13 November 2006

Case T-186/01 GlaxoSmithKline v Commission (27 September 2006)
All change for Article 81?


In this groundbreaking judgment, the Court of First Instance explains: that not all agreements that segment the internal market infringe Article 81(1); that dynamic efficiencies may be considered in the application of Article 81(3); and finally that the aim of Article 81 is to protect the welfare of the end user. Even though the Commission lost the case, calling for it to seriously rethink its policy against pharmaceutical firms’ attempts to segment the market, these general points confirm much of the Commission’s recent policy on the application of Article 81.

The facts
GSK notified its sales policy for a number of medicines sold to wholesalers on the Spanish market. This provided (in clause 4) for a low sale price for medicines that the wholesalers resold on to the Spanish health authorities, and a higher price for medicines that the wholesalers were likely to sell abroad. The rationale behind this price discrimination is that medicines are bought by national health authorities at fixed prices. The price fixed in Spain is less than that fixed for the UK, and this creates an incentive for Spanish wholesalers to buy extra medicines and ship them to the UK (parallel imports). This causes GSK losses on the British market. Clause 4 was designed t thwart parallel imports. (This is a more direct strategy than that which another pharmaceutical company (Bayer) had tried a few years back, where it reduced the numbers of medicines sold to Spanish wholesalers. In this instance the Commission charged Bayer with entering into an agreement with Spanish wholesalers containing an export ban, but the CFI and ECJ had found that there was no agreement between the parties, only a unilateral move by Bayer).
It is little wonder that the Commission should seek to prevent this behaviour, as it partitions market, thus contrary to the creation of a single market.

The CFI’s assessment of Article 81(1): restrictions by object
The Commission found that the agreement had as its object the restriction of competition. Based on previous case law, this finding seemed unremarkable but the CFI disagreed.
The court said (118) that the objective of Art 81(1) is to prevent firms from restricting competition between themselves or with third parties from reducing the welfare of the final consumer.
The CFI cites three authorities in support of this assertion. It is important to read these closely to see how the CFI supports this statement.
First the CFI cites paragraph 115 of its judgment in Cases T-213 and 214/01 of 7 June 2006 where it said that the competition rules as a whole are there to improve the welfare of the consumer.
Second, it cites Consten and Grundig at page 493 (this page reference must be an error because the law report does not run to that page).
Third, it cites Case 28/77 Tepea where the ECJ ruled that an agreement that restricted intra-brand competition infringes Article 81(1) by harming consumer interests. The CFI generalises this ruling.
So no precedent seemed to support this conclusion directly but the CFI’s statement must surely be welcomed as in line with today’s economic approach to competition law. It is however curious that the court states that protecting consumer welfare is the objective of Article 81(1), and not of Article 81 as a whole (so there is a small but significant difference between this case and T-231/01). The significance is that if Art 81(1) is there to protect consumer welfare then is Art 81(3) there to protect some other interest?

The Court went on to say that not every agreement which partitions the market can be found to have an anticompetitive object: ‘while it is accepted that an agreement intended to limit parallel trade must in principle be considered to have as its object the restriction of competition, that applies insofar as the agreement may be presumed to deprive final consumers of those advantages.’ (para.121) Which means that sometimes agreements which partition the market might not harm consumers and s escape Article 81(1). This is a U-turn from the well-understood rule that these kinds of agreement are ‘hard-core restraints’. The court explained that in this sector, parallel trade was not going to lead to price competition and lower prices for consumers because their prices are fixed under national law. At the end of its discussion the court reminds us that this factual matrix was an unprecedented situation (147), which suggests that in most other goods and services, a ban on parallel trade will e deemed to have an anticompetitive object.

The implication of the court’s reasoning is that the anticompetitive object of an agreement must be looked at in its factual context. But then what is the difference between object and effect?


The CFI’s assessment of the effects of the clause
This part of the decision is peculiar because the court finds that the agreement might harm consumers… by transforming the definition of consumer. Recall how in the object category the emphasis was on the patient. Here (at 184) the court discovers an additional consumer – the sickness insurance schemes who, it says, suffer loss as a result of a diminution in intra brand competition. That is, the ability of Spanish wholesalers to export would have led to these consumers paying slightly lower prices, and in some Member States this might have led to lower prices for patients too.

This is puzzling. In the ‘object’ discussion the consumer is only the patient and the court says that his losses are not self evident after analysing the facts. Then in the ‘effect’ discussion, the consumers are the health insurance bodies and the patient, and both benefit from intra brand competition. This blurs the line between object and effect so much that it is invisible.

Article 81(3)
This segment focuses n the first condition ‘improving the production or distribution of goods or promoting technical and economic progress.’

Two interesting points here.
First, the court accepts the validity of GSK’s argument: restrictions on parallel trade can serve to allow it to raise money to finance research, and in this market, the key to inter-brand competition is innovation. The Commission was bound to trade off the improvements in efficiency stemming from innovation against the modest losses that some buyers suffered as a result of higher prices caused by the restriction in competition.

Second, in setting out some preliminaries we get this passage (244)
The Commission has, in particular, a margin of discretion which is subject to a restricted judicial review, in the operation consisting, once it has been ascertained that one of the criteria on which Article 81(3) EC makes provision for an exemption was satisfied, in weighing up the advantages expected from the implementation of the agreement and the disadvantages which the agreement entails for the final consumer owing to its impact on competition, which takes the form of a balancing exercise carried out in the light of the general interest appraised at Community level.

What is the general interest? (I doubt the court was quoting from Rousseau) does it mean that in 81(3) we look at more than just consumer benefits? This passage may be an aberration, later on the court ( like the Commission) reads the first condition of 81(3) as being about efficiencies.

The remaining questions

What does this case mean for the split between 81(1) and 81(3)?
(1) Maybe Odudu is right in suggesting that under Art 81(1) one shows a loss in allocative efficiency and in Art 81(3) one proves a gain in productive efficiency. See his recent book The Boundaries of EC Competition LawThe Scope of Article 81. However I don’t find this distinction helpful because an improvement under 81(3) is often one that improves allocative efficiency also, given that there must be a benefit to consumers for the improvement to count. Moreover, the court here suggests that GSK’s arguments were about increasing inter-brand competition. This kind of competition yields allocative efficiency.
(2) Maybe (the view I favour) the Commission looks for any loss of consumer welfare under Art 81(1) and then demands that the other party show any gains in consumer welfare. This allows the Commission to operate a quick look approach when some basic facts suggest a risk to competition.
(3) Maybe (and this is the view I would like to favour) under Art 81(3) we look at wider perspectives than just consumer welfare. After all, if you read paras 118 and 244 one after the other, consumer welfare is the objective for 81(1) and the general interest is the objective under Art 81(3). Public policy in Article 81(3)?

What is the difference between object and effect?
I have no idea after this case. Before, we had a fairly good idea that restrictions on parallel trade had an anticompetitive object. Now it seems that one has to have a perfunctory look at the market before being able to presume that any practice has an anticompetitive object.

Wasn’t notification such a good idea?
Note that notification/exemption (the system that the Commission killed off on 1 May 2004) is alive and kicking for this case and also O2 where the parties have been allowed to re-notify. This proves that a system where controversial cases can be notified ex ante is helpful in competition law. It had been mismanaged under Regulation 17, but a more modest scheme limiting notification to some agreements would have been beneficial.

Will GSK get the exemption now?
No predictions here but three indicators: first there seems to be considerable lobbying in support of clearance; second there seems to be considerable evidence that GSK has a point; but third the Commission has a margin of discretion. The worse thing that could happen is that the case gets settled informally, so that the Commission avoids publishing an uncomfortable precedent. This is the likely outcome.