When is an infringement of competition law - obvious? Considering the state of the “object” infringement following the latest EU Court judgments

It is well-established that under EU competition rules, anticompetitive arrangements are categorised as either “by object” or “by effect”.  This categorisation is significant because, in the case of by object infringements, there is no requirement on the enforcing authority to establish anticompetitive effects in the relevant markets to find an infringement – and impose potentially severe penalties.  Recent cartel investigations in sectors as diverse as TV/computer monitor tubes, automotive bearings, and interest rate derivatives have led to very high levels of penalties, with total cartel fines imposed by the EU reaching just under €2 billion in each of 2012, 2013 and 2014.

By contrast, in by effect cases, the authority must demonstrate that the arrangements in question led to an actual anticompetitive effect in the relevant market, for example through price increases which would not have occurred absent the arrangements in question (i.e., in what is commonly referred to as the counterfactual).

The justification for this dichotomy is that there are procedural inefficiencies in requiring the authority to establish detailed economic effects arising from conduct which is clearly anticompetitive.  In such cases the authority should be able to cut through the process and impose a penalty in order to punish the infringing parties and to deter similar future behaviours, without being waylaid by complicated arguments around establishing the actual effects of the conduct in question.

It has long been recognized that the most serious types of competition law infringement - such as price fixing, market or customer allocation, and bid-rigging – will be treated as amounting to by object infringements.  Indeed, the EU has issued guidelines confirming that these categories of behaviour will be treated as infringements by object – and so their actual effects are not relevant to the existence of a competition law infringement.

So far so good.

Over the years, the case law of the European courts, and the decisional practice of the European Commission, has been required on many occasions to identify whether particular behaviours qualify as by object infringements, or whether they are insufficiently obvious infringements of competition law such that they should be treated as illegal only if an anticompetitive effect can be demonstrated.  This is to be expected as there will always be a need to define boundaries to any legal concept.

Of particular importance to businesses is the point at which exchanges of information - i.e. discussions around market conditions or gathering of market intelligence, either directly between competitors or indirectly via suppliers or customers - might be considered so obviously anticompetitive as to constitute an infringement by object.  The categorisation of conduct in this area is critical, as by object arrangements are treated as tantamount to cartels, incurring the most serious penalties and having no ability to argue that there was an absence of competitive harm in their defense.  However, it is equally recognised that there will be necessary and productive exchanges on certain points between competitors (e.g. common innovations to improve industry standards) which should not be prevented through fear of competition sanctions.

Arrangements which fall outside the “by object box” face a higher evidential hurdle from the regulators’ perspective in that the competition authority must show an anticompetitive effect as a direct result of the conduct in question.  Consequently, conduct which falls under the by effect categorisation has proven less likely to be investigated at all given the greater challenges the competition authority faces in demonstrating an anticompetitive effect.

All of this goes to the critical question for modern business of the correct compliance standards to impose:  legal uncertainty as to what constitutes a by object infringement will lead to either overly restrictive or overly permissive compliance standards, neither of which are of broader benefit to economic welfare.

Two European Court judgments over the past year are important to consider - Cartes Bancaires and, more recently, Bananas.  We consider below what light these cases shine on the definition of the object infringement - and the lessons for businesses in terms of the level of interaction which is permissible with competitor entities.

It was the 1966 Consten & Grundig judgment of the European Court of Justice (ECJ) which first established that it was not necessary to demonstrate the effects of an anticompetitive arrangement once its anticompetitive object had been established.  This delineation has been followed in numerous cases over the years.  Advocate General Kokott in the T-Mobile case provided the oft-quoted analogy that object infringements are akin to drunk driving, i.e. inherently wrong, and merit sanction even where there has been no actual harm as a consequence of the actions in question.

A number of leading cases over the year have given further insights into conduct which should be treated as an infringement by object.  The Beef Industry Development Society (BIDS) case is a prime example which established that a Government-sponsored initiative to manage industry overcapacity would fall into the object category – the case concerned an initiative by an Irish beef processing association comprising ten companies to address overcapacity in the industry by reducing the number of processors by 25 per cent.  BIDS asked some members to exit the industry in Ireland for at least two years, disposing of equipment and decommissioning land.  Those exiting would be compensated by those staying in the market during the course of the agreement.  BIDS argued that the agreement would not adversely affect competition, but aimed to improve competitiveness in the beef industry. The ECJ disagreed, holding that through a coordination of market outcome – which prevented the natural selection of market players – the arrangement indeed had as its object the restriction of competition.  Importantly, the ECJ made clear that in determining the existence of an infringement by object, the lack of any subjective intention of the parties to distort competition is irrelevant.  From a policy perspective, this raises challenges:  should parties be punished for cartel-like behaviour when they genuinely did not believe the arrangements they were entering were anti-competitive?

The case of Allianz Hungaria further clouded the edges of the “object box”:  this case involved an arrangement between the insurer, Allianz, and the Hungarian National Association of Automobile Dealers, under which Allianz agreed that repair charges could be increased by dealers in proportion to the volume of Allianz insurance policies sold by the dealers (acting as agents for Allianz).  The ECJ concluded in this case that the existence of a remuneration link between repair services and volume of insurance contracts sold did not of itself constitute a by object restriction.  However, when considered in the relevant legal and economic context, the provisions did amount to an object infringement.  This ruling suggested it is necessary to look to the nature of the goods or services, and the “real conditions of the functioning and structure of the market” to determine an infringement by object.  This raises real uncertainty – there are types of conduct which may be treated as the most serious “by object” infringements, but only when considered in their proper economic context. Does this not entail considering their effects?  And if this is the case, how should such types of behaviour be distinguished from the less “obvious” competition infringements where a full “by effect” analysis is required?

The Cartes Bancaires judgment has been broadly viewed as seeking to restore a more balanced and certain approach to what constitutes a by object infringement.  The case involved a system set up in 1984 by the principal French banking institutions to manage payments and withdrawals using issued cards of the member banks. Card users would be able to use an ATM of any of the member banks and to pay for trades using the card.

In 2002, the group notified the European Commission of its plan to make a series of changes to pricing policies. The Commission found these arrangements to breach competition rules on the basis that they included a mechanism which encouraged exclusion of other potential card offerers and provided for the member banks to increase charges for use of bank cards.  One of the questions then considered on appeal was whether this system – which was essential to the operation of bank cards within France – was so obviously a restriction of competition that it should be treated as a by object infringement.

Ahead of the final judgment, Advocate General Wahl urged the ECJ to “refine its much debated case law on the concept of restriction by object”, calling for a more restrictive interpretation. In his view, a clear list of categories of behaviour amounting to by object infringements was the correct approach as it would provide (1) legal certainty; (2) deterrence of future infringement; and (3) procedural economies by avoiding detailed investigation in obvious “by object” cases. Wahl concluded that the measures implemented in the Cartes Bancaires case were not sufficiently obvious, and related to financial contribution by members to the operating costs of the bank card system, and protections to prevent free-riding through an incentives scheme. The question of whether there was a negative impact on competition was therefore necessarily effects-based.

This opinion was absorbed in the ECJ’s final decision, which returned to the traditional reliance on the assessment of the “nature” and “sufficiency” of harm, rather than the broad analysis used in Allianz Hungaria which required consideration of the relevant legal and economic context.  The decision was broadly welcomed by the competition law community in providing a clearer framework as to what behaviours amount to a “by object” infringement – although the Commission’s response was that the Cartes Bancaires decision was in any event in line with its approach and previous case law.

The next big decision on the question of “by object” infringements was in March 2015.  This case first concerned a finding by the Commission in 2008 of a price-fixing cartel in Northern Europe between banana importers, Chiquita, Dole, and Weichert, imposing fines of EUR 60.3 million (although Chiquita received immunity for whistleblowing under the Commission’s leniency scheme).

The facts centered on the defendants’ weekly setting and announcement of “quotation prices” for bananas for the coming week. Prior to setting these weekly quotation prices, the parties had regularly exchanged information by way of telephone calls, discussing a range of market factors relevant to the level of quotation prices although not agreeing the quotation prices themselves. These factors included, at the extreme, apparently innocuous information such as weather predictions and how these might impact upcoming banana crops.

The parties would then formally exchange price quotations after these had been determined, allowing for them to monitor the correlation between pricing decisions and pre-pricing discussions.  The parties argued that the factors discussed were not sufficiently proximate to their actual quotation prices to amount to an obvious “by object” infringement, and that consideration of the effects of the arrangements should have been undertaken.  The parties also argued that quotation prices did not correlate to actual prices subsequently agreed in negotiations with customers, that quotation prices related to green bananas, rather than yellow bananas which were ultimately sold to customers, and that operation of banana import quotas (i.e. a de facto supply limitation) meant that such discussions were incapable of distorting competition in any event.

The ECJ rejected the parties’ appeal in its entirety, stating that there are categories of conduct which are so likely to distort competition that any analysis of the effects, or construction of a counterfactual, is unnecessary.  The judgment is significant in its treatment of information exchange - it held that communications between competitors around quotation prices were relevant to the market, as even where discussions do not relate directly to price, it is possible from such signals, market trends or indications to infer the intended development of prices – and that such exchanges should therefore be treated as “by object” infringements. This echoes the 2009 T-Mobile case, in which information exchange was considered to be “tainted with an anticompetitive object” for being “capable of removing uncertainties concerning the intended conduct of the participating undertakings”. In T-Mobile, too, the Court had seen as irrelevant the consideration of actual effects in its finding of liability.

While the debate around the delineation of the “by object” infringement will continue, it is to be hoped that the European Courts, when judging whether a practice between market players is anticompetitive by object, will continue to tend towards a category-based approach which limits the “by object” framework to truly “obvious” competition law infringements.  As per the opinion of AG Wahl in Cartes Bancaires, there are clear benefits of this approach: legal certainty will allow business to more easily regulate its conduct when armed with a pre-determined list of clearly infringing behaviours.

However, the concern remains that the case law demanding that behaviour be considered in its context allows a wider application of the “by object” category.  In particular, where this case law is relied upon to capture complex arrangements which have not been considered in previous cases, or information exchanges which are not obviously related to pricing or competitive behaviour, there is a legitimate concern that the “by object” category is being abused.  The consequence of this will be businesses taking a more cautious approach to competition compliance than is actually necessary and not participating in arrangements which may well in fact have positive economic outcomes for both the parties involved and consumers more broadly – for example information sharing which facilitates innovation or efficiencies.

Developments in this area will continue to be watched closely in any event.