CJEU, 18 December 2025, Case C-245/24, Lukoil Bulgaria
The judgment given by the Court of Justice of the European Union on 18 December 2025 in the case between Lukoil Bulgaria and Lukoil Neftohim Burgas and the Bulgarian competition authority deserves particular attention, as it provides substantial clarification on the application of Article 102 TFEU to refusals of access to essential facilities resulting from privatisations. The case arose from the Bulgarian competition authority’s finding against two companies of the Lukoil group for having refused, between 2016 and 2021, to give third parties access to their fuel transport and storage infrastructure, made up of onshore depots, marine depots and pipelines. This infrastructure, originally developed by the Bulgarian public authorities as part of a State petrochemical complex created in 1963, had been privatised from 1995 onwards and acquired as to a majority by the Lukoil group in 1999.
The central question raised by this judgment concerns the conditions for applying the test laid down in the 1998 Bronner case law where the essential facility controlled by the dominant undertaking does not result from its own investments, but comes from a privatisation of public assets. The Court recalls that the Bronner judgment had set three cumulative conditions for a refusal of access to constitute an abuse of a dominant position: the facility must be indispensable to carrying on business on the market concerned, the refusal must be likely to eliminate all effective competition, and there must be no objective justification. These strict conditions were justified by the need to strike a balance between the requirements of undistorted competition and the protection of freedom of contract and of the dominant undertaking’s right to property. More fundamentally still, they reflect the legitimate concern not to discourage dominant undertakings from investing in the development of efficient facilities, at the risk of seeing their competitors benefit free of charge from the fruits of those investments without having to bear the associated costs and risks.
Yet the situation examined in the Lukoil case differs appreciably from that in the Bronner case. The fuel storage and transport infrastructure had not been developed by the Lukoil group for the purposes of its own activities, but was a legacy of the Bulgarian State economic system, financed by public funds and operated for several decades by a State-owned enterprise before its privatisation. This circumstance raises a fundamental question: do the concerns relating to the protection of incentives to invest, which underpin the strict application of the Bronner test, remain relevant where the dominant undertaking did not bear the initial costs of developing the facility? The Bulgarian competition authority had specifically taken the view that the conditions laid down in the Bronner judgment did not apply to facilities that a dominant undertaking had not itself created or whose construction cost it had not borne, particularly where those facilities had been acquired through a privatisation.
The Court of Justice takes a nuanced position on this point that avoids any automatic outcome. It begins by reaffirming the cumulative nature of the two criteria for applying the Bronner test: first, the facility must have been developed by the dominant undertaking for the purposes of its own activities and, second, that undertaking must own the facility. The cumulative nature of these conditions means that it is sufficient for one of the criteria to be absent for the strict Bronner conditions to become inapplicable. As regards the ownership criterion, the Court states that it is intended to confine the application of the Bronner test to situations in which particular attention must be paid to the right to property, or to the equivalent degree of control enjoyed by the dominant undertaking, which enables it to refuse access to third parties. Where a dominant undertaking does not have full decision-making autonomy over access to the facility it operates, because of powers or obligations imposed by law, regulation or contract that prevent it from refusing access to third parties, it cannot be regarded as genuinely owning that facility within the meaning of the Bronner test.
This analysis leads the Court to affirm an essential principle: where a dominant undertaking does not have full decision-making autonomy over access to the facility, the cumulative nature of the criteria for applying the Bronner test makes it irrelevant whether the facility was developed for the purposes of the undertaking’s own activities. In other words, the absence of full decision-making autonomy is in itself sufficient to rule out the application of the Bronner test, irrespective of the historical origin of the facility. This position is fully justified: the fair balance between competition and the rights of the dominant undertaking must necessarily reflect the limitations on those rights resulting from intervention by the public authorities. Here the Court follows on from its 2023 Lietuvos geležinkeliai case law (Case No. C-42/21 P), which had already held that the Bronner conditions were not intended to apply where a dominant undertaking was subject to a regulatory obligation to give access to its facility.
But it is on the reverse question that the judgment makes its most innovative contribution. What happens where the dominant undertaking does in fact have full decision-making autonomy over access to a facility originally developed by the public authorities and subsequently acquired through a privatisation? The Lukoil companies argued that the position should be different where a facility over which the dominant undertaking admittedly has full decision-making autonomy had nevertheless been built and developed by public authorities using public funds, then acquired by that undertaking at a competitively set price, and in which the undertaking had made substantial subsequent investments. The Court rejects this reasoning in its general principle, holding that, whatever the size of the price paid and of the investments made by the dominant undertaking in the facility concerned, as long as that undertaking does not have full decision-making autonomy over access to that facility, its situation cannot be regarded as analogous to that which gave rise to the Bronner case.
On the other hand, the Court introduces a decisive criterion where full decision-making autonomy exists: the mere fact that a facility was built or developed by public authorities or with public funds is not sufficient, in any event, to exclude the application of the conditions laid down in paragraph 41 of the Bronner judgment. The EU court draws a subtle but decisive distinction here.
Where a public facility has been acquired by the dominant undertaking at a price and on terms resulting from a competitive procedure, it is comparable to a facility built or developed by that undertaking itself. In such a configuration, the justifications underlying the Bronner test become relevant again: the undertaking agreed to pay a market price to acquire the facility, thereby bearing a real economic cost that places it in a situation comparable to that of an undertaking that developed its assets itself. The economic logic is clear: if the acquisition took place under genuinely competitive conditions, the undertaking internalised the value of the facility in the price it agreed to pay, and it would therefore be illegitimate to deprive it of the prerogatives normally attached to ownership on the ground that the facility was originally created by the public authorities.
But the Court immediately adds an essential qualification: this assimilation applies only if the privatisation procedure was actually capable of guaranteeing the competitive nature of the price and of the terms of acquisition. Otherwise, if the privatisation procedure was not such as to ensure a market price, such assimilation must be ruled out. This criterion is of considerable practical importance. It requires a concrete examination of how the privatisation was carried out: was it an open and transparent tender procedure? Was there genuine competition? Did the price obtained reflect the real value of the assets sold, or did it indicate a manifest undervaluation of the privatised facilities? This last scenario is not purely theoretical, as the user rightly points out. Privatisations carried out in the countries of Central and Eastern Europe during the 1990s were frequently characterised by conditions that departed significantly from the standards of a fully developed market economy. In some cases, strategic infrastructure may have been sold at prices significantly below its real economic value, whether because of political constraints, a lack of expertise in valuing complex assets, or the dominant negotiating position of potential buyers in a context of weak competition for those assets.
The judgment thus introduces a new evidentiary requirement: where a competition authority intends to apply the strict Bronner conditions to a privatised facility, it will be for that authority to show that the privatisation procedure actually took place under conditions capable of guaranteeing the competitive nature of the price and of the other terms of acquisition. This burden of proof is far from trivial. It involves a retrospective analysis of the economic and procedural conditions of the privatisation, which may prove complex where it dates back several decades. In the Lukoil case, the privatisation of the Burgas petrochemical complex had taken place in 1999, more than twenty years before the conduct complained of. Assessing the competitive nature of a privatisation procedure carried out in that particular historical context, marked by the transition from a planned economy to a market economy, raises delicate methodological questions.
The Court extends the same reasoning to situations in which the dominant undertaking, without owning the facility concerned, holds exclusive rights that give it full decision-making autonomy over access to that facility. Such a facility must be treated as a facility owned by that undertaking. Here again, where those exclusive rights were acquired by the dominant undertaking at a price and on terms resulting from a competitive procedure, the facility is comparable to a facility built or developed by the undertaking itself. This clarification is particularly relevant for public service concessions and public-private partnerships, in which the undertaking does not become the owner of the facilities but obtains exclusive control over them for a fixed period.
On the first question referred for a preliminary ruling, concerning the possibility of finding a single infringement rather than separate infringements, the Court takes a pragmatic position that reflects a fine understanding of how anticompetitive strategies work in practice. It holds that a competition authority is not required to find separately that the conditions of Article 102 TFEU are met in respect of each of the categories of conduct identified, provided that it can establish that those conditions are met in respect of the overall abusive conduct attributed to the undertaking. This approach is justified by the recognition that two companies belonging to the same undertaking may jointly contribute to an abuse of a dominant position, each refusing access to installations under their respective control but forming part of the same essential facility controlled by the undertaking. Competition law focuses on the unity of conduct on the market, and the formal separation between various companies resulting from their distinct legal personalities cannot prevent that unitary assessment. This position prevents circumvention strategies consisting in artificially fragmenting control of an essential facility among several legal entities in order to escape a finding of abuse.
The contribution of this judgment must be measured against the economic and social stakes raised by the control of essential facilities by dominant undertakings. Energy transport and storage infrastructure is a strategic asset whose control gives its holders considerable market power. Where such infrastructure results from substantial public investment made over several decades and has then been transferred to private operators on terms that did not necessarily reflect its real economic value, the question whether those operators may legitimately invoke the protective conditions of the Bronner judgment takes on a dimension that goes beyond the legal framework alone and touches on fundamental political choices concerning the organisation of essential services. The solution adopted by the Court has the merit of not drawing an automatic distinction based on the public or private origin of the facility, while introducing a concrete assessment criterion based on the actual conditions in which the privatisation took place. This case-by-case approach preserves the flexibility needed to deal with the diversity of concrete situations, while providing competition authorities and economic operators with sufficiently precise criteria to guide their action.
It will be worth watching closely how national competition authorities and national courts implement this new criterion relating to the competitive nature of privatisation conditions. Assessing the competitive nature of a privatisation procedure raises substantial methodological questions: what are the relevant indicators for determining whether the price paid matched the market value of the assets? How can the value of complex assets be assessed retrospectively in a context of economic transition marked by uncertainty and the absence of stable reference markets? To what extent should the political and macroeconomic conditions prevailing at the time of the privatisation be taken into account in that assessment? The answers to these questions will largely determine the effectiveness of the control of abuses of dominant positions in the sectors affected by privatisations of essential facilities.
