3.1 Unbundling
551/2021

3.1 Unbundling

3.1.1 An outline of unbundling

The idea of 'unbundling' the management, legal corporate status and even ownership of the energy network is a core element of the EU's energy legislation. Most of the ex-incumbent electricity and gas companies were typically vertically integrated, which can create difficulties for liberalizing these markets:

[t]hey have an inherent interest in retaining their customers, market share, and thus profitability. When competition is introduced, the ex-monopolists hold a 100% market share. Thus, any gain in market share by new competitors means a loss in market share by the ex-incumbent. It is perfectly natural that the ex-incumbent will endeavour to prevent any loss of market share. Where the ex-incumbent owns the network, it has a natural incentive to make third party access to it as difficult as possible.(1)C. Jones (gen. ed.), EU Energy Law – Volume I: The Internal Energy Market (Leuven: Claeys & Casteels, 3rd edn., 2010), 10.

In essence,(2)For details, see Johnston & Block, ch. 3. unbundling seeks to:

  • introduce competition where possible within the system, including through trade from other countries, thus enhancing the system's responsiveness to changes (on matters such as input costs, etc.);

  • reduce incentives to cross-subsidise up- or downstream business using profits garnered from control over natural monopoly assets in transmission and distribution (as illustrated by Diagram 1, above) or otherwise favour such other parts of the business (e.g. via sharing market-sensitive information): this should also encourage the network operator to focus on its own issues and performance, rather than being run to serve the interests of associated up- or downstream parts of a vertically integrated business;

  • encourage investment and innovation across the system;

  • ease the supervisory tasks entrusted to the National Regulatory Authority ('NRA'), concerning issues like tariffs, market monitoring and transparency;

  • enable - if this were thought desirable by the relevant Member State - privatisation of (elements of) the energy supply system.

Alongside these potential benefits, unbundling in general, and ownership unbundling in particular, also impose various costs upon system operators, users and customers; for some, these costs may well outweigh the benefits to be gained from unbundling.(3)See, e.g., M. Mulder, V. Shestalova and M. Lijesen, ‘Vertical separation of the energydistribution industry’ (CPB No 84, 2005) and B. Baarsma et al, ‘Divide and Rule. The Economic and Legal Implications of the Proposed Ownership Unbundling of Distribution and Supply Companies in the Dutch Electricity Sector’ (2007) 35 Energy Policy 1785. These costs may include:

  • one-off transaction costs on asset sales and/or structural reorganisation;

  • replacing internal processes with a series of contracts (time delays, ongoing transaction costs);

  • the need for regulation of natural monopoly assets, which itself imposes costs on society or at least users of the system;

  • more generally, unbundling models stopping short of full ownership unbundling require policing the limits of such other approaches, which imposes further regulatory oversight costs, and compliance costs on the part of the undertaking;

  • the loss of (easy?) government ability to achieve policy goals through the energy system.

There are also arguments for moving beyond functional and legal separation to require the full ownership unbundling of the transmission system operator, which essentially reside in improving or enhancing various elements of the benefits outlined above.(4)See, further, M. Pollitt, ‘The Arguments For and Against Ownership Unbundling of Energy Transmission Networks’ (2008) 36 Energy Policy 704.

Under the current EU legislative framework, Member States are required to use of one three basic models for Transmission System Operators:

(1) the ownership unbundling model (which is the basic principle and the default model(5)See, e.g., the Third IEM Directives, recitals 11 (Elec.) and 8 (Gas). from the EU Commission's perspective), was first introduced in Article 9 of the Third Electricity and Gas IEM Directives, and is now included in Article 43 of Electricity Directive (EU) 2019/944 with respect to the electricity sector. This requires complete separation of the ownership of the transmission business from other levels up- and/or downstream (generation, distribution, supply, etc.);

(2) the independent system operator (ISO), provided for by Articles 44 and 45 of the (EU) 2019/944 (previously Articles 13 and 14 of Electricity Directive 2009/72/EC) and Articles 14 and 15 of the Gas Directive 2009/73/EC. Ownership of the network can still be held by the vertically integrated entity, but the transmission network must itself be managed by an independent system operator, which must be entirely separate from the vertically integrated company and which is to perform all network operator functions; or

(3) the independent transmission operator (ITO), detailed in Articles 46 to 51 of the Electricity Directive (EU) 2019/944 (previously Articles 17 to 23 of the Electricity Directive 2009/72/EC) and Articles 17 to 23 of the Gas Directive 2009/73/EC. Here-under, separation of the transmission activities must be achieved through the establishment of an ITO, which must be responsible for the maintenance, development, and operation of the networks, even though those networks remain the property of the vertically integrated company.

Once a Member State has adopted the first model, under the Directives it is not permitted to 'regress' back to a weaker unbundling model in future. It should be emphasised that these Directives provide a significant amount of detail concerning the national rules which will be required to govern the ISO and, in particular, the ITO models, in an attempt to ensure that these options "provide the same guarantees regarding independence of action of the network in question and the same level of incentives on the network to invest in new infrastructure that may benefit competitors".(6)Commission, ‘Proposal for the Third Package Directives’, COM(2007) 195 (19 September 2007). The concomitant of providing these alternatives is that the NRA will have a significant role to play in ensuring the respect of these detailed rules by the transmission system operator:(7)For detailed discussion, see Johnston & Block, paras. 3.32-3.94. from a fundamental rights perspective, this is significant, in that it may render the NRA the appropriate defendant if any of its regulatory activity is found to be disproportionate in its effects upon that operator.

The reason for providing some detail concerning unbundling is that its goals and detailed regulation will prove crucial in any analysis of the fundamental rights implications of EU or national rules which establish or further extend the unbundling principles: prima facie, rules which strongly control the enjoyment of property (i.e. the transmission business) held by a company, even to the extent of requiring that property to be sold and specifying certain key characteristics of those allowed to buy it, amount to a restriction upon rights to free enjoyment of property and possessions under Article 1 of the First Protocol to the ECHR and/ or Article 17(1) of the Charter.

Diagram 2: Flowchart on legal issues raised by national ownership unbundling measures

The flowchart reproduced in Diagram 2 (above) tries to locate the relevance of such fundamental rights arguments within the EU law firmament, noting the potential impact of fundamental rights upon the EU's law-making process and competence as well as their relevance to national implementation of EU law and national level law-making, where EU fundamental rights law may operate as a constraint upon national competence and autonomy.(8)This was developed from A. Johnston, ‘Ownership Unbundling: Prolegomenon to a Legal Analysis’, ch, 23 in M. Bulterman, L. Hancher, A. McDonnell & H. Sevenster (eds.), Views of European Law from the Mountain – Liber Amicorum Piet Jan Slot (Alphen aan den Rijn: Kluwer Law Intetnational, 2009).

3.1.2 Ownership unbundling and fundamental rights

Despite the relative paucity of case law to date, the issue remains one of real significance: strong views have been expressed(9)See, e.g.: J-C Pielow, G Brunekreeft, and E Ehlers, ‘Legal and Economic Aspects of Ownership Unbundling in the EU’ (2009) 2(2) Journal of World Energy Law & Business (see the comment in reply by K. Talus & A. Johnston, (2009) 2(20 Journal of World Energy Law & Business 149); and E. Ehlers, Electricity and Gas Supply Network unbundling in Germany, Great Britain and The Netherlands and the Law of the European Union: A Comparison (Antwerp: Intersentia, 2010). that the far-reaching implications of unbundling in general, and ownership unbundling in particular, require strong and cogent justifications if the intrusion upon property rights is to be found proportionate. Praduroux and Talus, on the other hand, have concluded that there does not appear to be a conflict between fundamental rights and general principles of EU law on the one hand and ownership unbundling on the other hand.(10)S. Praduroux and K. Talus, ‘The third legislative package and ownership unbundling in the light of the European fundamental rights discourse’ (2008) 9 Competition and Regulation in Network Industries 3-28. The key fundamental right in question is likely to be the right to property laid down in Article 1 of the First Protocol to the ECHR (and the corresponding terms of Article 17 of the EU's Charter of Fundamental Rights). In short, provided the transmission assets are sold off, thus ensuring that their current owners receive some compensation in return for their inability any longer to own such assets, it seems that this should amply satisfy the proportionality requirements imposed by the ECHR under this provision.(11)It can be noted that the rationale underlying the fundamental rights analysis under the ECHR (mirrored in many national systems) is very similar to the basis upon which claims to recover stranded costs have been developed and subsequently analysed under EC law in the State aids field. See Commission Communication relating to the methodology for analysing State aid linked to stranded costs (26 July 2001), which document is available on the internet at: https://ec.europa.eu/competition/state_aid/legislation/stranded_costs_en.pdf. See further the brief article by B. Allibert, ‘A methodology for analysing State aid linked to stranded costs, and first cases’, (2001) Competition Policy Newsletter, Number 3, October 2001, pp. 25-27, discussing the Decisions taken by the Commission on the applications by Austria, Spain and the Netherlands. A brief discussion is required to justify this assertion, which is based upon the broader case law under the ECHR, given that the issue has yet to be addressed directly under EU (fundamental rights) law.(12)The same structure of discussion could also apply to rules mandating third party access (TPA) to energy networks at the transmission and distribution levels, although we are not aware that it has been utilised in practice.

3.1.2.1 'Deprivation of property'?

The main test for 'deprivation' of property under Article 1 of the First Protocol is the extinction of the owner's rights in the property, usually by means of a legal transfer of those rights to another by operation of law or the exercise of a legal power to do so. Ownership unbundling mandated by EU law would appear to conclude that the only way to promote competition would be to force current incumbents to transfer certain companies or assets to new market entrants. Such a move would be a State act and would no doubt be laid down in the relevant legal framework (thus satisfying the basic conditions for such a deprivation).(13)It should be noted that the reference to ‘the general principles of international law’ as a condition for such deprivation of property has been held by the European Court to be relevant only in the situation where the party claiming interference with his possessions is not a national of the expropriating state: see James v. U.K. (1986) 8 EHRR 123, confirmed in Lithgow v. U.K. (1986) 8 EHRR 329, at (inter alia) para. 115. However, given the approach of the Court to compensation in deprivation cases (considered briefly below), the inapplicability of the public international law principle (requiring compensation to be given to non-nationals for deprivation of their property) is unlikely to make much difference in practice. However, even in this hypothetical situation, the typical method would be to force the sale of such assets, thus ensuring some form of compensation for the incumbent operator. The adequacy of the compensation that such a method might provide falls to be considered below.

3.1.2.2 Justifying an infringement?

In all situations where an infringement by means of some interference with possessions has been shown, the state must show that this interference was justifiable to escape a finding that its conduct has been unlawful. There are separate elements(14)Clearly, ownership unbundling rules under EU law meet the criterion of being ‘conditions provided by law’ which is necessary for any justifiable infringement of Convention Rights. This basis in law must be accessible, sufficiently certain and must provide protection against arbitrary abuses. Thus, it is not only a requirement to be able to point to a positive legal provision empowering the body in question to take the action of which the applicant complains; there is also an element of the ‘Rule of Law’ about this requirement. These criteria seem satisfied in the case of ownership unbundling under EU law. to be considered here, but it should not be forgotten that there is an essential link between how the public interest is defined and the shape of the proportionality argument that follows. The question of compensation is part of that proportionality analysis, but given its centrality to the ownership unbundling scenario, it will be highlighted separately in what follows.

3.1.2.2.1 Public interest/General interest

Any justification for an infringement upon the right to the peaceful enjoyment of possessions must state the grounds upon which that interference is to be made. The Strasbourg Court has tended to be deferential to the Member States' definitions and explanations of why a certain restriction was necessary: for example, leasehold enfranchisement legislation in the U.K. was held to be a policy calculated to enhance social justice within the community and therefore was 'properly described as being "in the public interest"'.(15)James v. U.K. (1986) 8 EHRR 123, at para. 49.

On the case law as it stands, therefore, it seems highly likely that the type of public/general interest ground that would be relied upon by the state in the ownership unbundling scenario (such as benefiting overall social welfare by the introduction of competition) would be difficult and perhaps impossible to characterise as not being acceptable under the Convention. However, while the ground of public interest may be legitimate, it must still be analysed whether the means chosen to fulfil that ground were proportionate to the benefit to be gained.

3.1.2.2.2 Proportionality

Although there is no express reference to a proportionality test in the wording of Article 1 of the First Protocol, it is clear from the Strasbourg Court's jurisprudence that such a requirement is inherent in that Article. Proportionality is a general principle of the Convention and requires there to be a 'reasonable relationship of proportionality between the means employed and the aim sought to be realised'.(16)James v. U.K. (1986) 8 EHRR 123, at para. 50. In the context of Article 1 of the First Protocol, the Strasbourg Court has developed a requirement that a 'fair balance' must be struck 'between the demands of the general interest of the community and the requirements of the protection of the individual's fundamental rights'.(17)Sporrong and Lönnroth v. Sweden (1982) 5 EHRR 35, para. 69. This approach is followed by the Court in all cases of infringement of Article 1 of the First Protocol, whether concerning deprivation, control of use, or more general interference with the enjoyment of possessions.

It is important to note that the intensity of the proportionality test applied will vary according to the severity of the infringement in question. 'Deprivation of property is inherently more serious than a control of its use',(18)See Gillow v. U.K. (1989) 11 EHRR 335 for a clear recognition of this point. thus suggesting that it will be more difficult to argue that the action of a public body in depriving a company of its property is a proportionate way to achieve the public interest goal at issue. In any application of the idea of fair balance, however, it is clear that two elements will be key: first, is there any entitlement for the property owner to compensation for the interference suffered? Second, is there any procedure open to the applicant to challenge the measure that has caused the interference with his possessions? In the parallel stranded costs situation, a good example of the procedural element is provided by Article 24 of Directive 96/92/EC,(19)[1996] O.J. L27/20. under which Member States were allowed to develop plans to compensate incumbent companies for stranded costs. These plans were then to be submitted to the European Commission within a certain period of time for their examination in accordance with the EU's State aid rules. Equally, the absence of any such procedure may well lead to a finding that the interference is a disproportionate one that fails to respect the balance to be struck between the competing interests at stake.(20)See Sporrong and Lönnroth, n. 51, supra for a good example, although here it was the combination of the failure to provide any means of compensation with the lack of any opportunity to challenge the measures which seemed to tip the balance overall. This illustrates the interlinked nature of the proportionality analysis in such cases, covering many different and yet connected issues.

3.1.2.2.3 Compensation

It would appear that there is no absolute right under the Convention to receive compensation in return for an interference with the right to the peaceful enjoyment of one's possessions. Rather, the availability and extent of any compensation falls to be considered as part of the overall analysis of the proportionality of the interfering measure. However, it is also accurate to state that the more serious the infringement of the right to peaceful enjoyment of one's possessions, the stronger the presumption that at least some compensation must be paid for the 'fair balance' of interests to be respected.

With regard to the deprivation of possessions and compensation, only in 'exceptional circumstances' will the taking of property without compensation be justifiable; otherwise, the protection afforded by Article 1 of the First Protocol 'would be largely illusory and ineffective'.(21)Lithgow v. U.K., n. 47, supra; see esp. paras. 80-83. However, while compensation should normally be an amount 'reasonably related to [the] value' of the property taken, there is no 'guarantee [of] a right of full compensation in all circumstances, since legitimate objectives of public interest, such as pursued in measures of economic reform may call for less than reimbursement of the full market value ...'.(22)Ibid. This seems to imply that there is a proportional relationship between the nature and extent of the public interest, on the one hand, and the individual burden to be borne, on the other. That is to say that 'the greater the public gain to be achieved by the legitimate aim, the greater the financial burden the property owner can be expected to bear. To this extent the state enjoys a wide margin of appreciation in calculating compensation terms'.(23)Rook, Property Law & Human Rights (London: Blackstone Press, 2001), p. 72. Generally, the defendant States have not been successful in arguing that their case falls within the 'exceptional circumstances' needed to escape the need to provide compensation.(24)D. Harris, M. O’Boyle & C. Warbrick, Law of the European Convention on Human Rights (London: Butterworths, 1995), p. 532 suggest that a possible example might be seizure of property during times of war (see now D. Harris, M. O’Boyle, E. Bates & C. Buckley, Harris, O’Boyle & Warbrick: Law of the European Convention on Human Rights (Oxford: OUP, 4th edn., 2018), while D. Rook, Property Law & Human Rights (London: Blackstone Press, 2001), p. 71, n. 2 suggests that a local authority landlord exercising the remedy of distress for rent might be another. However, there are examples where the Court has been rather deferential to the terms upon which compensation has been calculated.(25)See Lithgow v. U.K., n. 47, supra, where the calculation of the compensation paid to a company which was to be nationalised was made on the basis of the value of its shares at a point before the announcement of the nationalisation plan, rather than on the basis of company assets held at the date of nationalisation. The Court acknowledged that such a broad public interest issue as nationalisation legislation involved the consideration of a very wide range of competing interests, which the Member State and its national authorities were best placed to assess. Overall, the Court found that adequate reasons did exist for the compensation criteria chosen and, as a result, held the U.K. to be within its margin of appreciation and thus found no violation of the Convention. Overall, therefore, it would appear that ownership unbundling would be likely to survive a challenge based upon fundamental rights under EU law, at least if the approach of the Strasbourg Court under the ECHR is any guide.

3.1.3 Ownership unbundling and Article 345 TFEU

The Grand Chamber judgment of the Court in Netherlands v. Essent is the only case where the Court of Justice has dealt with a privatisation ban related to the sale of shares in electricity and gas DSO organisations under Article 345 TFEU governing national systems of ownership rights.(26)Joined cases C-105/12 to C-107/12, ECLI:EU:C:2013:677. The Court found that the privatisation ban fell within the scope of Article 345 TFEU, but that the prohibition nevertheless constituted a restriction on the free movement of capital pursuant to Article 63 TFEU. The judgment is, however, more ambiguous concerning the potential influence of the principle in Article 345 TFEU on considering legitimate justification grounds.(27)K. Haraldsdottir, ‘The nature of neutrality in EU law: Article 345 TFEU’ (2020) 45(1) E.L. Rev. 2020 3-24. Haraldsdottir argues that the role of the neutrality principle enshrined in Article 345 TFEU must be viewed in relation to the specific merits of each case, where the application of the principle may depend on the social function or strategic importance of the property at issue.(28)Ibid. Based on this reasoning, the decision in Netherlands v. Essent may also be seen as not contradicting the reasoning of the EFTA Court in Hjemfall, where the Court noted that Norway could pursue a system of public ownership for its hydropower resources, provided the objective is pursued in a non-discriminatory and proportionate manner, with reference to the equivalent provision to Article 345 TFEU in Article 125 EEA.(29)Case E-02/06, EFTA Surveillance Autority v. Norway (judgment of 26 June 2007). Finally, it is notable that the CJEU judgment in Netherlands v. Essent did not refer even once to fundamental rights protection in general, the ECHR, EU or national fundamental rights law, focusing instead solely upon Article 345 TFEU and the free movement of capital under EU law.