’s resource needs, also complicating planning. This entails a risk of insufficient resources being available in ACER, which could potentially reduce the quality of output. ⁸⁸ Considering how several ACER-adopted TCMs have been legally challenged, extending the adoption process is no guarantee of a quality of output that everyone is content with or that is upheld in court. ⁸⁹

Although not systematically examined in this study, differences were observed in the amount of discretion and flexibility provided to Member States in the adopted European TCMs. This suggests variation in the harmonisation attained at the end of what are rather extensive rule-making processes. This is also striking considering that this process of actual rule-making occurs after the adoption of implementing acts. Future research should examine the effects of this additional layer of delegated rule-making that exists within electricity market regulation for EU integration and harmonisation.

Delegated rule-making procedures that engage national actors collectively on the European scene could affect the relationship between the EU and Member States. Extensive delegated rule-making, through implementing acts or EU agency decisions, effectively reduces Member States’ discretion when implementing and applying EU legislation. Delegated rule-making involves Member State representatives, such as national ministries or agencies, adopting implementing or delegated acts that fill out the details of EU directives or regulations, usually after the latter’s adoption or revision. In contrast, double-delegated rule-making below implementing acts on electricity, as observed, by design involves a larger group of national actors in European rule-making. The significantly higher number of TCMs compared to implementing acts means more frequent engagement of these actors. By raising contestation amongst different parts of the electricity sector to the EU level, rule-making at this level could potentially become more important for balancing various sector interests. Thus, double-delegated rule-making on electricity market issues could expand the EU-level regulatory regime and pave the way for further EU integration. ⁹⁰

The TCM procedure is a potentially unique case of “double”-delegated rule-making in the EU. In effect, further delegation was provided for via implementing acts that themselves represent a form of delegated rule-making. This raises broader questions regarding conformity with the constitutional framework of delegation in the EU. The Meroni doctrine holds that discretionary power cannot be delegated to EU agencies, as this would distort the institutional balance amongst the EU institutions as laid down in the treaty. ⁹¹ Recent modification through case law allows EU agencies to adopt decisions under specified and narrowly restricted conditions and where such decisions are subject to judicial review. ⁹² Further research should examine whether double-delegated rule-making, even if constituting rule-making de facto rather than de jure, remain within these limits.