Introduction

In its 2018 Enerjisa decision,(1) the Competition Board imposed administrative fines amounting to a total of TL143 million on three retail electricity sales companies (RESCs) (ie, AYESAS, BASKENT and TOROSLAR – together, the RESCs) and one electricity distribution company (ie, AYEDAS), all of which were controlled by Enerjisa Enerji AS (Enerjisa), for abusing their dominant positions in various relevant markets.

The above subsidiaries all applied for a judicial review of the board's decision. In its respective judgments, the Ankara 13th Administrative Court upheld the fines imposed on the RESCs(2) but annulled the fine imposed on AYEDAS.(3)

Background

The conduct that the board characterised as abuse of a dominant position in Enerjisa can be categorised as:

  • the RESCs' exclusionary practices; and
  • AYEDAS's exclusionary practices.

According to the board, both types of practice had the effect of actually or potentially restricting competition in the downstream markets for the retail sales of electricity and different types of consumer residing in the respective electricity distribution regions.

RESCs' practices

In its assessment, the board examined a number of unilateral practices on an individual basis and eventually ruled that all of these practices should be deemed parts of a single strategy to exclude competitors. There were significant technical differences between the unilateral practices scrutinised. In general, the board highlighted that the RESCs, which previously held a legal monopoly to sell electricity in their respective distribution regions before market liberalisation and still retained the exclusive right to sell electricity to those 'ineligible'(4) customers, used this as an advantage to foreclose their competitors from the retail electricity sales market by manipulating their customers and leveraging practices (ie, using their dominant position in one relevant market to exclude competitors in another neighbouring market).

Hence, the board established two main theories of harm based on the manipulation of consumers and leveraging practices. The first theory of harm was built on the findings of behavioural economics. The board argued that the RESCs had benefited from the general low awareness levels of those consumers who had subsequently passed the thresholds and become 'eligible' to purchase electricity from their competitors, and prevented them from changing suppliers via certain practices, which were not de jure exclusivity requirements per se, but had the same effect in practice due to the manipulation of these customers' cognitive biases.

As for the second theory of harm, the board claimed that the RESCs had:

  • certain advantages as the exclusive suppliers to all ineligible customers in their respective distribution regions; and
  • used these advantages to the detriment of their competitors, especially by way of making certain offers to customers that may not be matched by the competitors.

Most importantly, the board refused to classify such practices as 'price-based' unilateral conduct and expressly refrained from conducting an 'equally efficient competitor test', by arguing that this test is irrelevant for 'non-price based' unilateral conduct.

AYEDAS's practices

AYEDAS held a legal monopoly in the electricity distribution market in the Anatolia electricity distribution region of Istanbul. The evidence that the board collected indicated that AYEDAS was delivering debt notices on behalf of AYESAS (the relevant RESC operating in the same distribution region), thereby providing it with a competitively significant cost advantage with regard to its competitors in the downstream market. Enerjisa argued that AYESAS's debt notices were actually being delivered to consumers by a third party (ie, EEDAS, which was also a wholly owned subsidiary of Enerjisa) under a contract that required AYESAS to pay for this notification service. Enerjisa further claimed that this company would also have provided the same services to all competitors, under equal terms, had there been such requests. However, the board refused to take this defence into consideration and held that the evidence at hand (an email communication) was sufficient to establish a competition law violation.

Anti-competitive effects of unilateral conduct

One of the most significant aspects of Enerjisa was that the actual effects of the unilateral conduct examined were quite ambiguous. In fact, some of the practices that were deemed to constitute part of a single violation had limited actual effects and some had not even been implemented. For example, the board found evidence that the RESCs collected certain forms (ie, the IA-02 forms)(5) from consumers with blank signature dates. The board argued that the RESCs thus had the opportunity to fill out the dates on the forms subsequently and create the impression that the signing had occurred at a much later date, which would give them a competitive advantage over their competitors. This is because the sector-specific regulations held that, where there existed more than one IA-02 form, signed by the same consumer but with different suppliers, the form with the latest date would prevail. Although the board found no evidence indicating that the RESCs had implemented such a strategy, this did not prevent the board from deciding that this constituted an abuse of dominance. Hence, in Enerjisa, the board adopted a broad interpretation of the term 'potential effects' and implied that this term also covers potential behaviour.

Judicial review of Enerjisa

Court decisions regarding RESCs

In three separate decisions regarding the appeal of the administrative fines imposed on the RESCs controlled by Enerjisa, the Ankara 13th Administrative Court rejected all of the claims raised by the RESCs and upheld the board's decision. While the specific judgments contain detailed explanations regarding different types of unilateral conduct, there are three aspects that might have significant implications for the judicial review processes of the board's decisions concerning the abuse of a dominant position going forward.

First, the court decided that the board did not err in law by focusing on the collective effects of various practices. The court decisions contain several statements which emphasised that only when a specific unilateral conduct is evaluated "together with other unilateral practices that are subject to examination" it is understood that the market could be foreclosed to competitors. The decisions include no further clarification as to how collective effects could be inferred from individual effects, or whether a combination of certain practices (none of which individually amount to a violation) could be deemed as an abuse of a dominant position under this approach.

Second, the court did not oppose the board's approach to the concept of potential effects. In its analysis concerning the IA-02 forms referred to above, the court noted that the fact that these forms contained no signature dates could lead to practices that would amount to an abuse of dominance, when evaluated together with the RESCs' other practices. Thus, the court implicitly confirmed that the 'possibility of abusive practices' may be deemed as 'potential effects', even in the absence of any evidence showing that undertakings had actually engaged in any abusive practices so far.

Finally, the court implicitly concurred with the board's position that it did not have to assess the relevant conduct's effects on equally efficient competitors to establish anti-competitive effects. The court expressed that the possibility of excluding 'some' downstream competitors may be sufficient to characterise the conduct in question as an abuse of a dominant position, without making any reference to the required efficiency levels of such competitors.

The court did not expressly mention the equally efficient competitor test or the relevance of the board's characterisation of the RESCs' conduct as non-price based. However, when the assessment of the 13th Chamber of the Council of State (CoS) – the highest administrative court in Turkey – in TTNET(6) is taken into consideration, it is probable that the court's assessments do not relate to price-based conduct. Indeed, in TTNET, it was indicated that the board must focus on the effects of the conduct in question on equally efficient competitors, while determining whether below cost-price rates charged in certain offers may amount to the abuse of a dominant position.

It is also possible that the court may have considered the specific circumstances surrounding Enerjisa (eg, the presence of a legal monopoly in the upstream market and certain parts of the downstream markets) while rendering its decision and, thus, deriving far-reaching consequences from these decisions might be a bit of a stretch. In any case, the court's decisions may still be subject to further judicial review by the CoS, if the parties decide to appeal.

Court decision regarding AYEDAS

With respect to the administrative fine imposed on AYEDAS, the court held that the board had failed to prove 'beyond any doubt' that AYEDAS actually leveraged its legal monopoly in the upstream market by delivering AYESAS's debt notices free of charge. The court stipulated that the board should:

  • not have relied solely on the email communications between AYEDAS and AYESAS; and
  • have further scrutinised whether AYESAS had actually made payments to EEDAS for its payment notice delivery services.

Notably, the court focused on the board's inability to prove the existence of such conduct, rather than its actual or potential effects. Still, this decision shows that the court considered that the approach adopted by the Ankara 6th Administrative Court in Sahibinden,(7) whereby it was held that the required standard of proof for exploitative abuse of dominant position is 'beyond all doubt', is also valid for exclusionary abuse of a dominant position.

While the administrative courts held, in Sahibinden and AYEDAS, that in cases concerning (exclusionary or exploitative) abuse of a dominant position, the existence of a violation must be established 'beyond all doubt', the CoS (in 12 Banks)(8) decided that the required standard of proof in the case at hand, which considered an alleged horizontal anticompetitive agreement, was 'beyond reasonable doubt'.

The abovementioned decisions do not clarify whether 'beyond all doubt' and 'beyond reasonable doubt' should be deemed as different standards of proof and, if so, what the differences between these standards are. Further, while it is noted in Sahibinden that excessive pricing (the most common category of exploitative conduct) constitutes an exceptional type of violation and that the relevant data must unequivocally set out that intervention would yield positive outcomes, none of these decisions expressly confine the applicability of the relevant standard of proof to a particular type of violation. Hence, the foregoing precedents do not suggest that the standards of proof to be satisfied by the board depend on the characterisation of the violation. Moreover, given that 'beyond reasonable doubt' is not an explicitly recognised standard of proof in Turkish public law (where all doubt must be eliminated to prove wrongdoing), it would not be unreasonable to argue that 'beyond all doubt' and 'beyond reasonable doubt' are essentially the same and that both refer to the highest standard of proof applicable in Turkish public law.

Implications for future

The court decisions regarding the RESCs may have significant consequences with respect to the concept of anti-competitive effects, especially to the extent that exclusionary unilateral practices are considered, in case these decisions are broadly interpreted to encompass all such practices, irrespective of the peculiar economic and legal conditions surrounding each case. However, such an expansive interpretation would not be in line with the precedents of the CoS, since the CoS laid down that the board should focus on the effects on equally efficient competitors (rather than the effects on 'any' competitor) and then evaluate whether foreclosure is likely as a result of the conduct being examined. These precedents were also in line with the Competition Authority's Guidelines on the Assessment of Abusive Conduct by Undertakings with Dominant Position.

In its abovementioned TTNET decision, the CoS noted that the equally efficient competitor test consists of two tiers. The first tier examines the relation between prices and costs, in order to determine whether the prices of the dominant firm may be replicated by equally efficient competitors that have similar cost structures. If it is established that the prices may not be replicated by such competitors, the second tier of the test comes into play, where the assessment is whether the examined practices may lead to market foreclosure. In TTNET, the CoS expressly set out that although the prices remained below the costs in some instances (ie, the conditions in the first tier were satisfied), the market was not foreclosed (ie, the conditions in the second tier were unsatisfied) as TTNET's competitors with similar cost structures (ie, the equally efficient competitors) were able to gain profits and increase their market shares when the examined practices were being implemented. Although it may be meaningful to argue that the first tier would not apply to non-price-based conduct (as there are no obvious prices and costs to compare), there is no reason to assume that the methodology adopted in the second tier should also be confined to price-based conduct. On the contrary, given that non-price-based exclusionary practices would be deemed anti-competitive only if they lead to market foreclosure, it should be recognised that the second tier would directly apply to these practices, as well.

In light of the foregoing, the court's findings regarding the RESCs may be interpreted more narrowly and it may be assumed that the peculiar economic and legal conditions surrounding Enerjisa was taken into consideration by the court in its reasoning, although this was not expressly spelled out. Still, if the parties appeal, it is critical that these issues are addressed in the judicial review of the relevant court decisions by the CoS, both to prevent an inconsistency between the precedents and an imposition of undue burdens on dominant firms by restricting their ability to compete effectively. If it is accepted that unilateral practices of dominant firms that have the potential to affect any competitors negatively, regardless of their efficiency, would constitute an abuse of a dominant position, this approach could lead to significant welfare losses due to the chilling effect, contrary to the aims of competition law.

An earlier version of this article was first published on Mondaq on 27 October 2020.

Endnotes

(1) Competition Board's decision dated 8 August 2018 and numbered 18-27/461-224.

(2) Court decisions dated 16 July 2020 and numbered E.2019/1969-K.2020/1318; dated 16 July 2020 and numbered E.2019/1970-K.2020/1319; dated 16 July 2020 and numbered E.2019/1956-K.2020/1317.

(3) Court decision dated 16 July 2020 and numbered 2019/660-2020/1315.

(4) 'Ineligible consumers' are consumers whose electricity consumption levels per annum are below a certain kWh threshold (as determined by the Energy Market Regulatory Authority) and, therefore, are unable to freely choose their energy providers.

(5) IA-02 forms were required documents for providing electricity to eligible consumers through bilateral electricity sales agreements.

(6) Decision of the 13th Chamber of the CoS dated 18 May 2016 and numbered E.2015/5104-K.2016/1849.

(7) Sahibinden, the Ankara 6th Administrative Court dated 18 December 2019 and numbered E.2019/946-K.2019/2625.

(8) 12 Banks, the 13th Chamber of the CoS dated 21 May 2019 and numbered E.2016/3513-K.2019/1777.